Using Design Research and History To Tackle A Fundamental Problem With School Algebra

History of Mathematics Education
Sinan Kanbir
M.A. (Ken) Clements
Nerida F. Ellerton
Using Design Research

and History to Tackle a
Fundamental Problem
with School Algebra
Series Editors
Nerida F. Ellerton
M. A. (Ken) Clements
More information about this series at http://www.springer.com/series/13545

“Responsible people have described the main goal of the first course in algebra as an
understanding of the properties of a field” (Meserve & Sobel, 1964, p. 160).
“You have a danger of people being limited throughout their lives by what math they
got early on—or didn’t. There’s a lot of stuff that uses Algebra 2, and students who
don’t take it may be unaware that they are limiting their options later on. On the other
hand, it’s much better to have someone who genuinely understands modeling and
quantitative reasoning and has a feeling for statistics than someone who took an
Algebra 2 class but is totally bewildered by it.”
Mark Green, quoted in Bressoud (2016, p. 1182)
“Algebra is powerful—but it can also be frightening. It demands a shift of attention

from signified to signifiers. It can then become a game in which signifiers are
exchanged with other signifiers. … Algebra creates an alternative world which may
be under our control, but in which some people feel that nothing is real” (Tahta,
1990, p. 58).
Sinan Kanbir • M. A. (Ken) Clements • Nerida F. Ellerton
Using Design Research

and History to Tackle
a Fundamental Problem
with School Algebra
Improving the Quality of Algebra Education
at the Middle-School Level
Foreword by A. Eamonn Kelly

Sinan Kanbir M. A. (Ken) Clements
Department of Mathematics Sciences Department of Mathematics
University of Wisconsin–Stevens Point Illinois State University
Stevens Point, WI, USA Normal, IL, USA
Nerida F. Ellerton
Department of Mathematics
Illinois State University
Normal, IL, USA
ISSN 2509-9736 ISSN 2509-9744 (electronic)

ISBN 978-3-319-59203-9 ISBN 978-3-319-59204-6 (eBook)
DOI 10.1007/978-3-319-59204-6
Library of Congress Control Number: 2017945773
© Springer International Publishing AG 2018

This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on
microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation,
computer software, or by similar or dissimilar methodology now known or hereafter developed.
The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply,
even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations
and therefore free for general use.
The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to be
true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, express or
implied, with respect to the material contained herein or for any errors or omissions that may have been made. The
publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Printed on acid-free paper
This Springer imprint is published by Springer Nature

The registered company is Springer International Publishing AG
The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland
Foreword
The learning of algebra has continued to attract scientific interest for over 80 years (e.g.,
Layton, 1932; Rittle-Johnson, Loehr, & Durkin, 2017; Stewart, 2017; Thorndike et al., 1923).
Today, students’ learning of algebra has implications for government policies and standards
of mathematical practice (e.g., Bartell et al., 2017).
This book considers past, present and possible future aspects of an important issue in
contemporary school education—specifically, if “algebra for all” is to become something
more than a slogan, then what kinds of algebra, algebra instruction, and algebra learning
should feature in intended, implemented, and received school mathematics curricula
(Westbury, 1980)? The first three chapters are concerned with analyses of antecedents (“How
did we get to where we are now?”); the “middle” chapters take up present-day key issues
(“Where are we now, and what are we doing to improve the situation?”); and the two final
chapters offer reflections on issues related to future research and policy (“What should we do
to improve the quality of curricula and the teaching and learning of algebra in the future?”).
The authors adopted a design-research approach, which had five essential elements:
1. Identifying the main problem to be investigated, and its historical, theoretical,
practical, and ethical dimensions;
2. Identifying and working with key contributors;
3. Developing a research plan involving the gathering and rigorous analyses of data;
4. Planning for, implementing, and evaluating, a sequence of events based on a “plan-
act-observe-reflect” action-research program;
5. Conducting, and reporting, research in a manner consistent with the study design.
Those five elements fit well with modern principles of design research. Kelly, Lesh and Baek
(2008), in their Handbook of Design Research Methods in Education, maintained that design
research in education “is directed at developing, testing, implementing, and diffusing
innovative practices to move the socially constructed forms of teaching and learning from
malfunction to function, or from function to excellence” (p. 3).
The design-research investigation described here incorporated three main design
elements. After the research team had identified the main problem (“Why do so many
middle-school and secondary-school students fail to learn algebra well?”), they considered
the theoretical, practical, and ethical dimensions of that problem. Clearly, the problem is
common—yet surprisingly, there is no agreement within the international mathematics
education community on how it might best be solved. The researchers decided to plan,
conduct and evaluate a study which would guide and illuminate the path for further studies.
The second key design element was to identify and work with the key stakeholders associated
with the research program. The third design element was the co-development and
implementation of a research plan which would allow for rigorous gathering and analyses of
relevant data—interpretation of which might generate at least a local solution to the problem.
The research team created a design by which both formative and summative data in
qualitative and quantitative forms would be gathered and analyzed, at various stages of the
project. It was also decided that this data collection and evaluation process should incorporate
a plan-act-observe-reflect action research cycle (Carr & Kemmis, 2004; Kelly & Lesh, 2008),
v
vi Foreword by A. Eamonn Kelly
with aspects of the plan being progressively modified in light of data gathered from pencil-
and-paper tests, interviews with students, observations of workshops, and student reactions to
material presented in class.
As Hjalmarson and Lesh (2008) noted, the development of the product and the
development of knowledge are intertwined throughout a design-research study. In this study,
student knowledge, in multiple forms, was progressively developed, and monitored. That
knowledge influenced later aspects of the design process and contributed to the researchers’
understanding of factors influencing the teaching and learning of school algebra.
From my perspective, this book is important for three main reasons. First, the authors
address a well-recognized problem in school education; second, the study engaged key
teachers, students, and school administrators so that everyone was actively involved in what
they deemed to be an important piece of research with likely benefit for all concerned; and
last, but not least, the research exercise demonstrated the value of a design-research approach
to those from other, and equally important, research traditions. I am happy to recommend this
book to anyone considering adopting a design-research orientation in education studies.
References
Bartell, T., Wager, A., Edwards, A., Battey, D., Foote, M. & Spencer, J. (2017). Toward a
framework for research linking equitable teaching with the Standards for Mathematical
Practice. Journal for Research in Mathematics Education, 48(1), 7–21.
doi:10.5951/jresematheduc.48.1.0007
Carr, W., & Kemmis, S. (2004). Education, knowledge, and action research. London, United
Kingdom: Routledge.
Hjalmarson, M., & Lesh, R. A. (2008). Engineering and design research: Intersections for
education research and design. In A. E. Kelly, R. A. Lesh & J. Y. Baek (Eds.), Handbook
of design research methods in education: Innovations in science, technology, engineering
and mathematics learning and teaching (pp. 96–110). New York, NY: Routledge.
Kelly, A. E., Lesh, R. A., & Baek, J. Y. (Eds.). (2008). Handbook of design research methods
in education: Innovations in science, technology, engineering and mathematics learning
and teaching. New York, NY: Routledge.
Layton, E. T. (1932). The persistence of learning in elementary algebra. Journal of
Educational Psychology, 23(1), 46–55. http://dx.doi.org/10.1037/h0071929
Rittle-Johnson, B., Loehr, A.M. & Durkin, K. (2017). Promoting self-explanation to improve
mathematics learning: A meta-analysis and instructional design principles. ZDM
Mathematics Education doi:10.1007/s11858-017-0834-z.
Stewart, S. (2017). And the rest is just algebra. Basel, Switzerland: Springer International.
Thorndike, E. L., Cobb, M. V., Orleans, J. S., Symonds, P. M., Wald, E., & Woodyard, E.
(1923). The psychology of algebra. Macmillan.
Westbury, I. (1980). Change and stability in the curriculum: An overview of the questions. In
H. G. Steiner (Ed.), Comparative studies of mathematics curricula: Change and
stability 1960–1980 (pp. 12–36). Bielefeld, Germany: Institut für Didaktik der
Mathematik-Universität Bielefeld.
Anthony E. Kelly
George Mason University
Fairfax, Virginia June 2017
Contents
Foreword (by A. Eamonn Kelly) .......................................................................................... v

Contents ................................................................................................................................ vii
List of Figures ........................................................................................................................ xi
List of Tables ....................................................................................................................... xiii
Overall Book Abstract, and Individual Chapter Abstracts ............................................. xv
Preface to the Series ............................................................................................................. xxi
Preface to the Book ............................................................................................................. xxiii
1. Identifying a Problem with School Algebra ................................................................... 1
A Fundamental Problem with School Algebra ................................................................... 1
Some Performance Data, and Associated Critiques of Practices in School Algebra .......... 2
The Background to How and Why this Book was Written ................................................. 5
Overview of this Book ......................................................................................................... 6
2. Historical Reflections on How Algebra Became a Vital Component
of Middle- and Secondary-School Curricula ................................................................ 11
Providing Historical Frameworks for Mathematics Education Research ......................... 11
Algebra in Secondary School Mathematics: The Debate Over Purpose ........................... 14
Is School Algebra a Unidimensional Trait? ...................................................................... 44
The History of Mathematics and the History of School Mathematics .............................. 47
3. Framing a Classroom Intervention Study in a Middle-School
Algebra Environment ...................................................................................................... 59
Applying Principles of Design Research in Mathematics Education ................................ 59
The Role of Theory in the Middle-School Algebra Investigation ..................................... 62
4. Document Analysis: The Intended CCSSM Elementary- and Middle-School
Algebra Curriculum ........................................................................................................ 71
Approaches to Introducing Algebra: Structure and Modeling .......................................... 71
Overview of the Main Study ............................................................................................. 72
Concepts of Algebraic Structure and Modeling, as Presented in the
Common-Core State Standards for Mathematics .......................................................... 73
Structural and Functional Approaches, as Presented in the Mathematics
Textbook Used by Participating Students ..................................................................... 80
The Problem as it Appeared to be at the Beginning of the Study ..................................... 83
5. Review of Pertinent Literature ...................................................................................... 87
The Design Research Foundation for the Study ................................................................ 88
Theoretical Bases for the Study ......................................................................................... 90
Signifiers, Objects, Interpretants, and Charles Sanders Peirce ......................................... 91
Cognitive Structures and Individual Learners ................................................................. 102
vii
viii Contents
Literature Which Helped Frame the Design of the Teaching Intervention ..................... 106
Research Questions .......................................................................................................... 108
6. Research Design and Methodology .............................................................................. 115

Setting up the Intervention ............................................................................................... 116
Professional Development in a Theoretical Context ....................................................... 121
Intervention Setting and Theoretical Base ....................................................................... 121
Procedures ........................................................................................................................ 123
Instrumentation ................................................................................................................ 124
Research Hypotheses, and Issues Related to the Quantitative Analysis .......................... 128
Issues Related to the Qualitative Analyses ...................................................................... 136
Concluding Comments ..................................................................................................... 138
7. Quantitative Analyses of Data ...................................................................................... 141

Overview of Quantitative Data ........................................................................................ 141
Analyses of Pre-Teaching Data ....................................................................................... 144
Calculation of Effect Sizes ............................................................................................... 152
Summary and Concluding Comments on the Quantitative Analyses .............................. 153
8. Qualitative Analyses of Data ......................................................................................... 155

Intended, Implemented and Received Curricula .............................................................. 155
Peirce’s Triadic Semiotic Position and Herbart’s Theory of Apperception .................... 156
Analyses of Qualitative Data Generated by the “Structure” Intervention ....................... 158
Analyzing Samples of Interview Data Relating to “Structure” ....................................... 160
Analyses of Qualitative Data Generated by the Modeling Intervention .......................... 166
Cognitive Growth in Modeling Related to the Subscript Notation for Sequences .......... 170
Cognitive Growth in Modeling Related to Generalizing for the nth Term ...................... 171
Analyses of Student Responses to the Six Interview Tasks ............................................ 174
Qualitative Analyses of the Implemented Curriculum .................................................... 186
Concluding Comments with Respect to the Qualitative Analyses .................................. 189
9. Answers to Research Questions, and Discussion ........................................................ 193

Answer to Research Question 1 ....................................................................................... 193
Possibilities for Future Related Research ........................................................................ 205
Final Comments on the Workshop “Lessons” ................................................................. 206
Comments on the Study by the Two Participating Teachers ........................................... 207
10. Postscript: Framing Research Aimed at Improving School Algebra ...................... 211
Sometimes the Most Appropriate Theories for Mathematics Education
Research will Come from the Past ............................................................................... 211
Contents ix
Giving Precedence to Peirce’s, Herbart’s, and Del Campo and Clements’ Theories ...... 212
Did the Intervention Improve the Participating Students’ Knowledge of,
and Ability to Generalize, and Apply, School Algebra? ............................................. 216
Moving Forward .............................................................................................................. 217
Author Biographies ............................................................................................................ 223
List of Appendices .............................................................................................................. 225
Appendix A: Protocol for Algebra Interviews with Seventh-Graders ............................. 227
Appendix B: Algebra Test (Three Parallel Versions Are Reproduced) ........................... 229
Appendix C: “Questionnaire” Completed by Seventh-Grade Students at School
W at the Beginning of the Algebra Workshops on “Structure” ................. 241
Appendix D: Statement of Instructional Aims for the Structure Workshops
with the Seventh-Grade Students at School W ........................................... 243
Appendix E: Detailed Lesson Plans for Four Workshops on “Structure” for
Seventh-Grade Students at School W, Including Homework
Challenges for Each Workshop .................................................................. 245
Appendix F: Detailed Plans for Group Tasks in the Modeling Workshops:
Finding Recursive and Explicit Rules for Patterns ..................................... 261
Appendix G: Classroom Observation Schedule ............................................................... 287
Appendix H: Pre-Teaching to Post-Teaching “Growth” with Respect to the Five
Basic Cognitive Structure Components ...................................................... 289
Appendix I: Generalization Categories (After Radford, 2006) ........................................ 295
Composite Reference List .................................................................................................. 299
Author Index ...................................................................................................................... 317
Subject Index ...................................................................................................................... 323
List of Figures
Figure 2.1 Page 10 from Ditton’s (1709) appendix. .............................................................. 19

Figure 2.2 Page 109 from Ditton’s (1709) appendix. ............................................................ 20
Figure 2.3 Page 65 of James Hodgson’s (1723) textbook for his
12- to-16-year-olds—note the use of algebra and the references
to “fluxions.” ........................................................................................................ 22
Figure 2.4 The first page on algebra in Alfred Andrew’s (1813)
algebra cyphering book. ....................................................................................... 30
Figure 2.5 Comments on a theory of relations (Vredenduin, 1962, p. 105.) ......................... 38
Figure 2.6 A geometric progressions task prepared by Zoltan Dienes
for middle-School students (from Dienes, 1968, p. 97). ...................................... 40
Figure 2.7 Different ways of “seeing” problems or situations which might
relate to mathematics (from Ellerton & Clements, 2014, p. 323). ....................... 47
Figure 5.1 Table associated with “Visiting Old Houses” task. ............................................. 98
Figure 5.2 A sequence of figures inviting a generalization of a relationship
between the number of a figure and the corresponding number
of matchsticks. ................................................................................................... 100
Figure 6.1 Summary of the design and time intervals for the study.................................... 123
Figure 6.2 Two structure questions (pre-teaching version of the Algebra Test). ................ 125
Figure 6.3 A modeling question (pre-teaching version of the Algebra Test). ..................... 126
Figure 7.1 Bar graphs, showing mean scores of the two groups at different
stages, on the structure and modeling subtests. ................................................. 144
Figure 7.2 Group 1 and Group 2 students’ pre-, mid-, post-teaching,
and retention mean scores on the structure subtest (maximum
possible score was 10)........................................................................................ 153
Figure 7.3 Group 1 and Group 2 students’ pre-, mid-, post-teaching,
and retention mean scores on the modeling subtest (maximum
possible score was 10)........................................................................................ 153
Figure 8.1 Student 2.5’s responses to the words “distributive property”
in the pre- and post-teaching interviews. ........................................................... 160
Figure 8.2 Student 2.6’s written pre- and post-teaching interview responses
to the request regarding the “distributive property.”.......................................... 161
Figure 8.3 Student 2.6’s written pre- and post-teaching interview responses
to a question concerned with the associative property for multiplication.......... 161
Figure 8.4 A student demonstrating a correct expressive understanding
of a property, without knowing its name. .......................................................... 162
Figure 8.5 Student 1.1 indicated that although she really liked the “crossing-the-river”
modeling task, the associative properties sometimes confused her. .................. 162
Figure 8.6 Learning to comprehend subscript notation for linear sequences. ..................... 168
Figure 8.7 Evidence for having learned to comprehend subscript notation. ....................... 168
Figure 8.8 Student 1.13’s post-teaching interview response
to the “table-of-values” Task. ............................................................................ 170
xi
xii List of Figures
Figure 8.9 Student generalizations in the pre- and post-teaching interviews. .................... 174
Figure 8.10 Seventh-grade students’ cognitive growth on two different
tasks inviting students to state the rule for the nth term. ................................... 174
Figure 9.1 Comparison of pre-teaching and mid-intervention structure means,
Group 1 and Group 2. ........................................................................................ 198
Figure 9.2 Comparison of pre-teaching and mid-intervention modeling means,
Group 1 and Group 2. ........................................................................................ 198
Figure 9.3 Bar graphs, showing mean scores of the two groups at pre-teaching
and post-teaching stages, on the structure and modeling subtests..................... 200
Figure 9.4 Bar graphs, showing mean scores of the two groups at post-teaching
and retention stages, on the structure and modeling subtests. ........................... 204
List of Tables
Table 1.1 Percentages Correct, 328 Mathematics Teacher-Education Students

on Four Equation/Inequalities Pairs (Ellerton & Clements, 2011) ....................... 5
Table 2.1 Six Distinguishable Purposes for Secondary School Algebra ............................ 16
Table 2.2 The National Mathematics Advisory Panel’s (2008) List of Major
Topics to be Covered in School Algebra (p. 16) ................................................. 25
Table 3.1 Messages of History on Language Aspects Associated with Five
of the Purposes of School Algebra ...................................................................... 64
Table 4.1 A Table of Values from Charles et al. (2004, p. 481) ......................................... 83
Table 5.1 Receptive and Expressive Modes of Communication (Del Campo
& Clements, 1987, p. 12) .................................................................................. 107
Table 6.1 Pre-Teaching Algebra Test Mean Scores, Standard Deviations (SD),
and Difference of SDs ....................................................................................... 120
Table 6.2 Summary of the Research Design and Timing in the Planned
Research Program ............................................................................................. 124
Table 6.3 Subscript Notations for Describing Class Means at Various
Points in the Study ............................................................................................ 129
Table 7.1 Group 1 Students’ Pre-Teaching, Mid-Intervention, Post-Teaching,
and Retention Scores for the Structure and Modeling Questions
(Out of 10, in Each Case) ................................................................................... 142
Table 7.2 Group 2 Students’ Pre-Teaching, Mid-Intervention, Post-Teaching,
and Retention Scores for the Structure and Modeling Questions
(Out of 10, in Each Case) .................................................................................. 143
Table 7.3 Effect Sizes for Four Sets of Intervention Workshops ...................................... 152
Table 8.1 Evidences for Qualities of Components of Students’ Concept
Images With Respect to the Associative and Distributive Properties ............... 159
Table 8.2 Summary of Data from 28 Interviewees in Relation to Concept
Images for the Associative Property for Addition ............................................ 164
Images for the Associative Property for Multiplication .................................... 165
Images for the Distributive Property for Multiplication Over Addition ........... 166
Table 8.5 Evidences for Qualities of Components of Students’ Concept
Images with Respect to Modeling Relationships .............................................. 167
Table 8.6 Summary of Data from 28 Interviewees in Relation to the Use
of the Subscript Notation in Modeling Tasks (Associated
with Linear Sequences) ..................................................................................... 171
Table 8.7 Summary of Data from 28 Interviewees in Relation to Generalizing
for the nth Term ................................................................................................ 171
xiii
xiv List of Tables
Table 8.8 Examples of Students’ Generalizations on Various Written Test Items ........... 173
Table 9.1 Effect Sizes for Intervention Workshops, Pre-Teaching
to Mid-Intervention ........................................................................................... 199
Table 9.2 Effect Sizes for Four Sets of Intervention Workshop Periods .......................... 200
Overall Book Abstract, and Individual Abstracts
for the Ten Chapters of the Book
Overall Abstract
This book is unusual in that it includes both a serious historical analysis of the
international history of school algebra and a description of a design-research investigation
whose aim was to improve the teaching and learning of middle-school algebra. It is argued,
at the outset, that it is important that strong decisions are needed to reach a situation in which
most middle- and secondary- school students will learn algebra better than they do at present.
The historical analysis identified the following six purposes for school algebra:
1. School algebra as a body of knowledge which prepares students for higher
mathematical and scientific studies;
2. School algebra as the study of generalized arithmetic;
3. School algebra as a gatekeeper for entry to higher studies;
4. School algebra as an integral component of mathematical modeling of real-world
contexts;
5. School algebra as a vehicle for generalizing numerical, geometrical, and other
mathematical structures; and
6. School algebra as the study of variables—symbols which can represent different
amounts of different quantities, and relationships between those quantities.
It is claimed that although these purposes intersect, under certain conditions, they are
nevertheless all conceptually separable from each other.
One of the conclusions from the historical analyses was that for 350 years there has been
a disconnect between the signifiers of school algebra (its signs, symbols, and pictures) and
the intended “signifieds” (the mathematical “objects” of the intended curriculum). This led the
authors to adopt some of the semiotic ideas of Charles Sanders Peirce for the theoretical
framework of the main study. However, it was recognized that there is a need to assist
learners to link the signifiers more readily with the intended curriculum, and that recognition
resulted in a decision to adopt, as part of the study frame, the “apperception” ideas of Johann
Friedrich Herbart and his followers. The Herbartians argued that students’ long-term
memories carry “cognitive structures”—made up of verbal knowledge, skills, images,
episodes (memories of relevant events), attitudes, and also idiosyncratic internal links
between these components. It is the teacher’s role to engineer learning environments which
will help her or his students to develop rich cognitive structures which will enhance learning.
With the idea of helping teachers to create richer algebra learning environments, the
theoretical position put forward by Gina Del Campo and Ken Clements which distinguished
between receptive and expressive understandings, but valued both, was incorporated into the
study. Workshops aimed at helping 32 seventh-grade students to enhance their cognitive
structures for algebra—with respect to the associate properties for addition and multiplication
and the distributive property—and to modeling real-world problems with linear sequences,
were developed, trialled, and carefully implemented as part of a successful design-research,
mixed-method, investigation. Details of the investigation are given. Analyses of data
suggested that the investigation was sufficiently successful that it would warrant replication.
xv
xvi Abstracts
Individual Chapter Abstracts

Chapter 1: Identifying a Problem with School Algebra
Abstract: The chapter begins by presenting data which suggest that there is a longstanding,
and fundamental, problem with school algebra. The problem is that many students who try
hard to understand the fundamental principles of algebra, fail to do so. But that statement
raises an important question—Why do so many school students find it difficult to learn the
subject well? The authors of this book set out to answer that question from three different
perspectives: the first relates to the question why students are asked to learn algebra.
Adopting a historical method of analysis, we identify six purposes which have been offered
as reasons for why school students should study algebra. The second perspective relates to
theories which might help explain why so many secondary-school students do not learn
algebra well. And, the third perspective offers a set of principles which might begin to
provide an answer to the fundamental problem. These principles are applied in an
intervention study, with seventh-grade students, which is described later in this book.
Chapter 2: Historical Reflections on How Algebra Became a Vital Component of

Middle- and Secondary-School Curricula
Abstract: The chapter begins by identifying, and placing in their historical contexts, the
main issues in a longstanding debate over the purposes of school algebra. The following six
purposes which have been attributed to school algebra by various writers over the past three
centuries are identified, and the emphases given to the purposes at different time are discussed:
(a) algebra as a body of knowledge essential to higher mathematical and scientific studies,
(b) algebra as generalized arithmetic, (c) algebra as a prerequisite for entry to higher studies,
(d) algebra as offering a language and set of procedures for modeling real-life problems,
(e) algebra as an aid to describing structural properties in elementary mathematics, and
(f) algebra as a study of variables. The question is then raised, and discussed, whether school
algebra represents a unidimensional trait.
Chapter 3: Framing a Classroom Intervention Study in a Middle-School Algebra

Environment
Abstract: It has become a tradition in the field of mathematics education that before a
researcher outlines the research design for a study he or she should outline a theoretical
framework for the investigation which is about to be conducted. Then, after research
questions are stated, and the design of the study is described, the investigation takes place.
The data gathering, data analyses, and interpretation are guided by the theoretical framework
and conclusions are couched in terms of, and seen in the light of, the theoretical framework.
There are many mathematics education researchers who regard this theory-based process as
sacrosanct, as absolutely essential for high-quality research. In the first part of this chapter it
is argued that the traditional “theoretical-framework” process just described is flawed, that it
can result in important aspects of data being overlooked, and that it can lead to incorrect, or
inappropriate, conclusions being made. It is argued that the first thing which needs to be done
in a mathematics education research investigation is to identify, in clearly stated terms, the
problems for which solutions are to be sought. Having done that, historical frameworks—
which have only occasionally been taken seriously by mathematics education researchers—
Abstracts xvii
should be provided. Then, having identified the problems and having provided a historical
framework, a design-research approach ought to be adopted whereby a theory, or parts of a
theory, or a combination of parts of different theories, are selected as most pertinent to the
problems which are to be solved. This chapter identifies three main problems: (a) “Why do
so many middle-school students experience difficulty in learning algebra?” (b) “What
theoretical positions might be likely to throw light on how that problem might be best
solved?” (c) “In the light of answers offered for (a) and (b), what are the specific research
questions for which answers will be sought in subsequent chapters of this book?”
Chapter 4: Document Analysis: The Intended CCSSM Elementary- and Middle-School

Algebra Curriculum
Abstract: Having identified the main problem (“Why do so many school students find it
difficult to learn school algebra well?”), and having made decisions on historical and
theoretical frameworks for the study, it was important that the intended algebra content, as
defined by the common-core mathematics curriculum and by the algebra content in textbooks
which had previously been used by participating students, and in the textbooks being used in
the seventh grade by the students, be identified and analyzed. The ensuing document
analyses, presented in this chapter, revealed that the seventh-grade students might have been
expected to know the associative and distributive properties for rational numbers and, given
tables of values, they might have been expected to be able to identify and summarize,
mathematically, the rules for uncomplicated linear sequences.
Chapter 5: Review of Pertinent Literature

Abstract: This chapter frames the main study described in this book in terms of the
theoretical positions of Charles Sanders Peirce, Johann Friedrich Herbart, and Gina Del
Campo and Ken Clements. Peirce’s tripartite position on semiotics (featuring signifiers,
interpretants, and signifieds), Herbart’s theory of apperception, and Del Campo and
Clements’s theory of complementary receptive and expressive modes of communication,
were bundled together to form a hybrid theoretical position which gave direction to the study.
The chapter closes with careful statements of six research questions which emerged not only
from consideration of the various literatures, but also from a knowledge of practicalities
associated with the research site, from our historical analysis of the purposes of school
algebra, and from our review of the literature.
Chapter 6: Research Design and Methodology

Abstract: The main study featured a mixed-method design, with complementary quantitative
and qualitative data being gathered and analyzed. Since random allocation of students to two
groups occurred, it was legitimate for null and research hypotheses to be formulated for the
quantitative analyses, and those hypotheses are carefully defined in this chapter. One of the
important challenges was to identify the population to which inferences would be made.
Details relating to the development of appropriate pencil-and-paper tests and an interview
protocol are also given, as are details relating to the calculation of Cohen’s d effect sizes.
xviii Abstracts
Chapter 7: Quantitative Analyses of Data

Abstract: Quantitative data from the main study are summarized and analyzed. Both the
structure and modeling workshops generated statistically significant performance gains.
Thus, after students had participated in both the workshops, their performances on both parts
of the Algebra Test—that is to say, on the questions concerning structure and on the
questions concerning modeling—were much improved. Cohen’s d effect sizes for each set of
workshops (the structure workshops and the modeling workshops) were large. The chapter
concludes by introducing two questions. First, although the performance gains were highly
statistically significant, and the effect sizes large, were they educationally significant? And,
second, “What was there about the interventions which generated such apparently impressive
results?”.
Chapter 8: Qualitative Analyses of Data

Abstract: Qualitative data from the main study are summarized, analyzed, and interpreted
from the perspective of Herbart’s theory of apperception and Del Campo and Clements’s
theory of receptive-expression modes of communication. For many of the students, there was
evidence of “significant growth,” but for some, there was “no evidence.” Findings from these
analyses complemented and supported findings from the quantitative analyses in Chapter 7.
Qualitative analyses of pre-teaching data suggested that the students remembered very little,
if anything, about structures and modeling that they had previously studied—despite the fact
that common-core expectations would be that they should have a strong grasp.
Chapter 9: Answers to Research Questions, and Discussion

Abstract: Answers to the six main research questions are given, and issues arising from the
answers are discussed. Both the quantitative and qualitative analyses have pointed to the
success of both the structure and the modeling workshops. Initially, the seventh-grade
participants had very little knowledge of the associative and distributive properties—they did
not know the definitions, and could not apply the properties to numerical calculations. A
similar situation was true so far as modeling was concerned—whereas, initially, some
students could identify recursive rules for simple linear sequences, none could identify
explicit rules. Relevant algebraic conventions and language were not known. As a result of
the students’ active engagement in workshops in which the students learned appropriate
language and conventions, and made generalizations in terms of variables, most of the
participating students—but not all of them—showed strong improvement in relation to
structure and modeling. Students’ knowledge of definitions and skills improved, they
developed appropriate imagery, and their self-confidence when asked to answer questions
relating to structure and modeling improved. The results are linked to the theories of Charles
Sanders Peirce, Johann Friedrich Herbart, and Gina Del Campo and Ken Clements.
Chapter 10: Postscript: Framing Research Aimed at Improving School Algebra
Abstract: This final chapter is written as a guide to persons wishing to carry out research
which aims to improve middle-school students’ understanding of school algebra to the point
where not only will the students be able to generalize freely, but will also be able to apply the
algebra that they learn. The first point made in the chapter is that mathematics education
Abstracts xix
researchers need to take the history of school mathematics more seriously, because the six
purposes of school algebra identified in the historical analysis presented in Chapter 2 of this
book were important not only in helping the research team identify the importance of
language factors in school algebra, but also in designing the study which would be carried
out. The second point made was that in a design-research study the theoretical frame is likely
to be not one single theory, but a composite theory arising from a bundle of part-theories that
are suggested by needs revealed in the historical analysis. The third, and final point is the
need for mathematics education researchers to remember that, ultimately, the aim of school
mathematics is to help students learn mathematics better, so that the students will be
competent and confident to use it whenever they might need it in the future. Research designs
should be such that tight assessments can be made with respect to whether the results of the
studies will help educators improve the teaching and learning of algebra in schools
Preface to the Series
From the outset it was decided that the series would comprise scholarly works on a
wide variety of themes, prepared by authors from around the world. We expect that authors
contributing to the series will go beyond top-down approaches to history, so that emphasis
will be placed on the learning, teaching, assessment and wider cultural and societal issues
associated with schools (at all levels), with adults and, more generally, with the roles of
mathematics within various societies. In the past, scholarly treatises on the history of
mathematics education have featured strong Eurocentric/American emphases—mainly
because most researchers in the field were scholars based in European or North or South
American colleges or universities. It is hoped that the books in the new series will be
prepared by writers from all parts of the world.
In addition to generating texts on the history of mathematics education written by
authors based in various parts of the world, an important aim of the series will be to develop
and report syntheses of historical research that have already been carried out with respect to
important themes in mathematics education—like, for example, “Historical Perspectives on
how Language Factors Influence Mathematics Teaching and Learning,” and “Important
Theories Which Have Influenced the Learning and Teaching of Mathematics.”
The mission for the series can be summarized as:
• To make available to scholars and interested persons throughout the world the
fruits of outstanding research into the history of mathematics education;
• To provide historical syntheses of comparative research on important themes in
mathematics education; and
• To establish greater interest in the history of mathematics education.
We hope that the series will provide a multi-layered canvas portraying the rich details
of mathematics education from the past, while at the same time presenting historical insights
which can support the future. This is a canvas which can never be complete, for today’s
mathematics education becomes history for tomorrow. A single snapshot of mathematics
education today is, by contrast with this canvas, flat and unidimensional—a mere pixel in a
detailed image. We encourage readers both to explore and to contribute to the detailed image
which is beginning to take shape on the canvas for this series.
Nerida F. Ellerton
M. A. (Ken) Clements
(Series Editors)
January, 2017
xxi
Preface to the Book
Recently, David M. Bressoud, a former President of the Mathematical Association of

America, wrote an important, and balanced, review of the book The Math Myth and Other
Stem Delusions, by Andrew Hacker (see, Bressoud, 2016, Hacker, 2016). In the book,
Hacker argued that too many U.S. high school students study Algebra II in which they are
expected to learn operations on polynomials and understand connections between their zeros
and factors, construct and compare linear, quadratic and exponential models, understand the
general role of functions in modeling a relationship between two quantities, and come to see
trigonometric functions as models of periodic phenomena. Bressoud pointed out that an
introductory knowledge and understanding of all of those topics could be useful for most
people, and is essential for those who would seek a STEM career. He acknowledged,
however, that a problem might arise if all secondary students were to be expected to take
Algebra II. For his part, Hacker was concerned that the content of Algebra II was not suited
to the present and future needs of many secondary-school students.
We believe that the purposes of algebra in middle- and secondary-school curricula have
never been subjected to careful scrutiny and that, in particular, a decent history of school
algebra has been lacking. In Chapter 2, in this book, we have taken up the challenge
presented by that statement and have offered the beginnings of a history, written from of a
global perspective, of school algebra. Six purposes of school algebra emerge from our
historical analysis. When algebra first entered secondary-school curricula, in the late 1600s,
the contents were chosen by high-level mathematicians, and the students were selected
because they had demonstrated that they were good at arithmetic. The fact that, initially,
algebra was a subject designed for just a few élite students would have enormous
ramifications for the future. Content and standards of the past began to be, and continue to
be, defended by those who are worried about declining standards in the schools, and by those
who are very concerned to ensure that beginning college students are well prepared for
rigorous mathematics courses that they might face.
This book tells of an attempt to show how the situation might be changed. First, the
main problem was identified, clearly articulated, and located within the history of school
education. Then, a decision was made to design a lighthouse investigation which might
suggest how ordinary middle-school students can be actively engaged in the learning of
important, curriculum-relevant algebra. The investigation was in the form of a design-
research study whose theoretical frame emerged from a consideration of what needed to be
done to solve the main problem. That theoretical frame was pieced together from three main
theories—the semiotic position of Charles Sanders Peirce, the theory of apperception by
Johann Friedrich Herbart, and the receptive-expression theory of lesson design by Gina Del
Campo and Ken Clements. The international and historical base of the theoretical framework
can be recognized by the fact that Peirce was an American philosopher who lived more than
a century ago, Herbart was a German philosopher and educator who lived two centuries ago,
and Del Campo and Clements developed their theoretical position in Australia in the 1980s.
The structure of this book differs from that of many other books. In the modern age,
chapters of books are sold separately (usually in electronic form) and, mindful of that fact, we
have prepared each chapter assuming that it might be read as a stand-alone document. Thus,
xxiii
xxiv Preface to the Book
at the beginning of each chapter there is an abstract and set of key words for that chapter.
Occasionally, facts and important points of view mentioned in earlier chapters are repeated in
the main text of a chapter—for the benefit of those without access to other chapters. At the
end of each chapter we have provided a reference list for the chapter. Before the first chapter
we have reproduced the abstracts of all 10 chapters as well as an “overall” abstract for the
book. And, toward the end of the book we have included a “composite reference list” which
brings together all the information in the reference lists for the individual chapters.
We want to thank, sincerely, Mr. X and Mr. Y, the two teachers at School W who were
our partners in the main study. They were enthusiastic and always mindful of the best
interests of their students. As for the students themselves, they too were wonderful
participants in the study—always giving of their very best, both in class and during the
research interviews. We felt very honored when Eamonn Kelly agreed to write the foreword
for this book, because it was Eamonn who stimulated our initial interest in design research.
Of course, we are grateful to George Seelinger, for his unwavering support, and to other
colleagues within the Department of Mathematics at Illinois State University, where we were
based when we carried out the main investigation described in this book. We should also
express our appreciation to the numerous persons working in archives in various parts of
Australia, Brunei Darussalam, Singapore, the United States of America and the United
Kingdom, who helped us in the historical component of the research reported in this book.
And, of course, we are extremely grateful to Melissa James and her assistant, Sara Yanny-
Tillar, and all the planning and production team at Springer, for their total cooperation with
us as we prepared this book.
We would be pleased to hear from anyone interested in replicating the main study,
summarized in Chapters 3 through 9 of this book.
References
Bressoud, D. M. (2016). Book review: The Math Myth and Other STEM Delusions. Notices
of the American Mathematical Society, 63(10), 1181–1183.
Hacker, D. M. (2016). The math myth and other STEM delusions. New York, NY: The New
Press.
Sinan Kanbir M. A. (Ken) Clements

Department of Mathematical Sciences Nerida F. Ellerton
University of Wisconsin Department of Mathematics,
Stevens Point, WI Illinois State University
June 2017
Chapter 1
Identifying a Problem with School Algebra
Abstract: The chapter begins by presenting data which suggest that there is a longstanding,
and fundamental, problem with school algebra. The problem is that many students who try
hard to understand the fundamental principles of algebra, fail to do so. But that statement
raises an important question—Why do so many school students find it difficult to learn the
subject well? The authors of this book answer that question from three different perspectives:
the first relates to the question why students are asked to learn algebra. Adopting a historical
method of analysis, we identify six purposes which have been offered as reasons for why
school students should study algebra. The second perspective relates to theories which help
explain why so many secondary-school students do not learn algebra well. And, the third
perspective offers a set of principles which might begin to provide an answer to the
fundamental problem. These principles were applied in an intervention study with seventh-
grade students which is described in this book.
Keywords: Mathematics education research, History of school algebra, Student difficulties

in learning algebra, Theories in mathematics education
A Fundamental Problem with School Algebra

In the opening chapter of the First Yearbook of the National Council of Teachers of
Mathematics (hereafter “NCTM”), David Eugene Smith (1926)—a former President of the
Mathematical Association of America and a prolific author of school mathematics
textbooks—claimed that algebra curricula in U.S. secondary schools had improved over the
period 1900–1925 and that, in particular, a large amount of “entirely useless and
uninteresting work that had cumbered up the inherited course” had been “struck out” (p. 10).
Smith’s optimism, however, was not shared by all of the leading educators of his time. In the
same First Yearbook, Raleigh Schorling, then NCTM’s President, pointed out, for example,
that Edward Lee Thorndike, a well-known education psychologist, believed that ninth-grade
algebra students “had mastery of nothing whatsoever” (quoted in Schorling, 1926, p. 65).
Eighty-five years later, in NCTM’s Seventieth Yearbook, Jeremy Kilpatrick and Andrew
Izsák (2012) began their chapter titled “A History of Algebra in the School Curriculum” with
the following quotation from an unnamed editorial writer of the 1930s: “It [i.e., algebra] has
caused more family rows, more tears, more heartaches, and more sleepless nights than any other
school subject” (p. 3). Kilpatrick and Izsák also stated that around 1910 “algebra was fast
becoming a major source of failure in school” (pp. 6–7), and that during the period from 1910
to 1950 algebra enrollments in U.S. high schools fell from 57 percent to below 25 percent. In
another chapter in NCTM’s Seventieth Yearbook, Daniel Chazan (2012) commented that just
20 years ago, “algebra was seen as abstract mathematics suitable only for students who were
developmentally ready and college intending” (p. 20). He added that, in fact, as late as the
1980s some U.S. students completed high school without ever having formally studied algebra.
© Springer International Publishing AG 2018 1

S. Kanbir et al., Using Design Research and History to Tackle a Fundamental Problem with School Algebra,
History of Mathematics Education, DOI 10.1007/978-3-319-59204-6_1
2 Ch 1: Identifying a Problem with School Algebra
He attributed that situation to two factors: first, it had never been made clear why all students
might benefit from studying algebra; and second, the subject was regarded, by many, as very
difficult. Many other similar comments, from a wide range of sources, could be quoted here—
there can be little doubt, in fact, that the longstanding problems associated with algebra in U.S.
secondary schools (Cajori, 1890) persisted throughout the twentieth century (House, 1988).
Some Performance Data, and Associated Critiques of Practices in School Algebra

The difficulties that school students experienced in learning algebra in the twentieth
century were certainly not confined to North America. Around 1990, Mollie MacGregor
(1991) asked 45 eleventh-grade students, in Melbourne Australia—who had all studied
algebra for at least four years—to write an equation relating the number of pupils in a school
and the number of teachers in the school, given the following tabular information:
Number of pupils 100 200 300 400 … 1000
Number of teachers 5 10 15 20 50
MacGregor reported that only 17 of the 45 students (38%) gave a correct answer. One might
say: “Surely, it should have been obvious to students who had been studying algebra for
more than four years that the number of pupils (P, say) was equal to 20 times the number of
teachers (T say). Hence P = 20T would be an appropriate response.”
MacGregor (1991) also asked 235 ninth-grade students—each of whom had been
studying algebra for at least two years in secondary schools in Melbourne—to answer a set of
pencil-and-paper tasks which included the following three questions:
1. “The number y is eight times the number z.” Write this information in
mathematical symbols.
2. s and t are numbers, s is 8 more than t. Write an equation showing the relation
between s and t.
3. The Niger River in Africa is y metres long. The Rhine River in Europe is z metres
long. The Niger is three times as long as the Rhine. Write an equation which shows
how y is related to z.
MacGregor reported that the percentages of correct responses were 34.5% (for Question 1),
28.1% (for Question 2) and 33.2% for Question 3. MacGregor also asked 19 eleventh-grade
students to attempt Question 1, and found that only 11 of them answered it correctly. With
respect to Question 1, MacGregor (1991) commented: “It was expected that all, or almost all,
students would get this right. Indeed, it is hard to imagine why anyone could be wrong”
(p. 50).
Pongchawee Vaiyavutjamai (2004, 2006) asked 231 ninth-grade secondary-school
students in Chiang Mai, Thailand, to respond to similar questions to Questions 1, 2, and 3
(shown above)—the main difference between her data and MacGregor’s was that in Thailand
the questions were presented in the Thai language. Analysis revealed that the students in
Thailand performed at even lower levels than had the students in Melbourne, Australia. A
similar finding was reported by Lim Ting Hing in his study involving tenth-grade students in
Brunei Darussalam. In Lim’s (2000) study the language of testing was English—which was
the language of instruction, but not the first language, for the students.
There have been many other studies in which analyses of data have revealed just how
difficult many students find school algebra (see, e.g., Booth, 1984, 1988; Fujii & Stephens,
Performance Data, and Critiques of Practices in School Algebra 3
2001; Hart, 1981; Kieran, 2007; Küchemann, 1981; Sfard, 1995). Certainly, the phenomenon
is not confined to a few nations. For example, data from the 2011 “Trends in Mathematics and
Science Study” (TIMSS) involving students from over 40 nations suggested that less than 50%
of eighth-grade students worldwide would give a correct answer to the question “If t is a
number between 6 and 9, then t + 5 is between what two numbers?” For the same 2011 TIMSS
study, only 43% of the eighth-grade sample gave a correct response to a question asking them
100
to find the value of 100 – when t is equal to 9 (Mullis, Martin, Foy, & Arora, 2012).
1+ t
Given these performance data, it becomes important to seek answers to a fundamental
question—Why do so many middle-school and secondary-school students experience
difficulty learning elementary algebra? We believe that it is not acceptable to adopt a head-
in-the-sand attitude by asserting that such data are unimportant because the students involved
were only beginning to learn algebra. If one does not expect middle-school and lower-
secondary school students to learn elementary algebra well, then why require them to study
it? In case the reader thinks that we are adopting an unduly negative, or “positivist,”
approach, it is important to state, at the outset, that there is much agreement among
mathematicians and mathematics educators across the world that many secondary-school
students are struggling to learn algebra well (see, e.g., Cai & Knuth, 2011, Ellerton &
Clements, 2011; Kiang, 2012; Kieran, 2007; Kilpatrick & Izsák, 2012; Wu, 2011). Furthermore,
the phenomenon is not new (see, e.g., Sfard, 1995; Porro, 1789). Indeed, in Chapter 2 of this
book we shall argue that ever since the introduction of algebra into secondary-school curricula,
in the seventeenth century, students have experienced difficulties with school algebra.
Not much attention will be given, in this book, to the learning of algebra by elementary-
school children, or by persons studying algebra at higher-level, post-secondary-school
education institutions such as universities and community colleges. Rather, our emphasis will
be on algebra in middle-school and lower-secondary-school classes. In Chapter 2 we will
provide an overview of historical factors which, it has been suggested, have caused school
algebra to be exceedingly difficult for many middle-school and lower-secondary-school
children (aged between about 10 and 15 years). We shall argue that for over three centuries
most students attending algebra classes in secondary schools struggled to understand the
meanings of the main symbols and signs of school algebra, and did not develop relational
understandings of the mathematics being signified by those symbols and signs. An important
part of our thesis will be that parents, education administrators, politicians, mathematics
teachers, mathematics educators, and mathematicians have never carefully identified the
dimensions of this educational problem, and that for centuries there has been an absence of
scholarly historical analyses of the history of school algebra at the secondary-school level.
Historically, the move towards including algebra in secondary school mathematics
curricula relied on the advice of, or textbooks of, outstanding mathematicians such as Isaac
Newton, Alexis-Claude Clairaut, and Leonhard Euler. In the seventeenth, eighteenth and
nineteenth centuries, most school students who studied algebra were in schools which were
selective in the sense that to gain admittance to them a student needed to show evidence that
he or she could do well in mathematics. An artificially high level of expectation for school
algebra was thereby created. Then, in the twentieth century when, gradually, a greater
proportion of children began to proceed to secondary schools, an important question arose—
which students should now be expected to study algebra, and should the subject be redefined
so that the new version would fit the needs and abilities of the new generation of students?
In the twentieth and twenty-first centuries, results from international comparative

performance studies (such as those conducted by the International Association for the
Evaluation of Educational Achievement, and the Organisation for Economic Co-operation
and Development), drew attention to the high performance in school mathematics of students
in Eastern Asian nations such as Japan, Korea, Singapore, Taiwan, and Hong Kong. What
has not been widely recognized is that this “superior” performance of Asian nations was,
arguably, a result, at least partly, of massive participation of students in those nations in a
“shadow education system” by which students received, on average, many hours of tuition
each week in mathematics outside of normal school hours (Bray & Kwo, 2014). For instance,
Toh (2008) reported that 97% of all school students in Singapore were participating in
outside-of-school tuition classes, and that mathematics was the most common subject dealt
with in those classes (see, also, Bray & Lykins, 2012).
The high performance of the East Asian nations has resulted in mathematicians and
education administrators in many European and American nations or states—where the
extent of shadow education tutorial classes is much less than in East Asia—believing that the
teaching and learning of algebra in their schools is poor. Thus, the high standards in school
algebra inherited from the past, when school algebra was reserved for an elite, are not only
being retained, but are being propagated as “normal” and desirable for all.
How Well Do Middle-School Teachers Understand Algebra?

A possible reason why so many beginning algebra students find it difficult to master the
subject is that many of their teachers do not have strong, relational understandings of the
subject and, as a result the students are not experiencing mathematically-strong teaching of
the subject (Kiang, 2012; Perel & Vairo, 1967; Wu, 2011). Several research studies have
generated data which support that contention. Ellerton and Clements (2011), who
investigated the algebra knowledge of 328 U.S. teacher-education students who were seeking
endorsement to become specialist middle-school mathematics teachers, reported data which
suggested that most of the prospective teachers had retained very little of what they had been
asked to learn about algebra at school. All 328 students had passed Algebra I and Algebra II
in U.S. high schools, and about one-third of them had taken calculus classes. All of them had
passed an elementary general mathematics course at university level. In fact, these
prospective middle-school mathematics specialists were taking their last algebra course
before becoming fully qualified teachers of mathematics. Yet, as entries in Table 1.1 reveal,
most of the students, did not have relational understandings of elementary algebra.
Table 1.1 summarizes Ellerton and Clements’s (2011) analyses of data, generated by
the 328 prospective teachers, with respect to four pairs of matching tasks. For every equation
there was a “matching” inequality (e.g., x2 > 4 was regarded as matching x2 = 9). The tasks
shown in Table 1.1 were designed for the purpose of checking whether the students—who
would soon be technically qualified to teach Algebra I—had learned to think holistically
about the meanings of equations and inequalities. For example, to what extent would they be
able to reason that the inequality x2 + 2 > 0 would be true if x were to be any real number?
Background to How and Why this Book was Written 5
Table 1.1
Percentages Correct, 328 Mathematics Teacher-Education Students on Four Equation/Inequalities
Pairs (Ellerton & Clements, 2011)
Equation (“State Acceptable Number (& %) Matching Acceptable Number (&
all real numbers Answer Correct Inequality Answer %) Correct
which would make (n = 328) (n = 328)
the following true.”)
x2 = 9 3, –3 74 (23%) x2 > 4 x > 2 or x < –2 16 (5%)
x2 + 6 = 0 No real 69 (21%) x2 + 2 > 0 All real 53(16%)

solution numbers
4(x + 1) = 4(x – 3) No real 173 (53%) 9(x + 1) > 9(x – 2) All real 77(23%)
solution numbers
(x – 3)(x – 2) = 0 2, 3 194 (59%) (x – 3)(x – 1) > 0 x < 1 or x > 3 2 (1%)
Handwritten reflections (submitted for an assignment titled “Where I went wrong and
why”) by the 328 prospective teachers in the Ellerton and Clements (2011) study confirmed
that in almost all cases they had not thought about the meanings of tasks. Thus, for example,
when asked which real-number values of x would make the statement x2 > 4 true, only 5% of
the prospective teachers gave a correct answer (see Table 1.1). The most common answer
was x > 2, and the second most common response was x > ±2. Although 59% of the
prospective teachers gave a correct response to the quadratic equation (x – 3)(x – 2) = 0, half
of those who were correct thought that the x in (x – 3) stood for “3” and, simultaneously, the
x in (x – 2) stood for “2.” In their written responses to inequality tasks, such as (x – 3)(x – 1) >
0, and in interviews, none of the 328 students sketched a graph.
Yet, these prospective teachers had studied school algebra during the so-called “NCTM
Standards” period—when meaningful learning of mathematics was supposed to have been a
matter of paramount importance. In their written reflections, students typically wrote that they
had “forgotten” what they had learned in school algebra classes. Even those who had studied
calculus tended to make that claim. If that was indeed the case, then one must ask—what was
the point of getting them to study algebra in the first place? What is even more disquieting is
that our experience has been that many experienced mathematics educators do not want to hear
about such data—they tend to use pejorative language such as “positivist” to describe any
interpretation which suggests that the data indicate that improvement is needed. Many do not
want to be reminded of Liping Ma’s (1999) comparative study of the mathematical knowledge
of elementary school teachers in China and the United States of America, which revealed that
U.S. elementary teachers have weaker knowledge of the structures to be associated with
elementary number properties than their mathematically less-qualified counterparts in China.
The Background to How and Why this Book was Written

Most of this book is concerned with describing three research studies on the teaching and
learning of algebra in five middle-school classes, conducted by a team of six researchers—
three school teachers and three mathematics education researchers—during the period
September 2014 through March 2016. The studies took place in two schools in a midwestern
state of the United States of America. So, at the outset, it is germane to ask—why then
should this book be part of a Springer series on the history of mathematics education? The
first three chapters of the book, and the last chapter (Chapter 10), have been written with the
express purpose of answering that question.
The investigations described in this book were originally intended to be basically a study
of whether seventh- and eighth-grade students in two middle schools, and their teachers,
were able to cope with the algebraic demands of sessions in which the students worked on
tasks which required them to apply algebraic ideas in the context of modeling, problem-solving,
and algebraic structural considerations. A pilot study (Kanbir, 2014) was aimed at developing
an instrument which would be suitable for use in subsequent interventions. This aspect of the
study seemed to be successful in that it generated an instrument which was regarded as valid
by the teacher and the three authors, and had a Cronbach-alpha reliability of 0.84.
In a second pilot study (Kanbir, 2016), the participants were the students in three seventh-
grade classes, their teacher, and the three authors of this book. The teacher participated in one-
on-one professional development sessions led by the three authors, pre- and post-intervention
pencil-and-paper test scores were obtained for all students, and pre- and post-intervention
one-on-one interviews took place with 18 selected students—six high achievers, six middle
achievers, and six low achievers. For the intervention, the teacher led students through a
series of questions which were not unlike those on the written tests used in the study, and the
tasks used in the one-on-one interviews. Although analyses of pre- and post-intervention test
and interview data revealed that after the intervention classes the students were able to give
correct answers to a higher proportion of questions than before, effect sizes for the
intervention were small. Furthermore, the post-intervention interview data revealed that
students still did not understand, in meaningful ways, the algebra that they had been taught.
The third study is described in Chapters 4 through 9 of this book. It, too, was an
intervention study. In this third study, 32 participating seventh-grade students were engaged
in small-group discussions and each contributed to group presentations to the whole class.
The effect sizes were large, and comparisons of pre- and post-intervention interview data
indicated that much higher levels of understanding were achieved.
Overview of this Book

This book has five sections. The first three of the sections have just one chapter each.
• The first section comprises this first chapter. It introduces readers to what we regard
as the basic problem of school algebra—namely, there is a need to find out why so
many beginning algebra students find algebra so difficult. It also summarizes the
pilot and main studies described in this book.
• The second section comprises the second chapter. It offers a summary of the history
of school algebra and, in particular, identifies six purposes which have been
attributed to school algebra at various times during the past 350 years. The reason
for offering such a serious overview of the history of school algebra is because we
believe the fundamental difficulty cannot be properly studied unless it is placed in
the context of purposes which have, in the past, been associated with the teaching
and learning of algebra in middle- and lower-secondary schools.
• The third section comprises Chapter 3. It discusses a design-research approach (Kelly
& Lesh, 2000; Kelly, Lesh, & Baek, 2008) which was devised and employed in an
Overview of this Book 7
attempt to respond realistically to the fundamental problem. The issue addressed is

this: What needs to be done if there is to be a good chance of solving the problem?
In answer to that question it is argued that the major problem has always been one of
getting students to be fluent in their receptive and expressive understandings of the
major signifiers of elementary algebraic concepts and principles. Then, a summary is
provided of what the present authors believe needs to be done in order that students
will become comfortable and competent in using those signifiers so that they will be
able to work on algebraic tasks which are deemed to be appropriate for them at their
stages of mathematical development. From this description, it should be obvious to a
reader steeped in the literature that a semiotic approach to the research was adopted,
the object being to help learners link commonly-used “signifiers” with desired
mathematical “signifieds.” Not one, but three, theoretical approaches were adopted, in
order to frame the research which would be carried out.
• The fourth section comprises five chapters (Chapters 4 through 8) and describes a
successful intervention study in which 32 beginning students were introduced to
basic concepts involving algebraic structures and mathematical modeling.
• The fifth section comprises the final two chapters in which research questions are
answered and comment is made on implications of the investigation for teaching
secondary-school algebra and for further research.
An unusual feature of this book is that, despite the fact that most of it is concerned with
details relating to the planning, implementing, and evaluation of a mixed-method research
study, it appears as part of Springer series on the history of mathematics education. The
reason behind the decision to include this book in the series was that the historical analysis
provided in Chapter 2 provided an important guide when decisions were being made about
the design of the main study. There is a sense in which the design-research approach to the
main study suggested that not one, but three, theoretical bases for the study might be
appropriate—Peirce’s (1931) triadic signifier-interpretant-signified theory, Herbart’s (1904)
theory of apperception, and Del Campo and Clements’s (1987) receptive-expression theory
of classroom discourse. Parts of those three theories were combined because the historical
analysis suggested that, when bundled together, the hybrid theoretical position would provide
the most suitable theoretical base for what was to be studied.
There is much evidence indicating that algebra is taught and learned in different ways
around the world. It is wrong to think, for example, that there is a single South-East-Asian way
of teaching algebra (Leung, Park, Holton, & Clarke, 2013). Even within the same school,
different algebra teachers can adopt different teaching strategies. However, no matter where
young people are asked to learn algebra for the first time, they will be faced with the challenge
not only of learning the chief signs and conventions of school algebra, but also of connecting
those signs and conventions to properties of numbers, to graphical representations, and to
posing, modeling, and solving real-world problems. Those who persist will be confronted
with the concept of a variable, and be expected to acquire a language which will facilitate their
attempts to generalize. In that sense, learning algebra should involve more than becoming
familiar with a syntax by which letters are manipulated according to well-defined rules. This
book is concerned with helping middle-school students come to grips with the “essence” of
school algebra—enabling them to learn, receptively and expressively, the key “signifiers”
and to connect those with mathematical objects (what one might call the “signifieds”), and
then be able to apply what they have learned in a range of mathematical and real-world contexts.
8 References for Chapter 1
It is not a matter for surprise that middle-school students find it difficult to learn the
syntax and semantics to be associated with the signifiers of school algebra, for it took the
mathematicians of history a long time before they arrived at modern algebraic notations for
what, from a historical perspective, were conceptually difficult “signifieds.” It was not until
the sixteenth and seventeenth centuries that the present symbols (involving x, y, a, b, x2, √x,
etc.) began to be used—before that, rhetorical and syncopated notations were adopted, even
when the object was to solve relatively simple linear or quadratic equations (Cajori, 1890,
1928; Sfard, 1995). Furthermore, the algebra to be associated with negative numbers, with
“imaginary” solutions to equations such as x2 + 1 = 0, as well as with the concepts of real
numbers, the real-number line, the Cartesian plane, the concept of a variable, and
relationships between variables, was only represented in modern notations from the
seventeenth century onwards (Cajori, 1890).
Thus, modern school algebra expects young children to learn, quickly, something
which took mathematicians a very long time to conceptualize and notate. Seen from that
vantage point, it is not at all surprising that the signifiers of school algebra, and their
associated signifieds, present major pedagogical challenges to teachers, and that young
learners struggle to learn “elementary” algebra. Ironically, many modern-day mathematicians
are among those who seem to think that the main content and themes of “high-school
algebra” should be easily acquired by most children aged between 10 and 16 years.
References
Booth, L. R. (1984). Algebra: Children’s strategies and errors. Windsor, Berks, UK: NFER-
Nelson.
Booth, L. R. (1988). Children’s difficulties in beginning algebra. In A. F. Coxford & A. P.
Schulte (Eds.), The ideas of algebra: K–12, 1988 Yearbook. Reston, VA: National
Council of Teachers of Mathematics.
Bray, M., & Kwo, O. (2014). Regulating private tutoring for public good: Policy options for
supplementary education in Asia. Hong Kong, China: Education Research Centre
(CERC), UNESCO.
Bray, M., & Lykins, C. (2012). Shadow education: Private supplementary tutoring and its
implications for policy makers in Asia. Manila, The Philippines: Asian Development
Bank.
Cai, J., & Knuth, E. (2011). A global dialogue about early algebraization from multiple
perspectives. In J. Cai & E. Knuth (Eds.), Early algebraization: A global dialogue from
multiple perspectives (pp. vii–xi). Heidelberg, Germany: Springer.
Cajori, F. (1890). The teaching and history of mathematics in the United States (Circular of
Information No. 3, 1890). Washington, DC: Bureau of Education.
Cajori, F. (1928). A history of mathematical notations. La Salle, IL: The Open Court
Publishing Co.
Chazan, D. (2012). The shifting landscape of school algebra in the United States. In C. E.
Greenes & R. Rubinstein (Eds.), Algebra and algebraic thinking in school mathematics
(pp. 19–33). Reston, VA: National Council of Teachers of Mathematics.
Del Campo, G., & Clements, M. A. (1987). A manual for the professional development of
teachers of beginning mathematicians. Melbourne, Australia: Association of
Independent Schools of Victoria.
References for Chapter 1 9
Ellerton, N. F., & Clements, M. A. (2011). Prospective middle-school mathematics teachers’

knowledge of equations and inequalities. In J. Cai & E. Knuth (Eds.), Early
algebraization: A global dialogue from multiple perspectives (pp. 379–408).
Heidelberg, Germany: Springer.
Fujii, T., & Stephens, M. (2001). Fostering an understanding of algebraic generalizations
through numerical expressions. In K. Stacey, H. Chick, & M. Kendal (Eds.),
Proceedings of the 12th Conference of the International Commission on Mathematical
Instruction: The future of the teaching and learning of algebra (Vol. 1, pp. 258–264).
Melbourne, Australia: The University of Melbourne.
Hart, K. M. (Ed.). (1981). Children’s understanding of mathematics, 11−16. London, UK:
Murray.
Herbart, J. F. (1904). The science of education. London, UK: Sonnenschein.
House, P. (1988). Reshaping school algebra: Why and how? In A. F. Coxford (Ed.), The
ideas of algebra, K–12 (pp. 1–7). Reston, VA: National Council of Teachers of
Mathematics.
Kanbir, S. (2014, November). Two approaches: Beginning algebra students’ variable concept
development. Professional project presented to the Group for Educational Research in
Mathematics at Illinois State University, Normal, IL.
Kanbir, S. (2016). An intervention study aimed at enhancing seventh-grade students’
development of the concept of a variable (Doctoral dissertation). Available from
ProQuest Dissertations and Theses database (Kanbir_ilstu_0092E_10787.pdf).
Kelly, A. E., & Lesh, R. A. (Eds.). (2000). Handbook of research design in mathematics and
science education. Mahwah, NJ: Lawrence Erlbaum.
Kelly, A. E., Lesh, R. A., & Baek, J. Y. (Eds.). (2008). Handbook of design research
methods in education: Innovations in science, technology, engineering and
mathematics learning and teaching. New York, NY: Routledge.
Kiang, L. Y. (2012). Mathematics K–12: Interview with Wu Hung-Hsi. Mathematical
Medley, 38(1), 2–15.
Kieran, C. (2007). Learning and teaching algebra at the middle school through college levels.
In. F. K. Lester (Ed.), Second handbook of research on mathematics teaching and
learning (pp. 707–762). Charlotte, NC: New Age Publishing, Reston, VA: National
Kilpatrick, J., & Izsák, A. (2012). A history of algebra in the school curriculum. In C. E.
Küchemann, D. E. (1981). Algebra. In K. M. Hart (Ed.), Children’s understanding of
mathematics: 11−16 (pp. 102–119). London, UK: John Murray.
Leung, F. K. S., Park, K., Holton, D., & Clarke, D. (Eds.). (2013). Algebra teaching around
the world. Rotterdam, The Netherlands: Sense Publishers.
Lim, T. H. (2000). The teaching and learning of algebraic equations and factorisation in O-
level Mathematics: A case study (Unpublished M.Ed. dissertation). Universiti Brunei
Darussalam.
Ma, L. (1999). Knowing and teaching elementary mathematics: Teachers’ understanding of
fundamental mathematics in China and the United States. Mahwah, NJ: Erlbaum.
MacGregor, M. E. (1991). Making sense of algebra: Cognitive processes influencing
comprehension. Geelong, Australia: Deakin University.
Mullis, I. V. S., Martin, M. O., Foy, P., & Arora, A. (2012). TIMSS 2011 international
results in mathematics. Chestnut Hill, MA: TIMSS & PIRLS International Study
Center, Boston College.
Peirce, C. S. (1931/1958). Collected papers (Vol. 1–VIII). In C. Hartshorne, P. Weiss, & A.
Burks (Eds.), Cambridge, MA: Harvard University Press.
Perel, W. M., & Vairo, P. D. (1967). New mathematics and old teachers. The Educational
Forum, 31(3), 345–348.
Porro, F. D. (1789). L’algèbre selon ses vrais principes. London, UK: Chez Barois l’Aîné.
Schorling, R. (1926). Suggestions for the solution of an important problem that has arisen in
the last quarter of a century. In C. M. Austin, H. English, W. Betz, W. C. Eells, & F. C.
Touton (Eds.), A general survey of progress in the last twenty-five years: The first
yearbook (pp. 58–105). Reston, VA: National Council of Teachers of Mathematics.
(Republished, 1995)
Sfard, A. (1995). The development of algebra: Confronting historical and psychological
perspectives. Journal of Mathematical Behavior, 14, 15–39.
Smith, D. E. (1926). A general survey of the progress of mathematics in our high schools in
the last twenty-five years. In C. M. Austin, H. English, W. Betz, W. C. Eells, & F. C.
(Republished, 1995).
Toh, M. (2008, June 17). Tuition Nation. Asiaone News.
http://www.asiaone.com/News/Education/Story/A1Story20080616-71121.html
Vaiyavutjamai, P. (2004). Factors influencing the teaching and learning of equations and
inequations in two government secondary schools in Thailand (Unpublished PhD
dissertation). Universiti Brunei Darussalam.
Vaiyavutjamai, P. (2006). Grade 9 students’ errors on gently challenging algebra tasks: The
good news and the bad news. In H. S. Dhindsa, I. J. Kyeleve, O. Chukwu, Lim Siew
Bee, Hjh Zaitun bte Hj Taha, A. Baimba, & S. Upex (Eds.), Shaping the future of
science, mathematics and technical education (pp. 212–221). Gadong, Brunei
Darussalam: Universiti Brunei Darussalam.
Wu, H. (2011). The mis-education of mathematics teachers. AMS Notices, 58(3), 372–384.
Chapter 2
Historical Reflections on How Algebra Became a Vital Component
of Middle- and Secondary-School Curricula
Abstract: The chapter begins by identifying, and placing in their historical contexts, the
main issues in a longstanding debate over the purposes of school algebra. The following six
purposes for school algebra, recognized by various writers over the past three centuries, are
then identified: (a) algebra as a body of knowledge essential to higher mathematical and
scientific studies, (b) algebra as generalized arithmetic, (c) algebra as a prerequisite for entry
to higher studies, (d) algebra as offering a language and set of procedures for modeling real-
life problems, (e) algebra as an aid to describing structural properties in elementary
mathematics, and (f) algebra as a study of variables. The question is then raised, and
discussed, whether school algebra represents a unidimensional trait.
Keywords: Algebra as a gatekeeper, Algebra as generalized arithmetic, History of

mathematics education, John Perry, Polynomial equations, Purposes of school algebra,
School algebra, Structural approach to algebra
Providing Historical Frameworks for Mathematics Education Research
There is a tradition in mathematics education of requiring graduate students who are

preparing a research dissertation to provide a clear theoretical framework for their study, and
in Chapters 3 and 5 such a framework will be provided for the main study which is described
in this book. However, there is no firmly established tradition requiring researchers to
provide a historical framework for their proposed research.
A strong—but nonetheless indirect—statement in favor of the need to provide historical
frameworks for mathematics education research projects was given in Springer’s Third
International Handbook of Mathematics Education (Clements, Bishop, Keitel, Kilpatrick, &
Leung, 2013). In that Handbook there were four major sections, each dealing with a different
theme, and each section was structured on the basis of past, present and future aspects of the
theme. The first chapter in each section was concerned with analyses of antecedents (“How
did we get to where we are now?”); the “middle” chapters provided analyses of present-day
key issues for the theme (“Where are we now, and what recent events have been especially
significant?”); and the final chapter in each section reflected on future policy (“What should
we do to improve the quality of the teaching and learning in the future?”).
Like the Third International Handbook, this book adopts a past-present-future
organizational structure. This chapter, and the next, offer historical and theoretical frameworks,
respectively, for the intervention study described in Chapters 3 through 9. The last chapter
looks to the future. The aim of the intervention study was to suggest how present practices in
relation to middle-school algebra might be improved. The intermediate chapters are based on

12 Ch. 2: Historical Reflections on the Purposes of School Algebra
a doctoral study conducted by the first author (Kanbir) and co-supervised by the second and
third authors (Clements and Ellerton). It is hoped that the historical framework provided in
this chapter will serve as a model for the kinds of historical analyses that one might
reasonably expect to find in documents in which mathematics education research studies are
described.
The National Council of Teachers of Mathematics’ (NCTM’s) Fifty-Sixth and Seventieth
Yearbooks (NCTM, 1988, 2012) were both devoted to algebra education, and each made many
references to difficulties which U.S. school students had experienced with algebra in the past.
The reasons offered were not always consistent. In the Seventieth Yearbook, for example, Cai
and Moyer (2012) claimed that “U.S. students persist in using concrete, arithmetic-based
problem-solving strategies, even when such strategies are inefficient … [but]… Chinese
students effectively use abstract, algebra-based strategies to solve the same problems that U.S.
students approach concretely with arithmetic-based strategies” (p. 178). Yet, many authors
(e.g., Chambers, 1964; Dienes, 1960) have recommended that concrete aids should be used to
teach young children algebra before arithmetic.
Our timeframe for discussing why many students fail to learn algebra well will be
much wider than that taken into account by most other writers on the subject. In the editors’
preface, provided at the front of this book (and at the front of other books in Springer’s
history of mathematics education series), Nerida Ellerton and Ken Clements wrote:
We hope that the series will provide a multi-layered canvas portraying the rich
details of mathematics education from the past, while at the same time presenting
historical insights which can support the future. This is a canvas which can never
be complete, for today’s mathematics education becomes history for tomorrow. A
single snapshot of mathematics education today is, by contrast with this canvas,
flat and unidimensional—a mere pixel in a detailed image.
In this chapter we will explore and seek to contribute to the detailed image which is
beginning to take shape on a canvas which is primarily concerned with the history of algebra
education in schools.
Recently-published summaries of aspects of the history of school algebra (see, e.g.,
Artigue, Assude, Grugeon, & Lengant, 2001; Cai & Knuth, 2011a; Chazan, 2012; Chorlay,
2011; da Ponte & Guimarāes, 2014; Schubring, 2011; Spielhagen, 2011), have been mainly
concerned with situations relating to school algebra in Europe and America. No
comprehensive history of school algebra, viewed from global standpoints, has ever been
published and although this book will not fill the void, this second chapter will offer a
framework for such a work. In this chapter, we summarize various emphases in school
algebra over the past 350 years. Chapter 3 draws attention to theories which might be useful
for explaining why school algebra has caused so much difficulty for so many learners. Most
of the other chapters describe an intervention study in which some middle-school students,
who were beginning to learn algebra, were able to learn fundamental algebraic concepts well.
The lack of a history of school algebra written from a fully internationalized perspective
is a serious matter given that the seventeenth and eighteenth centuries were marked by massive
colonization programs, whereby the colonizers (mainly European nations) tended to introduce
school mathematics textbooks into their colonies, and the languages used in most of those
textbooks were those of the colonizing powers. The chief authors of the textbooks were, almost
always, based in Europe, and textbooks were written which seemed to suggest that school
mathematics should be a culture-free exercise. Even for students in the European homelands,
the textbooks were designed to suit the perceived needs of children of elites. The first school
Historical Reflections on the Purposes of School Algebra 13
algebra textbooks used in the “colonies” were often written from high mathematics vantage
points and, we would argue, were entirely unsuited to the needs of indigenous children, or of
children of slaves, or of other children whose command of the spoken and written language
used by mathematics teachers or authors of mathematics textbooks was not strong (Clements,
Grimison, & Ellerton, 1989).
There were attempts to change the situation—in 1855, in Victoria, Australia, for
example, two professors (William Hearn and William Wilson) of the recently-established
University of Melbourne, wanted to create university-entrance regulations which were
different from those of the old “home” universities in Europe. Hearn argued:
There are many parents who wish their sons to enter life at an early age but would
gladly send them to the University if they could obtain there the amount of quality
of education which they wish them to acquire. Such persons think that the study of
the classics or the higher mathematics is a needless expenditure of time, and that
these subjects, while they have no direct bearing upon their children’s future
occupations, tend to distract young men from, and give them a distaste for, more
practical pursuits. The soundness of such views is not the question. If such an
opinion exists, and it is prevalent at home, and probably still more so here, the
making these studies a sine qua non for a degree would amount to a practical
exclusion of the class to whom I have referred. (University of Melbourne, 1855)
However, the colonial conservatives who would administer the yet-to-be-opened University
of Melbourne did not approve of such a radical point of view, and they decided that passes in
Greek, Latin, Arithmetic, Algebra and Euclid, at the University’s matriculation examination,
would be required of all persons wishing to take degree courses. Although the course
prescribed for matriculation Algebra only went as far as “quadratic equations in one
unknown” (Clements et al., 1989), any idea that algebra should be “for all children” was not
part of the thinking of those who administered the University.
The above University-of-Melbourne episode draws attention to the need to recognize
that, from its beginnings, school algebra was, by design, not intended for everyone.
Secondary education, of which algebra would become an integral part, was for “the chosen.”
That was the intention, and any respectable history of school algebra should make that clear.
It took centuries before the idea of “algebra for all” would be put forward with any degree of
conviction (Lawson, 1990; Steen, 1992), and even when that did occur, the challenge of
unravelling the forms of algebra education which, over the centuries, had taken root as
“normal” was something which society had to face—usually against staunch opposition from
those who wanted to maintain the status quo. The subconsciously-held traditions of what was
normal led to forms of algebra being prescribed for school algebra programs which were not
suited to the needs and backgrounds of many students. Furthermore, issues associated with
forms of classroom organization which might be appropriate for school algebra programs
were rarely considered. Therefore, in order to study the history of school algebra adequately,
one needs not only to take account of the intended curriculum (as summarized in textbooks,
and in formal curriculum statements prepared by local, state, or national education
authorities) but also the implemented curriculum (as represented in algebra cyphering books,
or workbooks, or what transpired in algebra classes), and the received curriculum (as
represented by student recollections, and data from tests and examinations) (Westbury,
1980).
Da Ponte and Guimarāes (2014) stated that in the seventeenth century, algebra “was
understood as the study of polynomial equations” and “the study of operations with quantities”
(p. 461). They argued that school algebra received a fillip when it began to be applied to the
study of geometry, using Descartes’ algebraic notations and also notations introduced by
Gottfried Leibniz, in his quest to create a universal science. Leibniz argued that although we
use natural language as a principal means of communication, in fact natural language is
sufficiently obscure, from a technical linguistic point of view, that it is difficult to use it to
reason logically (Loemker, 1969). Accordingly, Leibniz emphasized the importance of
creating symbols which could serve as well-defined signs for representing not only important
mathematical “objects,” but also relationships between those objects (Cajori, 1928).
Algebra in Secondary School Mathematics: The Debate Over Purpose

For at least three centuries there has been much controversy within and between
mathematicians and mathematics educators over what school algebra should embrace. In
1789, for example, François Porro a French commentator, claimed that the principles of
algebra as laid down in the past were erroneous and the source of many faults. Porro (1789)
complained bitterly about the faulty use of statements such as “minus by minus gives plus”
(p. 1). However, such statements have continued to be used in school classrooms, and slowly
but surely more and more students have been required to study the subject (see, e.g.,
Kilpatrick & Izsák, 2012). In 1895, the National Education Association, in the United States
of America, called for the introduction of algebra into the seventh- and eighth-grade curricula
of U.S. schools, as a “transitional step to algebra proper” in the junior high school. Seven
years later, Eliakim Hastings Moore, President of the American Mathematical Society, urged
the adoption of a problem-solving, laboratory approach to school mathematics, with a heavy
emphasis on the role of graphs and the concept of function (Jones & Coxford, 1970; Roberts,
2012). Moore’s ideas were based on an approach to school mathematics which had been
developed by John Perry, a British engineer and applied mathematician. Although Perry’s
views on mathematics education were becoming well known around the world (e.g., in
Australia—see Clements, 1979), they were opposed by David Eugene Smith, a prolific writer
of school mathematics textbooks and Professor of Mathematics at Teachers College,
Columbia University, New York (Clements, Keitel, Bishop, Kilpatrick & Leung, 2013).
Ironically, Smith would become a friend of Felix Klein, the German mathematician who
argued that a “function approach” to school algebra was desirable (Clements, Keitel et al.,
2013; Kilpatrick & Izsák, 2012; Roberts, 2012).
The dispute between Moore and Smith, over what should be the role of algebra in school
curricula, revealed that the content of school algebra courses had become part of a political
skirmish in which unlikely alliances held sway. Smith’s rejection of Perry’s laboratory
approach to school algebra, an approach which emphasized the concept of a function, would be
reflected in the relative lack of emphasis on functions in many of Smith’s algebra textbooks.
Obviously, Smith had much to lose, because if Perryism achieved popularity across the United
States of America then he would have had to rewrite his numerous algebra textbooks—for if he
did not then they would have become outdated. It was not surprising, then, that Smith
maintained that Perry’s approach was extreme, and was something which was not accepted by
mathematics teachers, particularly those in the Eastern states of the United States of America
(Clements, Keitel et al., 2013; Husky, 2011). Despite the fact that Perry, Moore, Smith, and
Klein were all key contributors to a debate over the purpose and form of school algebra,
the result was hardly enlightening. Fundamental disagreements over the purposes of school
algebra were expressed, and neither unity nor clarity was achieved (Cajori, 1890; Roberts,
Algebra in Secondary School Mathematics: The Debate Over Purpose 15
2012). A similar lack of clarity of purpose would be characteristic of debates over the purposes
of school algebra that would occur in many nations throughout the twentieth century.
In this section we offer a summary of the modern history (post-1670) of school algebra.
The summary will not be fully documented because, as da Ponte and Guimarāes (2014) have
stated, although there is an abundant literature on the history of algebra, “the history of the
teaching of algebra is largely unwritten” (p. 459). Furthermore, most published histories of
algebra are seriously deficient so far as acknowledgement of non-European roots of the
subject are concerned (Al-Khalili, 2011; Cajori, 1890; Colebrooke, 1817/2013), and the
relatively few attempts to offer a history of school algebra have rarely looked beyond what
transpired in European and American schools. So far as the history of algebra education is
concerned, there is far too much to be adequately condensed into one summary chapter, and
this present chapter should be regarded as an attempt to build on, and complement, da Ponte
and Guimarāes’s (2014) work. Here, our emphasis will be on a consideration of the purposes
of school algebra, as seen from historical perspectives based on data from numerous
countries.
Table 2.1 summarizes the following six purposes of school algebra which we have
identified as having been assumed to be true by persons—certainly not all persons—
interested in secondary school education at various times during the past 350 years. We
would maintain that although each of the six purposes is conceptually separable from the
other five, there are, nevertheless, intersections.
1. School algebra as a body of knowledge which prepares students for higher
mathematical and scientific studies;
2. School algebra as the study of generalized arithmetic;
3. School algebra as a gatekeeper for entry to higher studies;
4. School algebra as an integral component of mathematical modeling of real-world
contexts;
5. School algebra as a vehicle for generalizing numerical, geometrical, and other
mathematical structures; and
6. School algebra as the study of variables—symbols which can represent different
quantities, and display relationships between those quantities.
We would also note, at the outset, that although all six purposes are still regarded as
important at the present time, there is much dispute over the relative importance of each.
According to Zalman Usiskin (1988), two issues “relate to the very purposes for
teaching and learning algebra” (p. 11)—first, the extent to which students should learn to
manipulate algebraic expressions; and second, the role of functions and the timing of their
introduction. Usiskin outlined the following four conceptions of algebra:
1. Algebra as generalized arithmetic;
2. Algebra as the study of procedures for solving certain kinds of problems;
3. Algebra as the study of relationships among quantities; and
4. Algebra as the study of structures.
Our six “purposes” emerged, for us, during our analyses of large data sets, including
handwritten and printed documents from many nations (Ellerton & Clements, 2012).
Table 2.1
Six Distinguishable Purposes for Secondary School Algebra
Period Assumed Key Writers or Comments
Purpose Players
1693– A body of René Descartes, Algebra can be seen as an important, beyond-arithmetic
2017 knowledge Humfrey component of mathematics. Mathematicians recognized
essential for Ditton, Isaac that knowledge of algebra could assist students to
higher Newton, comprehend and express concepts and principles within
mathe- Gottlieb higher mathematics (e.g., in conic sections, trigonometry,
matical and Leibniz, calculus, and mechanics). This idea of expanding school
scientific Sylvestre mathematics curricula through algebra would receive
studies François decisive fillips when Cartesian graphs, algebraic
structures, and calculus began to be included in intended
Lacroix
school mathematics curricula.
1700– Generalized Louis Bourdon, Emphasis is on the language, syntax, and, ultimately, the
2017 arithmetic Etienne Bézout, semantics of elementary algebraic operations. With this
Nicolas Pike, focus, solving an equation came to be equivalent to
Warren Colburn “finding the unknown number(s).” An axiomatic approach
to algebra, similar to that of Euclidean geometry, was
sometimes adopted.
1800– A pre- Persons defining During the 19th century many countries required school
2017 requisite for university- students who intended, subsequently, to enter prestigious
entry to entrance pre- universities to succeed in a school subject designated
higher requisite “Algebra.” This tended to change after the 1960s, although
studies subjects it remained the case in the United States of America.
1870– A language James Hodgson, Since the 1870s this has often been called the “functional
2017 for Sylvestre F. approach to algebra.” Students should learn to describe
modeling Lacroix, sequences recursively and explicitly, to prepare tables of
real-life John Perry, values, and to plot and interpret Cartesian graphs (which
problems depict relationships). This emphasis received a major fillip
Felix Klein,
when, across the world, calculus began to be introduced
Eliakim H.
into school curricula.
Moore
1870– An aid for Felix Klein, In the 1950s and 1960s, Caleb Gattegno and Zoltan Dienes
2017 describing “Bull” argued that elementary-school students should begin to
basic Wentworth, learn algebra before arithmetic, and that the emphasis
structural Nicolas should be on structural properties. But students struggled to
properties Bourbaki, recognize that in a statement like “for all real numbers a, b,
c, it must be true that a + (b + c) = (a + b) + c,” the a, b, c
Caleb Gattegno,
were, in fact, being used as variables. In the 1870s Klein
Zoltan Dienes
applied function concepts, structural ideas, and associated
symbolisms, to geometry.
1960– A study of SMSG authors, To solve an equation (or an inequality) was to be seen as
2017 variables Robert Davis, finding values of a variable which would make an open
Daniel Chazan sentence true, and which would make it false (and not
merely finding “unknown” numbers). Structural properties
(e.g., the distributive property), stated in algebraic
language, were to be seen as statements involving
variables. Tables of values and Cartesian graphs were to be
regarded as depicting relationships between variables.
When we first prepared Table 2.1 we had not read Usiskin’s (1988) chapter, and
therefore it became a matter of interest to us to compare Usiskin’s four conceptions with our
six purposes. The two purposes in Table 2.1 which were not considered in Usiskin’s list were
“a body of knowledge essential for higher mathematical and scientific studies” and “a
prerequisite for entry to higher studies.” The first of these two additional purposes relates to
preparing students, cognitively, for higher studies, and the second to whether successful
participation in school algebra can, legitimately, provide a useful means of evaluating
whether students are ready to proceed with higher studies. The two are similar but clearly
separable. Our analyses of historical data suggest that both have been very significant in the
history of school algebra.
Another seemingly-related list was offered by Filloy and Sutherland (1996), who
identified the following six “criteria” which could “act as a framework from which all
aspects of algebra curriculum development” could be analyzed:
• Mathematical concepts and related sign systems;
• Mathematical cognition and cognitive tendencies;
• Teaching and learning and abstraction processes;
• The relationship between algebra and practical knowledge;
• New technologies for teaching and learning algebra; and
• Mathematical modeling—the analytical and instrumental tool of algebra within
other areas of knowledge. (p. 140)
Filloy and Sutherland’s (1996) list was directed at establishing criteria for curriculum
development—which was not our goal when we identified the six “purposes” listed in Table
2.1. Missing from Filloy and Sutherland’s statement, and associated text, was any attempt to
place their criteria within a reasonably tight historical framework. Of course, they are not the
only scholars to have written about algebra education, especially algebra in schools, without
considering historical contexts. The time has come, we believe, for mathematics education
scholars to attempt to place all decent research in their field within well-researched historical
frameworks (Clements, Bishop et al., 2013).
If, indeed, the six purposes listed in Table 2.1 are separable—as we claim they are—
then important curriculum questions arise. Have all six purposes always been deemed
important in planning school algebra curricula? If the answer is “No,” then which have been
regarded as appropriate for which students, in which schools, and when, and why? Is school
algebra a unidimensional field of study and, if it is not, what meaning should we give to the
term “algebra”? Do current school curricular statements and algebra textbooks take sufficient
account of the six purposes?
But, why six purposes, and not five? Or seven, or eight? There could be more than six
purposes—for instance, new purposes might be associated with function concepts, or the
availability and use of new technologies. Then again, there could be less than six purposes.
For example, should algebra continue to be regarded as a gatekeeper for entry to higher
studies? And, perhaps, there might be only one overarching purpose (such as “providing
students with a language which will help them to learn to generalize”), which would
synthesize the six purposes. Although the purposes identified in Table 2.1 are important from
a mathematics curriculum perspective, it could be the case that there is no obvious best way
of describing them adequately. One wonders which forms of words, and which conceptual
structures, would be needed so that a description would be adequate.
We now comment more fully on each of the six purposes identified in Table 2.1.
Algebra as a Body of Knowledge Essential for Success in Higher Mathematical and Scientific
Studies
Ellerton and Clements (2017) have argued that the first secondary school at which
algebra was part of the mathematics curriculum was the Royal Mathematical School
(hereafter “RMS”) within Christ’s Hospital, in central London, England. RMS was
established in 1673, with the specific mission of preparing boys aged between 12 and 16 to
become mathematically-competent apprentices in the Royal Navy or the merchant marine.
For the first 20 years of RMS’s existence, algebra was not formally part of its intended
mathematics curriculum, but in 1693 a request by Edward Paget the Cambridge-trained RMS
master, that the School should publish an algebra textbook, written in the Latin language by
Johannis Alexandri, a Swiss mathematician, was approved by school authorities. This book
was republished by the school in 1709, only this time two different versions appeared—one
in English and the other in Latin (Alexander, 1709; Alexandri, 1693). The 1709 English-
language version included lengthy appendices written by Paget (who was no longer at
Christ’s Hospital) and by Humfrey Ditton, the master responsible for a “New Mathematical
School” (NMS), which co-existed with the RMS at Christ’s Hospital at that time (Alexander,
1709).
Although Ditton’s NMS was supported by Isaac Newton, it existed for 10 years only,
and was notoriously unsuccessful. Students at Christ’s Hospital simply did not want to be
part of the New Mathematical School, but RMS—of which James Hodgson was the master—
was able to secure the requisite number of students each year. Ditton (1709) stated that he
had written the appendix (which occupied 128 pages) to the 1709 Christ’s Hospital edition of
Alexander’s (1709) book with the intention of making the book “still more useful to young
beginners” (p. ii). Isaac Newton had approved the publication of the book, and Newton’s
former Trinity College protégé, Edward Paget, formerly RMS master at Christ’s Hospital and
someone whom Ditton described as an “excellent and everyway learned mathematician”
(p. 105), had contributed 23 pages, within an appendix, to the “construction of solid
problems” (p. 105). However, when Ditton died in 1715 NMS was abandoned and would
never be re-established (Ellerton & Clements, 2017).
Experiences of the three authors of this book in teaching mathematics to learners of all
ages, in many nations, has left us in no doubt that Ditton and his supporters, Isaac Newton
and Edward Paget, seriously over-estimated what RMS and NMS students would be capable
of learning so far as algebra was concerned. Figure 2.1 shows page 10 of Ditton’s (1709)
appendix—the algebra is concerned with whether surd quantities are “commensurable” or
“incommensurable.” Given that RMS and NMS students would have been aged from 12 to
16 years, the level of abstraction is such that we are confident that most of the students would
not have been able to comprehend it.
A similar comment would apply to the mathematics shown in Figure 2.2, which is from
page 107 of Ditton’s (1709) appendix. This page was prepared by Edward Paget, but was
based on material in a geometry textbook by Descartes. The diagram in Figure 2.2 was
headed “A general way of constructing all solid problems reduc’d to an equation of three or
four dimensions” (p. 106). Once again the level of abstraction, and the remoteness of the
topic under consideration, would have meant that hardly any of the RMS or NMS boys
would have been able to understand the text.
Figure 2.1. Page 10 from Ditton’s (1709) appendix.

Figure 2.2. Page 109 from Ditton’s (1709) appendix. This page was actually prepared by
Edward Paget, a former RMS master.
This first example of school algebra would draw attention to a major problem which
would be found with respect to all six major purposes listed in Table 2.1. The 1693 RMS
version of the textbook which summarized the intended algebra curriculum was written in
Latin, and subsequent events at Christ’s Hospital would show that RMS boys understood very
little Latin (despite their having studied that language for more than four years). The RMS
students certainly did not know enough Latin to be able to comprehend Alexandri’s (1693)
algebra textbook (Ellerton & Clements, 2017). Not only was the algebra written in a foreign
language in the 1693 textbook, but, language aside, the mathematics was too difficult for boys
aged between 12 and 16. The symbols used in the text related to sophisticated concepts that the
boys had never previously studied. Isaac Newton supported the idea that the RMS boys should
study algebra from Alexandri’s book (Ellerton & Clements, 2017), but, like his Cambridge
protégés, Paget and Ditton, Newton over-estimated the level of algebra that 12- to 16-year-olds
were capable of learning. Not only was there a serious mismatch between the language of the
text and the learning capacities of the students, but one also had to ask whether the
mathematics was relevant to the future needs of future midshipmen and navigators.
James Hodgson, who was RMS master between 1709 and 1746, caused to have
published in 1723 a massive two-volume book titled A System of Mathematics. On a page
dedicated to King George I, Hodgson stated that his book was expressly “designed for the
use of the Royal Mathematical School in Christ’s Hospital,” and in his preface he stated that
he could not “recollect that there is one thing left undemonstrated that is capable of it.”
Throughout the book he used algebra, often at sophisticated levels. The page shown in Figure
2.3, for example, makes use of Isaac Newton’s form of calculus (“fluxions”). Despite the fact
that RMS students had not learned algebra before entering the RMS program, they were
expected to follow what would have been, for them, obscure algebraic notations and
reasoning. Nowhere in Hodgson’s (1723) book was there a specific section on elementary
algebra.
Throughout the eighteenth century almost all RMS students—most of whom were aged
between 12 and 16—were allowed less than two years to study mathematics beyond
arithmetic. Hardly any of them—possibly none of them—would have comprehended what
Hodgson had written on the page shown in Figure 2.3. Although one might applaud Hodgson
for being willing to use algebra, and for his desire to present mathematics fully and
accurately, the RMS students would, nevertheless, have gained the impression that algebra
was something that they had virtually no chance of ever understanding—despite the fact
that their teacher used it freely. Little wonder, then, that the RMS students were reduced to
copying information from Hodgson’s text (Ellerton & Clements, 2017, see especially
Chapter 5). The implicit message was: “Copy the algebra into your cyphering books, but
there is no need to understand it.”
If Christ’s Hospital was, indeed, the first regular school to teach algebra to 12- to 16-
year-olds then it is clear that the ways it was presented to the students by Edward Paget,
Humfrey Ditton, and James Hodgson meant that algebra as part of the secondary-school
implemented curriculum had got off to the worst possible start—the message was, “Algebra is
something which is important for mathematicians, but school students should not really expect
to understand it.” Given the unsatisfactory launching of algebra into the secondary-school
curriculum it is interesting to note that during the early years of the nineteenth century, more
than 100 years after Christ’s Hospital students had first been asked to learn algebra, RMS
students were still struggling to learn algebra well (Ellerton & Clements, 2014, see Chapter
10).
Figure 2.3. Page 65 of James Hodgson’s (1723) textbook for his 12- to-16-year-old RMS
students—note the use of algebra and the references to “fluxions.”
From the beginnings of school algebra, then, a mismatch between the type and
complexity of algebraic symbolism and the readiness of learners to cope with algebraic
language often arose because textbook authors, teachers, and curriculum developers asked
students to grapple with sophisticated algebraic language in order to understand concepts that
would be met in topics like logarithms, conic sections, trigonometry, differential and integral
calculus, and abstract algebra.
During the seventeenth and eighteenth centuries, students attempting to learn algebra for
the first time often experienced difficulty understanding what they were being asked to learn.
One reason was that many of the authors of textbooks were gifted mathematicians (e.g.,
Étienne Bézout (1794) and Alexis-Claude Clairaut (1746), in France; Christian Wolfe (1739)
in Germany; and Thomas Simpson (1755), in England) who had not had much experience
teaching algebra to students aged between 11 and 16 years. These authors directed their
books at students who were in higher-education institutions, including colleges where future
military personnel were being trained, and assumed that their textual expositions were
sufficiently clear that capable school-aged children would benefit from using them.
Sometimes, the algebra would be hidden within areas of mathematics which were not
obviously related to algebra—consider, for example, the following question taken from a
University of Melbourne matriculation examination for December 1902:
sin A sin 2 A + sin 3 A sin 6 A + sin 4 A sin 13 A
Simplify .
sin A cos 2 A + sin 3 A cos 6 A + sin 4 A cos13 A
This was part of Question 8 on a “Geometry and Trigonometry” university-entrance paper,
and although it was obviously in the domain of trigonometry, a level of algebraic dexterity
was needed in order to arrive at an appropriately correct answer. The University of
Melbourne set separate pass and honors matriculation “Algebra” papers, but many of the
questions on the “Geometry and Trigonometry” paper assumed a strong knowledge of
elementary algebra. A similar statement could be made about questions on mechanics,
elasticity, hydrostatics and heat which appeared on the University of Melbourne’s
matriculation Physics paper for 1902 (Clements, 1979). The point being made here is that an
unstated, but nevertheless important, purpose of secondary school algebra was to prepare
students for higher studies, and not only higher studies in algebra. That was true in most parts
of the world—see, e.g., definitions of courses for courses in Germany in Bolton (1900), and
definitions of courses in England and Wales in the Oxford and Cambridge Examination
Board’s (1899) prospectus.
During the eighteenth and nineteenth centuries it became increasingly important for
students wishing to be well prepared for higher studies in mathematics and science to acquire a
knowledge of elementary algebra (Ellerton & Clements, 2017). However, the formal language
by which algebra was presented in textbooks was often difficult for the students. That was
literally the case when Christ’s Hospital students of the 1690s had to study the subject from a
textbook written in Latin, but it continued to be the case when the textbooks were written in the
students’ first languages—because the signs and symbols (the “signifiers”) of elementary
algebra were introduced at such a rapid pace, and in such a way, that students were often
reduced to trying to memorize rules and cases, in order that they might cope with recitations
and with written tests and examinations. Such has been the emphasis on intended curricula,
as opposed to implemented and received curricula, that the mismatch between what was
stated in textbooks and what was being taught and, more seriously, what was (and was not)
learned, has often gone largely unnoticed. Algebra cyphering books, and not textbooks,
provide documentary evidence of implemented curricula, and past examination papers and,
better still, examination scripts prepared by students, testify to the received curriculum.
The British research project, Concepts in Secondary Mathematics and Science [CSMS],
conducted in the late 1970s and early 1980s, linked students’ levels of understanding of
algebraic letters to Piagetian stages of cognitive development and to IQ scores (Hart, 1981).
It was concluded, in Project publications, that most of the 13- to 15-year-old participants did
not cope with items which required them to interpret letters as generalized numbers or as
specific unknowns. During the years following the CSMS project it has been widely accepted
that, by itself, cognitive level does not offer a sufficient explanation for the ways in which
algebraic notations are misinterpreted—although, inappropriate matching of curricular
materials to cognitive levels may be part of an appropriate explanation. In this chapter we
argue that difficulties in learning to use algebraic notation have several origins, including:
• Incorrect intuitive assumptions by authors and curriculum developers with respect to
students’ rate of learning of the meanings of unfamiliar notational systems; and
• Poorly-designed teaching materials (including material using modern technologies).
More recently, a U.S. National Mathematics Advisory Panel (2008), which was made
up of leading mathematicians and mathematics educators, reviewed the algebra topics that, it
recommended, should be emphasized in U.S. schools. The Panel’s list is shown in Table 2.2.
According to the Panel, the topics listed should form the basis of Algebra I and Algebra II
courses offered in all U.S. high schools. Noticeably absent in the Panel’s list was any
consideration of what research was saying about what algebra the majority of U.S. middle-
school and secondary-school children were ready to learn or capable of learning at different
ages. The Panel’s main criterion for including a topic in its list was that it represented part of
that body of knowledge which was deemed to be logically ordered and essential for success
in higher mathematical and scientific studies. That same motivation has always been
important—for example it was the prime consideration in the seventeenth century when
algebra was first made part of school curricula at Christ’s Hospital (Ellerton & Clements,
2017).
The National Mathematics Advisory Panel (2008) stated that its list of major topics
represented an agreement after its members had consulted the following five sources:
1. The current state standards for Algebra I and Algebra II courses and for integrated
curricula;
2. Current textbooks for school algebra and integrated mathematics;
3. The algebra objectives in NAEP’s 2005/6 Grade 12 mathematics assessment;
4. The American Diploma Project’s benchmarks for a high school exit test; and
5. The algebra standards in Singapore’s mathematics curriculum for Grades 7–10. (p.
15)
The Panel contended that the major topics to be included in the list should be those in well-
regarded existing curricula, and especially those in the school mathematics curricula of
nations whose students performed best in international test programs.
National Mathematics Advisory Panel’s Topics for School Algebra 25
Table 2.2
The National Mathematics Advisory Panel’s (2008) List of Major Topics to be Covered in
School Algebra (p. 16)
Symbols and Expressions
• Polynomial expressions
• Rational expressions
• Arithmetic and finite geometric series
Linear Equations
• Real numbers as points on the number line
• Linear equations and their graphs
• Solving problems with linear equations
• Linear inequalities and their graphs
• Graphing and solving systems of simultaneous linear equations
Quadratic Equations
• Factors and factoring of quadratic polynomials with integer coefficients
• Completing the square in quadratic expressions
• Quadratic formula and factoring of general quadratic polynomials
• Using the quadratic formula to solve equations
Functions
• Linear functions
• Quadratic functions—word problems involving quadratic functions
• Graphs of quadratic functions and completing the square
• Polynomial functions (including graphs of basic functions)
• Simple nonlinear functions (e.g., square and cube root functions, absolute value; rational
functions; step functions)
• Rational exponents, radical expressions, and exponential functions
• Logarithmic functions
• Trigonometric functions
• Fitting simple mathematical models to data
Algebra of Polynomials
• Roots and factorization of polynomials
• Complex numbers and operations
• Fundamental theorem of algebra
• Binomial coefficients (and Pascal’s Triangle)
• Mathematical induction and the binomial theorem
Combinatorics and Finite Probability
• Combinations and permutations, as applications of the binomial theorem and Pascal’s
Triangle
Algebra as Generalized Arithmetic

According to da Ponte and Guimarāes (2014), in the seventeenth century “algebra
began to constitute a generalization of the methods of arithmetic, enabling broader classes of
problems to be solved, and it also generated its own problems, transforming itself into a theory
of solving polynomial equations” (p. 460). Isaac Newton, described algebra as “symbolical
arithmetic” (quoted in Turnbull, 1961, p. 363), and commented that solving equations ought
to be an important object of any school mathematics program which aimed to go beyond
computational arithmetic. The immediate aim was to get students to the point where they
could solve lower-degree polynomial equations, and the scope of a course might be stated as,
for example, “equations of the second degree” (Davies, 1837, p. 137), or with more
advanced, or older, students, as “cubic and higher equations” (Webster, 1808, p. 330).
Teachers, and textbook authors were expected to assist students to deal with the syntax and
semantics of a system of signs that summarized general properties of numbers. The early
chapters of algebra textbooks were concerned with definitions and symbols that students
would need to know, and rules by which they would need to operate.
Authors of algebra textbooks tended to believe that students could learn the language
and modes of operations simply by thoughtfully reading their books. Thus, for example, an
advertisement for an English translation of Leonhard Euler’s (1797) Elements of Algebra
stated that “the object of the celebrated author was to compose an elementary treatise by
which a beginner, without any other assistance, might make himself a complete master of
algebra” (p. xxxi). But, anyone who has interviewed students attempting to learn the signs,
terminologies, and wider meanings of elementary algebra would not agree with the
advertisement’s claim that any beginner could grasp the meaning of what Euler intended to
communicate. Take for example, the following passage (which, probably, Euler would not
have regarded as difficult):
But when it is required to multiply √a by √b, the product is √ab; because … if a
square has two or more factors, its root must be composed of the roots of those
factors. Wherefore we find the square root of the product ab, which is √ab, by
multiplying the square root of a or √a, by the square root of b or √b. It is evident
from this, that if b were equal to a, we should have √aa for the product of √a by
√a. Now √aa is evidently a, since aa is the square of a. (p. 61)
For a mathematically well-trained reader the meaning of this passage is clear and the
mathematics associated with the fact that if a and b are positive real numbers (or either is
zero), the product of √a and √b, equals √ab is not conceptually challenging or difficult to
verbalize. But for a young secondary school student the above passage is replete with
potentially confusing words and sentences. Terms such as “root,” “factors,” “product,”
“square,” and semantically complex sentences such as “if a square has two or more factors,
its root must be composed of the roots of those factors,” would cause difficulty for many
young readers. Many mathematicians would not be aware of the difficulties that such a
passage could present to persons who are beginning to learn algebra.
What was true of Euler’s book was true of other books written by well-known
mathematicians (Cajori, 1890). Thus, for example, in a later edition of Elémens d’ Algèbre a
l’Usage de l’Ecole Centrale des Quatre-Nations, Sylvestre François Lacroix (1815) explained
how 4a + 9b – 2c, 2a – 3c + 4d, and 7b + c – e could be added in the following passage:
La somme indiquée sera … 4a + 9b – 2c + 2a – 3c + 4d + 7b + c – e. Mais les
termes 4a et + 2a étant formés de quantités semblables, se réunissent en un seul
égal à 6a. De même les termes + 9b, + 7b donnent 16b. Les termes – 2c et – 3c,
tous deux soustractifs, produisent dans le total le même effet que la soustraction de
5c; et comme, en vertus du terme + c, on aura d’une autre part à ajouter c, il restera
seulement à retancher 4c. La somme des expressions proposes sera donc ramenée à
6a + 16b – 4c + 4d – e. (p. 34)
Algebra as Generalized Arithmetic 27
Lacroix did not comment on why, for example, 6a + 16b should not be written as 22ab—
although most teachers of beginning algebra students know that such an error is very
common. Mathematicians like Lacroix did not necessarily have the pedagogical content
knowledge that made them aware of such common misconceptions. Likewise, there was no
direct comment on why the sum of 4d and – e should remain as 4d – e, and not be further
simplified—although Lacroix did use the expression “quantités semblables.” the English
version of which (“like terms”) would become common in beginning algebra classes
wherever the subject was taught using the English language. But for centuries (and still
today), beginning algebra students would struggle to appreciate why, for example, 3ab was
not “like” 3a + b.
Lacroix’s books “introduced a new arithmetic-algebra-geometry sequence, based on the
idea that algebra is nothing more than universal arithmetic” (da Ponte & Guimarāes, 2014, p.
464). Algebraic notations also came to be used in geometrical proofs and exercises (Bourdon,
1831). A view of algebra as universal arithmetic was translated into the educational
institutions of other European nations (e.g., Venema, 1714), and from there across the
Atlantic into the colonies, or former colonies (e.g., Brazil), of leading European nations (e.g.,
Portugal). During the period 1780–1840, there was a strong French influence on the teaching
and learning of algebra in colleges in the United States of America (Cajori, 1890), with
highly formal textbooks by Lacroix and Bourdon being adopted (see, e.g., Davies, 1837). But
those books had a stultifying effect on the teaching and learning of algebra in schools, and
ultimately the French “scientific discussions” were seen as too theoretical for U.S. school
children. A new generation of U.S. textbook authors were moved to introduce the more
“practical methods of the English school,” especially those by John Bonnycastle (Davies,
1837, p. iv). Bonnycastle’s An Introduction to Algebra, was much used—between 1806 and
1847 there were 19 editions of the book published in the United States of America
(Karpinski, 1980).
The disconnect between what authors of widely-used elementary textbooks on algebra
wrote and what beginning algebra students would understand readily was never more evident
than in the section on algebra in the famous North American textbook by Nicolas Pike
(1788). Prior to the Revolutionary War (1775–1783), most arithmetic textbooks used in the
North American British colonies had been written by British authors (e.g., Cocker, 1677;
Dilworth, 1762). These books did not mention algebra. Immediately after the Revolutionary
War, however, there was a surge of activity in American publishing for schools and colleges
(Monroe, 1917), and some of the most popular arithmetic textbooks included sections on
algebra. Nicolas Pike's (1788) portentous 512-page A New and Complete System of
Arithmetic Composed for Use of Citizens of the United States, the first major mathematics
textbook written by a North American author, included a 33-page chapter on algebra (pp.
469–501).
As well as Pike’s (1788) original textbook, a 1793 Abridgement of the New and Complete
System of Arithmetic was published and was attributed to Pike. The Abridgement (Pike, 1793),
which had 371 pages, was significantly smaller than the New and Complete System of
Arithmetic and omitted, entirely, the section on algebra that Pike had included in his 1788
textbook—despite the fact that the 1793 abridgement was particularly aimed at schools. “Old
Pike,” as Pike’s (1788) original textbook (which included the section on algebra) would
come to be known, would go through six editions between 1788 and 1843 (Albree, 2002;
Karpinski, 1980; Thomas & Andrews, 1809), but editions published after 1822 did not
include the specific section on algebra.
The following is an unedited transcript of the first page on algebra in Pike’s (1788) book:
1. Like quantities are those which consist of the same letters.
2. Unlike quantities are those which consist of different letters.
3. Given quantities are those whose values are known.
4. Unknown quantities are those whose values are unknown.
5. Simple quantities are those which consist of one term only.
6. Compound quantities are those which consist of several terms.
7. Positive or affirmative quantities are those to be added.
8. Negative quantities are those to be subtracted.
9. Like signs are all + or all –.
10. Unlike signs are + and –.
11. The co-efficient of any quantity is the number prefixed to it.
12. A binomial quantity is one consisting of two terms; a trinomial of three terms;
and a quadrinomial of four terms, &c.
13. A residual quantity is a binomial, where one of the terms is negative. In the
computation of problems, the first letters of the alphabet are put for known
quantities, and the last letters for those which are unknown.
AXIOMS
1. If equal quantities be added to, subtracted from, multiplied or divided by equal
quantities, the wholes, remainders, products and quotients will be respectively
equal.
2. The equal powers of equal roots of quantities are equal.
3. Two quantities respectively equal to a third, are equal to each other.
4. The whole is equal to all its parts taken together.
ADDITION
CASE 1. To add quantities which are alike and have like signs (when a leading
quantity has no sign before it, + is always understood; and a quantity, without any
co-efficient prefixed to it, is supposed to have unity, or 1).
RULE: Add all the co-efficients together, and to their sum adjoin the letters
common to each term, prefixing the common figu (sic.) …
(Pike, 1788, p. 469)
After graduating from Harvard College in 1766, Pike (1743–1819), a native of New
Hampshire, taught school mathematics for two decades. At the time he prepared the first
edition of his book he was an experienced teacher of mathematics, and the language he used
for his definitions resembled those of a teacher trying to explain difficult ideas to students
who had just begun their first course in algebra. In fact, however, most of the “definitions”
were mathematically inadequate, and virtually meaningless as stand-alone statements. What,
for example, did “Algebra is the art of computing by symbols” really mean? Pike stated that
“Like quantities are those which consist of the same letters,” but did that mean that in algebra
abc was “like” a + b + c? What did Pike mean by “Positive or affirmative quantities are those
to be added”? Or, by “Negative quantities are those to be subtracted”? Or by “Like signs are
all + or all –”? Or, in the “axioms,” what was the meaning of “The equal powers of equal
roots of quantities are equal”?
Algebra as Generalized Arithmetic 29
The “definitions” were so poorly worded that one wonders if the text of Pike’s book
was checked by anyone other than Pike himself. Pike adopted an “Introduction-Rule-Case-
Example-Exercise” genre (which Ellerton and Clements (2012) have abbreviated to IRCEE)
and clearly he expected students to learn algebra through a cyphering approach (Ellerton &
Clements, 2012, 2017). There is considerable evidence that before 1820 the cyphering
tradition controlled the teaching and learning of beginning algebra in the United States of
America—the Ellerton-Clements collection of 450 North American cyphering books
includes 27 “algebra cyphering books,” each of which was prepared in a different school at
some time between 1775 and 1860. Our analyses of these algebra cyphering books has
indicated that at least as early as the eighteenth century (and, we believe, probably well
before then), the emphasis on algebra as generalized arithmetic was being propagated in
North American secondary schools, known as “academies.” After about 1700 this approach
was part of the implemented algebra curriculum wherever school students were asked to
learn algebra.
Figure 2.4 shows a page from an algebra cyphering book prepared by Alfred Andrews
in Connecticut, in the United States of America in 1813. Case 1st for “addition” was stated as
“When like quantities have like signs,” and in the Figure one can see six examples when like
terms are added to like terms. The next 11 pages of the book dealt with subtraction,
multiplication, and division of algebraic expressions. Then followed 7 pages on algebraic
fractions. Alfred’s adoption of the IRCEE genre is evident in Figure 2.4.
The rules and cases, and exercises in Figure 2.4 were identical to statements and
exercises in Samuel Webster’s (1808) Mathematics, and very similar to the introductions to
“Algebra” in Nicolas Pike’s (1788) textbook, or in Consider and John Sterry’s (1790) book.
In fact, almost every introductory section on algebra in textbooks and in algebra cyphering
books began with descriptions of how to add, subtract, multiply and divide algebraic
expressions; then would come sections on algebraic fractions, and “surds” (sometimes within
sections titled “involution” and “evolution”). Soon after, would follow a section on linear
equations, and then, perhaps, one on quadratic equations. The more descriptive, and
advanced, continental European texts began in the same way but often proceeded to more
general considerations related to polynomial equations of the third and fourth degrees, and
perhaps, to arithmetical and geometrical progressions (see, e.g., Porro, 1789).
From the outset, the. authors of those algebra texts used in early European and U.S.
schools put forward the idea that algebra was best thought of as generalized arithmetic, and
that that was something which should be firmly planted in the minds of all beginning algebra
students (Cajori, 1890). This expectation assumed that the students would quickly learn the
language of algebra. Thus, for example, the Pestalozzi-inspired Warren Colburn, in his book
on elementary algebra, quickly introduced very complicated rules defining correct syntax and
semantics, and that approach continued throughout the book (see, e.g., Colburn, 1825, p. 170,
for notes on “completing the square”). Noticeably, Colburn did not claim in his Preface that
he had successfully used the book with students. Indeed, he seemed to think that most
students would have no difficulty following his book, and that those who did not could
rectify the situation with “one hour’s study of some treatise” (p. 4).
Figure 2.4. The first page on algebra in Alfred Andrew’s (1813) algebra cyphering book.
School Algebra as a Gatekeeper 31
The aim of all authors of textbooks on elementary algebra was to lead students along a
path that would ultimately enable them to solve lower-degree polynomial equations
(Chateauneuf, 1929). The equations, however, were of the find-my-number variety; students
were not asked to investigate which values would make the equations true and which would
make them false. That same “fill-in-the-blank” or “find-the-number” emphasis remains in
most elementary-school algebra courses today. In only a few modern algebra textbooks is
attention given to the idea that letters in the equations are to be regarded as variables. The
emphasis is rarely on finding values of the letters which will make the equations, or the
inequalities, true, and which, false. After carrying out additions, subtractions, multiplications
and divisions of algebraic terms, and after solving linear and, perhaps, quadratic equations,
students are asked to solve algebraic word problems, but these are not often related to
genuine real-life situations.
The U.S. National Mathematics Advisory Panel (2008) recommended that by the end
of Grade 6 all students should be introduced to “the use of symbolic notation and the concept
of generality, both being integral parts of algebra” (p. 18). This, the National Panel stated,
was consistent with the findings of an analysis of curricula of Grades 1–8, of “the highest-
performing countries on TIMSS (Singapore, Japan, Korea, Hong Kong, Flemish Belgium,
and the Czech Republic), sometimes called the “A+ countries” (p. 17).
In the first chapter of this book we cited data which suggested that standard approaches
to introducing algebra to middle-school students (that is to say, to students aged between 10
and 14 years) have not been very successful because many students fail to learn even the
most elementary algebraic language, concepts and skills (MacGregor & Stacey, 1997).
Authors of school algebra textbooks in the eighteenth and nineteenth centuries were not
usually aware of the extent of significant differences between the intended, implemented, and
received algebra curricula (Clements & Ellerton, 2015). In the twentieth century educators
have steadily become more aware of such differences, and that has led some of them to
reconsider how algebra should be introduced to beginning algebra students (see, e.g., Cai &
Knuth, 2011c).
School Algebra as a Gatekeeper for Entry to Higher Studies
In the course of criticizing the common-core mathematics sequence (National

Governors Association Center for Best Practices, & Council of Chief State School Officers,
2010—hereafter written as CCSSM, 2010), R. James Milgram (2013), Emeritus Professor of
Mathematics for Stanford University, told those present at a conference held at Notre Dame
University, in Indiana:
For the past 150 years the high schools have acted as a buffer between K–8 and
our colleges. … For years, until the 1990s, we were defended, by the high
schools. … [But now there has been a decline in the quality of the preparation of
students for colleges]. We have algebra without conics and logarithms—neither
one of which I’m sure anyone in the audience cares very much about—but believe
me, if you’re doing engineering, if you’re doing the hard sciences, if you’re doing
almost any of the basic fields which depend on real data analysis, these are
critical. ... Right now over two-thirds of our graduate students in the hard sciences
and engineering [at Stanford University] are foreign-born. … Common core
defines college readiness as having “passed Algebra II.” That is a very weak
standard.
Milgram, who described Stanford University as an institution for the “elite,” maintained that
school algebra standards had slipped dramatically since the mid-1980s, and that most
university-based mathematicians and scientists believed that much higher standards needed
to prevail in school algebra across the nation. From Milgram’s perspective, there was nothing
inherently wrong with the idea that school algebra should act as a gatekeeper to higher
university studies in mathematics, engineering, or the physical sciences, and that more
attention needed to be paid to calculus in U.S. secondary schools. Higher science,
engineering, and mathematics departments at U.S. universities needed to be guarded, he said,
from students who were not capable of becoming part of a mathematical elite.
From the beginning of the nineteenth century, and in some nations even before then,
higher education authorities recognized that many children found it difficult to learn algebra.
That reality raised two possibilities—either not to include algebraic knowledge as something
required of prospective students, or to include it. Many institutions took the second option on
the grounds that that would, somehow, assure quality in their incoming students (Pappano,
2012; Rech & Harrington, 2000). Thus, for example, in Australia in 1900 university-
matriculation requirements of the Universities of Adelaide, Melbourne, Sydney, and
Tasmania, included a pass in a subject called Algebra at the university’s entrance
examination. The courses for Algebra were defined in the following terms (Carslaw, 1914;
Clements, 1979):
University of Adelaide: To quadratic equations, surds;
University of Melbourne: The fundamental operations, highest common factor,
lowest common multiple, fractions, simple equations, quadratic equations of one
unknown, simultaneous equations of the first degree not involving more than three
unknowns, problems involving the above.
University of Sydney: To quadratic equations involving one unknown quantity.
University of Tasmania: Definitions and explanations of algebraic signs and terms,
proof of arithmetic and algebraic laws, symbolic expressions, addition, subtraction,
multiplication and division of algebraic quantities, factors.
Passes in the university-entrance examinations for these lower-level algebra subjects were
needed by students wishing to enter any course in the universities. For entry to certain
engineering or science-related degree courses, and certain other higher-status courses, passes
on examinations in higher-level algebra were required. Thus, for example, in 1900, persons
wishing to take up courses in Law, Medicine, Engineering or Science at the University of
Sydney had first to pass a written examination that required knowledge of “the three
progressions, the binomial theorem for a positive index, and the properties and use of
logarithms.” At the University of Melbourne, the honours matriculation Algebra examination
demanded knowledge of the remainder theorem, quadratic equations of two or more
unknowns, elimination, surds, indices, square root, ratio, proportion and variation, the three
progressions, systems of numeration, permutations and combinations, and the binomial
theorem for positive integral index (University of Melbourne, 1900).
In the Australian states around 1900, most students entered the universities when they
were aged about 17 or 18, and a similar situation prevailed in many other nations. Algebra
courses for university entrance in Scotland, for example, were similar to those in Australia,
as were courses for 17-year-olds in England and in Germany (Clements, 1979). In England
and Germany, however, students could enter the prestigious universities only after they had
completed higher-level courses. Between the ages of 17 and 19 university-bound students in
School Algebra as a Language for Modeling Real-Life Problems 33
those nations had to prepare for algebra examinations pitched at higher levels than the
examinations aimed at 17- or 18-year-olds in Australia and Scotland. In the United States,
the highest level of algebra studied in high schools was probably at a lower standard than that
required of university-bound 17-year-olds in Australia and Scotland, and certainly at a much
lower standard than was expected of 19-year-olds in England and Germany (Olney, 1885).
David Eugene Smith (1905) claimed that teachers in the eastern states of the United
States of America wanted to see how much of the spirit of German gymnasium mathematics,
with its pure, as opposed to applied, mathematical tradition (Jahnke, 1983), could be
transplanted into American schools. He acknowledged that his position was essentially a
conservative one, but emphasized that that was the position favored by teachers of
mathematics in the eastern states. For Smith, improving high-school mathematics had
become synonymous with writing pure mathematics textbooks that would better prepare
students for examinations which would select the best students wishing to proceed to higher
studies in mathematics.
This emphasis on written examinations for sorting students was, at least in Western
nations, a nineteenth-century development (Cremin, 1956; Ellerton & Clements, 2012; Keitel,
2006; Spielhagen, 2011). Although it could be argued that it provided “bright” children in
poor families with a means of accessing higher mathematics, it could also be interpreted as
providing an “objective” means by which children from mainly well-to-do families were
selected for higher mathematical studies (Spielhagen, 2011). Algebra was being used as a
gatekeeper. In that sense, the bias in favor of an elite based on social class was replaced by a
different, but highly related, bias based on “readiness to learn,” or “giftedness.” Expressions
such as “ability to learn” came to be used to justify procedures whereby the study of higher
mathematics was preserved for the few (Clements, 1992).
School Algebra as a Language for Modeling Real-Life Problems

Elementary school algebra texts have traditionally asked students to solve problems
which have the appearance of being real-life applications but, on reflection, barely qualify as
belonging to that category. Take for example, the following question posed in Charles
Davies’ (1846) Elementary Algebra Embracing the First Principles of the Science:
A person has two horses, and a saddle worth 50 pounds. Now, if the saddle be put
on the back of the first horse, it will make his value double that of the second; but,
if it be put on the back of the second, it will make his value triple that of the first.
What is the value of each horse? (p. 107)
Such a question might provide useful practice for getting students to define variables
carefully, and to set up equations which summarize the situation and enable a solution to be
reached. But the question addressed by the task is not one that would be asked, even by an
avid horse rider. Paradoxically, genuine problems which could have been solved relatively
easily by algebra were classified as being part of arithmetic, and were dealt with by cleverly
conceived rules which avoided the use of algebra. Examples of such problems were those
considered to belong to topics bearing names such as “rule of three,” “double rule of three,”
“inverse rule of three,” “equation of payments,” “loss and gain,” “barter,” “discount,”
“simple interest,” “compound interest,” “alligation,” “fellowship,” “practice,” “arithmetical
progressions,” and “geometrical progressions” (Ellerton & Clements, 2012; Swetz, 1992).
Isaac Newton (1739) and Pierre-Simon Laplace (1799) were among numerous eminent
seventeenth- and eighteenth-century mathematicians to model real-life problems with algebra,
but although the questions they put forward as relevant to school students arose in practical
contexts, those contexts would not have been of great interest to many school students.
Furthermore, often the algebra needed to model the associated real-life situations was
difficult and would have been beyond the capabilities of most school students. James
Hodgson (1723) used algebraic symbolism and concepts when discussing and applying the
concepts of latitude and longitude in his A System of the Mathematics, and that book was
written for 14- to 16-year-old boys at the Royal Mathematical School (RMS) within Christ’s
Hospital in London (Ellerton & Clements, 2017). Although those concepts should have been
of interest to RMS students, who were training to be navigators, it is doubtful whether the
students would have been sufficiently mature, from an algebraic developmental point of
view, to cope with the level and language of algebraic reasoning in Hodgson’s text.
During the nineteenth century, those responsible for designing and implementing
school mathematics programs increasingly embraced more active forms of learning and
teaching. Starting with Johann Heinrich Pestalozzi (1746–1827), the great Swiss educator,
and followed by Warren Colburn in the United States, there gradually emerged support for
the idea that young children, from about six years of age, including girls, should study
arithmetic, and that whole-class instruction in which the teacher used inductive methods
might profitably be used. Part of Pestalozzi’s theoretical position was that the questions
which teachers directed at their students should have been related to contexts familiar to, and
of interest for, the students. However, even Warren Colburn (1825), in his book An
Introduction to Algebra Upon the Inductive Method of Instruction, failed to offer examples
which would have been of genuine interest and relevance to beginning algebra students. A
typical example in Colburn’s book was:
A market woman sold at one time 7 eggs, 12 apples and a pie for 26 cents; at
another time 12 eggs, 18 pears and 3 pies for 69 cents; at a third time 20 pears, 10
apples and 17 eggs for 69 cents; and at a fourth time, 7 pies, 18 apples and 10
pears for 66 cents. Each article was sold, at every sale, at the same price as at first.
What was the price of each article? (p. 109)
Although such a question had a real-life context, the question itself was not one that that
would be likely to arise in the lives of students.
Colburn’s (1825) algebra textbook was written for use in schools, but it was not until
well into the nineteenth century that educators began to wonder whether school algebra might
profitably be studied by more than a small proportion of very capable students. Certainly, from
the 1820s onwards, algebra was part of the intended and implemented curricula of high schools
in the United States (Kilpatrick & Izsák, 2012), but only a small proportion of American
schoolchildren proceeded to the high schools from the elementary common schools (Snyder,
1993; Snyder & Dillow, 2011). For much of the nineteenth century, algebra was seen as
something that only high-achieving students—specifically, those who would be likely to pass
entrance examinations set for colleges—would ever get to study. Those who did get to study it
were presented with a very formal course of study taught in a traditional way. The emphasis
was on memorizing definitions, rules and cases, and then copying out model examples. Then
would come written exercises and, finally, written tests and examinations. Calculus was rarely
studied in the early high schools in the United States of America.
When a much larger proportion of children began attending high schools in the twentieth
century, the question arose whether all secondary-school students should study algebra and, if
so, which form of algebra (Bestor, 1956; Cremin, 1964). Whereas academic mathematicians
and physical scientists tended to favor a rigid curriculum with a strong emphasis on
School Algebra as a Language for Modeling Real-Life Problems 35
manipulation of algebraic expressions and linear and quadratic equations, many teachers and
academic educators felt that that would not be consistent with the ideal of a liberal form of
education which was relevant to all students (Rosskopf, 1970).
The concept of algebra as a generalized form of arithmetic prevailed for much of the
nineteenth century (Cajori, 1890), but the value of that emphasis was challenged in the last
quarter of the century. In particular, between 1875 and 1910, John Perry, an Irish-born
engineer and applied mathematician, popularized the concept of mathematics laboratories in
which problem-based approaches to mathematics teaching and learning took advantage of
technological developments associated with the use of squared paper for graphs, and the use
of personal slide rules (Brock, 1981, Roberts, 2012). Perry began this work when based in
Japan between 1875 and 1879, and expanded it when he became Professor of Engineering
and Mathematics at the City and Guilds of London Technical College. His greatest influence
came between 1896 and 1914 when he was Professor of Mathematics and Mechanics at the
Royal College of Science, in South Kensington, London. He worked hard to popularize a
form of school and college mathematics education which linked mathematics, the physical
sciences, engineering, architecture, and manual work (Perry, 1899, 1902, 1912). Perry
envisaged a form of mathematics education in which students from all levels of society
would constantly gather data in laboratory sessions, graph them on Cartesian planes, and then
attempt to make generalizations and predictions based on patterns which seemed to be
present. He urged the regular use of squared paper, with the mathematical concept of
function providing the key integrating theme (Brock & Price, 1980; Price, 1986, 1994).
Many of Perry’s students were artisans studying mathematics in evening classes. In
1901, Perry delivered a major address on the teaching of mathematics to a meeting of the
British Association held in Glasgow (Perry, 1902), In 1902 his ideas were advocated in the
United States by the President of the American Mathematical Society, Eliakim H. Moore, of
the University of Chicago, and the laboratory approach to teaching and learning mathematics
was much discussed in the mid-western states of the United States of America (Moore, 1903;
Roberts, 2012). In fact, in the early years of the 20th century there was a vanguard of
optimistic mathematicians and mathematics teachers in many nations who were prepared to
give “Perryism” a chance in their schools. Perry’s influence extended to Australia, France,
Germany, Italy, Japan, the United Kingdom, and the United States of America (Borel, 1904;
Clements, 1992; Giacardi, 2009; Nordgaard, 1928; Roberts, 2012; Ruthven, 2008). Perry’s
ideas became a cause célèbre among many Chicago-based mathematicians and mathematics
teachers (Moore, 1903, 1906; Roberts, 2012; Young, 1914). Everywhere, it seemed, those
involved in the previously well-defined world of middle- and high-school mathematics were
pausing and holding their collective breaths, as they contemplated the possibility of adopting
a laboratory, integrated curriculum approach at all levels of school mathematics.
But, analysis of documents held in the David Eugene Smith collection at Columbia
University reveals that Felix Klein and Smith combined to stem the flow of this form of
radicalism. In 1904 Felix Klein had proposed that “the function idea, graphically represented,
should form the central notion of mathematical teachings” (quoted in Nordgaard, 1928, p. 81),
but although Smith publicly stated that he supported the idea that lower secondary school
mathematics curriculum should be reconstructed using the function concept as its new base,
in reality he did very little to propagate that idea. He wrote many school mathematics textbooks,
including algebra textbooks, but in those books a very formal, generalized-arithmetic view of
algebra was presented. There is evidence within Smith’s correspondence with Klein—held in
Box 62 within the David Eugene Smith collection at Columbia University—that Smith and
Klein deliberately set out to limit the influence of Perryism.
In fact, Smith (1905, 1913) viewed Perryism as an extreme position held by an influential
minority in England and in the midwestern states of the United States of America. He believed
that educators in the eastern states of the United States, in cities like New York, Philadelphia
and Boston, regarded Perry’s ideas as an aberration in the education process. Smith (1905)
summarized the attitudes of teachers in the eastern states to Perryism in the following way:
1. Any effort to introduce physical experiments into the classes in mathematics
has no support whatever from either the teachers of mathematics or those of
physics. …
2. Any effort to seek the applications of mathematics chiefly in physics or in
science generally, has not met with favor, and is not likely to find advocates.
The consensus of opinion is that the number of applications of algebra to
physics, for example, is exceedingly small, those to business being
considerably larger, even though these are not numerous. …
3. The attempt to have algebra and geometry appear to the pupil as having any
considerable application to science or to business aside from a few special
propositions will not be made. (p. 207)
Smith added that teachers in the eastern states not only wanted their students “to love
mathematics for its own sake” (p. 208), but also wanted them to be well prepared for the
rigid system of public examinations. He saw no inherent contradiction in those aims.
According to Smith, teachers in the Eastern states would disagree strongly with the
proposition that “no equation should be given without a genuine application, that no problem
be assigned without a genuine application, that no problem should be assigned without a
physical context, that no topic shall be considered save as it bears upon life” (p. 208). Smith
added that teachers in the eastern states wanted to develop “pure mathematics” laboratories
in which pupil activity took place in the mind rather than with physical apparatus.
Smith did little to incorporate functions and their applications fully into the numerous
textbooks he wrote on school mathematics. For example, the textbook Academic Algebra,
which he co-authored with the mathematician Wooster Woodruff Beman (see Beman &
Smith, 1902), contained no graphs and no discussion of functions, despite the fact that the
preface claimed that the book would prepare students for college mathematics.
In London, Perry succeeded in persuading examining boards to offer alternative forms
of examinations in which problems linked to laboratory methods were asked, and by 1910
these examinations were widely used by schools. After Perry’s retirement in 1914, however,
the alternative examinations were dropped (Clements, Keitel et al., 2013). In the United
States of America there were early signs that a more practical, graphical approach to algebra
might be manifest in the secondary schools (Schultze, 1909), but the lack of support from
influential textbook authors such as David Eugene Smith resulted in such considerations
being sidestepped during the first half of the twentieth century (Roberts, 2012).
Thus ended, at least for the moment, a promising movement in mathematics education
which had begun in Japan, had blossomed in the United Kingdom, and had spread its influence
as far as the United States of America, France, Italy, and Australia (Borel, 1904; Clements,
1992; Giacardi, 2009; Ruthven, 2008). The preferred approach to school mathematics
across the world was, at that time, in the balance—and some thought it was tilting in the
direction of applied mathematics by which most, if not all, students would become acquainted
Algebra for Describing Basic Structural Properties of Real Numbers 37
with real-world problem solving. Conservative mathematics education leaders, like David
Eugene Smith, rejected this view as extreme and supported the growing influence of written
examinations and printed textbooks. That influence spread quickly across the world, even to
nations like Japan (Ueno, 2006), so that “mathematics for all” was increasingly interpreted as
preparing all students for externally-set tests and examinations.
Early in the twentieth century another language for modeling real-life problems at the
school level suddenly gained popularity, as more and more nations moved to include calculus
in the courses of study in their secondary schools. When, around 1910, civil service and
university authorities in England decided to include calculus in their entrance examinations,
British schools quickly included calculus in their mathematics curricula (Godfrey, 1913), and
within a period of 15 years, calculus had become part of the mathematical studies for senior
mathematics classes in many British Commonwealth and Continental European nations—
including all the Australian states (Carslaw, 1914; Clements et al., 1989; Grimison, 1988;
McQualter, 1975). In the United States of America, however, only a very small proportion of
students got to study calculus while still at school.
During the New Math(s) period there were renewed calls for secondary-school algebra
to place a much greater emphasis on “functions and relations,” but the formal set-language
notations and symbolisms tended to be stressed, and not the modeling of real-life problems
which, some thought, would have been of greater interest to students (see, e.g., May & Van
Engen, 1959). The language factor associated with textual materials was often foreboding
and, once, again, an opportunity for important curricular advancement was lost. Consider, for
example, the statement about a theory of relations shown in Figure 2.5 (taken from
Vredenduin, 1962, p. 105). This is from a book which was supposed to focus on the relation
between arithmetic and algebra in the mathematical education of children up to the age of 15
years. The mathematics in the Figure is fine, but any expectation that teachers of middle-
school students would be willing to cope with the level of formalism introduced was
unrealistic. Around 1900, John Perry applied the concept of a function in pedagogically
sound ways, but 60 years later it had been over-formalized. Such was the new mathematics.
Algebra as an Aid for Describing Basic Structural Properties of Real Numbers

The formalization of the concept of a group by mathematicians in the 1840s (Kleiner,
1986; Sfard, 1995) stimulated mathematicians to give more attention to fundamental
structures of “analysis,” a branch of mathematics which made heavy use of algebraic
notations. In 1872 Felix Klein, in his Erlanger program, elaborated structural properties of
mappings of the plane. But, in the nineteenth-century most authors of school algebra
textbooks paid little or no attention to such structural matters. Although, occasionally, one
could find reference to structural properties of real numbers (see, e.g., Day, 1814, p. 42; Ray,
1848, p. 48; Wentworth, 1881, p. 17), those authors rarely offered systematic summaries of
the formal structure of real numbers. Even those who wrote about how school algebra should
be taught rarely drew attention to structural matters (see, e.g., Raub, 1888).
Fletcher Durell and Edward Robbins’s (1901) Grammar School Algebra provided a
notable exception to the situation described in the last paragraph. They not only offered a
“formal statement of laws of arrangement” (p. 22), in which the commutative and associative
properties, and the distributive property for “the arrangement and grouping of the symbols for
quantity in algebra” were listed, but they also discussed the “meanings” of the properties (pp.
22–25). Durell and Robbins (1901) were clearly ahead of their time with their recognition of
the importance of algebraic structures for school students. In most countries it was not until the
period of the new mathematics (1957–c.1970) that serious thought was given to using
algebraic notations to summarize, for secondary-school students, fundamental structural
properties of real numbers and of mappings of the plane (Davis, 1964; UNESCO, 1973).
Figure 2.5. Comments on a theory of relations (Vredenduin, 1962, p. 105).
During the second half of the 1930s, and in the 1940s and the 1950s, a group of mainly
French mathematicians, who published collectively under the pseudonym “N. Bourbaki,”
generated a series of high-level publications on the “architecture” of mathematics. It was not
until the late 1950s, however, that this would have an effect on school algebra curricula
(Moon, 1986). Jean Dieudonné, a leading Bourbakist, played an important role in framing the
direction of a Conference held at Royaumont, France, in 1959, which propagated the “new-
math(s)” idea that school mathematics curricula around the world were antiquated and were
no longer providing a good preparation for higher studies in mathematics. The Royaumont
Conference concluded that school mathematics needed to give much more attention to
structural properties of real numbers—particularly to group and field properties (Clements,
2003; Dienes, 1960; Moon, 1986; Phillips, 2014). A UNESCO (1973) statement even
concluded that in school mathematics, algebraic structure “must be at the very core of all that
we teach” (p. 14).
In the 1960s, and in the early 1970s, during the period of the new mathematics, it was
common for secondary-school students to be asked, by authors of algebra textbooks, to justify
each line of the solution for an equation with a reference to an appropriate “field” property
(see, e.g., Davis, 1964; Exner & Rosskopf, 1970). Thus, for example, in going from 3x + 5 =
26 to x = 7, it was expected that students would recognize that what was really happening was:
• -5 was added to both sides;
• (3x + 5) + -5 could be rewritten as 3x + (5 + -5) because of the associative property
for addition;
• 5 + -5 was equal to 0 because 5 and -5 were additive inverses;
• 3x + 0 was equal to 3x, because 0 was the additive identity element;
• On the right-side of the equation, 26 – 5 was equal to 21;
• With the (equivalent) equation 3x = 21, both sides could be pre-multiplied by 1/3;
1 1
• (3x) could be rewritten as ( × 3)x, because of the associative property for
3 3
multiplication;
1 1
• ( × 3)x could be rewritten as 1x, because and 3 were multiplicative inverses;
3 3
• 1x was equal to x, because 1 was the multiplicative identity element;
• Hence, x = 7, and so the truth set was {7}.
It was hardly surprising that by the mid-1970s many practicing teachers had rejected such
extreme formalism, mainly because it made little sense—either to them (Perel & Vero, 1967)
or to their students. Most of them could not respond adequately to tasks involving equations or
inequalities which demanded that their students demonstrate an understanding of the formal
“rules” which constituted the field and order properties. It was much easier for all concerned to
stick with the traditional “subtract 5 from both sides, and then divide both sides by 3.”
In the 1960s, in particular, many math(s) textbooks were written by mathematicians
who seemed to assume that their superior knowledge of mathematics entitled them to advise
school teachers how to teach mathematics (see, e.g., Davis, 1964). Zoltan Dienes, a strong
mathematician and an outstanding teacher of mathematics—even with very young children
(Clements, 2003)—developed many textual materials which had a strong emphasis on
algebraic structures (e.g., Dienes, 1967, 1968). A difficulty arose, however, when persons
with less mathematical knowledge, and with less teaching ability, than Dienes attempted to
translate Dienes’s very creative approaches to algebraic structure into their own classrooms.
During the period of the new mathematics both teachers and students struggled to understand
the mathematics written in textbooks inspired by high-level mathematicians—the writing of
which was often supported by large research grants.
In 1960 and 1961 Dienes worked with Jerome Bruner in the Harvard Mathematics
Learning Project. Such was Dienes’ reputation as a teacher that his ideas and material
were translated into schools in many parts of the world, including Australia and
Papua New Guinea (Clements, Keitel et al., 2013). Figure 2.6 reproduces a task, on
geometric progressions, prepared by Dienes (1968, p. 97), who claimed that the task was
suitable for fourth-, fifth-, and sixth-grade children (p. v). Dienes might have been able to
teach such material successfully to middle-school students, but one doubts whether
anyone else could have done so.
Figure 2.6. A geometric progressions task prepared by Zoltan Dienes for middle-school
students (from Dienes, 1968, p. 97).
Among many mathematicians (though certainly not all of them—see, for example,
Kline, 1973), the period of the new mathematics was a time of optimism so far as school
mathematics was concerned. In a Report on the Relations Between Arithmetic and Algebra in
Mathematical Education up to the Age of 15 Years, edited by “Dr. Hans Freudenthal,
Professor of Pure and Applied Mathematics at the University of Utrecht” (Freudenthal,
1962a, p. 1), Freudenthal (1962b) argued that “Cauchy-Cantor sequences rather than
Dedekind cuts” should be referenced in lower-secondary schools because “real numbers have
to be introduced as infinite decimal fractions” (p. 26). Despite over-optimistic, questionable,
statements like that, Freudenthal (1962b) nevertheless had some wise words to say about
how issues associated with the language of school algebra had been disregarded. He
particularly referred to “the semantical problem of the letters in algebra,” and was concerned
with how, at the school level, “meaning” was aimed at “a relation between a name and the
thing named by it” (p. 37).
To his credit, Freudenthal (1962b) was also concerned with elementary structural
considerations, especially as they related to school algebra. He wrote:
The student is not told why he has to write 2a rather than a2, 2ab rather than a2b,
ab rather than ba. He is tacitly allowed to believe that the “forbidden” expressions
are wrong. This taboo prevents the student from realizing that the allowed and the
forbidden expressions are equal. This is particularly dangerous where the taboo
urges that the bracket which can be missed out can be dropped. Forbidden
expressions of the type 2(ab) occur with so small a frequency that the students does
not learn that they are equal to 2ab. Errors of the type 2(ab) = (2a)(2b) where a and
b stand for more involved expressions, are due to this didactical negligence. (p. 36)
In later years Hans Freudenthal (1905–1990) would acquire a reputation as an outstanding
mathematics educator. The above passage shows clearly that he recognized the divide which
often existed between signs used with the intention of signifying mathematical objects and
the fact that many middle-school and lower-secondary school students did not learn the
mathematics-appropriate meanings of those signs. Such matters are dealt with in the field of
semiotics, and will be a subject of further consideration in the next chapter of this book.
Of interest to education historians is whether the New Math(s) emphasis on structure
permanently affected the teaching and learning of school mathematics, and if so whether the
effect was uniform within and between nations. There is some evidence that the importance
of structural considerations was better appreciated in some nations than others (see, e.g.,
Ding, 2016b; Education Department of Victoria and Australian Council for Educational
Research, 1966). Two decades ago, Liping Ma (1999) reported research which investigated
U.S. and Chinese elementary school teachers’ knowledge and understanding of the
fundamental properties of real numbers. Ma, who specifically investigated the teachers’
grasp of properties such as the commutative, associative, and distributive laws for rational
numbers, concluded that the Chinese teachers’ knowledge and understanding surpassed that
of the American teachers—despite the fact that the U.S. teachers tended to have studied
mathematics longer than had their Chinese counterparts. Ma’s findings were widely reported,
and have often been cited—by both mathematics educators and mathematicians.
During the first decade of the twenty-first century, the National Council of Teachers of
Mathematics (NCTM) emphasized the importance of fundamental structures within
elementary and middle-school curricula in important publications such as the Standards
(NCTM, 2000) and the Focal Points (NCTM, 2006), and through its support of the national
common-core sequence (National Governors Association Center for Best Practices, &
Council of Chief State School Officers, 2010). Analysis of common-core documents clearly
indicated that it was to be expected that by the end of sixth grade most students would be
able to summarize the fundamental properties and apply them in appropriate contexts. The
University of California mathematician, Hung-Hsi Wu (2007), for example, claimed that “by
the sixth grade most students already know about the associative and commutative laws of
addition and multiplication” (p. 4). Four years later, in his common-core-related textbook on
numbers in elementary school mathematics, Wu (2011) made clear that he had not changed
his mind on what he had written in 2007.
The results of recent research reported by Meixia Ding and her associates (see, e.g.,
Ding, 2016a, b; Ding & Li, 2010, 2014; Ding, Li, & Capraro, 2013) has suggested that within
the United States of America most prospective middle-school teachers have an inadequate
understanding of the fundamental properties of rational (and real) numbers. Thus, for
example, Ding et al., (2013), in reporting a study which examined preservice elementary
teachers' (PT’s) knowledge for teaching the associative property (AP) of multiplication,
concluded:
Results reveal that PTs hold a common misconception between the AP and
commutative property (CP). Most PTs in our sample were unable to use concrete
contexts (e.g., pictorial representations and word problems) to illustrate AP of
multiplication conceptually, particularly due to a fragile understanding of the
meaning of multiplication. The study also revealed that the textbooks used by PTs
at both the university and elementary levels do not provide conceptual support for
teaching the AP of multiplication. (p. 36)
One might expect that a similar lack of understanding would be found in middle-school
students, and that expectation led the first author of this book to conduct two investigations
into the algebraic knowledge and understandings of 80 seventh- and eighth-grade students at
a middle-school in a rural city in a midwestern state of the United States of America (Kanbir,
2014, 2016). In both studies, the students’ knowledge and understandings of the associative
and distributive properties for rational numbers were explored by analyses of data generated
by written tests and one-on-one interviews. During the interviews it became clear that the
students knew virtually nothing of the fundamental properties of rational numbers—yet their
mathematics teacher was well qualified, enthusiastic and highly regarded by all concerned.
One of the most important findings of Kanbir’s studies was that when they were asked
to “do” elementary arithmetic tasks involving several operations, all the students relied
heavily on the PEMDAS (“Please Excuse My Dear Aunt Sally”) mnemonic for order of
operations. On the other hand, none of the students could state the associative properties for
addition and multiplication, or the distributive property, accurately, and none readily knew
when they were applying those properties in elementary arithmetic and algebra exercises.
School Algebra as the Study of Variables

Some readers might think that this sixth purpose—school algebra as the study of
variables—has always been the central role of school algebra, because algebra would not
really be algebra without it. But, from about 1960 the idea that an equation was not so much
a “find-my-number” task but rather an activity for finding those numbers which would make
an open sentence—usually an equation or inequality—true and those which would make it
false, began to be made explicit (see, e.g., Davis, 1964; Ellerton & Clements, 2011). The
emergence of this refined view of elementary algebra during the late 1950s and in the 1960s
was recollected, in 2007, by Henry Pollak, a distinguished applied mathematician with a
demonstrated interest in mathematics education, and someone who served as a consultant to
one of the School Mathematics Study Group’s (SMSG’s) curriculum development teams
during the period of the new mathematics. Pollak recalled that, like other SMSG writers, he
found himself confronted with the fundamental question of what it meant to “solve” an
equation or an inequality (Karp, 2007). According to Pollak, SMSG team members were
surprised when they realized that there did not seem to be any universally-accepted
agreement on such a basic matter.
Pollak stated that the team which wrote SMSG’s ninth-grade Algebra texts agreed to adopt
the definition that to “solve” an “open sentence” involving a variable (usually an equation, or
an inequality) is to find all acceptable values of the variable which will make that sentence true.
The word “acceptable” was important, because if x were constrained to being a real
number then some apparently simple equations—like, for example, x 2 + 1 = 0—would have
School Algebra as the Study of Variables 43
no solution. However, the open sentence x2 + 1 > 0 would have infinitely many real-number
solutions. One might say that if the replacement set for x were the set of real numbers
(hereafter denoted R), then the truth set for x2 + 1 = 0 would be ∅, the null set. On the other
hand, with the same constraints operating, the truth set for x2 + 1 > 0 would be R.
The SMSG team viewed equations and inequalities as part of the same conceptual
framework. As team members, Allen, Douglas, Richmond, Rickart, Swain, and Walker
(1965), explained:
We write open [original emphasis] sentences which involve variables and for
which the notion of a truth set becomes important. It is essential that the student
consider both equations and inequalities as sentences, as objects of algebra with
equal right to our attention. (p. 44)
The SMSG approach brought to the forefront what it meant to “solve” an equation or
inequality. It emphasized the need to state the replacement set for the variable. It did not
directly use the language of functions, but what the team members decided could easily have
been expressed in function terminology if that had been deemed to be necessary. It enabled
the link between finding solutions to open sentences and Cartesian graphs to be made—for
example, the inequality 3x – 6 = 4x – 8 could be associated with the intersection of the sets
{(x, y): y = 3x – 6} and {(x, y): y = 4x – 8}, and, also with the values of the variable x for
which 3x – 6 was equal to 4x – 8.
However, although the idea of linking the equation 3x – 6 = 4x – 8 with the x-co-
ordinate of a point on a Cartesian plane would appeal to mathematicians, and to many
secondary-school teachers of mathematics, it was, in fact, conceptually dense and something
which many school students struggled to understand. Undoubtedly, the symbolism used in
the early 1960s to signify the mathematical object under consideration was, for most
beginning algebra students, extremely difficult. Although mathematicians find no difficulty
with a notation like
{(x, y): y = 3x – 6} ∩ {(x, y): y = 4x – 8},
a notation such as that was off-putting for many young learners.
The concept of a variable was extended to school geometry during the “new
mathematics” era when textbook authors thought of mappings of a plane as involving the
process of mapping sets of points on a plane to sets of points on the same plane, by
transformations such as translations, reflections, rotations, dilations, etc. (see, e.g., Clements,
Evans, Green, Smith, & Watterson, 1967). In this context, the replacement set became a set
of points on the plane, and sets of points became the domains of variables (which could be
mapped into other sets of points by well-defined rules). Mathematically, the idea was
elegant, but educationally it was difficult for school children. The approach was not new, for
it had been emphasized by Felix Klein in his Erlanger program—and set out in his
Vergleichende Betrachtungen über Neuere Geometrische Forschungen (Klein, 1872; Sfard,
1995). Since the 1960s, there has always been some emphasis on this variables approach to
geometry, and between 2010 and 2017 it was a cornerstone of the geometry sections of the
common-core sequence adopted in many states within the United States of America.
Over the past four decades, or so, the conception of school algebra as the study of
variables has received a fillip as a result of technological advances. In the 1970s and 1980s,
Logo software emphasized the position of variables in elementary computing (Papert, 1980),
and geometrical software such as Cabri-Géomètre (Laborde, 2001) and Geometers’ Sketchpad
(Battista, 2002) were used to assist school children to study geometry from a transformation,
mappings-of-the-plane, approach which emphasized a variables perspective (Banchoff,

2012). Still, in schools, there was much resistance to this variables approach to school
algebra. Algebraic equations are still being treated as “find-the-number” exercises, and there
is still much emphasis on simplifying algebraic expressions by making use of “like terms.”
The emphasis on the concept of a variable has also arisen in relation to structural
properties of real numbers. For example, when the associative property for multiplication is
stated as: “For arbitrary real numbers a, b, c, the value of a × (b × c) is equal to (a × b) × c,”
the letters a, b, and c, are, in fact, being used to represent variables. That is curious, because
in most contexts associated with beginning school algebra the early letters of the alphabet (a,
b, and c) are often used to represent constant values, and it is the final letters of the alphabet
(x, y, and z) which are most commonly used to represent variable quantities. For a beginning
algebra student this can be confusing. Also, with a statement like “the value of a × (b × c) is
equal to (a × b) × c, for all real numbers,” the meaning of the word, “all” can cause
difficulty, because, in fact, it does not clearly convey the idea that the “a” in a × (b × c) must
always take the same value as the “a” in (a × b) × c, etc. Mathematicians have learned the
notation, and the overall meaning of the symbols which make up the signifier, but many
beginning algebra students have not, and they therefore find it difficult to learn important
structural properties with understanding.
Ellerton and Clements (2011) reported that about 50 percent of 328 prospective middle-
school mathematics specialists in the United States of America believed that when they gave
the answer a, b as solutions to the quadratic equation (x – a)(x – b) = 0, for them the x in the
first parentheses, (x – a) was equal to a, and simultaneously the x in (x – b) was equal to b.
That fundamental misconception of a variable became evident in a check when students
agreed that since 0 × 0 = 0, it followed that a and b were both solutions. Lim (2000),
Vaiyavutjamai (2004), and Vaiyavutjamai, Ellerton, and Clements (2005) had previously
noted that secondary school students in Brunei Darussalam, Thailand, and the United States
of America who had completed lessons on quadratic equations were likely to harbor the same
serious misconception of the variable. Recent research (e.g., Didis & Erbas, 2015; López,
Robles, & Martínez-Planel, 2015) has shown that that and other misconceptions with
variables used in quadratic equations can be found lurking in the cognitive structures of
secondary students in many nations.
It seems to be the case that although algebra has been part of high-school mathematics
curricula for centuries, it is only in recent times that research has begun to unravel difficulties
that students experience in understanding the fundamental concept of a variable. In the
seventeenth, eighteenth, and nineteenth centuries, and for much of the twentieth century,
those who defined the curricula for “algebra in schools” by writing school algebra textbooks,
tended to be mathematicians or senior secondary teachers of mathematics who had very little
idea of the conceptual difficulties which many, perhaps most, students beginning to learn
algebra would experience (see, e.g., Klein, 1949, 2004). “Bottom-up” forms of research were
needed to complement the much more common “top-down” prescriptions and advice.
Is School Algebra a Unidimensional Trait?

In the 1960s and 1970s, when the new mathematics movement was sweeping the world,
many nations whose secondary-school systems had previously included arithmetic, algebra,
geometry, trigonometry, calculus and mechanics as separate subjects in their curricula decided
to unify their mathematical offerings by creating composite subjects—often called, simply,
“Mathematics” (or descriptive titles such as “Mathematics I,” “Mathematics II,” “Basic
Is School Algebra a Unidimensional Trait? 45
Mathematics,” “Calculus,” and “Advanced Mathematics”) (Clements, 2003). This did not
occur in all nations, however—for example, in the United States of America the titles
Algebra I, Algebra II, and Geometry remained commonplace in all states.
Even in the late 1980s, it was still the case that many students, in some nations, did not
formally study algebra at all (Clements, Keitel et al., 2013). In the second half of the 1980s in
the United States of America, for example, about 30 percent of secondary-school students
had not studied algebra and would never study it while still at school. Of those who had,
many had studied it for only a small period of time. Yet, in some other nations, formal
algebra was studied by most students in grades 7 and 8 (Clements, 2003; Usiskin, 1987), and
often right through grades 7 through 12.
The list of six separable purposes (in this chapter) of school algebra arising from our
analysis of its history—see Table 2.1—has raised the question whether the term “school
algebra” now has a definite meaning. In the United States of America, algebra is studied as a
subject, yet, such is the power of labeling that many U.S. students who have completed
courses in calculus have told us that they do not think that algebra is involved in calculus
(Ellerton & Clements, 2011).
The U.S. national common-core sequence for mathematics—which, since 2010, has
been adopted in many U.S. states—includes a content section called “operations and
algebraic thinking” which is intended to refer to content covered throughout all K–12 grades.
There is an often-expressed belief that although children in grades K–2 may not be studying
algebra, directly, “we should expect students in early grades to think algebraically” (Cai &
Knuth, 2011b, p. 3). Such is the strength of this idea that there have been articles recently
published in the Journal for Research in Mathematics Education titled “The Development of
Children's Algebraic Thinking: The Impact of a Comprehensive Early Algebra Intervention
in Third Grade” (Blanton et al., 2015) and “A Learning Trajectory in 6-Year-Olds’ Thinking
About Generalizing Functional Relationships” (Blanton, Brizuela, Gardiner, Sawrey, &
Newman-Owens, 2015). There can be no doubting the large impact of this new thinking
about the nature and scope of algebraic thinking in school mathematics curricula on some
educators in the United States of America and in other parts of the world (see, for example,
chapters in Cai & Knuth, 2011c).
All of the above squarely raises at least four important questions, which are mainly
curricular in their nature:
• What is school algebra?
• What aspects of school mathematics embrace algebra?
• What is algebraic thinking?
• Does the term “school algebra” refer to a unidimensional cognitive trait?
Our historical overview in this chapter has drawn attention to six separable purposes for
school algebra, and all of them would still appear to be relevant to today’s discussions on
curriculum.
Should algebra be a gatekeeper to higher-level studies (and not only to higher-level
mathematical studies)? One might ask whether persons wishing to study Law, or Philosophy,
or …, should be required to have succeeded in school algebra before they can gain entry to
their preferred courses. Undoubtedly, it is true that functional thinking—by which variable
quantities are related to each other—is now important in many areas of life, and the possession
of algebraic skills can help one make good progress in areas of life outside of science and
mathematics. In that sense, school algebra is more important today than it was at any earlier
time. So, what does that imply for today’s school algebra?
There can be no doubt that algebra as generalized arithmetic has been, is, and will
continue to be, important. That said, it would be difficult to counter an argument that the first
steps in secondary-school algebra should not be to learn to operate effectively with algebraic
symbols and other representations of varying quantities. Given the great difficulty which
many beginning algebra students have always had with the language and written symbols of
algebra, one must ask whether it is wise to insist that all seventh-grade (or eighth-grade, …)
students should be expected to learn to recognize and use the main signifiers which point
toward more sophisticated algebraic knowledge and skills. It would be interesting to ask
adults across the world today, in various walks of life, whether they think that all students
should be required to study secondary-school algebra.
The view of algebra as functional thinking, as essential for being able to model real-life
situations, is often emphasized by mathematicians, mathematics educators, and scientists.
But, one wonders how often most adults, even those who are well versed in algebra, use
algebra to model practical problems—when was the last time you solved a quadratic equation
outside of the classroom? On the other hand, graphical representations of data are common in
most forms of media, and are more easily presented these days as a result of the march of
technology. These representations often summarize relationships between varying quantities,
and carry an invitation to express relationships between the variables. The challenge is for
secondary teachers to devise ways and means of getting students to define adequately
variables relating to situations which are real and of interest to the students, to devise learning
contexts in which they generate data, and to assist the students to move towards making
generalizations about how the data can be related. John Perry and Eliakim Hastings Moore
strongly favored that approach more than century ago, but their efforts were thwarted by,
among others, the conservative but highly influential textbook-author, David Eugene Smith—
who, in 1920, served as President of the Mathematical Association of America.
It seems to us that an enduring purpose of school algebra is the need to assist students
to develop their number sense by becoming familiar with the structural features of rational
and real numbers. Too much of the history of school algebra has followed from initiatives of
outstanding mathematicians, like Isaac Newton, Felix Klein, and Jean Dieudonné, who have
had little experience in teaching “ordinary” school children aged between 10 and 17 years.
The period of the new mathematics (between about 1957 and 1975) generated textbooks
which emphasized structure, but the language and symbolisms used in textbooks were so
sophisticated that only a relatively small percentage of school students were prepared to
engage fully with the texts provided. In recent years those who constructed the common-core
algebra sequence seemed to think that structural properties (such as the commutative,
associative and distributive properties) will be well known by middle-school students, but
our experience and research indicates that that is definitely not the case—the study
summarized in Chapters 4 through 9 of this book provides further evidence for what we are
saying. There still remains the question, through: Are algebraic structures too difficult for
most middle- and high-school students? Tzanakis (1991) is one among many to have argued
that the answer is “Yes.”.
We are not convinced that school algebra, as it is now being considered in various parts
of the world, represents a unidimensional trait. Studying algebra from a structural point of view
may be quite a different thing from studying it from a modeling perspective. Certainly, the two
can be combined, but the issue is whether students, teachers, and researchers, who are engaged
The History of Mathematics and the History of School Mathematics 47
in efforts to bring together the different strands recognize that although the two approaches
may be mutually supportive, they may not be representatives of “the same thing.”
A major conclusion from the historical overview provided in this chapter is that since
the introduction of algebra into the secondary school curriculum there has always been,
among middle-school and lower-secondary-school students, a disconnect between the
signifiers (signs and symbols) and the signifieds (the mathematical objects which are being
considered). We believe that the main issue is a semiotic one, and the study which is
presented later in this book has proceeded from the assumption that students need to learn not
only to understand what the authors of textbooks write about algebra, but also how to use
appropriate language and to make generalizations of their own (Clements, 1980; Clements &
Ellerton, 1996).
The History of Mathematics and the History of School Mathematics

Figure 2.7, which is from Ellerton and Clements (2014—see p. 323), distinguishes
between three types of mathematics—research mathematics, service mathematics, and
mathematics education—and emphasizes that the three forms of mathematics are developed
within societies in which ethnomathematical forces shape, use, and sometimes modify
existing forms of both mathematics and practices which belong to mathematics education.”
The advent of secondary school mathematics, in the seventeenth century, never prevented
ethnomathematical contexts from varying enormously, both within and between nations.
Advances in mathematics, and in applications of mathematics, interacting with the cultures
and needs of evolving communities, have generated changes to what Ian Westbury (1980)
referred to as “intended,” “implemented,” and “received” school mathematics curricula, but it
is difficult to establish a link between local changes in school mathematics curricula and the
work of research mathematicians.
Research
Mathematics Service
Mathematics
Mathematics
Education
Ethnomathematical Contexts,
Including Family, Community, and Work
Figure 2.7. Different ways of “seeing” problems or situations which might relate to
mathematics (from Ellerton & Clements, 2014, p. 323).
There is a natural tendency for research mathematicians to believe that they are the only
“true” mathematicians, and for “applied mathematicians” to believe that they are the only
ones who “have their feet on the ground,” doing work which permits mathematics to become
useful to society. This carries over to those who have written histories of mathematics,
especially histories focusing on the past 350 years, who have displayed a tendency to
eulogize the great achievements of mathematicians during that period (see, e.g. Struik, 1948).
So far as the history of school mathematics over the past 350 years is concerned, there has
been a tendency to assume that because there were such great achievements in mathematics
among European research mathematicians (see, e.g., the second volume of Struik, 1948)
then, somehow, it is reasonable to assume that the best research into the history of school
mathematics should have come from European scholars, especially European mathematicians
or mathematics educators—and if one looks at the list of authors who wrote chapters for
Karp and Schubring’s (2014) edited collection, Handbook on the History of Mathematics
Education, then one might be persuaded that that assumption was correct. But, such has been
the high performances of students in schools in Singapore, Japan, Taiwan, Hong Kong,
China, India, and Vietnam, in numerous comparative studies over the past 50 years, that any
assumption that, overall, mathematics educators in any single part of the world other than
Eastern Asia have the best knowledge on matters concerning school mathematics should be
questioned.
Studying the history of school mathematics can provide many scholarly challenges that
may not be much related to Western mathematical knowledge. Take, for example, the situation
in Papua New Guinea and Oceania where, half a century ago, there were some 1000 different
languages spoken and 1000 different indigenous counting systems used (Owens, Lean, Paraide,
& Muke, 2017). A deep understanding of Western mathematical knowledge is not really a strong
qualification for being able to understand, and tell, the story of how the Western version of
school mathematics has been introduced to Papua New Guinea’s schools over the past 70
years, or so.
In the 1960s, Zoltan Dienes, who held a doctorate in Pure Mathematics and had a hard-
won, international reputation as a brilliant teacher of mathematics at the school level, tried to
introduce his internationalized version of school mathematics, with a strong emphasis on
algebraic structures, into the community schools of Papua New Guinea. However, neither the
nation’s students nor their teachers could cope with Dienes’s very structured, quite
sophisticated, but heavily-funded program, and the overall result was not one which
enhanced the overall quality of mathematical learning in the nation’s schools (Clements,
Keitel et al., 2013).
Although school mathematics is an international phenomenon, local versions of school
mathematics need to take account of economic, cultural and linguistic contexts if school
children are to learn mathematics well. Those who wish to study the history of school
mathematics should not assume that a superior knowledge of mathematics is the most
important lens through which one should study the history of school mathematics. More
bottom-up approaches (e.g., Hertel, 2016), which value local cultures and traditions, need to
become standard fare for those who will research the history of school mathematics.
In 1977 the National Council of Mathematics, in the United States of America, published
a book with the title Calculus: Readings from the Mathematics Teacher (Grinstein & Michaels,
1977). The book had 11 sections, the first of which carried the summary title “Historical
Overview,” and the second, “Pedagogical Overview.” From a history-of-school-mathematics
perspective the book is interesting for it offers much information on the history of calculus,
and on how calculus concepts and applications were spread abroad during the period 1670–
1820. However, there was virtually nothing in the book on the history of the teaching and
learning of calculus in schools. The first chapter, which was written, in 1946, by the
mathematician Carl B. Boyer, was aimed at celebrating the first calculus textbook, by
Marquis de L’Hôpital (1696), but there was no mention in the book—which was published
by NCTM—of how textbooks on calculus for schools came to be written, on how books on
calculus were used in schools, and whether such books, when written, were suited to the
needs and mathematical development of most school learners.
In its own way, the 1977 NCTM publication on Calculus epitomizes the lack of
attention that has been accorded the history of school mathematics during the period 1670–
1820—a major focus has been on algebra textbooks, but there is little more than silence on
whether there was strong evidence that the availability of the textbooks generated better
teaching and better learning of algebra. We need to know if, and how, in the first half of the
nineteenth century, algebra textbooks were used in ways which made them part of the
cyphering tradition (Ellerton & Clements, 2012). We also need to know how, following the
demise of the cyphering tradition, algebra textbooks came to be used in whole-class teaching
situations. We need to know the extent to which school children understood the algebra that
they were asked to learn. That important historical agenda has been largely neglected by
mathematics education researchers—even the relatively few of them interested in the history
of school mathematics. Hopefully, the situation is about to change.
Within Southeast-Asian nations, algebra is typically taught in whole-class
environments, and such has been the success of students in some nations that there has been a
rush to investigate why the students appear to be doing so well. There have been claims that
teachers are extremely well trained, and that they often ask their students to respond to high-
level questions (see, e.g., Stigler & Hiebert, 1999). However, there have been other studies,
also carried out in Southeast Asia (see, e.g., Vaiyavutjamai, 2004, 2006), which have
revealed that many secondary students in Asian schools do not learn much algebra from their
algebra lessons, and that teachers only rarely ask their students to respond to high-level
questions. A rush to over-generalization needs to be avoided.
The main issue appears to be not what makes up the intended curriculum, but rather
what constitutes the implemented curriculum. In the study described in this book, there is
much emphasis on linking intended, implemented and received algebra curricula, at middle-
school levels—the need for which is getting much recent attention among mathematics
education researchers (see, e.g., Cai, Morris, Hohensee, Hwang, Robinson & Hiebert, 2017).
References
Albree, J. (2002). Nicolas Pike’s Arithmetic (1788) as the American Liber Abbaci. In D. J.
Curtin, D. E. Kullman, & D. E. Otero (Eds.), Proceedings of the Ninth Midwest History
of Mathematics Conference (pp. 53–71). Miami, FL: Miami University.
Alexander, J. (1709). A synopsis of algebra, being the posthumous work of John Alexander,
of Bern ... To which is added an appendix by Humfrey Ditton ... (translated from the
Latin by Samuel Cobb). London, UK: Christ’s Hospital.
Alexandri, J. (1693). Synopsis algebraica, opus posthumum. London, UK: Christ’s Hospital.
Al-Khalili, J. (2011). The house of wisdom: How Arabic science saved ancient knowledge
and gave us the Renaissance. New York, NY: The Penguin Press.
Allen, F. B., Douglas, E. C., Richmond, D. E., Rickart, C. E., Swain, H., & Walker, R. J.
(1965). First course in algebra: Teacher’s commentary, Part 1. Stanford, CA: School
Mathematics Study Group.
Andrews, A. (1813). Cyphering book (handwritten manuscript). Cheshire, CT: Cheshire
Episcopal Academy.
Artigue, M., Assude, T., Grugeon, B., & Lengant, A. (2001). Teaching and learning algebra:
Approaching complexity through complementary perspectives. In H. Chick, K. Stacey,
J. Vincent, & J. Vincent (Eds.), The future of the teaching and learning of algebra
(Proceedings of the 12th ICMI Study) (pp. 21–32). Melbourne, Australia: The
University of Melbourne.
Banchoff, T. (2012). Algebraic thinking and geometric thinking. In C. E. Greenes & R.
Rubinstein (Eds.), Algebra and algebraic thinking in school mathematics (pp. 99–110).
Reston, VA: National Council of Teachers of Mathematics.
Battista, M. T. (2002). Shape makers: A computer environment that engenders students’
construction of geometric ideas and reasoning. Computers in Schools, 17(1/2), 105–
120.
Beman, W. W., & Smith, D. E. (1902). Academic algebra. Boston, MA: Ginn.
Bestor, A. (1956). The restoration of learning. New York, NY: Alfred A. Knopf.
Bézout, É. (1794). Cours de mathématiques à l’usage des gardes du pavillon et de la maine.
Paris, France: Baudelot & Eberhart.
Blanton, M., Brizuela, B. M., Gardiner, A. M., Sawrey, K., & Newman-Owens, A. (2015). A
learning trajectory in 6-year-olds’ thinking about generalizing functional relationships.
Journal for Research in Mathematics Education, 46(5), 511–558.
Blanton, M., Stephens, A., Knuth, E., Gardiner, A. M., Isler, I., & Kim, J. S. (2015). The
development of children's algebraic thinking: The impact of a comprehensive early
algebra intervention in third grade. Journal for Research in Mathematics Education,
46(1), 39–87.
Bolton, F. E. (1900). The secondary school system of Germany. London, UK: Edward
Arnold.
Borel, E. (1904). Les exercices pratiques de mathématiques dans l’enseignement secondaire.
Revue Générale des Sciences Pures et Appliqués, 15, 431–440.
Bourdon, L. P. M. (1831). Elements of algebra, by Bourdon, translated from the French for
the use of colleges and schools. Boston, MA: Hilliard, Gray, Little and Wilkins.
Brock, W. H. (1981). The Japanese connection: Engineering in Tokyo, London, and Glasgow
at the end of the nineteenth century. The British Journal of the History of Science,
14(3), 227–243.
Brock, W. H., & Price, M. H. (1980). Squared paper in the nineteenth century: Instrument of
science and engineering, and symbol of reform in mathematical education. Educational
Studies in Mathematics, 11, 365–381.
Cai, J., & Knuth, E. (2011a). A global dialogue about early algebraization from multiple
Cai, J., & Knuth, E. (2011b). Preface to Part I. In J. Cai & E. Knuth (Eds.), Early
algebraization: A global dialogue from multiple perspectives (pp. 3–4). Heidelberg,
Germany: Springer.
Cai, J., & Knuth, E. (Eds.). (2011c). Early algebraization: A global dialogue from multiple
perspectives. Heidelberg, Germany: Springer.
Cai, J., Morris, A., Hohensee, C., Hwang, S., Robinson, V., & Hiebert, J. (2017). Making
classroom implementation an integral part of research. Journal for Research in
Mathematics Education, 48(4), 342–347.
Cai, J., & Moyer, J. (2012). Developing algebraic thinking in earlier grades: Some insights
from international comparative studies. In C. E. Greenes & R. Rubinstein (Eds.),
Algebra and algebraic thinking in school mathematics (pp. 169–180). Reston, VA:
National Council of Teachers of Mathematics.
Publishing Co.
Carslaw, H. S. (1914). The teaching of mathematics in Australia: Report presented to the
International Commission on the Teaching of Mathematics. Sydney, Australia: Angus
& Robertson Ltd.
Chambers, C. E. (1964). The Cuisenaire Gattegno method of teaching mathematics. Reading,
UK: Educational Explorers.
Chateauneuf, A. O. (1929). Changes in the content of elementary algebra since the beginning
of the high school movement as revealed by the textbooks of the period (Unpublished
Ph.D. thesis) University of Pennsylvania.
Chorlay, R. (2011). The multiplicity of viewpoints in elementary function theory: Historical
and didactical perspectives. In V. Katz & C. Tzanakis (Eds.), Recent developments on
introducing a historical dimension in mathematics education (pp. 57–66). Washington,
DC: Mathematical Association of America.
Clairaut, A.-C. (1746). Elémens d'algèbre. Paris, France: Guérin.
Clements, M. A. (1979). Relationships between the University of Melbourne and the
secondary schools of Victoria, 1890–1912 (Unpublished Ph.D. thesis). The University
of Melbourne.
Clements, M. A. (1980). Analysing children’s errors on written mathematical tasks.
Educational Studies in Mathematics, 11(1), 1–21.
Clements, M. A. (1992). Mathematics for the minority: Some historical perspectives on
school mathematics in Victoria. Geelong, Australia: Deakin University.
Clements, M. A. (2003). An outsider’s view of North American school mathematics
curriculum trends. In G. M. A. Stanic & J. Kilpatrick (Eds.), A history of school
mathematics (Vol. 2, pp. 1509−1580). Reston, VA: National Council of Teachers of
Mathematics.
Clements, M. A., Bishop, A. J., Keitel, C. Kilpatrick, J., & Leung, F. (Eds.). (2013). Third
international handbook of mathematics education. New York, NY: Springer.
Clements, M. A., & Ellerton, N. F. (1996). Mathematics education research: Past, present
and future. Bangkok, Thailand: UNESCO.
Clements, M. A., & Ellerton, N. F. (2015). Thomas Jefferson and his decimals 1775–1810:
Neglected years in the history of U.S. school mathematics. New York, NY: Springer.
Clements, M. A., Evans, K. M., Green, L., Smith, G., & Watterson, G. (1967). Mathematics
for today and tomorrow. Melbourne, Australia: Sun Books.
Clements, M. A., Grimison, L. A., & Ellerton, N. F. (1989). Colonialism and school
mathematics in Australia 1788–1988. In N. F. Ellerton & M. A. Clements (Eds.),
School mathematics: The challenge to change (pp. 50–78). Geelong, Australia: Deakin
University.
Clements, M. A., Keitel, C., Bishop, A. J., Kilpatrick, J., & Leung, F. (2013). From the few to
the many: Historical perspectives on who should learn mathematics. In M. A. Clements,
A. J. Bishop, C. Keitel, J. Kilpatrick, & F. Leung (Eds.), Third international handbook
of mathematics education (pp. 7–40). New York, NY: Springer.
Cocker, E. (1677). Cocker's arithmetick: Being a plain and familiar method suitable to the
meanest capacity for the full understanding of that incomparable art, as it is now
taught by the ablest school-masters in city and country. London, UK: John Hawkins.
Colburn, W. (1825). An introduction to algebra upon the inductive method of instruction.
Boston, MA: Hilliard, Gray & Co.
Colebrooke, H. T. (1817/2013). Algebra with arithmetic and mensuration from the Sanscrit
of Brahmegupta and Bhascara (translated by Colebrooke). New York, NY: Cambridge
University Press.
Cremin, L. A. (1956). The problem of curriculum-making: An historical perspective. In
Association of Supervision and Curriculum Development (Ed.), What shall high
schools teach? (pp. 6–26). Washington, DC: Association of Supervision and
Curriculum Development.
Cremin, L. A. (1964). The transformation of the school: Progressivism in American
education 1876–1957. New York, NY: Vintage Books.
da Ponte, J. P., & Guimarāes, H. M. (2014). Notes for a history of the teaching of algebra. In
A. Karp & G. Schubring (Eds.), Handbook on the history of mathematics education
(pp. 459–472). New York, NY: Springer.
Davies, C. (1837). Elements of algebra translated from the French of M. Bourdon. New
York, NY: Wiley & Long.
Davies, C. (1846). Elementary algebra: Embracing the first principles of the science. New
York, NY: A. S. Barnes and Co.
Davis, R. B. (1964). Discovery in mathematics: A text for teachers. Reading, MA: Addison-
Wesley.
Day, J. (1814). An introduction to algebra, being the first part of a course in mathematics,
adopted to the method of instruction in the American colleges. New Haven, CT: Howe
& Deforest.
de L’Hôpital, G. F. A. (1696). Analyse des infiniment petits, pour l’intelligence des lignes
courbes. Paris, France: L’Imprimerie Royale.
Didis, M. G., & Erbas, A. K. (2015). Performance and difficulties of students in formulating
and solving quadratic equations with one unknown. Educational Sciences: Theory &
Practice, 15(4), doi: 10.12738/estp.2015.4.2743.
Dienes, Z. P. (1960). Building up mathematics. London, UK: Hutchinson & Co.
Dienes, Z. P. (1967) Groups and co-ordinates. Paris, France: Hutchinson Educational.
Dienes, Z. P. (1968). Powers, roots and logarithms. New York, NY: Herder and Herder.
Dilworth, T. (1762). The schoolmaster’s assistant. Being a compendium of arithmetic, both
practical and theoretical (11th ed.). London, UK: Henry Kent.
Ding, M. (2016a). Developing preservice elementary teachers’ specialized content
knowledge for teaching fundamental mathematical ideas: The case of the associative
property. International Journal of STEM Education, 3(9), 1–19.
Ding, M. (2016b). Opportunities to learn: Inverse operations in U.S. and Chinese elementary
mathematics textbooks. Mathematical Thinking and Learning, 18(1), 45–68.
Ding, M., & Li, X. (2010). A comparative analysis of the distributive property in U.S. and
Chinese elementary mathematics textbooks. Cognition and Instruction, 28(2), 146–180.
Ding, M., & Li, X. (2014). Transition from concrete to abstract representations: The
distributive property in a Chinese textbook series. Educational Studies in Mathematics,
87, 103–121.
Ding, M., Li, X, & Capraro, M. M. (2013). Preservice elementary teachers’ knowledge for
teaching the associative property of multiplication: A preliminary analysis. The Journal
of Mathematical Behavior, 32, 36–52.
Ditton, H. (1709). Appendix to the algebra of John Alexander. London, UK: Christ’s
Hospital.
Durell, F., & Robbins, E. R. (1901). Grammar school algebra. Harrisburg, PA: R. L. Myers
& Co.
Education Department of Victoria and Australian Council for Educational Research. (1966).
Background in mathematics: A guidebook to elementary mathematics for teachers in
primary schools. Melbourne, Australia: Authors.
Ellerton, N. F., & Clements, M. A. (2012). Rewriting the history of school mathematics in
North America 1607–1861. New York, NY: Springer.
Ellerton, N. F., & Clements, M. A. (2014). Abraham Lincoln’s cyphering book and ten other
extraordinary cyphering books. New York, NY: Springer.
Ellerton, N. F., & Clements, M. A. (2017). Samuel Pepys, Isaac Newton, James Hodgson
and the beginnings of secondary school mathematics: A History of the Royal
Mathematical School within Christ’s Hospital, London 1673–1868. New York, NY:
Springer.
Exner, R. M., & Rosskopf, M. F. (1970). Proof. In M. Rosskopf (Ed.), The teaching of
secondary school mathematics: Thirty-third yearbook (pp. 196–240). Washington, DC:
Filloy, E., & Sutherland, R. (1996). Designing curricula for teaching and learning algebra. In
A. J. Bishop, M. A. Clements, C. Keitel, J. Kilpatrick, & C. Laborde (Eds.),
International handbook of mathematics education (pp. 139–160). Dordrecht, The
Netherlands: Kluwer Academic Publishers.
Freudenthal, H. (Ed.). (1962a). Report on the relations between arithmetic and algebra in
mathematical education up to the age of 15 years. Groningen, The Netherlands: J. B.
Wolters.
Freudenthal, H. (1962b). Logical analysis and critical survey. In H. Freudenthal (Ed.), Report
on the relations between arithmetic and algebra in mathematical education up to the
age of 15 years (pp. 20–41). Groningen, The Netherlands: J. B. Wolters.
Giacardi, L. (2009). The school as “laboratory”: Giovanni Vailati and the project to reform
mathematics teaching in Italy. International Journal for the History of Mathematics
Education, 4(1), 5–28.
Godfrey, C. (1913). The teaching of calculus. Mathematical Gazette, 7, 233–240.
Grimison, L. A. (1988, July). The introduction of calculus into the secondary mathematics
curriculum—How, when and why? Paper presented to the annual conference of the
Mathematics Education Research Group of Australasia, Geelong, Australia.
Grinstein, L. S., & Michaels, B. (1977). Calculus: Readings from the Mathematics Teacher.
Murray.
Hertel, J. T. (2016). Investigating the implemented curriculum of navigation cyphering
books. For the Learning of Mathematics, 36(3), 4–10.
Hodgson, J. (1723). A system of the mathematics. London, UK: Thomas Page.
Husky, M. E. (2011). David Eugene Smith and the algebra curriculum (Unpublished M.Sc.
thesis). Illinois State University.
Jahnke, H. N. (1983). Origins of school mathematics in early nineteenth century Germany.
Journal of Curriculum Studies, 18(1), 85–94.
Jones, P. S., & Coxford, A. F. (1970). First steps toward “revision,” 1894–1920. In P. S. Jones
& A. F. Coxford (Eds.), A history of mathematics education in the United States and
Canada (pp. 36–45). Washington, DC: National Council of Teachers of Mathematics.
Kanbir, S. (2014, November). Two approaches: Beginning algebra students’ variable
concept development. Professional project presented to the Group for Educational
Research in Mathematics at Illinois State University, Normal, IL.
Kanbir, S. (2016, April 12). Three different approaches to middle-school algebra: Results of
a pilot study. Paper presented at the 2016 Research Conference of the National Council
of Teachers of Mathematics, held in San Francisco, CA.
Karp, A. (2007). Interview with Henry Pollak. The International Journal for the History of
Karp, A., & Schubring, G. (Eds.). (2014). Handbook on the history of mathematics
education. New York, NY: Springer.
Karpinski, L. C. (1980). Bibliography of mathematical works printed in America through
1850. New York, NY: Arno Press.
Keitel, C. (2006). Perceptions of mathematics and mathematics education in the course of
history—A review of Western perspectives. In F. K. S. Leung, K-D Graf, & F. Lopez-
Real (Eds.), Mathematics education in different cultural traditions—A comparative
study of East Asia and the West: The 13th ICMI Study (pp. 81–94). New York, NY:
Springer.
Klein, F. (1872). Vergleichende betrachtungen über neuere geometrische forschungen
(“comparative review of recent researches in geometry”), which was reprinted in 1893
in Mathematische Annalen, 43, 63–100. An English translation by Mellen Haskell
appeared in 1893 in the Bulletin of the New York Mathematical Society, 2, 215–249.
Klein, F. (1949). Elementary mathematics from an advanced standpoint: Arithmetic, algebra,
analysis (translated from the German by E. R. Hedrick & C. A. Noble). New York,
NY: Dover Publications.
Klein, F. (2004). Elementary mathematics from an advanced standpoint: Geometry
(translated from the German by E. R. Hedrick & C. A. Noble). New York, NY: Dover
Publications.
Kleiner, I. (1986), The evolution of group theory: A brief survey. Mathematics Magazine,
59(4), 195–215.
Kline, M. (1973). Why Johnny can’t add: The failure of the new math. New York, NY: St.
Martin’s Press.
Laborde, C. (2001). Integration of technology in the design of geometry with Cabri-
Geometry. International Journal of Computers for Mathematical Learning, 6, 283–317.
Lacroix, S. F. (1815). Complément des elémens d’algébre, à l’usage de l’école centrale des
quatre-nations (3rd ed.). Paris, France: Chez Courcier.
Laplace, P.-S. (1799). Traité de mécanique céleste. Paris, France: J. B. M. Duprat.
Lawson, D. R. (1990). The problem, the issues that speak to change. In E. L. Edwards (Ed.),
Algebra for everyone (pp. 1–6). Reston, VA: National Council of Teachers of
Mathematics.
level Mathematics: A case study (Unpublished M.Ed dissertation). Universiti Brunei
Darussalam.
Loemker, L. (Ed.). (1969). Leibniz: Philosophical papers and letters. Dordrecht, The
Netherlands: D. Reidel.
López, J., Robles, I., & X Martínez-Planel, R. (2015). Students’ understanding of quadratic
equations. International Journal of Mathematical Education in Science and
Technology. doi: 1080/0020739X.2015.119895
MacGregor, M., & Stacey, K. (1997). Students’ understanding of algebraic notation: 11–13.
Educational Studies in Mathematics, 33, 1–19.
May, K. O., & Van Engen, H. (1959). Relations and functions. In H. P. Fawcett, A. M. Hach,
C. W. Junge, H. W. Syer, H. Van Engen, & P. S. Jones (Eds.), The growth of
mathematical ideas, grades K–12: Twenty-fourth yearbook (pp. 65–110). Washington,
DC: National Council of Teachers of Mathematics.
McQualter, J. W. (1975). Some influences on mathematics curriculum making: A historical
review. Australian Mathematics Teacher, 30(1), 15–26.
Milgram, R. J. (2013). Common Core’s effect on math. Retrieved from
https://www.youtube.com/watch?v=MuexIMWBOQY
Monroe, W. S. (1917). Development of arithmetic as a school subject. Washington, DC:
Government Printing Office.
Moon, B. (1986). The “new maths” curriculum controversy. London, United Kingdom:
Falmer Press.
Moore, E. H. (1903). On the foundations of mathematics. Science, 17(428), 401–416.
Moore, E. H. (1906). The cross-section paper as a mathematical instrument. School Review,
14, 317–338.
National Council of Teachers of Mathematics. (2000). Principles and standards for school
mathematics. Reston, VA: Author.
National Council of Teachers of Mathematics. (2006). Curriculum focal points for
Prekindergarten through Grade 8 mathematics: A quest for coherence. Reston, VA:
Author.
National Governors Association Center for Best Practices, & Council of Chief State School
Officers. (2010). Common Core State Standards for Mathematics. Washington, DC:
Authors. (Also cited as CCSSM (2010)).
National Mathematics Advisory Panel. (2008). Foundations for success: The final report of
the Mathematics Advisory Panel. Washington, DC: U.S. Department of Education.
Newton, I. (1739). Philosophiæ naturalis principia mathematica (Vol. 1). Geneva,
Switzerland: Barrillot & Filii.
Nordgaard, M. A. (1928). Introductory calculus as a high school subject. In J. R. Clark & W.
D. Reeve (Eds.), Selected topics in the teaching of mathematics (Third Yearbook of the
National Council of Teachers of Mathematics) (pp. 65–101). New York, NY: Teachers
College, Columbia University.
Olney, E. (1885). First principles of algebra: An introduction to the author’s complete, or
university algebra. New York, NY: Sheldon & Company.
Owens, K. D., Lean, G. A., Paraide, P., & Muke, C. (2017). History of number: Evidence from
Papua New Guinea and Oceania. New York, NY: Springer.
Oxford and Cambridge Schools Examination Board. (1899). Oxford-Cambridge schools
agreement, 1899. Cambridge, UK: Author.
Papert, S. (1980). Mindstorms: Children, computers, and powerful ideas. New York, NY:
Basic Books.
Pappano, L. (2012). The algebra problem. Harvard Education Letter, 28(3), 1–2.
Perel, W. M., & Vero, P. D. (1967). New mathematics and old teachers. The Educational
Forum, 31(3), 345–348.
Perry, J. (1899). Practical mathematics. London, UK: Department of Science and Art of the
Committee of Council on Education.
Perry, J. (Ed.). (1902). Discussion on the teaching of mathematics. London, UK: Macmillan.
Perry, J. (1912). Practical mathematics. Nature, 90, 34–35.
Phillips, C. J. (2014). The new math: A political history. Chicago, IL: The University of
Chicago Press.
Pike, N. (1788). The new complete system of arithmetic, composed for the use of the citizens
of the United States. Newburyport, MA: John Mycall.
Pike, N. (1793). The new and complete system of arithmetic composed for the use of the
citizens of the United States (abridged for the use of schools). Newbury-Port, MA: John
Mycall, Isaiah Thomas.
Price, M. H. (1986). The Perry movement in school mathematics. In M. H. Price (Ed.), The
development of the secondary curriculum (pp. 103–155). London, UK: Croom Helm.
Price, M. H. (1994). Mathematics for the multitude? Leicester, UK: Mathematical
Association.
Raub, A. N. (1888). Methods of teaching: Including the nature, object, and laws of
education, methods of instruction, and methods of culture. Philadelphia, PA: E. L. Raub
& Co.
Ray, J. (1848). Ray’s algebra, Part First: On the analytic and inductive methods of
instruction. Cincinnati, OH: Winthrop B. Smith & Co.
Rech, J. F., & Harrington, J. (2000). Algebra as a gatekeeper: A descriptive study at an urban
university. Journal of African American Studies, 4(4), 63–71.
Roberts, D. L. (2012). American mathematicians as educators, 1893–1923. Boston, MA:
Docent Press.
Rosskopf, M. F. (1970). Mathematics education: Historical perspectives. In M. Rosskopf
(Ed.), The teaching of secondary school mathematics: Thirty-third yearbook (pp. 3–29).
Washington, DC: National Council of Teachers of Mathematics.
Ruthven, K. (2008). Mathematical technologies as a vehicle for intuition and experiment: A

foundational theme of the International Commission on Mathematical Instruction, and a
continuing preoccupation. International Journal for the History of Mathematics
Education, 3(2), 91–102.
Schubring, G. (2011). On how n and i turned out to become indices in mathematical
sequences and formulae. Educational Studies in Mathematics, 77(2–3), 331–350.
Schultze, A. (1909). Graphic algebra. New York, NY: The Macmillan Company.
Simpson, T. (1755). A treatise of algebra (2nd ed.). London, UK: John Nourse.
Smith, D. E. (1905). Movements in mathematical teaching. School Science and Mathematics,
5(3), 205–209.
Smith, D. E. (1913). Laboratory methods in mathematics. In P. Monroe (Ed.), A cyclopaedia
of education (Vol. 4, pp. 159–160). New York, NY: Macmillan.
Snyder, T. D. (1993). 120 years of American education: A statistical portrait. National
Center for Educational Statistics, Washington, DC: U.S. Government Printing Office.
Snyder, T. D., & Dillow, S. A. (2011). Digest of education statistics 2010. Washington, DC:
National Center for Educational Statistics, U.S. Government Printing Office.
Spielhagen, F. R. (2011). The algebra solution to mathematics reform (completing the
equation). New York, NY: Teachers College Press.
Steen, L. A. (1992). Does everybody need to study algebra? Basic Education, 37(4), 9–13.
Sterry, C., & Sterry, J. (1790). The American youth: Being a new and complete course of
introductory mathematics, designed for the use of private students. Providence, RI:
Authors.
Stigler, J. W., & Hiebert, J. (1999). The teaching gap. New York, NY: The Free Press.
Struik, D. J. (1948). A concise history of school mathematics (2 vols.) New York, NY: Dover
Publications.
Swetz, F. (1992). Fifteenth and sixteenth century arithmetic texts: What can we learn from
them? Science & Education, 1, 365–378.
Thomas, I., & Andrews, E. J. (1809). Preface to the third octavo edition of A new and
complete edition of arithmetick, composed for the use of citizens of the United States:
Seventh edition for the use of schools. Boston: Author.
Turnbull, H. W. (Ed.). (1961). The correspondence of Isaac Newton, 1688–1694 (Vol. 3).
Cambridge, UK: Royal Society/Cambridge University Press.
Tzanakis, C. (1991). Is it possible to teach abstract algebraic structures in high school? A
historical approach. Euclides, 11(2), 24–34.
Ueno, K. (2006). From Wasan to Yozan: Comparison between mathematical education in the
Edo period and the one after the Meiji restoration. In F. K. S. Leung, K.-D. Graf, & F.
Lopez-Real (Eds.), Mathematics education in different cultural traditions—A
comparative study of East Asia and the West: The 13th ICMI Study (pp. 65–79). New
York, NY: Springer.
UNESCO (1973). New trends in teaching mathematics (Vol. 3). Paris, France: Author.
University of Melbourne (1855, March 24). Registered correspondence, W. E. Hearn to E.
G. Mayne (held in the University archives).
University of Melbourne (1900). Matriculation examination regulations (held in the
University archives).
Usiskin, Z. (1987). Why elementary algebra can, should, and must be an eighth-grade course
for average students. Mathematics Teacher, 80(6), 428–438.
Usiskin, Z. (1988). Conceptions of school algebra and uses of variables. In A. F. Coxford &
A. P. Schulte (Eds.), The ideas of algebra, K–12, 1988 yearbook (pp. 8–19). Reston,
VA: National Council of Teachers of Mathematics.
inequations in two government secondary schools in Thailand (Unpublished Ph.D.
good news and the bad news. In H. S. Dhindsa, I. J. Kyeleve, O. Chukwu, Lim Siew Bee,
Hjh Zaitun bte Hj Taha, A. Baimba, & S. Upex (Eds.), Shaping the future of science,
mathematics and technical education (pp. 212–221). Gadong, Brunei Darussalam:
Universiti Brunei Darussalam.
Vaiyavutjamai, P., Ellerton, N. F., & Clements, M. A. (2005). Students’ attempts to solve
two elementary quadratic equations: A study in three nations. In P. Clarkson, A.
Downton, D. Gronn, M. Horne, A. McDonagh, R. Pierce, & A. Roche (Eds.), Building
connections: Research, theory and practice (pp. 735–742). Sydney, Australia:
Mathematics Education Research Group of Australasia.
Venema, P. (1714). Een kort en klare onderwysinge in de beginselen van de algebra ofte
stel-konst: Waar in niet alleen de voornaamste quetien Van Willem Bartjens en Martin
Wilkens see sinlijk Verhandeld, maar ook t’effens byna alle de stelkonstige question uit
de algebra Van Abraham de Graaf: ales in een seer nette ordre en geschiktheid na
malkanderen zijn gesteld en opgelost. Amsterdam, The Netherlands: Author.
Vredenduin, P. G. J. (1962). Functions and relations. In H. Freudenthal (Ed.), Report on the
relations between arithmetic and algebra in mathematical education up to the age of 15
years (pp. 99–110). Groningen, The Netherlands: J. B. Wolters.
Webster, S. (1808). Mathematics, compiled from the best authors, and intended to be the
text-book of the course of private lectures on these sciences in the University of
Cambridge (2nd ed.). Cambridge, MA: William Hilliard.
Wentworth, G. A. (1881). Elements of algebra. Boston, MA: Ginn, Heath, & Co.
Wolfe, C. (1739). A treatise of algebra, with the application of it to a variety of problems in
arithmetic, to geometry, trigonometry and conic sections. London, UK: A. Bettesworth
and C. Hitch.
Wu, H. (2007, September 13). “Order of operations” and other oddities in school
mathematics. Retrieved from https://math.berkeley.edu/~wu/order5.pdf
Wu, H. (2011). Understanding numbers in elementary school mathematics. Providence, RI:
American Mathematical Society.
Young, J. W. A. (1914). The teaching of mathematics in the elementary and the secondary
school. New York, NY: Longmans, Green, and Co.
Chapter 3
Framing a Classroom Intervention Study in a Middle-School
Algebra Environment
Abstract: It has become a tradition in the field of mathematics education that before a
researcher outlines the research design for a study he or she should outline a theoretical
framework for the investigation which is about to be conducted. Then, after research
questions are stated, and the design of the study is described, the investigation takes place.
The data gathering, data analyses, and interpretation are guided by the theoretical framework
and conclusions are couched in terms of, and seen in the light of, the theoretical framework.
There are many mathematics education researchers who regard this theory-based process as
sacrosanct, as absolutely essential for high-quality research. In the first part of this chapter it
is argued that the traditional “theoretical-framework” process just described is flawed, that it
can result in important aspects of data being overlooked, and that it can lead to incorrect, or
inappropriate, conclusions being made. It is argued that the first thing that needs to be done
in a mathematics education research investigation is to identify, in clearly stated terms, the
problems for which solutions are to be sought. Having done that, historical frameworks—
which have only occasionally been taken seriously by mathematics education researchers—
should be provided. Then, having identified the problems and having provided a historical
framework, a design-research approach ought to be adopted whereby a theory, or parts of a
theory, or a combination of parts of different theories, are selected as most pertinent to the
problems which are to be solved. This chapter identifies three main problems: (a) “Why do
so many middle-school students experience difficulty in learning algebra?” (b) “What
theoretical positions might be likely to throw light on how that problem might be best
solved?” (c) “In the light of answers offered for (a) and (b), what are the specific research
questions for which answers will be sought in subsequent chapters of this book?”
Keywords: Charles Sanders Peirce, Design research, Historical frameworks for mathematics
education research, Johann Friedrich Herbart, Modeling in school algebra, Structure in
school algebra, Traditions in mathematics education research
Applying Principles of Design Research in Mathematics Education

Two decades ago, Erich Wittman (1998), a German mathematics educator, in
summarizing what he called the “core of mathematics education,” drew attention to the
following eight points which, he said, require attention:
• Analysis of mathematical thinking and of mathematical ways of thinking;
• Development of local theories (for example, on mathematizing, problem
solving, proof and practicing skills);
• Exploration of possible contents that focus on making them accessible to
learners;

60 Ch. 3: Classroom Intervention Study in a Middle School
• Critical examination and justification of contents in view of the general goals

of mathematics teaching;
• Research into the pre-requisites of learning and into the teaching/learning
processes;
• Development and evaluation of substantial teaching units, classes of teaching
units and curricula;
• Development of methods for planning, teaching, observing and analyzing
lessons; and
• Inclusion of the history of mathematics education. (Wittman, 1998, p. 88)
The title of Wittman’s paper in which this list appeared was “Mathematics Education as a
Design Science,” and for the investigation described in Chapters 4 through 8 of this book we
adopted a design-science approach. Although, at the time the investigation was being designed,
the authors did not focus on Wittman’s list of eight characteristics which, he argued, defined
the “core of mathematics education,” his eight characteristics would appear to be consistent
with our approach to the investigation which will be described. We were particularly pleased
that Wittman included the history of mathematics education as one of his core features.
What is Design Research in Mathematics Education?

Design research in mathematics education might be profitably thought of as having five
essential elements:
1. Identifying the main problems to be investigated, and their theoretical, practical,
and ethical dimensions;
2. Identifying and working with key persons who might help solve the problem(s);
3. Developing a research plan which provides for the gathering and rigorous analysis
of data;
4. Planning for, implementing, and evaluating, a “plan-act-observe-reflect” action-
research program.
5. Conducting, and reporting, the research.
Before we comment on how these five design elements applied to the investigation described
in Chapters 4 through 8, we make some further preliminary comments about design research.
Kelly, Lesh and Baek (2008), in their Handbook of Design Research Methods in
Education, maintained that design research in education “is directed at developing, testing,
implementing, and diffusing innovative practices to move the socially constructed forms of
teaching and learning from malfunction to function, or from function to excellence” (p. 3). That
statement fitted well with the main problem that we had identified for our investigation—why is
it the case that over the past three centuries, secondary-school students have found it so difficult
to learn algebra well. From our perspective, the idea that universal, culture-free laws (“grand
theories”) exist in mathematics education settings around the world is unsound (Thomas, 1997).
Education research in school mathematics ought to be sufficiently flexible to take account of
unique cultural components. That statement should not be seen as meaning to imply that
researchers should reject the need to identify theories, or parts of theories, that might be
particularly helpful as they try to solve problems that they have identified.
In commenting on four major design elements associated with the research investigation
we would emphasize that fuller commentary will be provided in the chapters which follow.
Here, the aim is to describe, briefly, how the term “design research” is interpreted.
Applying Principles of Design Research in Mathematics Education 61
Design element 1: Identify the main problems, and their theoretical, practical, and
ethical dimensions. Although we had identified the main problem we wished to solve (“Why
do so many middle-school and secondary-school students fail to learn algebra well?”), we also
needed to give careful consideration to the dimensions of that problem. Clearly, the problem
was universal—in the sense that it has been seen to be present in all nations—but we did not
have the resources or time needed to conduct a large-scale study across many nations. What we
decided to do was to plan, conduct and evaluate a study which, we hoped, might serve as a
lighthouse study—an investigation which would guide and illuminate the path for further
studies which, taken together, might solve our problem in greater detail than we could offer.
Because the investigation would involve middle-school students and teachers, the
normal ethics approvals would need to be obtained from district and school officials, from
parents or guardians of participating students, from participating teachers, and from relevant
university authorities. The design of the study which we wished to conduct was complex,
involving random allocation of students to different groups. It was important that all
participants in the study were willing to be involved. In the submissions for ethics clearances,
details were required with respect to who the key stakeholders in the research would be, and
how participants’ beliefs and values might be affected.
Design element 2: Identify and work with key players in the research process. This
design element called for us to reflect on, and make decisions about, who should be the major
participants in our proposed investigation, and who would be the key stakeholders. Our
responses to these personnel issues were constrained by the fact that the main study would be
conducted as a doctoral study, and described in the form of a doctoral dissertation. The first
author of this book would prepare his doctoral dissertation on the study (Kanbir, 2016a), and
the second and third authors would be joint-chairs of the dissertation committee. It was
regarded as an important feature of the design of the study that the participating teachers should
feel free to teach the lessons, which would be an important part of the intervention, in whatever
ways they deemed best (provided, of course, the main planned themes were covered).
It should be noted, here, that at the design stage there was no intention that details of
the study would be made into a book—the decision to do that came after the study had been
completed and defended, and the dissertation completed and approved.
Design element 3: Develop a research plan which provides for rigorous gathering
and analyses of data. From the outset the first author of this book determined to create a
design by which both formative and summative data in qualitative and quantitative forms
would be gathered and analyzed, at various key stages of the project. The design of the main
study was not finalized until after two large pilot studies had been completed (Kanbir, 2014,
2016b). Analyses of data from the pilot studies were taken into account in the design of the
main study and in the preparation of evaluation instruments for the main study.
Design element 4: Plan for, implement and evaluate a “plan-act-observe-reflect”
action-research mentality. As indicated in the notes on “design element 2,” the study which
is described in later chapters of this book was first activated in the form of a doctoral
dissertation by the first author of this book (Kanbir, 2016a). As such, the normal checks and
balances which pertain to the preparation and evaluation of most U.S. doctoral dissertations
applied. In particular, in addition to joint dissertation-chairs, there were two other academic
educators on the dissertation committee, and at the proposal stage the proposed research was
approved after a public presentation on the intended procedures and instruments. The design
of the study had to be approved by school authorities and by participating teachers at the school
where the study took place. After the study was completed, the final dissertation was approved
by the doctoral committee after it been defended at a public meeting which was attended by 25
scholars, and at which numerous questions were asked of, and answered by, the first author of
this book. Throughout the study, a plan-act-observe-reflect action research mentality was
adopted (Carr & Kemmis, 2004; Kelly & Lesh, 2000), with aspects of the plan being
progressively modified in light of data gathered from pencil-and-paper tests, interviews with
students, observations of workshops, and student reactions to material presented in class.
As Hjalmarson and Lesh (2008) noted, the development of the product and the
development of knowledge are intertwined throughout a design-research study. In the case of
the study described in later chapters of this book, student knowledge, in multiple forms, was
progressively developed, and monitored. That knowledge influenced later aspects of the
design process and contributed to the authors’ developing understanding of factors
influencing the teaching and learning of middle-school algebra. Design research asks for the
development, through iterative applications of local models, of strategies aimed at solving the
research problem(s). With school algebra, a single course of study, or textbook, or teaching
method, or examination, is not likely to be equally suitable across a wide range of cultures.
Design element 5: Conducting and reporting the research. This is to be done in
ways which are consistent with the above four design elements.
The Role of Theory in the Middle-School Algebra Investigation

Theory can play an important role in any education research investigation in which
design-research principles are being followed. Having identified the fundamental problem(s),
the question arises how that problem, or problems, can best be solved. It is at that point, in
particular, that the development of a theoretical frame can become important.
Probably the easiest way to recognize why theory can be important in a design-research
investigation is to consider an analogy. If, for example, architects and engineers faced the
challenge of designing and building a bridge which would span a large gap over a deep, and
wide, ravine whose sides were at very different heights, then careful collaboration between
experts would obviously be called for. Otherwise, serious consequences of poor design or poor
construction might need to be faced. All relevant architectural and engineering theories for
designing and building the bridge would need to be taken into account. The analogy can be
taken further—if, on one side of the ravine there was a jungle-type environment, through which
there were many narrow paths, often used by persons and animals, but on the other side there
were freeways on which cars and trucks regularly traveled at high speeds, then not only would
differences in physical conditions need to be considered but also differences in social practices
and cultural conditions. Because of the physical and cultural differences pertaining to life on
the two sides of the proposed bridge, serious thought might need to be given by those
designing the bridge to issues such as whether lanes for slow-moving “traffic” might be needed
across the bridge, and if so how many. Any differences in fundamental beliefs of those living
on either side of the bridge would need to be considered, too, particularly if those on one or
both sides did not want their way of life to be greatly affected by the new bridge.
No analogy is perfect, but it is interesting to reflect on the situation which arose when
seventh-graders were asked to participate in an algebra-intervention study in a small rural
school in midwest North America. Most of the seventh-graders had only minimal background
The Role of Theory 63
so far as formal algebraic thinking was concerned. The students might be considered to be on
one side of a divide. They were not familiar with elementary algebraic terminology,
symbolisms and methods of manipulating symbols, and were certainly not familiar with the
deeper mathematical objects which were in the forefront of the minds of some of those who
constructed the algebraic sequences in the common-core sequence (who might be thought of as
being on the other side of the divide). On the other side, too, were the curriculum designers, the
textbook authors, the test constructors and, perhaps, the teachers. To further complicate the
situation, the teachers knew that some of their students were much more ready than others to
deal with the cognitive challenges which were likely to arise in their seventh-grade algebra
studies. Given such a divide, what might an effective teaching strategy look like?
In algebra education, many existing theories might be regarded as “fuzzy.” Although,
much has been written about how middle-schools should be introduced to algebra, some of
what had been written may not have been relevant for the proposed main study because of
the circumstances surrounding that study. For example, the theories and writings of scholars
who have emphasized the need to introduce modeling tasks, involving “functional thinking,”
to middle-school students may not be directly applicable in a situation in which seventh-
graders were not familiar with terminologies and notations used in “explicit” and “recursive”
representations of sequences.
Lessons from History
The historical analysis presented in Chapter 2 resulted in six purposes for school algebra
being identified, and each was deemed to be relevant to our investigation. Of the six purposes,
only one—school algebra as a gatekeeper for those responsible for admitting students to higher
academic institutions—was “administrative” in nature. The other five had direct implications
for what was to be learned, how it would be learned, and effects that algebra education should
have on the direction of the mathematical development of participating middle-school students.
Closer examination of the five non-administrative purposes identified in Chapter 2
suggested that there is a common factor influencing students’ development in school algebra
knowledge, skills, and application. That common factor is learning to recognize, and use
expressively, the language of algebra. By “the language of algebra” we mean the
terminologies, symbolisms, imageries, and relationships in the forms of algebra to be
introduced. The kind of things to which we are alluding are summarized in Table 3.1. The
table makes it clear that in order to learn middle- and secondary-school algebra, students
need to learn not only to understand highly specific terminologies and symbolisms which
will be introduced by educationally significant other persons—e.g., textbook authors,
teachers, and persons who create externally-set tests of algebraic knowledge—but they will
also need to learn to create and use their own terminologies and symbolisms for specific
tasks.
In Chapter 1, we reported revealing data, from three nations, associated with
secondary-school students’ replies to the following task:
“The number y is eight times the number z.” Write this information in
mathematical symbols.” (MacGregor, 1991, p. 50)
Readers not steeped in the literature on how language factors affect mathematical learning
might find it difficult to imagine why students who had been attending algebra classes on a
regular basis for several years did not give an appropriate response to this “y = 8z” task.
Table 3.1
Messages of History on Language Aspects Associated with Five of the Purposes of School
Algebra
Assumed Comments on Language (Terminology and Symbols) Associated with
Purpose: School that Purpose in School Algebra
Algebra as …
A body of During secondary-school mathematics, algebra is used even in areas like
knowledge trigonometry, analytic geometry, and calculus. One cannot do well in higher
essential for higher studies in mathematics, the natural and biological sciences, information
mathematical and technology, or engineering unless one develops receptive and expressive
scientific studies. mastery of the symbolisms, terminologies, and graphical representations
increasingly used in school algebra.
Generalized Once students enter secondary-school classes they are progressively
arithmetic. confronted with the importance of natural numbers, integers, and rational, real
and complex numbers. Important aspects of these sets of numbers are
expressed algebraically, and even apparently simple concepts such as even
numbers and odd numbers are often expressed in algebraic terms. All students
need to learn how to use letters and equations to summarize important
formulae and relationships. The most elementary aspects of the terminologies,
symbolisms, and relationships are usually dealt with in middle-school and
lower-secondary-school algebra. Research has suggested that many students
struggle to learn the instrumental “rules” for elementary algebra and, more
seriously, are unable to recognize or develop patterns and generalizations, using
algebra.
A language for The “functional thinking” approach requires receptive and expressive
modeling real-life understanding of many forms of signification. The concept of a function is
problems. associated with specific terminology such as domain, range, rule, graph,
coordinate, table of values, and independent and dependent variables. Students
need to learn to construct and interpret tables of values associated with real-life
data, to draw Cartesian and other graphs, and to use sophisticated terminology
and symbolism. They need to link that terminology and symbolism to reality.
An aid for describing The structural-thinking approach to school algebra introduces unusual
basic structural terminology such as closed, commutative, associative, identity element,
properties. inverse elements, distributive property, and field. The field properties are
expressed using algebraic terminology which many middle-school students
find difficult to interpret, and use expressively. A statement like “if a, b
represent any real numbers then a + b = b + a” is easily understood by
mathematically-trained persons but its full meaning may not be understood by
middle-school students.
A study of The idea that if x represents any real number then the open sentence 3x + 1 =
variables. 3x – 1 is never true, but the inequality 3x + 1 > 3x – 1 is true for all real
numbers, is difficult for many middle-school students—because they believe
they have been taught that open sentences have one solution. More generally,
the idea that on a Cartesian plane all points on a graph will have coordinates
which make the rule true, but those not on the graph will have co-ordinates
which will make it false, is also something which is likely to confuse middle-
school students.
Language Aspects with Five Purposes of School Algebra 65
Any person doubting the confusion which secondary school students who have been
studying algebra for some years can have with the “y = 8z” task should read the analysis of the
performance of 231 ninth-grade Thai students on the Thai-language version of the task, and
also associated interview data, presented by Vaiyavutjamai (2004, 2006). In Vaiyavutjamai’s
study, the students were asked to respond to the task immediately before and after 13 lessons
on linear equations and inequalities, and then, once again, six months after the lessons. At the
pre-teaching stage, 23 percent of the 231 responded correctly; at the post-teaching stage, the
percentage increased to 27%; and at the retention stage, the percentage was 25%.
MacGregor’s (1991) task would appear to be similar to the following famous “students-
and-professors” problem (Clement, Lochhead, & Monk, 1981):
Write an equation using S and P to represent the following. There are six times as
many students as professors at this university. Use S for the number of students
and P for the number of professors.
Numerous scholarly papers have been published concerning the surprisingly high error rates
associated with the students-and-professors problem. Lochhead and Mestre (1988) attributed
much of the confusion associated with that problem to the order of key symbols used in the
statement of the problem. The word “six” is closer in the question to the word “students”
than it is to the word “professors,” and that might have led persons attempting the task to link
the 6 with the S, resulting in the “reversal,” P = 6S.
The same kind of faulty reasoning might explain why many students responded
incorrectly to MacGregor’s “y = 8z” task (for which the most common error was “z = 8y”).
From our perspective, however, the main lesson to be learned from data from the “y = 8z”
task is that the language of algebra is much more difficult to learn than many appreciate.
Many students, when asked to solve algebraic word problems, resort to dangerous key-word
or order-of-word approaches because they have become accustomed to adopting such
mechanistic approaches in response to mathematics word problems (Lean, Clements, & Del
Campo, 1990). They cannot understand the meanings of questions because they are
overwhelmed by the complexity of the language, and this explains their instrumental
reactions to text involving algebraic ideas. The “y = 8z” task represents the kind of language
one will often hear a teacher using in middle-school and lower-secondary algebra classes.
Technically, there is nothing wrong with that language—but the main educational issue is
this: it is semantically more complicated than might, at first sight, appear to be the case.
A most important caveat is that in many algebra classes it is often the case that students
do not realize that they do not understand what their teachers are saying, and often teachers
do not know that many of their students have not understood what they have just taught the
class, verbally, or what have they written on blackboards or on printed handouts. The root of
the problem is that proper links between the “signifier” (for example, a word problem) and
the mathematical object which the teacher intends to signify (i.e., the “signified”) are not
always easily established—and data associated with the “y = 8z” task suggests that hardly
anybody seems to be aware of the magnitude of the problem.
An answer to the fundamental problem that we have identified with middle-school
algebra (“Why do so many middle-school students experience difficulty in learning algebra
well?”) has emerged from our summary of the history of school algebra. Many students do not
easily learn to use, receptively and expressively, the language found in school algebra texts.
They literally do not understand what their teacher (or the textbook author) has attempted to
communicate to them. Another way of saying the same thing is that their knowledge of the
language of elementary algebra is not sufficiently developed for them to link algebraic
“signifiers” with the mathematical objects (the “signifieds”) which their teachers, or the authors
of their algebra textbooks, wish them to do. That is the conjecture we framed as we surveyed
the history of school algebra in nations across the world, and that conjecture is consistent with
entries in Table 2.1 and Table 3.1. The conjecture was tested in the study which will be
described in Chapters 4 through 8 of this book.
What has been stated in the last paragraph should make clear that in the study which
will be described in the following chapters we were guided by theories in the realm of
semiotics—an area of scholarship which focuses on issues to do with meaning-making, with
studying signs and processes which make for meaningful communications. The main theorist
from whose writings our approach to semiotics was adopted was Charles Sanders Peirce
(1839–1914), who is regarded as having pioneered many important ideas in modern
semiotics. Peirce was chosen as the main theorist for the study because his tripartite
emphasis, signifier → interpretant → signified, was deemed to be especially relevant to the
study which we would conduct. For Peirce (1992, 1998), a sign was something which
represented an object, or a practice, or an idea which someone was intending to communicate
to someone else. In that sense, there would be a relationship between the “sign vehicle” (or
the “signifier,” that is to say, the specific form of the sign) and the sign object (the
“signified,” that is to say the aspect of the world—physical object, practice, or idea—that the
sign was intending to communicate). The “interpretant” was the meaning of the signifier as
understood by an interpreter—which may or may not be the same thing as the signified.
More will be said in relation to Peirce’s theory of semiotics in later chapters of this book.
Bundling Theories to Achieve a Satisfactory Solution to a Problem

Those conducting the study described in later chapters of this book took Peirce’s
theory—of how signifiers, interpretants and signified are related, and the effects of instruction
in middle-school algebra on the relationships—as fundamentally important. However,
reflecting a design-research mode of thinking, Peirce’s theory was not the only theory deemed
to be relevant to the study. Rather, Peirce’s theory was one in a bundle of three main theories
that guided the study design, the data collection, and the interpretation of data. The two other
theories were (a) the theory of “apperception,” put forward by Johann Friedrich Herbart (1776–
1841), a German philosopher and educator, and by his followers, the so-called “Herbartians,”
in the second half of the nineteenth century (Dunkel, 1970); and (b) the receptive/expressive
language theory advocated by Del Campo and Clements (1987) in the late 1980s.
Herbart’s concept of apperception. The study described in this book set out to show
that seventh-grade students could learn school algebra well, in the sense that they could learn to
talk about it in mathematically-acceptable ways, be able to apply it to curriculum-appropriate
tasks, and be able to retain what they learned. From the outset it was recognized that the 32
seventh-grade participants in the study would, throughout the duration of the project, have 32
different cognitive structures with respect to the algebraic themes which would be
emphasized—those themes would be algebraic structures, and recursive and explicit
formulations of mathematical models. In order to guide the investigations into changes in the
participating students’ cognitive structures, it was decided to make use of the theory of
“apperception” developed and described by Johann Friedrich Herbart (1776–1841), a German
philosopher and educator whose theories were revived, and modified, across the world during
the period 1870–1910, when groups of scholars, known as “Herbartians” became influential in
Bundling Theories 67
various nations (Dunkel, 1970). In fact, there were differences between Herbart’s (1904a,
1904b) educational ideas and those of the Herbartians, and it was to the latter that we looked
especially for guidance in our intervention study.
Ellerton and Clements (2005) have argued that within Herbartianism lay the seeds of
many of the twentieth-century’s major mathematics education theories. Herbart (1904a,
1904b) emphasized the need to recognize that with respect to any topic to be learned in any
education setting, the students come into the learning environment with qualitatively
different cognitive structures. It is the teacher’s task to design and engineer maximally-rich
learning environments—that, for Herbart, provided the main challenge for teachers.
The Herbartians’ solution (see Dunkel, 1970) was to recommend that lesson plans
should feature the following “five steps”:
1. Specify clearly what it was desired the children would learn (“concentration”).
2. Prepare lessons which take account of the students’ prior knowledge and
differences (“presentation”).
3. Aim at getting students to reflect on how the material was related to what they
already knew (“association”).
4. Aim at getting students to reflect on how the material which has been learned might
be generalized to wider areas of knowledge (“generalization”).
5. Help the students to apply the major ideas that have been learned (“application”).
More will be said in later chapters of this book about Herbart and the Herbartians, and about
how Herbartian ideas can be found within many modern theories used in mathematics
education (Ellerton & Clements, 2005). Here it suffices to note that not all the major aspects
of Herbartianist theory were taken into account in the study which will be described—for
more details on modern interpretations of, and the modern-day revival of interest in,
Herbart’s writings, see Suzuki (2012). Only those parts of the theory deemed to be especially
relevant to the design and implementation of the intervention and to an evaluation of its
effects on students’ thinking, and especially their knowledge of important algebraic concepts,
skills and principles, were considered. In the investigation, an attempt was made to identify
key elements of the cognitive structures with respect to school algebra for each of the 32
participating students, at different phases of the intervention. In other words, changes in the
students’ cognitive structures were monitored, and the effects of changes on the students’
algebraic thinking were evaluated.
Like Peirce, Herbart emphasized the importance of language—both the language used
by the teacher and the language developed by the students—on learning (Herbart, 1904a,
1904b). For Herbart it was important that instruction should be directed at helping students to
“welcome” new ideas into their “souls”—that is to say, into the student’s whole circle of
thinking. This was to be achieved by linking the new ideas with “every point, forward,
backwards, sideways” (Herbart, 1904a, pp. 55–56). What was of special interest was how
incoming ideas interacted with aspects of a student’s existing cognitive structures, and how
that interaction affected that student’s future knowledge, concepts and principles.
The receptive/expressive emphasis in lesson planning. The third component of our
bundle of part-theories was concerned with whether middle-school and lower-secondary
algebra students are regularly given the opportunity not only to read clear descriptions of the
content that they will be expected to learn, or to hear expositions from their teachers, but also
to engage in guided discussions with others about the content. In other words, this third
component provided guidelines on how students can be engaged in both receptive and
expressive communications directly concerned with the algebra under consideration.
The theory chosen here was that put forward by Del Campo and Clements (1987), in
their Manual for the Professional Development of Teachers of Beginning Mathematicians.
Del Campo and Clements researched the effects of “expanding the modes of communication”
in mathematics classrooms by consciously providing students with opportunities to be
actively involved in both receptive and expressive modes of learning. Not only should the
students read what others had written (receptive), they should also be given the opportunity
to write, and to speak, using their own words, about the mathematics that they have been
studying (expressive). Not only should they hear their teachers talking about mathematics
(receptive) but they should also engage in small-group discussions about it, and make verbal
reports to the whole class (expressive). Not only should they view and copy diagrams drawn
by others (receptive), but they should also be given the opportunity to create their own
diagrams and communicate them to others in their class (expressive). They should not only
observe their teacher and other students “acting out” aspects of mathematics (receptive), but
they should also create their own gestures and actions in their attempts to describe the
mathematics that they are learning (expressive). They should also be encouraged to visualize.
They should be invited to create visual images and to create and pose problems as well as to
solve problems created by others (expressive). The conjecture that this combination of
receptive and expressive modes of communication in mathematics classrooms would assist
student learning was supported by the results of research (Clements & Del Campo, 1987; Del
Campo & Clements, 1990). There are many other theories on classroom discourse and on
argumentation (e.g., Toulmin, 1969), but for this study the Del Campo and Clements theory
seemed to be most relevant for a study in which the aim was to engage students not only in
receptive learning, but also actively in expressive learning modes.
More will be said in the following chapters about the hybrid nature of the theoretical
base employed for the intervention. It is emphasized, here, though, that there was no attempt
to take account of, and use, all of the components of the complicated theories propounded by
Peirce, Herbart and the Herbartians, and Del Campo and Clements. Rather, the idea was to
form a bundle of part-theories which, when combined and applied, would be likely to yield
the best possible outcome for the intervention which would be conducted. It is also worth
saying, that in our own thinking about how the part-theories would be combined we never
felt the need to use the kind of complex language, often used by persons discussing semiotic
theories, or Herbartian theory, or classroom-interaction theory. With the study described in
this book we aimed at keeping our descriptions of what we did, and what other study
participants did, as simple and as direct as possible.
One final comment will be in order. In this chapter we have drawn attention to the fact
that the key theories used for the study were proposed by Peirce, Herbart, and Del Campo and
Clements, quite a long time ago. We could have chosen any of a large number of other
“competing,” more “modern,” theories, but we chose the three theories because we believed
they were the most pertinent. There is a need for a scholarly book on the history of theories,
and theory development, in the field of mathematics education, and hopefully this chapter
might have drawn attention to that need.
References of Chapter 3 69
References
Carr, W., & Kemmis, S. (2004). Education, knowledge, and action research. London, UK:
Routledge.
Clement, J., Lochhead, J., & Monk, G. S. (1981). Translation difficulties in learning
mathematics. American Mathematical Monthly, 88, 286–290.
Clements, M. A., & Del Campo, G. (1987). Fractional understanding of fractions: Variations
in children’s understanding of fractional concepts across embodiments, Grades 2 through
5. In J. Novak (Ed.), Proceedings of the Second International Seminar on
Misconceptions and Educational Strategies in Science and Mathematics (Vol. 3, pp. 98–
110). Ithaca, NY: Cornell University.
teachers of beginning mathematicians. Melbourne, Australia: Catholic Education
Office of Victoria/Association of Independent Schools of Victoria.
Del Campo, G., & Clements, M. A. (1990). Expanding the modes of communication in
mathematics classrooms. Journal für Mathematik-Didaktik, 11(1), 45–99.
Dunkel, H. B. (1970). Herbart and Herbartianism: An educational ghost story. Chicago, IL:
University of Chicago Press.
Ellerton, N. F., & Clements, M. A. (2005). A mathematics education ghost story:
Herbartianism and school mathematics. In P. Clarkson, A. Downton, D. Gronn, M.
Horne, A. McDonagh, R. Pierce, & A. Roche (Eds.), Building connections: Research,
theory and practice (pp. 313–320). Sydney, Australia: Mathematics Education
Research Group of Australasia
Herbart, J. F. (1904a). Outlines of educational doctrine. New York, NY: Macmillan.
Herbart, J. F. (1904b). The science of education. London, UK: Sonnenschein.
education research and design. In A. E. Kelly, R. A. Lesh, & J. Y. Baek (Eds.),
Handbook of design research methods in education: Innovations in science, technology,
engineering and mathematics learning and teaching (pp. 96–110). New York, NY:
Routledge.
Kanbir, S. (2016a). An intervention study aimed at enhancing seventh-grade students’
Kanbir, S. (2016b, April 12). Three different approaches to middle-school algebra: Results
of a pilot study. Paper presented at the 2016 Research Conference of the National Council
methods in education: Innovations in science, technology, engineering and mathematics
learning and teaching. New York, NY: Routledge.
Lean, G. A., Clements, M. A., & Del Campo, G. (1990). Linguistic and pedagogical factors
affecting children's understanding of arithmetic word problems. Educational Studies in
Mathematics, 21(2), 165–191.
Lochhead, J., & Mestre, J. P. (1988). From words to algebra: Mending misconceptions. In A.
F. Coxford & A. P. Schulte (Eds.), The ideas of algebra, K–12, 1988 yearbook (pp.
127–135). Reston, VA: National Council of Teachers of Mathematics.
McGregor, M. E. (1991). Making sense of algebra: Cognitive processes influencing
Peirce, C. S. (1992). The essential Peirce: Selected philosophical writings (Vol. 1, 1867–
1893). Bloomington, IN: Indiana University Press.
Peirce, C. S. (1998). The essential Peirce (Vol. 2). Bloomington, IN: Indiana University
Press.
Suzuki, S. (2012). The Kyoto School and J. F. Herbart. In P. Standish & N. Saito (Eds.),
Education and the Kyoto School of philosophy: Pedagogy for human transformation
(pp. 41–54). Dordrecht, The Netherlands: Springer.
Thomas, G. (1997). What’s the use of theory? Harvard Educational Review, 67(1), 75–104.
Toulmin, S. (1969). The uses of argument. Cambridge, UK: Cambridge University Press.
dissertation). Universiti Brunei Darussalam, Brunei Darussalam.
Wittman, E. (1998). Mathematics education as a “design science.” In A. Sierpinska & J.
Kilpatrick (Eds.), Mathematics education as a research domain: A search for identity
(pp. 87–103). Dordrecht, The Netherlands: Kluwer Academic Publishers.
Chapter 4
Document Analysis: The Intended CCSSM Elementary- and
Middle-School Algebra Sequence
Abstract: Having identified the main problem (“Why do so many school students find it
difficult to learn school algebra well?”), and having made decisions on historical and
theoretical frameworks for the study it was important that the intended algebra content, as
summarized by the common-core mathematics sequence and by the algebra content in
textbooks which had previously been used by participating students, and in the textbooks
being used in the seventh grade by the students, be identified and analyzed. The ensuing
document analyses, presented in this chapter, revealed that the seventh-grade students might
have been expected to know the associative and distributive properties for rational numbers
and, given tables of values, they might have been expected to be able to identify and
summarize, mathematically, the rules for uncomplicated linear sequences.
Keywords: Common-core algebra, Intended curriculum, Middle-school algebra, Middle-

school algebra textbooks
Before planning the investigation which will be described in this, and later chapters (and
which, hereafter, will usually be referred to as the “main study”), the first author conducted
two pilot studies with seventh- and eighth-grade students at a school in a rural city in a
midwestern state of the United States of America (Kanbir, 2014, 2016a). In both pilot studies
the first author investigated students’ knowledge and understandings of algebra content
which, according to the school’s syllabus documents, the students might have been expected
to know.
Each of the three authors interviewed students who participated in the pilot studies, and
during interviews we quickly reached the conclusion that the students knew very little about
algebra. Hardly any of the interviewees knew the meanings of common signs used in those
sections in their elementary algebra texts which had already been dealt with in class. Most of
them did not know what expressions like “the associative property for multiplication” or “the
distributive property” meant, and they struggled to identify and to describe explicit rules
which might apply to even relatively simple numerical patterns such as 1, 4, 7, 10, 13, …
The first author’s analyses of the pilot-study data prompted him to carry out the main
study—but at a different school from the one where the pilot studies took place. Of special
interest was whether the lack of knowledge of elementary algebraic notations, skills and
principles among seventh- and eighth-grade students in the pilot-study school would also be
found among seventh-graders in another school.
Approaches to Introducing Algebra: Structure and Modeling

Researchers (e.g., see, e.g., numerous chapters in Cai & Knuth, 2011a; also Kieran,
2006, 2007) have drawn attention to the following three forms of algebraic reasoning:

72 Ch. 4: Document Analysis: The Middle-School Algebra Curriculum
1. The recognition, formalization, and use of structures within sets of numbers. This can
involve reasoning about operations and structural properties with respect to sets of
numbers—like, for example, reflecting on whether the associative or commutative
properties hold for different operations when they are applied to different sets of
numbers.
2. Generalizing numerical and visual patterns to describe relationships between
variables. This involves exploring and expressing regularities in, for example,
growth patterns concerned with the sum of odd natural numbers.
3. Modeling real-life situations by developing appropriate equations and inequalities.
In the transition from arithmetic to algebra, students are introduced to a new language for
expressing numerical structures, for representing patterns, and for modeling real-life
situations. According to the common-core mathematics sequence (National Governors
Association Center for Best Practices, & Council of Chief State School Officers, 2010—
hereafter cited as “CCSSM, 2010”), students should be introduced to formal symbolic algebra
in the sixth grade, and in grades 7 through 10 they should use this symbolism when exploring
school-algebra concepts. The present study explored what had been laid down as the intended
algebra curriculum (Westbury, 1980) for the elementary- and middle-school classes of the
participating students. Analyses and interpretation of data presented in Chapters 5 through 8
will provide commentary on what proved to be successful, and not successful, so far as the
intervention with a cohort of students at one school, for the main study, was concerned.
Overview of the Main Study

In his doctoral dissertation, the first author of this book (Kanbir, 2016b) described the
“main study,” which was an investigation into the developing understandings of algebraic
knowledge, skills and concepts of 32 seventh-grade students at a rural public middle school in
a midwestern state of the United States of America—hereafter, the school will be referred to
as “School W.” In particular, a teaching intervention was described in which most of the
seventh-grade students at School W participated in a series of workshops in which they
explored elementary algebraic structures and applied algebraic concepts in real-world
modeling contexts.
The participating students’ growth in their knowledge, understanding and use of
algebraic concepts associated with the terms “associative property of addition,” “associative
property of multiplication,” “distributive property,” “recursive rule,” and “explicit rule,” was
investigated. Two teachers from School W also participated in the study—and henceforth
these teachers will be referred to as “Mr. X” and “Mr. Y.” Both were well-qualified and
experienced middle-school mathematics teachers. Further details relating to the design of the
main study will be given in Chapter 6 of this book.
Among other things, the main study was designed to investigate:
• The seventh-grade students’ knowledge and understanding of structural properties of
the set of rational numbers, especially of associative properties (for addition and
multiplication), and the distributive property for multiplication over addition, before
and after they were reminded of those properties in the seventh-grade workshops.
• The extent to which, at the pre- and post-intervention stages (that is to say, before
and after participating in the intervention workshops), the seventh-grade students
Algebra for Structure and Modeling: CCSSM 73
recognized when it might be useful to apply the above-mentioned structural

properties of rational numbers when making calculations.
• The extent to which the students were able, at the pre- and post-intervention stages,
to use recursive and explicit algebraic language and concepts to describe patterns
associated with linear sequences.
• The extent to which the students learned, during the intervention, to use elementary
ideas concerned with the concept of a variable so that they were enabled to identify
and formulate patterns which model relationships between variables, and to apply
elementary mathematics to solve relevant visual and non-visual problems.
In later chapters of this book, details of the design, implementation, and results of the
intervention will be provided. It will be sufficient here to state, though, that in the main study
Mr. X and Mr. Y led workshops in which structural and modeling approaches were
introduced and consolidated. Almost all of the seventh-grade students at School W were
active participants in those workshops.
Concepts of Algebraic Structure, and Modeling, as Presented in the Common-Core

State Standards for Mathematics
The mathematics component of the common-core sequence (CCSSM, 2010)
emphasized the importance of developing algebraic reasoning in the elementary and middle
grades. CCSSM (2010) defined proficiency standards in two major strands: mathematical
content and mathematical practice. The content standards were organized by grade (until
eighth-grade) and comprised a list of facts, skills, symbols, procedures, relationships, and so
on, which students were expected to know, understand and apply.
CCSSM (2010) also described eight “standards for mathematical practice” (hereafter
“SMP”). These focused on problem solving, reasoning, constructing argument, modeling,
using tools, attending to precision, looking for structure, and expressing regularity in repeated
reasoning. The practice standards emphasized the need for students to develop deep
understandings of the relevant concepts and procedures. Gavin and Sheffield (2015)
suggested that the SMP should ask teachers to teach in ways most likely to enable their
students to develop their thinking in ways which resembled how mathematicians think.
Structural Thinking in Algebra, Across the Grades

Those who prepared the K–8 common-core standards for mathematics regarded the
“field properties” for rational and real numbers as so important that they listed them in a
“glossary” (CCSSM, 2010, Table 3). However, CCSSM did not provide specific details on
how the properties should be taught, or on what it meant to understand them.
The distributive property of multiplication over addition was one of the structural
properties emphasized in the common-core. That property becomes a powerful algebraic tool
when one considers an expression like “8x + 4x.” By the distributive property, 8x + 4x equals
x(8 + 4) or x12, or 12x. Indeed, it is the distributive property which is behind (or justifies) the
popular language, used in many school algebra texts, which insists that although one can add
or subtract “like terms,” one should not attempt to simplify the addition or subtraction of
“unlike terms.” Hence, 8x + 5x equals 13x, but 8x + 5 cannot be further simplified. And, one
should not say that 8x + 4x2 always equals 12x2. The same distributive property, but thought
of in the “reverse direction,” can be called upon to justify statements like 5(2x – 1) = 10x – 5.
This study specifically examined the extent to which seventh-grade students learned to
notice, to state, and to apply the distributive property and the associative properties of
addition and multiplication for rational numbers. Because the students had not yet studied
irrational numbers—other than to become aware of the existence of pi (π) and square roots—
it was decided, at the outset, that the emphasis would be on the distributive property and the
associative properties of addition and multiplication for rational numbers.
Before outlining introductory algebra topics (many of which are introduced in its notes
for sixth and seventh grades), CCSSM (2010) formalized, for grades K–5, the subdomain of
“Operation and Algebraic Thinking” within its “Number Strands.” The interconnections
between numbers (arithmetic) and symbols (algebra) were important features of CCSSM, and
elementary and middle-school teachers were urged to encourage their students to pay special
attention to the meaning of the “equals sign,” and to terms such as “variable” and “graph.”
Algebraic Structure as it Appears in the Common-Core State Standards for

Mathematics
Table 3 in the “mathematics glossary” of the K–8 common-core State Standards listed
the following nine “properties of operations”—specifically, the associative and commutative
properties for addition and multiplication, the additive identity property of 0, existence of
additive inverses, the multiplicative identity property of 1, existence of multiplicative
inverses, and the distributive property of multiplication over addition. These properties are
formally stated in terms of “arbitrary numbers” a, b, and c.
An important aspect of any curricular planning aimed at assisting children to become
familiar with structural properties with respect to numbers is the fact that the properties are
often summarized using algebraic notations. Thus, for example, the associative property for
addition often appears as “(a + b) + c = a + (b + c).” What is not emphasized in CCSSM, or in
associated textbooks, is that in such a statement, a, b, and c are being used as variables in the
sense that (a + b) + c will equal a + (b + c) no matter which numerical values are assigned to
a, b, and c—provided, of course, if a has a particular value in (a + b) + c then a must have the
same value in a + (b + c), etc. For mature mathematicians, or for inexperienced but otherwise
well-qualified middle-school mathematics teachers, the meaning of that proviso will be clear,
but for seventh-grade students a correct interpretation of the complex sign “for arbitrary real
numbers, a, b, c, (a + b) + c will always equal a + (b + c)” may not be obvious.
Part of the problem, from an educational perspective, lies in the meaning many children
give to the word “arbitrary” in the statement “For arbitrary numbers a, b, c in a given number
system, (a + b) + c = a + (b + c).” A seventh-grade child might think: “‘Arbitrary’ means I
can use ‘any numbers.’ I’ll let a = 7, b = 2, and c = 4 on the left side, and a = 5, b = 3, and c =
10 on the right side. … But … that wouldn’t make a true statement. The rule’s wrong!”
After examining the CCSSM (2010) document, the researchers decided that it would not
be wise to attempt to teach the Grade 7 students who participated in the current study all of
the field properties during the course of the investigation. An understanding of the
mathematical poignancy of those properties is something to be built up carefully over years,
and in this study it was decided to pay special attention to just three of the properties—
namely, the associative property for addition, the associative property for multiplication, and
the distributive property (by which multiplication is distributed over addition or subtraction).
The wisdom of the decision to limit the scope of the investigation in that way should become
Algebraic Structure in CCSSM 75
apparent in the light of the following examination of what the CCSSM document implies in
relation to those three chosen properties.
The Associative Properties for Addition and Multiplication, and the Distributive
Property, in Common-Core Mathematics
Grade 1. The associative property is first specifically mentioned in first-grade CCSSM
mathematics content, when the following terminology is used:
Grade 1, CCSS.
Math Content.1.OA.B.3
Apply properties of operations as strategies to add and subtract. Examples: If 8 + 3
= 11 is known, then 3 + 8 = 11 is also known (commutative property of addition).
To find 2 + 6 + 4, the second two numbers can be added to make a ten, so 2 + 6 + 4
= 2 + 10 = 12 (associative property of addition). (CCSSM, 2010, p. 15)
The form of words used suggests that first-grade teachers might be expected to help their
students realize that in order to calculate 2 + 6 + 4 it might be better to add the 6 and the 4
first, rather than add the 2 and the 6 first, because 6 + 4 equals 10, and young learners get to
know about properties of 10 early. But young children are being taught to read printed text
from left to right, and therefore it is unlikely that many of them would think of adding the 6
and the 4 first. Whether it would be wise to expect first-graders to carry out such an analysis,
and the role of a teacher in facilitating that, needs to be determined by research.
Grade 3. In their “glossary,” those who developed the CCSSM guidelines referred to
the “associative property of addition,” the “associative property of multiplication,” and to the
“distributive property of multiplication over addition.” In a section on “Operations and
Algebraic Thinking” for Grade 3, the common-core document stated:
Apply properties of operations as strategies to multiply and divide. Examples: If
6 × 4 = 24 is known, then 4 × 6 = 24 is also known (commutative property of
multiplication). 3 × 5 × 2 can be found by 3 × 5 = 15, then 15 × 2 = 30, or by 5 × 2
= 10, then 3 × 10 = 30 (associative property of multiplication). Knowing that 8 × 5
= 40 and 8 × 2 = 16, one can find 8 × 7 as 8 × (5 + 2) = (8 × 5) + (8 × 2) = 40 + 16
= 56 (distributive property). (CCSSM, 2010, p. 23)
In CCSSM’s Grade 3 measurement section, one finds the intriguing expectation: “Use tiling
to show in a concrete case that the area of a rectangle with whole-number side lengths a and
b + c is the sum of a × b and a × c. Use area models to represent the distributive property in
mathematical reasoning” (CCSSM, 2010, p. 25).
Grade 4. For the fourth grade, CCSSM (2010) advised that letters could be used to
represent unknown quantities. The specific statement in the CCSSM document was: “Solve
two-step word problems using the four operations. Represent these problems using equations
with a letter standing for the unknown quantity” (p. 23). In CCSSM’s section on fourth-grade
“Operations & Algebraic Thinking,” teachers were advised to help students to use drawings,
equations and letter symbols when representing problems. Also, students were to learn to
justify standard short multiplication and long multiplication algorithms by referring to
distributive and associative properties, and the sense of these algorithms was to be illuminated
by using equations, rectangular arrays, and/or area models.
Grade 5. The section on “Operations & Algebraic Thinking” stated that fifth-grade
students were to “write and interpret numerical expressions” without necessarily evaluating
them. According to CCSSM, students might profitably express the calculation “add 8 and 7,
then multiply by 2” as “2 × (8 + 7),” and be asked to “recognize that 3 × (18932 + 921) is
three times as large as 18932 + 921, without having to calculate the indicated sum or product”
(p. 35). It was expected that the distributive property be used, and understood, by the end of
fifth grade.
In the Measurement section of CCSSM’s (2010) Grade 5 sequence, students were
expected to learn to find the volume of a right rectangular prism with whole-number side
lengths by packing it with unit cubes, and then to show that the volume was the same as
would be found by multiplying the edge lengths, or equivalently by multiplying the height by
the area of the base. Finally, and most pertinent to thinking about structure, the students were
to learn to “represent threefold whole-number products as volumes, e.g., to represent the
associative property of multiplication” (p. 37).
An interesting side-comment on the idea that the associative property of multiplication
can be represented by referring to the volume of a rectangular prism (that is to say, a cuboid)
is that it assumes that by the fifth grade, students are able to “conserve volume with solid
objects.” In other words, the teacher should recognize that (a × b) × c represents the volume
measure of the cuboid where the base is regarded as a rectangle with side length measures a
and b, and the expression a × (b × c) represents the volume measure of the cuboid where the
base is regarded as a rectangle with side length measures b and c, and be able to reason that
because it is the same cuboid, (a × b) × c must equal a × (b × c). Jean Piaget indicated that a
significant proportion of fifth-grade students are not able to conserve volume with solid
objects, and that conclusion has been supported by recent research (Twidle, 2006). If that is
indeed the case, then the suggested use of the volume measure of a cuboid to support, for
fifth-graders, the associative property for multiplication may lead to confusion.
Grade 6. In the sixth grade, students were expected to make noticeable advances in
algebra. Thus, for example, they should learn to describe 2(8 + 7) as a product of two factors
and view (8 + 7) as both a single entity and as a sum of two terms. They should also be able to
apply the distributive property to expressions such as 3(2 + x), to produce the equivalent
expression 6 + 3x and, in the reverse direction, for expressions such as 24x + 18y, to produce
the equivalent expression 6(4x + 3y). They should also learn to apply properties of operations
to y + y + y to produce the equivalent expression 3y.
Details in the last paragraph make clear the CCSSM expectation that, by the end of the
sixth grade, students should be able to use the distributive property for both “expanding
parentheses” and for factoring (using the “common factor”). Furthermore, the apparently
simple y + y + y = 3y is really to be justified from the distributive property (because in “y + y
+ y” the y is common and, therefore, by the distributive property y + y + y = y(1 + 1 +1) =
y × 3 or 3y—if one assumes the commutative property for multiplication, and knows that 3y
is a shorthand for 3 × y).
Grade 7. By this stage, students would be expected “to apply properties of operations as
strategies to add, subtract, factor, and expand linear expressions with rational coefficients”
(CCSSM, 2010, p. 49); and, hence, the distributive property was to be used to justify
assertions such as “one can add and subtract like terms, but not unlike terms.” For example,
3(2x – 1/3) + 5x + 7 can be written as 6x – 1 + 5x + 7 (having applied the distributive property
and a multiplicative inverse property), which can be written as 11x + 6 (having applied
Associative and Distributive Properties: CCSSM 77
commutative, associative, and distributive properties). But, 11x + 6 must not be written as 17x
because there is no common factor.
Also, the Grade 7 CCSSM section on “Expressions & Equations” stated that “students
are to learn to use properties of operations to generate equivalent expressions,” and are to
“understand that rewriting an expression in different forms in a problem context can shed
light on the problem and on how the quantities in it are related.” The following example was
given: “a + 0.05a = 1.05a” means that “increase by 5%” is the same as “multiply by 1.05.”
Part of the main study was concerned with the extent to which the participating seventh-
grade students’ succeeded in developing knowledge and understanding of the associative and
distributive properties for rational numbers, and the extent to which this helped them not only
to formalize the concept of a variable, but also to develop a better understanding of what is
traditionally regarded as elementary algebra. Lacking from the CCSSM’s statement is
reference to research confirming that the content which was suggested for different grade
levels was something that most children who are in those grade levels could manage.
Research findings based on, say, hypothetical learning trajectories developed by interviewing
small numbers of carefully selected students cannot be regarded as coming close to proving
that the hypothesized learning trajectories are suitable for children at different grade levels.
The existing algebra-education literature has not paid much attention to how structural
aspects of the set of real numbers involve the concept of a variable. But, for example, the
distributive property asserts that if a, b, and c represent arbitrary real numbers then a × (b + c)
will equal (a × b) + (a × c). One can vary the values of a, b, and c as much as one likes, but if
a, b, c represent real numbers then this relationship will always be true. Similarly, a + (b + c)
will equal (a + b) + c and a × (b × c) will equal (a × b) × c. However, a – (b – c) will not
usually be equal to (a – b) – c; and it will not generally be true that a ÷ (b ÷ c) will equal
(a ÷ b) ÷ c. Such statements draw attention to links between structural number properties and
the concept of variable.
There is evidence that despite CCSSM’s (2010) strong emphasis on field properties, and
on the distributive property in particular, some recent researchers have preferred to train
teachers to engage school students in a “separate-multiply-add” (SMA) “strategy” for
multidigit multiplication. Thus, for example, in a 38-page paper, Whitacre and Nickerson
(2016) stated, in a footnote, that it was “important to note the distinction between strategies
and properties,” and that “SMA is distinct from the distributive property” (Whitacre &
Nickerson, 2016, p. 283). That was the only time the term “distributive property” was
mentioned in the paper.
In the main study there was a series of six workshops dedicated to algebraic structure
which were expected to occupy a total of six weeks with the seventh-graders. The aim for
those workshops was to involve all of the students actively and expressively in the learning
process so that they would achieve an understanding not only of the “variables” aspect of
statements such as (a + b) + c = a + (b + c), (a × b) × c = a × (b × c), and a × (b + c) = a × b
+ a × c, but also of how those properties are vitally important in the development of standard
ways of operating in elementary algebra (for solving equations and inequalities, and for
creating equivalent algebraic expressions). They also provided justification for “clever” ways
of doing “mental arithmetic” calculations (such as finding the value, mentally, of 803 + 798,
or finding the value of 25 × (4 × 19), or finding the cost of 11 pens at $1.05 each). The
workshop notes for the structure workshop are reproduced as Appendix E to this book.
Modeling Across the Grades in CCSSM Algebra

Three seventh-grade classes participated in the second pilot study—one of the classes
had lessons which focused on visual-modeling tasks, another had lessons in which students
solved non-visual, modeling problems, and the third had lessons on structural aspects of
rational numbers (Kanbir, 2016a). The main study responded to the CCSSM expectation that
middle-school students should use algebraic thinking in both structure and modeling contexts.
Because of time limitations, and of implications arising from our analyses of pilot-study data,
the modeling component incorporated both visual and non-visual aspects of modeling,
particularly those associated with patterns associated with linear sequences.
One of the challenges of the workshops on modeling was to get the students to the stage
where they could understand the idea of an nth term in a sequence, where n could represent an
unspecified natural number (and, sometimes, zero as well). It was hoped that the study would
generate findings which would contribute to the literature concerning the growth of middle-
school students’ abilities to generalize. Terms such as “recursive rule” and “explicit rule”
would be employed in the modeling workshops, and the subscript sequence notation, tn, would
also be introduced and used. It was a matter of some interest to the researchers whether the
Grade 7 students would learn to cope with the subscript notation (which is commonly
employed by mathematicians in the specification of sequences, and is expected of Algebra 1
students in some middle schools and in high schools).
Grade 3. In the “Operations & Algebraic Thinking” section of the common-core
document it was stated that students should learn to “solve problems involving the four
operations, and identify and explain patterns in arithmetic” (CCSSM, 2010, p. 23). Notice the
specific use of the term “patterns” in that statement. Furthermore, Grade 3 students were
expected to learn to use letters to represent unknown quantities when attempting to solve two-
step word problems.
Grade 4. The following statement in a section in the common-core document (CCSSM,
2010) on “Generate and analyze patterns” makes it clear that fourth-grade students were to be
expected to learn to describe patterns both explicitly and recursively:
Generate a number or shape pattern that follows a given rule. Identify apparent
features of the pattern that were not explicit in the rule itself. For example, given
the rule “Add 3” and the starting number 1, generate terms in the resulting sequence
and observe that the terms appear to alternate between odd and even numbers.
Explain informally why the numbers will continue to alternate in this way. (p. 29)
Grade 5. Fifth-grade students were to “analyze patterns and relationships.” They were
to generate numerical patterns from two given rules, to identify apparent relationships
between corresponding terms, to form ordered pairs consisting of corresponding terms from
the two patterns, and graph the ordered pairs on a coordinate plane. For example, they might
be given the rule “Add 3” and the starting number 0, and also the rule “Add 6” and the
starting number 0, and be asked to generate terms in the resulting sequences, and observe that
the terms in one sequence are twice the corresponding terms in the other sequence. They
could be asked to explain informally why this was the case (CCSSM, 2010, p. 35).
Grade 6. The terms “dependent variable” and “independent variable” were to be
introduced to sixth-graders, who were to learn to use them to represent and analyze
quantitative relationships between variables (CCSSM, 2010). Students were to “use variables
Modeling Across the Grades in CCSSM Algebra 79
to represent two quantities in a real-world problem that change in relationship to one another;
write an equation to express one quantity, thought of as the dependent variable, in terms of the
other quantity, thought of as the independent variable” (p. 44). They were to be invited to
investigate relationships between dependent and independent variables using graphs and
tables, and to relate these to the equation. For example, in a problem involving motion at
constant speed, they would be expected to list and graph ordered pairs of distances and times,
and to write the equation d = 65t to represent the relationship between distance and time.
An interesting reflection on the information in the last paragraph is that whereas in
Grade 5 the domain for the variable was to be the set of counting numbers (with, perhaps,
zero as well), in Grade 6 the domain for the variable could be subsets of the set of real
numbers (corresponding, for example, to measures of time and distance). By the end of Grade
6, students were expected to be able to translate between real-life modeling and algebraic
expressions. They should also create algebraic expressions which describe a pattern.
Grade 7. By the end of Grade 7, students would be expected to model situations
involving linear equations and inequalities. They should have learned to solve word problems
leading to inequalities of the form px + q > r or px + q < r, where p, q, and r represent specific
rational numbers. Ideas developed from modeling will appear frequently in problems—for
example: “As a salesperson, you are paid $50 per week plus $3 per sale. This week you want
your pay to be at least $100. Write an inequality for the number of sales you need to make,
and describe the solutions” (CCSSM, 2010, p. 49). Notice that in that particular example, the
independent variable would be a natural number, or zero.
Whereas the common-core mathematics document specifically asked for fifth- and
sixth-grade students to become acquainted with the concepts of explicit and recursive rules
for sequences, the issue whether students should be expected to learn the subscript notation
for sequences was not addressed. However, algebra students in the ninth grade would be
expected to have learned the notation (CCSSM, 2010).
Over the past few decades, the idea that modeling activities should be more important at
all grade levels in school algebra than has traditionally been the case, has become increasingly
popular among mathematics educators. However, as Kieran (2007) pointed out, the meaning
of the term “modeling” has not always been made clear. It has been interpreted as applying to
a broad range of situations (e.g., with respect to word problems, physical situations, numerical
and visual patterns, etc.)—the commonality in all of these, though, is that students are invited
to capture and generalize regularities by using the mathematical language of formulas.
The seventh-grade participants in the main study would be introduced to the subscript
notation for sequences, and it was a matter of interest whether they would be able to cope
with that notation. Two issues were identified: first, would they be able to interpret statements
in which someone else used the notation; and second, could they themselves become fluent in
using the notation to describe sequences?
Mathematicians typically regard a functional relationship in which the domain is the set
of natural numbers (with, perhaps, zero also included) as a special type of function, called a
sequence. Part of the current study was concerned with the seventh-graders’ investigations of
sequence-like functional relationships, and although developments in the students’ “functional
thinking” were studied, the emphasis was mainly on sequences, and therefore the type of
function was restricted. From that perspective, it could be argued that a restricted, and
therefore perhaps mathematically narrow, sense of a functional relationships might have been
portrayed to, and acquired by, the student participants.
The other main content area with which the seventh-grade participants in the current study
grappled involved algebraic properties for which the variables were not restricted to being
natural numbers. In a statement like “if a, b, c denote arbitrary rational numbers then it must
be true that a + (b + c) is equal to (a + b) + c,” the early-alphabet letters are being used to
denote variables. In the workshops on the associative and distributive properties in the main
study, the types of numbers which a, b, c could represent were deliberately varied. In some
tasks they represented natural numbers; in other tasks they represented integers which could
include negative numbers or zero; and in other tasks they represented common fractions.
From that perspective, the structure component of the current study presented a different view
of school algebra from the component which concentrated on students dealing with
sequences, in which the domain was usually the set of natural numbers, perhaps with zero.
Cai and Knuth (2011b) drew attention to two dominant perspectives in algebra
education research. The first related to a perceived need to develop students’ algebraic
thinking so that the students will be able to make strong connections between arithmetic and
algebra. The second related to the importance of supporting teachers’ efforts to foster the
development of students’ algebraic thinking. The main study described in this book was
concerned to investigate both of those perspectives.
Structural and Functional Approaches as Presented in the Mathematics Textbook Used

by Participating Students
In this section, we present content and some samples of text showing the main
approaches to algebraic structure and modeling in the textbook which the seventh-grade
students at School W had been using. For some years, the seventh-graders had been using a
textbook, authored by Charles, Branch-Boyd, Illingworth, Mills and Reeves (2004). School W
has class sets of the book, which is titled Mathematics Course 2 and was published by
Pearson Prentice Hall.
Charles et al. (2004) defined a variable—rather poorly—as “a symbol that represents
one or more numbers” (p. 71). Immediately after that definition was given, an “algebraic
expression” was defined as “a mathematical phrase with at least one variable” (p. 71). Three
examples of variables (g, z, and x) were given, and three examples of algebraic expressions
(13 – g, z/2, and 3x + 8) were provided. Readers were informed that the “value of an algebraic
expression varies or changes, depending on the value of the variable” (p. 71). Later in the
book, an equation was defined as “a mathematical sentence with an equal sign” (p. 122), and
a solution was defined as “a value for a variable that makes an equation true” (p. 122). An
inequality was “a mathematical sentence which contains <, >, ≤, ≥, or ≠” (p. 123), and a
solution to an inequality was “any number that makes the inequality true” (p. 123). That
definition was, perhaps, unfortunate for it could cause a reader to believe, for example, that 5
was the solution to the inequality 2x + 1 > 10.
The introduction to algebra given in Charles et al. (2004) is standard, and similar to
introductions found in many textbooks used in middle schools during the past 50 years.
Among other things, the main study investigated how well the 32 participating students grew
to understand the concept of a variable as it was presented in the treatments of the associative
property for addition of rational numbers, the associative property for multiplication of
rational numbers, the distributive property of multiplication over addition or subtraction, and
in situations and contexts involving modeling from numerical and visual patterns.
Structural and Functional Approaches to School Algebra (Textbook) 81
Textbook Treatment of the Associative Property for Addition

The first mention of the associative property for addition of real numbers in Charles et
al. (2004) was on page 12, where the property was described by the statement “changing the
grouping of the addends does not change the sum.” Then, examples from arithmetic and
algebra were given:
Arithmetic. (2.5 + 6) + 4 = 2.5 + (6 + 4)
Algebra. (a + b) + c = a + (b + c)
A model example was then given under a heading “Why It Works”:
0.7 + 12.5 + 1.3 = 0.7 + 1.3 + 12.5 (commutative property)
= (0.7 + 1.3) + 12.5 (associative property)
= 2 + 12.5
= 14.5 (p. 13)
Then, exercises in which the associative property for addition could be used were given
(pp. 13–15). The fact that an associative property for subtraction of rational numbers does not
hold was not mentioned.
Textbook Treatment of the Associative Property for Multiplication

The associative property for multiplication of real numbers was first mentioned in
Charles et al. (2004) on page 18, where the property was summarized by the statement
“changing the order of the factors does not change the product.” Then, examples from
arithmetic and algebra were given:
Arithmetic. (3 ⋅ 2) ⋅ 5 = 3 ⋅ (2 ⋅ 5)
Algebra. (a ⋅ b) ⋅ c = a ⋅ (b ⋅ c)
A model example was then given under a heading “Why It Works”:
0.25 ⋅ 3.58 ⋅ 4 = 0.25 ⋅ 4 ⋅ 3.58 (commutative property of multiplication)
= (0.25 ⋅ 4) ⋅ 3⋅58 (associative property of multiplication)
= 1 ⋅ 3.58
= 3.58 (p. 18)
Then, exercises in which the associative property for multiplication could be used were
given (pp. 19–22). The fact that an associative property for division of rational numbers does
not hold was not mentioned.
Textbook Treatment of the Distributive Property for Multiplication Over Addition

The distributive property for multiplication over addition of rational numbers was first
mentioned in Charles et al. (2004) on page 51, where the property was summarized. Then,
examples from arithmetic and algebra were given:
Arithmetic. 9(4 + 5) = 9(4) + 9(5) and 5(8 – 2) = 5(8) – 5(2)
Algebra. a(b + c) = a(b) + a(c) and a(b – c) = a(b) – a(c)

After a model example was given, under a heading “Mental Math,” the distributive
property was used to find 6(53), which was said to be equal to 6(50) + 6(3), or 300 + 18, or
318. Then the reader was asked whether 6(50 + 3) must equal (50 + 3)6. Overall, there was
surprisingly little attention given to the distributive property in the book, and in most of the
cases which were mentioned the context was numerical rather than specifically algebraic.
Textbook Treatment of Modeling and Related Problem Solving

On pages 72 and 73 of the textbook, attention was given to writing algebraic
expressions corresponding to real-world situations. The theme was introduced by the
statement:
You can use algebraic expressions to describe data. Suppose your heart beats 72
times in one minute. You can write an expression for the number of beats in any
number of minutes.
Let m represent the number of minutes. → 72m the algebraic expression
represents the number of heartbeats in m minutes. (Charles et al., 2004, p. 72)
Then an example was given, with the temperature being increased from t degrees to 5 degrees
more than that. It was stated that the new temperature could be summarized by t + 5. Then
followed several pages of practice exercises in which algebraic expressions were to be found
for given expressions such as “four shirts more than s shirts” (p. 73).
Chapter 9 of the textbook was entitled “Patterns and Rules,” and had sections on
“Patterns and Graphs,” “Number Sequences,” “Patterns and Tables,” and “Using Tables,
Rules and Graphs.” Although terms like “arithmetic sequence,” “function,” “geometric
sequence,” and “conjecture” were often used, the emphasis was on informal rather than
formal mathematics, and only occasionally was attention given to making generalizations.
There were many tables of values in the chapter, and perhaps the most interesting, from a
pedagogical perspective, was that shown in Table 4.1, which appeared on page 481. This was
the only table in Charles et al. (2004) which included the symbol “…” (a notation which was
intended to invite readers to make a cognitive leap from specific cases to the general case). No
explanation of the meaning of … was given, and it was a matter of interest whether the
seventh-grade students in the main study would be aware of what was required of them. In the
two pilot studies conducted by the researcher (Kanbir, 2014, 2016a), it was found that
seventh- and eighth-grade students at a school different from School W were not aware of this
convention, and those students, if shown Table 4.1, would have been inclined to place “48”
under n, rather than “8n.” In fact, the three dots (…) convention has rarely been discussed in
school algebra texts—although it was used extensively in Robert Davis’s (1964) Discovery in
Mathematics, where its meaning was defined by the questionable statement, “the use of the
final three dots indicates that we mean that the list goes on forever without stopping” (p. 16).
In fact, “…” can have different meanings in different contexts.
Textbook Treatment of Modeling 83
Table 4.1
A Table of Values from Charles et al. (2004), p. 481
Number 1 2 3 4 … n
×8 ×8 ×8 ×8 ×8 ×8
Value of 8 16 24 32 … ?
Sequence
On reading the treatment of modeling in Charles et al. (2004), we were surprised that
there were so few examples of sequences in which students were invited to formulate explicit
rules for nth terms.
The Problem as it Appeared to be at the Beginning of the Study

Having analyzed and interpreted pencil-and-paper test and interview data during the
pilot studies, and having observed middle-school algebra classes during the pilot studies, we
began to wonder whether the expectations implicit for middle-school algebra in the common-
core mathematics document (CCSSM, 2010) and in Charles et al. (2004) were unduly
optimistic. During the pilot studies we gained the impression that most of the students knew
very little about the associative properties for addition and multiplication, or about the
distributive property. They did not know the meanings of the formal language “associative
property for addition,” “associative property for multiplication,” and “distributive property for
multiplication over addition.” More seriously, they did not seem to have an intuitive
understanding of how these properties might be useful for simplifying calculations like “17
times 11” or “(172 × 5) × 1/5” or “502 + 499.” Likewise, other than being able to identify
recursive rules, like “add 3,” for simple pattern tasks, the students seemed to struggle to
identify and describe patterns or to engage profitably in modeling tasks.
Initially, the first author wondered whether data generated in the pilot studies pointed to
something unique about the school at which the pilot studies had taken place. But, on
reflection, he recognized that that school seemed to be very well organized, the teacher who
was teaching the pilot-study mathematics classes was qualified, enthusiastic, and dedicated,
and the students were cooperative and well behaved. He conjectured that the pilot-study
school had not given much attention to early algebra or to patterns in its K–6 mathematics
program, but inquiries indicated that that was not the case. This led him to suspect that a
statement by the University of California mathematician, Hung-Hsi Wu (2007), that “by the
sixth grade most students already know about the associative and commutative laws of
addition and multiplication,” might be seriously wrong. His suspicions hardened into belief
when he read a paper by Ding, Li and Capraro (2013) which reported data indicating that
preservice elementary school mathematics teachers within the United States of America did
not know the difference between associative and commutative properties of multiplication.
Thus, the first author felt it was desirable to check if the situation regarding middle-
school algebra which was found at the pilot-study school was typical of what would be found
with students in other middle schools. That prompted him to conduct the main study in a
middle school other than the pilot school. The second and third authors of this book thought
that they knew an appropriate school, and thus it came to be that the three authors approached
the Principal and middle-school mathematics teachers at School W.
Two of the authors of this book (Clements and Ellerton) had supervised previous
doctoral research involving students and teachers at School W, and knew the school well.
They knew that it had high mathematics standards and that in previous years they had
examined data which suggested that middle-school students from School W tended to perform
slightly better than middle-school students in neighboring schools on pencil-and-paper tests of
middle-school mathematics. They also knew that in the past the school had been very willing
to participate actively in mathematics education research projects that it deemed to be well
designed and consistent with its curriculum. After talking with the Principal and with middle-
school teachers at School W, it was agreed that the main study would be based there—
provided, of course, the necessary assents and consents from students and parents, from
district and school administrators, and from the University where the first author was
enrolled, could be obtained.
Contemporary thinking about middle-school algebra has tended to emphasize the
importance of “functional thinking.” Amy Ellis (2011) is among many recent researchers to
have referred to “the value of organizing algebra around a functions approach” (p. 218).
However, Ellis (2011) stated that endeavors to implement an agenda based on that approach
had “proved difficult” (p. 218) and that “students emerge from middle school and high school
algebra classes with a weak understanding of function” (p. 219). Despite acknowledging past
failures with the functional approach, Ellis nevertheless called for middle-school students to
be introduced to functional relationships through “meaningful situations” (p. 235). Miriam
Lüken (2012) pointed out that although patterning competences and algebraic thinking have
been described on a theoretical basis, the sense and practicality of expecting elementary and
middle-school children to deal with these have not yet been shown empirically. Such a
statement was consistent with Ferdinand Rivera’s (2013) position that unless students receive
more support in the development of their structural thinking with respect to patterning
activities—in order to develop their “functional thinking”—they may not become proficient
with their symbolic algebraic thinking.
Carolyn Kieran (2007) pointed out that although field and order properties of real
numbers have always been an implicit part of symbol manipulations, children’s acquisition of
those properties has rarely been addressed by algebra education researchers. She maintained
that future research needed to be carried out concerning the interactions with and the
distinctions between the function approach, which emphasized modeling, and an equation
approach, which emphasized algebraic structures. This argument supported Blanton,
Stephens, Knuth, Gardiner, Isler, and Kim’s (2015) arguments in favor of a multilayered
approach to algebra instruction. Blanton et al. (2015) argued that such an approach could have
significant effects on students’ ability to generalize, represent, justify, and reason with
mathematical structures.
In the main study, middle-school students at School W were engaged in six workshops
involving their participation in tasks concerned with functional relationships and six other
workshops on algebraic structures. The truth of the often-repeated assertion that in early
algebra and in middle-school algebra a functional-thinking approach is likely to be more
beneficial for learners than a structure approach, was not taken as granted. The effects of each
approach, as well as a combination of the two approaches, on student knowledge,
understandings, and attitudes, were studied.
Identifying the Fundamental Problem 85
Having identified the problem for the research study, the question arose how it should
be framed, theoretically. Almost all the seventh-grade students in the pilot studies had not
recognized the meanings or implications of important signs often used in middle-school
algebra. They did not use, or know the meanings, of terminology that the common-core
sequence suggested they should have known. Their thinking with respect to structure was
dominated by a perceived need to apply the “Please Excuse My Dear Aunt Sally” (PEMDAS)
mnemonic, and their approaches to modeling were hampered by a lack of knowledge of
symbolic conventions which subtly apply in tasks involving tables of values.
The first author chose the major semiotic theory developed by Charles Sanders Peirce as
part of the theoretical base for the current study. Peirce emphasized the triadic “signifier,
interpretant, signified” relationship, and in the pilot studies it had become clear that signifiers
for algebra were not being linked, by the students, to the appropriate, curriculum-related,
“signifieds.” In order to complement Peirce’s theoretical position, Johann Friedrich Herbart’s
theory of apperception—which emphasized the need to take account of what was already “in
the minds” of learners before and during mathematics lessons—was also chosen as part of a
bundle of three theories for the main study. The third theory was the “expanding-the-modes-
of-communication” theory put forward by Del Campo and Clements (1987). This third theory
emphasized the importance of not only getting students to learn to comprehend the language
used by the teacher and textbook authors, but also to provide opportunities for them to use
that language expressively.
More will be said about Peirce’s, Herbart’s and Del Campo and Clements’s theoretical
positions in the next chapter.
References
Blanton, M., Stephens, A., Knuth, E., Gardiner, A. M., Isler, I., & Kim, J. S. (2015). The
development of children’s algebraic thinking: The impact of a comprehensive early
46(1), 39–87.
Cai, J., & Knuth, E. (Eds.). (2011a). Early algebraization: A global dialogue from multiple
Cai, J., & Knuth, E. (2011b). A global dialogue about early algebraization from multiple
CCSSM. (2010). Common core state standards for mathematics. Washington, DC: Authors.
[Also cited under National Governors Association Center for Best Practices, & Council
of Chief State School Officers. (2010).]
Charles, R. I., Branch-Boyd, J. C., Illingworth, M., Mills, D., & Reeves, A. (2004).
Mathematics course 2. Needham, MA: Pearson Prentice Hall.
Wesley Publishing Co.
Ellis, A. B. (2011). Algebra in the middle school: Developing functional relationships through
quantitative reasoning. In J. Cai & E. Knuth (Eds.), Early algebraization: A global
dialogue from multiple perspectives (pp. 215–238). Heidelberg, Germany: Springer.
Gavin, M. K., & Sheffield, L. J. (2015). A balancing act: Making sense of algebra.
Mathematics Teaching in the Middle School, 20(8), 460–466.
Kanbir, S. (2014, November). Two approaches: Beginning algebra students’ variable concept
development. Professional project presented to the Group for Educational Research in
Mathematics at Illinois State University, Normal, IL.
Kanbir, S. (2016a, April 12). Three different approaches to middle-school algebra: Results of
Kanbir, S. (2016b). An intervention study aimed at enhancing seventh-grade students’
Kieran, C. (2006). Research on the learning and teaching of algebra. In A. Gutiérrez &
P. Boero (Eds.), Handbook of research on the psychology of mathematics education:
Past, present and future (pp. 11–49). Rotterdam, The Netherlands: Sense Publishers.
Kieran, C. (2007). Learning and teaching algebra at the middle school through college levels:
Building meaning for symbols and their manipulation. In F. K. Lester (Ed.), Second
handbook of research on mathematics teaching and learning (pp. 707–762). Charlotte,
NC: Information Age Publishing; and Reston, VA: National Council of Teachers of
Mathematics.
Lüken, M. M. (2012). Young children’s structure sense. Journal für Mathematik-Didaktik, 33,
263–285.
Officers. (2010). Common core state standards for mathematics. Washington, DC:
Authors. (Also cited under CCSSM (2010.)
Rivera, F. D. (2013). Teaching and learning patterns in school mathematics: Psychological
and pedagogical considerations. Dordrecht, The Netherlands: Springer.
Twidle, J. (2006). Is the concept of conservation of volume in solids really more difficult than
for liquids, or is the way we test giving us an unfair comparison? Educational Research,
48(1), 93–109.
H. G. Steiner (Ed.), Comparative studies of mathematics curricula: Change and stability
1960–1980 (pp. 12–36). Bielefeld, Germany: Institut für Didaktik der Mathematik-
Universität Bielefeld.
Whitacre, I., & Nickerson, S. D. (2016). Prospective elementary teachers making sense of
multidigit multiplication: Leveraging resources. Journal for Research in Mathematics
Education, 47(3), 270–307.
Chapter 5
Review of Pertinent Literature
Abstract: This chapter frames the main study described in this book in terms of the
theoretical positions of Charles Sanders Peirce, Johann Friedrich Herbart, and Gina Del
Campo and Ken Clements. Peirce’s tripartite position on semiotics (featuring signifiers,
interpretants, and signifieds), Herbart’s theory of apperception, and Del Campo and
Clements’s theory of complementary receptive and expressive modes of communication,
were bundled together to form a hybrid theoretical position which gave direction to the study.
The chapter closes with careful statements of six research questions which emerged not only
from consideration of the various literatures, but also from a knowledge of practicalities
associated with the research site, from our historical analysis of the purposes of school
algebra, and from our review of the literature.
Keywords: Charles Sanders Peirce, Design research, Expressive communication, Johann

Friedrich Herbart, Modeling in school algebra, Receptive communication, Structure in school
algebra
In Chapter 4 we drew attention to some of the problems faced by middle-school teachers,

textbook authors, and curriculum designers in regard to middle-school algebra. One of the
problems was that although the National Governors Association and the Council of Chief State
School Officers’ Common Core State Standards for Mathematics (hereafter CCSSM), 2010)
has recommended that elementary and middle-school mathematics curricula should have
strong algebra components, the meaning of “algebra” has not been made clear. Certainly,
though, CCSSM identified two main strands of thought with respect to middle-school
algebra—those associated with (a) the structure of rational numbers, and (b) a functional-
thinking approach by which learners are engaged in patterning and modeling activities which
link independent and dependent variables (Cañadas, Brizuela, & Blanton, 2016; Kieran, 2006,
2011; Taylor-Cox, 2003). In this chapter these will be called the “structure” approach and the
“modeling” or “functional-thinking” approach. Interestingly, those two strands of thought
correspond to two of the “purposes” identified in Table 2.1 in Chapter 2 of this book.
The research literature has not established whether our two main strands for middle-
school algebra require the learning and application of the same kinds of skills, concepts, and
principles. Although desirable teaching methods have often been discussed for the functional-
thinking approach, that has not been the case with the structure approach. Furthermore, the
amount of classroom time in middle schools to be allocated to each has not been resolved.
Should both approaches receive the same amount of attention, or should one of the two be
given more time than, and priority over, the other? The research literature has not established
how the two forms should be combined in order that students will be best prepared for
applications in real-life situations, or for the study of higher mathematics and science.
Advocates of the structure approach emphasize the need to assist middle-school students
to learn to identify and use key structural properties which carry the names “closure,”

88 Ch. 5: Review of Pertinent Literature
“associative,” “commutative,” “identity,” “inverse,” and “distributive” (CCSSM, 2010). These

structural properties were obviously regarded as very important by those who developed the
CCSSM document, and that view is shared by many mathematicians (see, e.g., Wu, 2001).
Nevertheless, the common-core document framed the structural properties in formal
mathematical language, and it is not clear whether they can be learned in that way by middle-
school students.
For instance, in the CCSSM sequence, the distributive property was specifically
mentioned for seventh-grade algebra, and the identity 0.05a + a = 1.05a was given as an
illustration of the property. The implication seemed to be that students should recognize that
the distributive property enables 0.05a + a to be written as a × (0.05 + 1), which can be
written as a × 1.05; then, by the commutative property for multiplication (and a knowledge
of some of the conventions of the syntax of elementary algebra), that could be written as
1.05a. Many middle-school teachers would say that such an approach is extremely bookish,
and any link with the idea of a “5 percent increase” will not be appreciated or understood by
most seventh-grade students.
The research literature for middle-school mathematics (e.g., Cai & Knuth, 2011) does not
provide detailed evidence with respect to what middle-school students might reasonably be
expected to know about the key structural properties of the set of rational numbers, and
hardly any attention is given to describing methods for engaging students actively and
profitably in learning about them.
By contrast, in recent years, considerable attention has been given in the algebra-
education research literature to the functional-thinking approach. In 2015, for example,
lengthy papers appeared in separate issues of the Journal for Research in Mathematics
Education in which some authors pushed, unambiguously, for the importance of that
approach in elementary and middle schools (Blanton, Brizuela, Gardiner, Sawrey &
Newman-Owens, 2015; Blanton, Stephens, Knuth, Gardiner, Isler & Kim, 2015). But other
commentators (e.g., Kieran, 2006, 2011) have stopped short of recommending that the
functional-thinking approach should receive greater emphasis than the structure approach.
It was the question whether middle schools should attempt to balance the amount of
classroom time given to the two approaches and, if so, whether it is possible to link the
approaches conceptually, which was at the heart of the research described in this book. We
decided to study how well seventh-grade students in a highly-regarded, medium-sized public
school in a rural area in a midwestern state within the United States of America—which we
have called “School W”—had learned both the structure and functional-thinking approaches
during their elementary and early middle-school years. We also decided to study whether it
would be possible, in a modest amount of evenly-distributed classroom time, to improve
significantly the middle-school students’ knowledge and application with respect to both of
those two approaches to elementary and middle-school algebra.
The Design Research Foundation for the Study

As indicated in Chapter 3, the planning for the investigation described in this book was
strongly influenced by the design-research literature (see, e.g., Brown, 1992; Clements, 2011;
Kelly & Lesh, 2000; Kelly, Lesh & Baek, 2008). For the benefit of readers who do not have
access to Chapter 3, the extent of that influence will now be briefly revisited. Design-research
approaches to education research and development have emerged from considerations of
The Design Research Foundation for the Study 89
planning and design methods used by teams of architects and engineers who were preparing to
carry out major design projects
The early-algebra issues outlined above were recognized by the two main teachers of
middle-school mathematics at School W (in Chapter 3 these teachers were called “Mr. X”
and “Mr. Y”). Agreement was reached that a research team should be formed comprising the
three authors of this book and the two teachers. With the approval of the Principal of School
W, and with relevant ethics approvals having been obtained, it was decided that the team
members would design a middle-school-algebra intervention program which would include:
1. The collection and analysis of initial, pre-intervention pencil-and-paper test data,
and also interview data.
2. Some initial pre-intervention teaching and observation of eighth-grade algebra
classes at School W. This would involve all five members of the research team.
3. A teaching intervention involving seventh-grade classes at School W. The teachers
of the seventh-grade students, Mr. X and Mr. Y, would be the classroom teachers
for the intervention.
4. The collection and analysis of middle- and post-intervention data, including middle-
and post-intervention pencil-and-paper test data and post-intervention interview data.
5. The collection and analysis of pencil-and-paper retention test data.
In the planning process, careful attention was given to each of the following three
design elements:
1. What were the dimensions of the task?
• What was the aim of the intervention program?
• What did we plan to do in order to achieve that aim?
• What were the most likely threats that would prevent us from achieving our aim,
and what did we need to do to prevent those threats from becoming realities?
2. Who would be the key players in the intervention program?
• With the permission of the Principal of School W, it was agreed that a research
and teaching team of five persons would be involved, and the students in the
main part of the study would be all those seventh-grade students at School W
who gave their assent to participate and whose parents or guardians gave their
consent for their children to be involved.
• Preliminary meetings involving all five team members took place, and at those
meetings shared responsibilities for important components of the intervention
program were agreed upon. More details will be provided in Chapter 6 of this
book.
3. How should the program be implemented, evaluated and reported?
• It was agreed at the preliminary meetings that the three authors of this book
should model the proposed workshops for seventh-graders with classes of eighth-
grade students at School W, and that Mr. X and Mr. Y would be present and
observe all the eighth-grade classes taken by the three authors. Then, with the
seventh-grade classes, Mr. X and Mr. Y would attempt to follow the approaches
and materials prepared by the three authors—with the proviso that Mr. X and Mr.
Y should feel free, at any time, to depart from the teaching approaches that they
had observed in the model workshops, and from the materials used in those
workshops, when and if either or both of them felt the need to do so. More details
of the teaching intervention will be provided in Chapter 6.
• It was also agreed that the three authors should develop the materials and
artifacts to be used in the seventh- and eighth-grade workshops, and also
prepare pre- and post-intervention interview protocols and pencil-and-paper
tests.
• The first author of this book would analyze all relevant data arising from the
study (with the assistance of the other two authors of this book).
• It was also agreed that the first author would prepare a doctoral dissertation based
on the investigation and that, in addition, the first author would also prepare two
“teaching” papers—one that would be co-authored with Mr. X and the other with
Mr. Y—with the intention of submitting the papers to teaching journals.
Theoretical Bases for the Study

For the education investigation described in this book, careful consideration was given
by the five members of the research team to the selection of the most appropriate theories and
also to how those theories should be taken into account at the planning, intervention, and
evaluation phases of the study. The theoretical bases for the study will now be outlined.
The research team decided that the planning and implementation of the study, and the
interpretation of data generated by the investigation, should be informed by the work of
several scholars who had put forward theories that team members deemed to be especially
relevant to the study. We paid particular attention to the ideas of those who had attempted (a)
to link semiotic theories to middle-school mathematics curricula and practices in school
mathematics; (b) to identify the importance of students’ cognitive structures with respect to
curricular themes in school mathematics; and (c) to identify how those structures might be
affected by instruction and patterns of classroom communication.
When making decisions on how the study should be framed theoretically we did not
assume that the positions of modern scholars should be preferred to those of earlier scholars.
Indeed, two of the most important influences on our thinking came from the early, but classic,
works of:
• Johann Friedrich Herbart (1776–1841), a German philosopher and educator whose
main writings on education appeared about 200 years ago;
• Charles Sanders Peirce (1839–1914), a U.S. philosopher and logician and the son of
a famous Harvard University mathematician, Benjamin Peirce. Charles Sanders
Peirce is often regarded as the father of pragmatism, and the research team believed
that his semiotic theory, which deals with signification, representation, reference,
and meaning, is important for understanding students’ thinking about algebra.
Peirce and Herbart were not the only theorists whose writings had an impact on our
thinking—that will become clear as this chapter progresses.
This study did not aim specifically to apply Herbart’s or Peirce’s ideas in defining the
investigation, or in interpreting data arising from it. The ideas of other theorists would be taken
into account—thus, for example, Luis Radford’s (2006) distinctions between various types of
generalizations (factual, arithmetic, algebraic, contextual, and symbolic generalizations) were
Charles Sanders Peirce’s Semiotics 91
noted. It would be an exaggeration, though, to say that Radford’s writings, or those of any
other modern theorist, were taken more seriously than Peirce’s or Herbart’s.
From the outset, then, we did not seek to frame the investigation around any particular
theory, or theorist. Rather, we decided to investigate several important issues associated with
algebra education involving seventh-grade students and their teachers and, having conducted
two pilot studies with middle-school students (Kanbir, 2014, 2016), we allowed the
observations and results from those pilot studies to influence us when we selected, extended,
and refined the theoretical base, and the wording of the research questions, for the main
study. The research questions are listed at the end of this chapter.
Signifiers, Objects, Interpretants, and Charles Sanders Peirce

The first set of theoretical ideas that we linked to the study were those in the literature
related to “semiotics.” During the course of conducting interviews and observing classes for
the second pilot study it became apparent to the three authors that most of the seventh-grade
students who participated in that study had not learned to interpret the symbols of middle-
school algebra, and often they were not aware of conventions used in texts seeking to convey
information and ideas related to algebra. This led the research team to focus part of the
search within the literature on writers who had concentrated on learners’ understandings of
signs, symbols and representations—and, not surprisingly, this decision led us to the
literature on “semiotics.” Although, writers in that field often express simple, but important,
ideas in very sophisticated and dense language, in this literature review we try to use as
simple and as direct language as possible.
Both Charles Sanders Peirce (1992, 1998) and Ferdinand de Saussure (1959) developed
theories in which they emphasized the importance of signs, and the ways those signs
influenced the thinking and cognitive development of people (Presmeg, 2014). They showed
how signs might be regarded as “signifiers” which communicate features of objects (or
“signifieds”) to people (“referents”). Thus, for example, the word “parallel” is a sign which
communicates something definite to adults who know the English language; adults who also
know that “parallel” means something different from “parallelogram,” will probably also
know that the concepts represented by the two words, or signs, are, somehow, related.
However, for a 10-year-old child growing up in a non-English-speaking family, the
“meanings” of the words “parallel” and “parallelogram” might not yet be differentiated. In
other words, biological, conceptual, social, and linguistic development can influence the
meanings that learners attribute to signs.
For Peirce, a sign relation defines three roles encompassing: (a) a sign, (b) the accepted
meaning of the sign, called its object, and (c) the meaning that someone attributes to the sign,
called an interpretant. A sign relation is an irreducible triadic relation—irreducible in the
sense that the full meaning of the sign relation will not be obtained by considering only two
of the three roles (Peirce, 1998). By contrast, for de Saussure (1959), the sign relation is
dyadic, consisting only the sign (or signifier) and its meaning (the signified). de Saussure saw
this relation as arbitrary, motivated mainly by social convention or by instruction.
With de Saussure’s model a signifier, for example the word “tree,” stands for something
which is “signified,” which in this case relates to the abstract concept of “a tree.” Thus, a sign
can direct someone’s thinking toward an abstract object. Charles Sanders Peirce went beyond
this dyadic model by adding an object representamen, which would stand for the object other
than the sign and an interpretant, the meaning given by someone to the sign. According to
Peirce (1992), “a sign is a third mediating between the mind addressed and the object
represented” (p. 281). Some other triads—such as firstness, secondness, and thirdness, and
icon, index, and symbol—were offered by Peirce in relation to this process of objectification.
In the current study, Peirce’s triadic semiotic theory was regarded as central, the view being
taken that there were three major curriculum aspects, intended, implemented, and received
(Westbury, 1980).
Semiotic Aspects of Structural Properties for Rational Numbers

What might semiotic ideas have to do with seventh-graders as they attempt to learn
algebra? We answer that question by considering a task which was used in the study and
which will often be mentioned in this book. Twenty-eight of the 32 participating seventh-
grade students were asked to engage with the task on two occasions—when they were
individually interviewed before and after the main teaching interventions.
During the interviews each interviewee was shown, and asked to react appropriately to,
a typed pencil-and-paper version of the following task:
1
Without using a calculator, find the value of 4 × ( × 128).
4
This stimulus can be thought of as a composite sign made up of various symbols. Each
interviewee was asked to read the problem and, although the 28 interviewees involved in the
1
study had no trouble doing that, they tended to read “4 × ( × 128)” in different ways—like,
4
for example, “four times, open parentheses, one-fourth times one hundred twenty-eight, close
parentheses,” and “four multiplied by a quarter times one twenty-eight.”
What was especially interesting is that before participating in any of the workshop
sessions on structure, all 28 interviewees interpreted the symbols incorporated in the composite
sign as an instruction first to multiply one-fourth by 128 and then to multiply the answer
1
obtained by four. Not one interviewee at the pre-teaching stage associated the 4 with the .
4
The students’ thinking tended to be dominated by the mnemonic PEMDAS (“Please Excuse
my Dear Aunty Sally”), which they had learned to use as a guide for the order of operations to
be performed in complex calculations. In PEMDAS, the “P” is meant to indicate that what is
“inside” the “parentheses” needs to be dealt with first; and, therefore, at the pre-teaching stage
1
all 28 interviewees believed that they were expected to find the value of × 128 before they
4
did anything else. Furthermore, many of the students had come to believe that to find the value
1 128
of × 128, the first thing they needed to do was to write the 128 as , and that is what they
4 1
tried to do in their minds.
After the students had participated in the structure workshops provided in this study, the
words in the problem tended to signify to the students something quite different from what they
had signified before the intervention. In the “post-teaching” interviews, a majority of the
students began their responses by linking the 4 and the ¼; they then reasoned that since 4 times
¼ equals 1, the answer is 1 times 128, which is 128. When asked why they could multiply 4 by
¼ before doing anything else, some would say that they had used the associative property for
Numerical Structures and Order-of-Operation Mnemonics 93
multiplication. Yet, before the teaching intervention none of the students used the property, or
knew the meaning of, the expression “associative property for multiplication.”
The composite sign used to define the task “Without using a calculator, find the value
1
of 4 × ( × 128),” was not just any, randomly-ordered and randomly-chosen set of symbols.
4
Clearly, it originated from a “mathematical mind” and was designed to invite application of a
mathematical property known as the associative property for multiplication. Inherent within
the composite sign, for the initiated, was an educational idea—if you have well-formed
mathematical knowledge then you will link, mentally, the 4 with the ¼, and realize that the
answer is the product of 1 and 128, which is simply 128. The interesting issue, from a
mathematics education perspective, is that according to the common-core sequence (CCSSM,
2010), all seventh-grade students should be expected to notice that the 4 and ¼ should be
linked, so that the answer 128 would be fairly easily obtained, mentally. The associative
property for addition is specifically mentioned in the first-grade statement of CCSSM (2010),
and there is much reference to associative properties for addition and multiplication
throughout the elementary and middle-school CCSSM content descriptions. In other, words,
using the language of Peirce (1998), Otte (1998, 2011), and de Saussure (1959), the
1
composite sign “find the value of 4 × ( × 128)” was something that a seventh-grade teacher
4
1
might have expected would have signified to students the need to take the out of the
4
parentheses and link it with the 4, with the object of simplifying the carrying out of the task.
But, despite PEMDAS not being mentioned in the task, all of the participating students
consciously chose to use it. Most fifth- and sixth-grade students across the United States are
taught the “Please Excuse My Dear Aunt Sally” (PEMDAS) mnemonic, which is well known
among U.S. middle-school teachers. Jeon (2012), however, has provided evidence that
although preservice teachers tend to know the mnemonic, they do not know why it works,
and many are not aware of additional caveats which need to be linked to it. One such caveat
is that if a calculation involves a combination of multiplications and divisions (and no other
operations) then PEMDAS does not apply—but the operations should be carried out in the
order in which they are listed, from left to right. For example, consider the sign 36 ÷ 4 × 9. If
a student followed PEMDAS then the multiplication would come first, and the answer would
be 1. But the additional caveat requires that the division be carried out first, so the correct
answer is 81. How can fifth- or sixth-graders be expected to cope with a mnemonic with
“extras” like that? Of course, there is a similar caveat for calculations involving addition and
subtraction only—for example, 9 – 4 + 5 equals 10, and not 0.
Such is the influence of the PEMDAS mnemonic on U.S. children’s thinking that if
students in grades 5, 6, or 7 are asked to find the value of 97 × 9 + 3 × 9 they are likely to
carry out the multiplications first and then the addition. If the calculations are done
accurately then a correct answer will be obtained, but students using that method have not
attended to the structure inherent in the calculation.
Without number structure sense, but having acquired a drilled propensity to use the
order of operations demanded by PEMDAS, students can not only get wrong answers for
calculations, but also fail to learn important structural principles. Yet, a common-core
progression document (Common Core Standards Writing Team, 2011) for “K, Counting and
Cardinality; K–5 Operations and Algebraic Thinking” has specifically stated that Grade 6
students should be able to “discuss their reasoning more explicitly by focusing on the
structures of expressions and using the properties of operations explicitly” (p. 35).
CCSSM’s Common Core Writing Team (2011) made it clear that they expect middle-
school students to get to know the structural properties of rational numbers. Its team
members also drew attention to the importance of parentheses when they wrote:
Parentheses are important in expressing the associative and especially the
distributive properties. These properties are at the heart of Grades 3 to 5 because
they are used in the Level 3 multiplication and division strategies, in multi-digit and
decimal multiplication and division, and in all operations with fractions. (p. 27)
The Writing Team (2011) went on to say:
Understanding and using the associative and distributive properties ... requires
students to know two conventions for reading an expression that has more than
one operation:
1. Do the operation inside the parentheses before an operation outside the
parentheses (the parentheses can be thought of as hands curved around the
symbols and grouping them).
2. If a multiplication or division is written next to an addition or subtraction,
imagine parentheses around the multiplication or division (it is done before
these operations). At Grades 3 through 5, parentheses can usually be used
for such cases so that fluency with this rule can wait until Grade 6. (p. 27)
Although the Writing Team clearly recognized and attempted to address most of the
issues which we have raised, it fell short of demonstrating a genuine understanding of the
considerable semiotic difficulties involved. That is unfortunate because there is a danger that
students whose teachers focus on the PEMDAS order-of-operations convention may fail to
develop the kinds of structural thinking which was expected by those who developed the
CCSSM document.
1
To reiterate, it is likely that, for many seventh-graders, the sign “4 × ( × 128)”
4
will not be recognized as anything to do with a structural property of rational numbers.
Rather, it will signify a quite different mathematical principle—that related to the mnemonic
PEMDAS, by which the order in which a composite calculation is be carried out is dictated by
the words Parentheses, Exponent, Multiplication, Division, Addition, Subtraction, whose first
1
letters can be combined to form PEMDAS. Since there are parentheses in the sign “4 × ( ×
4
128),” the popularity of the mnemonic explains why seventh-graders might decide that they are
expected to carry out the calculation within the parentheses first. In other words, they will
attempt to find ¼ of 128 mentally, while holding the first 4 aside (mentally) for the moment.
1
Then, if they succeed in reducing × 128 to 32 (which should not be taken for granted
4
because calculations with fractions, even apparently simple ones, are notoriously difficult for
many middle-school students), then there is still the task of remembering the stored “4” and the
fact that this has to be multiplied by 32. Then there is the mental task of carrying out that
What is an Algebraic Task for Middle-School Students? 95
multiplication and obtaining 128. For most seventh-graders the cognitive load associated with
this method is overwhelming.
The above discussion suggests that signs used in middle-school mathematics may
signify different mathematical objects to different students. For many seventh-graders, the
1
sign 4 × ( × 128) signifies the need to carry out a series of calculations for which the first
4 1
calculation has to be × 128; but for some students, and hopefully for the teacher, the same
4
sign signifies the need to apply a mathematical property by which the first mental decision is
1
to associate the 4 and . Of interest is whether instruction aimed at getting seventh-grade
4 1
students to decide to proceed by the second route (that is to say, to link the 4 and , first)
4
will have the desired effect for most students.
What is an algebraic task for elementary and middle-school students? Some
mathematics educators distinguish sharply between what they call “arithmetic” and “algebra.”
Luis Radford (2011), for example, has stated that “to make algebraic thinking appear, and to
make it accessible to the students, some pedagogical conditions need to be created” (pp. 308–
309). According to Radford (2011), “what characterizes thinking as ‘algebraic’ is that it deals
with indeterminate quantities conceived of in analytic ways” (p. 310, original emphases). “In
other words,” Radford added, “you consider the indeterminate quantities (e.g., unknowns or
variables) as if they were known and carry out calculations with them as you do with known
numbers” (p. 310). Such a viewpoint seems to be quite traditional, despite Radford’s
subsequent attempt to argue that one does not really need to introduce a letter (or some other
symbol) to be in the realm of algebra. Like many mathematics educators (e.g., Blanton &
Kaput, 2011; Cai, Moyer, Wang & Nie, 2011), Radford tends to see algebra with elementary
and middle-school students as being most beneficial, educationally, if it is placed in the context
of sequences or modeling tasks in which the aim is to help students make decisions based on
perceived patterns between varying quantities.
Kaput (2008) argued for a wider view of algebra when he described algebraic thinking
as having either or both of two ingredients:
1. Making and expressing generalizations in increasingly formal and conventional
symbol systems; and
2. Reasoning with symbolic forms, including the syntactically-guided manipulations of
these symbols.
From this, Kaput (2008) called for school algebra curricula which, among other things,
would be based on the premise that algebra is the study of (a) structures and systems
abstracted from computations and relations, and (b) functions, relations and joint variation. In
the main investigation described in this book both of those aspects were inherent in the
assessment instruments which were developed and used in the teaching interventions which
were devised and used with the participating seventh- and eighth-grade students.
Some mathematics education researchers have argued that in addition to traditional,
alphanumerical forms of algebra there are non-traditional forms, not involving letters (see,
e.g., Britt & Irwin, 2011; Radford, 2011, 2015). In his recent research on algebraic thinking,
Radford (2011, 2015) has mostly focused on non-alphanumerical thinking and its progressive
transition to symbolic thinking in which letters are used to represent quantities or numbers.
He has maintained that if a student generates a formula merely by guessing and checking,
then that kind of naïve inductive thinking would not correspond to algebraic thinking
(Radford, 2006, 2015).
In commenting on their work for the New Zealand Numeracy Project, Murray Britt and
Kay Irwin (2011) took a different stance from Radford. According to Britt and Irwin,
students’ awareness of numbers and operations structure can illustrate their algebraic
thinking, even though pronumerals are not involved. Borrowing the term “quasi-variable”
from Toshiakira Fujii and Max Stephens (2001), they argued that often students’
explanations of their thinking revealed that they were treating the numbers as if they were
variables—provided they were being used in signs which pointed toward particular
structures, the numbers might be regarded as “quasi-variables” (Britt & Irwin, 2011, p. 138).
Some readers might complain that although the investigation described in this book is
supposed to be concerned with algebra education, a task such as “Without using a calculator,
1
find the value of 4 × ( × 128)” is in the domain of arithmetic and not of algebra. If one
4
accepts the first of Kaput’s (2008) ingredients, listed above, then that complaint is invalid. It
can be argued that it is not necessary for a written mathematical task to involve manipulation of
1
pronumerals to be an algebraic task. Recognizing that 4 × ( × 128) has the same value as (4
1 4
× ) × 128 demands knowledge of the essence (if not the formal name) of the associative
4
property for multiplication of rational numbers (Sriraman & Lee, 2011). There are various
kinds and degrees of algebraic knowledge and, like Britt and Irwin (2011), Fried (2008), Fujii
and Stephens (2001), Hewitt (1998), and de Saussure (1974), we believe that algebraic thinking
can occur in contexts in which pronumerals (e.g., x, y, a, b, etc.) are not employed.
We also recognize, though, that many seventh-grade students who can carry out an
associative transformation for multiplication do not know that the property that they have used
is called the “associative property for multiplication.” However, in our view, a student who
does not know the name of the property may still be regarded as “thinking algebraically.”
In the main study described in this book an attempt was made to see whether seventh-
grade students, before participating in workshops emphasizing associative and distributive
properties of rational numbers, had already learned to make associative and distributive
transformations. A second focus of the study was to examine the extent to which seventh-
grade students would improve their ability to recognize and make such transformations as a
result of participating in workshops aimed at helping them to do that. A third focus was to
ascertain whether some of the students would learn to state the formal names of the
associative and distributive properties of rational numbers, confidently and accurately.
Attention was given, therefore, to whether, and when, the participating seventh-grade
students could generalize sufficiently to able to recognize, almost instantly, that, for example: 9
1 1
× ( × 280) equals 1 × 280; that (29 × 15) × equals 29 × 1; and that similar
9 15
transformations can occur with addition (e.g., 288 + (12 + 453) is equal to (288 + 12) + 453).
Students who carry out such mathematically astute transformations without being prompted to
do so by another person have learned to generalize, even though their generalizations may not
Signifiers, Interpretants and Signifieds in Middle-School Algebra 97
involve pronumerals, and may not be associated, in their cognitive structures, with formal
mathematical names for the properties they use. In this book, this kind of generalizing from
arithmetic will be regarded as a form of algebraic thinking. Although relatively few seventh-
graders would be expected to be able to give, from memory, a formal verbal statement of, for
example, the associative property for multiplication of rational numbers—“for all rational
numbers, a, b, and c, the value of a × (b × c) is equal to the value of (a × b) × c,”—a decision
1 1
by a student to transform 4 × ( × 128) to (4 × ) × 128 would represent a kind of thinking
4 4
which might reasonably be classified as exhibiting “algebraic thinking.”
Signs do not signify the same thing to everyone. It is obviously wrong to think that a
sign will signify the same object to all persons. From that perspective, although an intended
content sequence, such as that provided by CCSSM (2010), might provide signposts which
are intended to guide teachers and school authorities toward key ideas for school
mathematics, readers of that content sequence may interpret signs in different ways. An
indication that those who developed the CCSSM mathematics sequence greatly valued
structural aspects of school algebra was that they included a list of the so-called field
properties for real numbers (the associative properties for addition and multiplication and the
distributive property, are three such properties) in a “glossary.” But they never made it
entirely clear exactly when students should be able to recognize and apply the different
properties, and when they should be able to state them, using formal mathematical language.
Furthermore, published research has not provided much guidance on such matters, either.
We do not know if it is reasonable, for example, to expect seventh-graders to recognize,
1 1
instantly, that 4 × ( × 128) is equal to (4 × ) × 128 or 1 × 128 or 128. Nor do we know the
4 4
proportion of seventh-grade students across the United States of America who would do that
without prompting, or the proportion of seventh-graders who would recognize that the sign
1
“Without using a calculator, find the value of 4 × ( × 128)” was inviting the application of
4
something known as “the associative property for multiplication of rational numbers.”
Furthermore, we do not know the proportion of such students who can state, verbally and
formally, the distributive property for multiplication over addition, for rational numbers.
Similar statements to those made in the last paragraph are true for all the field properties
for rational numbers. Although the authors of the common-core sequence have recognized that
the field properties are important in elementary and middle-school mathematics, they have not
made clear the orders in which such properties might reasonably be expected to be acquired by
schoolchildren—and at which levels of formalism. From that perspective, the research
described in this book will break ground that has become hardened as a result of neglect over
the years. The research will concentrate on seventh-graders’ knowledge and understandings of
the associative properties for addition and multiplication, and the distributive property for
multiplication over addition (and subtraction)—but, obviously, much more research is needed,
involving various educational aspects of all the field properties.
The writers intend to adopt some of the language used by semiotic theorists such as
Peirce (1992, 1998), de Saussure (1959), and Radford (2006). However, not all of the often
highly complex, linguistic distinctions and terminologies introduced and advocated by those
theorists have been regarded as relevant to the study.
Semiotic Aspects of the Functional-Thinking Approach to Early Algebra

In the study described in this book all of the seventh-grade students participated in two
separate sub-interventions—one involved a structure approach to middle-school algebra and
the other a functional-thinking or modeling approach. In this section, semiotic aspects of the
modeling approach will be discussed.
We do not know whether most adults are able to give appropriate meanings to the
signs, symbols, and conventions used in “tables of values” like the one shown in Figure 5.1.
However, many algebra educators believe that before elementary and middle-school children
can learn to think algebraically they will need to learn to interpret, and even to create, such
tables. Thus, it was important to investigate middle-school students’ thinking with respect to
tasks like “Visiting Old Houses,” which we now consider from a semiotic perspective.
The “Visiting Old Houses” task. In the interviews for the main study, 28 seventh-
graders were asked to imagine that they went on a tour of old houses. They were told that it
would cost $5 to register for the tour, and $2 extra for every old house visited—and were
shown the table in Figure 5.1.
Number of Old Houses 0 1 2 3 4 ... n

Visited
Total Cost (Dollars) $5 $7 $9 $11 ? ... ??
Figure 5.1. Table associated with “Visiting Old Houses” task.
The students were then asked two questions:

1. What number should we place under the 4 in the table?
2. What do you think we should put under the n?
The first and most obvious aspect of the composite sign represented by the table in
Figure 5.1 is its horizontal form. In the study described in this book the assumption that
seventh-grade students would realize that the numerals in the “cells” to the right of the words
“Number of Old Houses Visited” need to be related to the numerals in the cells immediately
below the upper-level cells was investigated. Would the seventh-graders realize that there is a
relationship between the numerical symbols in the upper cells, and the symbols representing
the costs (in dollars) in the corresponding lower cells? That might be especially difficult for a
child whose teacher had preferred to use tables of values which were presented in vertical
rather than horizontal form. In such a case there could be confusion between the signifier and
the signified.
In the top and bottom rows of the table in Figure 5.1 there are cells which have three
dots (...) and, by convention, these dots are often intended to convey the idea that there is to
be a “leap” toward generality. Thus, it is expected that the question mark beneath the “4”
could be replaced by the symbol “$13,” and the pair of question marks beneath the “n” by
$(5 + 2n), or something like that. It was also hoped that students would be able to offer a
reason (beyond guessing and checking) for why that should be the case.
Recursive and Explicit Rules for Modeling 99
At the pre-intervention stage, most of the participating seventh-graders did not know
the convention represented by the three dots, and thought that the n (in the last upper cell)
represented 6 (“because 4 + 1 + 1 equals 6”). In a similar way, they thought that the symbol
“??” represented $17 (“because 11 + 2 + 2 + 2 equals 17”).
There are educationally significant assumptions built into numerical relationships
represented in tabular form. Even if the conventions are known, the question whether
seventh-graders are ready to make the cognitive leap needed to go from particular values
associated with a recursive recognition of the sequence pattern to an explicit, generalized
representation, like $(5 + 2n), or $(2n + 5) in Figure 5.1, is something which demands much
relevant educational research. Is it reasonable to expect most seventh-graders to be able to
make that leap? What kind of teaching will assist students to learn to make such leaps?
It should also be emphasized that the composite sign that makes up a table of values like
that in Figure 5.1 comprises not just any randomly-ordered and randomly-chosen symbols.
Clearly, Figure 5.1 originated from a “mathematical mind,” and was designed to invite
someone to make a generalization. No clue was given how the task was to be completed,
except that a possible real-life scenario was well defined. This was an educational task, and
those who developed it—the three authors of this book—were interested in finding out if
seventh-grade students could make sense of the composite sign, and ultimately whether they
could identify the mathematical object which was in our minds when we developed the task.
There is also an issue about developing appropriate notations to represent
correspondences between variables. If the cost for visiting n old houses is denoted by $Cn,
would it possible for most seventh-graders to learn to comprehend and write, Cn = 2n + 5?
Which is a better notation, from a curriculum perspective—the function notation C(n), or the
sequence notation, Cn? Or, would it be better merely to write something like C = 2n + 5?
And, do the answers to those questions depend on the ages or grade levels of students?
For this study it was decided that there would be an emphasis on helping seventh-graders
to learn to use the subscript (sequence) notation. Could seventh-graders learn to give
appropriate meanings to the subscript notation, and could they be expected to develop the
explicit equation Cn = 2n + 5? Would it be reasonable to expect seventh-graders to develop and
comprehend this subscript notation to such an extent that they could describe the relationship
between the number of old houses visited and the cost recursively in the form Cn + 1 = Cn + 2,
with C1 = 7 (or, perhaps, C0 = 5)? As far as the writers are aware, despite the large amount of
research reported on young children’s development of so-called “functional thinking,” and
despite recursive and explicit relationships being present in CCSSM algebra specifications,
researchers have not provided answers to such fundamental curricular questions. The study
described in this book will begin to provide answers, but much more research will be needed.
Radford’s Comments on “Objectification” for Middle-School Algebra

Luis Radford’s (2003, 2004, 2006, 2011) discussions of semiotic means of
objectification have been part of his attempt to provide a theoretical base with respect to how
early-algebra students give meanings to symbolic algebraic expressions as they proceed from
arithmetic operations to algebraic abstractions in tasks of the pattern and modeling variety.
According to Radford (2003), the process of objectification takes place through three
successive generalization processes which finally produce what he calls “symbolic
mathematical objects.” Radford (2006) called his three processes “factual generalization” (p.
9), “contextual generalization” (p. 12), and “symbolic generalization” (p. 13). Of those three
processes, he regarded the first two as pre-symbolic generalizations, and the third as true
generalization. We now look at the meanings Radford attached to the three processes.
Factual generalization. With factual generalization, students do not go beyond
particular figures, but realize that there are some commonalities between consecutive figures.
Radford (2006) called a generalization of this kind an “arithmetic generalization” (p. 9). For
example, in the pattern shown in Figure 5.2, a factual generalization would allow the students
to find the number of matchsticks in any particular figure (e.g., Figure 10, or Figure 25, etc.)
without counting the matchsticks one after the other.
Figure 5.2. A sequence of figures inviting a generalization of a relationship between the

number of a figure and the corresponding number of matchsticks.
Although these types of actions are abstracted in the form of a numerical scheme,
Radford (2003) does not regard them as being in the realm of algebra. A process of
generalization has begun, but thinking is still “in the realm of arithmetic” (Radford, 2006, p.
10), and a mathematical object has not been recognized and described. Thus, for the number
of matches in Figure 10 say, a student might say, erroneously: “There will be 10 triangles,
and since there are 3 matches for each triangle, there will be 30 matches altogether.” Factual
generalization will not always allow a student to explain accurately how to find the number
of matchsticks in any figure or to find a formula for calculating the number of matchsticks in
Figure n (where n is some unspecified natural number or, depending on the sequence, zero).
Contextual generalization. According to Radford (2006), contextual generalization
occurs when symbols are given more concentrated meanings and there is a tendency to use
fewer signs (words). For example, with respect to Figure 5.2, a student might say: “If you
multiply the number of the figure by 2 and then add 1 you’ll get the number of matches.”
Such a description includes a mixture of mathematical symbols and natural language.
According to Radford (2006), with contextual generalization a sequence is identified verbally
and a relationship between the number of the figure and the number for the relevant property
is identified.
Symbolic generalization. Radford (2006) maintained that at this third stage, students
engage in multi-semiotic activities (i.e., making verbal statements or drawings, developing
formulas, etc.), as they attempt to create symbolic generalizations. A student’s thinking can
shift between recursive thinking and explicit thinking, and it is at this stage that
generalizations step into the realm of algebra. Radford (2006), after calling this process of
noticing, “objectification,” argued that it corresponds to an attempt to identify and describe a
mathematical object (pp. 5–8). In its etymological meaning, “objectification becomes related
to those actions aimed at bringing or throwing something in front of somebody or at making
something apparent” (p. 5). With the matchsticks problem (see Figure 5.2), students can
Scholars’ Preference for the Functional-Thinking Approach 101
formalize a commonality and construct a standard algebraic syntax. They manage to write a
formula representing the number of matchsticks in a “general” figure—as, for example,
n + (n + 1), where n is the number of the figure in the sequence, or as (n + n) + 1. At this
level, learners may not be able to arrive at different symbolic representations, and even if
they can, they might not consider the two expressions as equal because they represent two
different sets of mental actions.
For Radford (2006), by its very nature a sign such as “n + n + 1” can be counted as
algebraic even though it is not expressed in standard form. This sign, Radford maintained,
can be treated as a general statement because the person who wrote it can see the general
through the particular situation. The person has used alphanumerical symbolism and by
doing so appears to be able to build an expression like “n + n + 3,” irrespective of whether
the person can transform that into “n × 2 + 3,” or “2n + 3.”
Post-symbolic generalization. In the main study described in this book, each of the
instruments, the interview questions, the pencil-and-paper testing, and the teaching
intervention called for higher levels of objectification in the context of modeling situations.
Two different types of algebraic syntax were used. The first asked students to identify and
use a recursive formula in order to generalize; and the second asked them to identify, to
notate, and to use an explicit formula. For both types, a subscript notation was required for
the specifications. For example, for the matchsticks problem (see Figure 5.2), students were
expected to learn to write an explicit specification with a subscript notation—like, for
example, Mn = 2n +1. For a recursive specification, they were supposed to identify the rule
(contextual generalization) and write the formula using signs and symbols with subscripts—
for example, a recursive specification might look like “Mn+ 1 = Mn + 2, with M1 = 3.”
Symbolic generalizations and notations usually are first introduced in formal U.S.
education settings in eighth- or ninth-grade algebra classes. The purpose of asking seventh-
graders to learn to comprehend and use complex symbolic notations in the study described in
this book was to investigate whether the students could cope with them at the same time as
they were being asked to make generalizations involving the use of algebraic symbols.
Some Scholars’ Preference for the Functional-Thinking Approach

Luis Radford (2006) is among many recent scholars to display a strong preference for
adopting what they call the “functional-thinking approach” to early algebra education (see
also Blanton, Brizuela, et al., 2015; Blanton, Stephens, et al. 2015; Cai et al., 2011; Carraher,
Schliemann, & Schwartz, 2008; Moss & McNab, 2011). This approach is consistent with the
common-core expectation that sixth-graders should learn to “use variables to represent two
quantities in a real-world problem that change in relationship to one another; write an
equation to express one quantity, thought of as the dependent variable, in terms of the other
quantity, thought of as the independent variable.” Students should learn to “analyze the
relationship between the dependent and independent variables using graphs and tables, and
relate these to the equation” (CCSSM, 2010).
Although the main investigation described in this book included a strong component
relating to the functional-thinking approach in early-algebra education, the study did not
confine itself to that interpretation of early algebra education. Adopting the ideas of Britt and
Irwin (2011), a structure approach to early algebra was also incorporated into the study. With
some tasks, algebraic concepts were introduced without variables denoted by letters—such as
n or x or y—being introduced. The emphasis in this approach was on assisting seventh-grade
students to learn to recognize and use (a) the associative properties for the addition and
multiplication of rational numbers, and (b) the two aspects (“expanding parentheses” and
“factoring”) of the distributive property for rational numbers. As stated earlier in this chapter,
although the structure aspect of early algebra was much emphasized in the common-core
sequence for elementary and middle-school grades (CCSSM, 2010), there have not been
many researchers who have focused on when, or how well, middle-school students learn
those properties.
This present study attempted to place both the functional thinking and the structure
approaches to early algebra under the semiotic umbrellas provided by both Peirce (1998) and
de Saussure (1959)—but especially by Peirce. It also took advantage of the cognitive-structure
literature developed by Johann Friedrich Herbart (1898), the Herbartians (e.g., Charles
DeGarmo, 1889) and by latter-day scholars such as David Ausubel (1968), Robert Gagné (see
Gagné & Merrill, 1990; Gagné & White, 1978), Shlomo Vinner (see, especially, Tall &
Vinner, 1981; Vinner & Hershkowitz, 1980; Vinner & Dreyfus, 1989), and others who have
drawn attention to the concept of “apperception” or “cognitive structure” or “concept image.”
We now outline an attempt to link semiotic theory and cognitive-structure theory with the two
dominant thrusts which can be found in the early-algebra and middle-school algebra literatures.
Cognitive Structures and Individual Learners
In one of his many definitions of a sign, Charles Sanders Peirce (1998) distinguished
between a sign, the object which is being signified, and an interpretant of the sign. He wrote:
I define a sign as anything which is so determined by something else, called its
object, and so determines an effect upon a person, which effect I call its
interpretant. (p. 478)
Peirce claimed that signs can be associated with three inter-related concepts—a sign, an
object, and an interpretant (Campos, 2010). We can think of the sign as the signifier, for
example, a written word or calculation, an oil painting, etc. The object, on the other hand, can
be thought of as being whatever was intended to be signified by whoever created the sign—
like, for example, the meaning of a written sentence. The interpretant, a distinctive feature of
Peirce’s account, is best thought of as the understanding that a particular person has of what
the sign is signifying.
Peirce’s introduction of the concept of an “interpretant” draws attention to the fact that a
sign can mean different things to different people who are attempting to interpret its meaning.
Thus, for example, on seeing 4 × (1/4 × 128) a seventh-grade student might immediately think
“PEMDAS—I’ve got to do what’s in the parentheses first.” But, another student, on seeing the
same sign, might immediately think: “I’ve got to associate the 4 and ¼, and when I multiply
those I’ll get 1, which will then be multiplied by 128.” A third student might think: “OK—the
teacher wants to see if I recognize that the associative property for multiplication should be
used.” Clearly, the second and third ways of thinking are, from a mathematical perspective,
preferable to the first, and that judgment introduces an educational aspect to semiotics. In the
realm of school mathematics, a curricular statement carries the expectation that learners will
learn to give appropriate meanings to signs, where the word “appropriate” is to be interpreted
as meaning “consistent with the goals of the curriculum.” Seen from that perspective, a major
task of teachers of mathematics is to assist learners to recognize signs, give the signs
Cognitive Structures and Individual Learners 103
appropriate meanings, and then do the mathematics demanded by what is signified in the tasks
in which the signs are present.
If we regard mathematics education as a sequence of processes by which students learn to
recognize signs, process their meanings appropriately, and then respond to tasks, and even
create tasks, in which the signs are used, then it becomes important to study how different
students process signs (that is to say, “cognitive inputs”). But that raises the question why
students process the same sign in different ways. The obvious answer to that, of course, is that
with respect to any concept it is likely that different students will have different “cognitive
structures,” and that fact will result in their interpreting the signs in different ways.
A historically influential string of theories, which greatly influenced the authors’
planning for the study, can be linked to the writings on “apperception” by Johann Friedrich
Herbart, the eighteenth- and early nineteenth-century German scholar (Herbart, 1904a,
1904b), and to theories which might be related to Herbart’s apperception put forward by
Jerome Bruner and David Ausubel (in the 1960s), Robert Gagné and his co-workers (during
the period 1975–1995), and Shlomo Vinner and his co-workers (in the 1980s).
According to Herbart (1898), a good teacher should, when planning lessons,
consciously try to take account of existing cognitive understandings and relationships that
can be assumed to exist in learners’ minds. For Herbart, a learner’s apperception with respect
to a stimulus has been generated by the pool of antecedent experiences which that learner has
had. The implication for school education is that a teacher should become fully acquainted
with the mental development of pupils, in order that advantage can be taken of what the
pupils already know, and of how they think. A problem arises, of course, from the fact that
there are many students present in any particular school classroom, and a teacher cannot be
expected to know how each of the pupils will think with respect to what is about to be taught.
Nevertheless, Herbart maintained, in addition to clarifying expected cognitive
outcomes for a lesson, or series of lessons, teachers should reflect on, and build into lesson
plans, the relevant knowledge, skills, images, beliefs, and principles that their students have
already acquired, so that their readiness to learn what is on the agenda for any particular
lesson will be enhanced. In the 1890s and early 1900s that idea was held to be of central
importance by Charles DeGarmo (1900), at Illinois State Normal University—which became
the center of Herbartian studies in North America (Dunkel, 1970). Many years later, some of
Herbart’s ideas were built into theories set out by Aleksei Leont’ev (see Bedney & Meister,
1997), Jerome Bruner (1963) and David Ausubel (1968)—although, there is no record that
any of Leont’ev, Bruner or Ausubel knew much about Herbart or his theory of apperception.
Thus, for example, Ausubel urged teachers to build “advance organizers” into lessons plans,
in order that appropriate cognitive structures might be created and synthesized in learners’
minds. Accordingly, Ausubel (1968) wrote: “The most important single factor influencing
learning is what the learner already knows. Ascertain this and teach him accordingly” (p. vi).
During the course of a mathematics lesson a learner can be confronted with an
abundance of signs which are intended to represent, or signify, curriculum-relevant
mathematical concepts, skills and principles. Herbart, with his theory of apperception,
contended that different learners will interpret the same signs in different ways, depending on
how their minds are initially structured with respect to the signs.
During the period 1975–1995, Robert Gagné and his co-workers (see, e.g., Gagné, 1985;
Gagné & Merrill, 1990; Gagné & White; 1978) maintained that a learner’s long-term memory
comprises a unique set of components which include:
• Verbal knowledge (e.g., the definition of an equilateral triangle);

• Intellectual skills (e.g., how to “complete the square” with quadratic expressions);
• Imagery (e.g., a visual image of an isosceles triangle);
• Episodes (memories of personal events—e.g., remembering a particular occasion
when the learner scored 0 out of a possible 20 on a test on fractions); and
• Attitudes toward aspects of the input.
According to Gagné, a learner’s processing of an input will depend critically on the interaction,
within the learner’s mind, of the input with that learner’s uniquely-configured cognitive
structure with respect to that input. As a result of the interaction of the input with cognitive
structure, an idiosyncratic restructuring of the mind can occur. In a class of 20 learners a
teacher might expect 20 different restructurings, depending on internal reactions to the sign
inputs (see, e.g., Mackenzie & White, 1982; Zhang, Clements, & Ellerton, 2015a, 2015b,
2015c).
In mathematics education in the 1980s, the related idea of a concept image (Vinner &
Hershkowitz, 1980) was accorded center stage. Vinner and his coworkers (Tall & Vinner 1981;
Vinner & Hershkowitz, 1980; Vinner & Dreyfus, 1989) contended that idiosyncratic and
unique links in an individual learner’s cognitive structure determine that learner’s concept
image with respect to any sign. Furthermore, if a teacher attempts to use the sign to convey a
mathematical concept or principle to students then it is likely that that sign will be interpreted
in as many ways as there are students in the class. That viewpoint was confirmed in a series of
mathematics education research investigations conducted in the 1980s (e.g., Dreyfus &
Eisenberg, 1982; Vinner & Dreyfus, 1989). In those studies, student thinking within important
content areas of school mathematics (e.g., fractions, functions, limits of functions) were
analyzed with respect to concept definitions and concept images (Clements, 2014).
Concept-image researchers used the term “concept image” in a way which was consistent
with Herbart’s (1898) theory of apperception. In the words of Tall and Vinner (1981):
We shall use the term concept image to describe the total cognitive structure that is
associated with the concept, which includes all the mental pictures and associated
properties and processes. ... As the concept image develops it need not be coherent
at all times. ... We shall call the portion of the concept image which is activated at a
particular time the evoked concept image. At different times, seemingly conflicting
images may be evoked. Only when conflicting aspects are evoked simultaneously
need there be any actual sense of conflict or confusion. (p. 152)
What might this mean in a seventh-grade mathematics classroom? Consider, for example,
the concept image that a seventh-grade student, John say, might evoke when exposed to the
1
sign “Without using a calculator, find the value of 4 × ( × 128).” Imagine that during the
4
initial exposure the following “inner” conversation took place in John’s mind: “Oh, it’s a
calculation, and there are parentheses involved. We had this kind of problem a couple of weeks
ago, and I got full marks on the test. I’ll have to use PEMDAS—which means that I’ll have to
work out what’s inside the parentheses first. Oh no, it’s a multiplication sum involving a
fraction. I hate fractions. How will I find one-fourth times 128? When I was in Grades 4 and 5
Mr. Jones and Mrs. Tomlinson taught me to think of ¼ as being one piece of a circle after it has
been divided into four equal pieces. Hmm ... we’ll need 128 of those smaller bits. How can I do
Components of Cognitive Structures 105
1
that? Oh, that’s right, I’ve got to multiply by 128 and if you want to do that the first thing
4 1 128
you should do is write the 128 as a fraction. So it’ll be times . How can I do that?
4 1
Should I cancel by dividing the 4 into the 128? ...”
Notice that in this imaginary inner conversation, John’s thinking was initially captured
by the sign (1/4 × 128). The mnemonic PEMDAS was not mentioned in the question, but the
parentheses prompted John to think about PEMDAS. This was linked to, indeed reinforced
by, John having successfully applied PEMDAS on a recent test—which is an example of
what Gagné and White (1978) called “episodic” thinking. Then the issue of skills arose in
1
John’s mind—how could he find times 128? This prompted John to recall another earlier
4
episode in which he had represented ¼ by a circular region—Gagné and White (1978) would
have regarded this as evoking a visual image—but then John struggled with the idea of
finding what 128 of the circular regions would be. Within his sequence of thought he had
cause to reflect that he “hated fractions.”
In the above description of an imagined scenario, the sequence and content of the
1
thinking stimulated by the sign “4 × ( × 128)” involved:
4
• Memory of verbal information (concerning PEMDAS);
• An attempt to recall appropriate skills (“How do I find the value of 1/4 × 128?”);
• Memory of a relevant past episode (getting a perfect score on a test);
• Imagery (evoking part of a circular region in the mind to represent a fraction);
• Expression of an attitude (“I hate fractions”).
Although the above discussion is not based on data generated by an actual student, in the
study described in this book similar data were, in fact, generated, and analyses of those data
enabled important features of seventh-grade students’ concept images, for both elementary
algebraic structures and for modeling real-world situations using algebra, to be identified.
Furthermore, the design of the study enabled salient features of concept images to be
identified for 28 of the 32 participating students, both before and after teaching interventions,
and with respect to both algebraic structure and functional thinking. One of the aims of the
study was to be able to describe effects of the teaching interventions on concept images.
Seen from the vantage point of observing students in seventh-grade algebra classrooms,
it is not difficult to reconcile the concept of apperception, and its application to analysis of
educational data, with semiotic theories. Marx Wartofsky (1979), a philosopher and historical
epistemologist, emphasized that what one notices, and what escapes one’s attention, depend
to a large extent on one’s cultural, educational, and episodic backgrounds. In other words,
what a learner notices in a particular context depends on the cognitive structure of the long-
term memory that the learner brings to the context. Thus, for example, with regard to the
second question in Figure 5.1, in which the “Visiting Old Houses” task was elaborated,
seventh-grade students who had never been exposed to tables in which they were expected to
make a cognitive leap to the “nth case” obviously wondered what they were being asked to
do. They wondered, with strong justification, why there were “spaces with three dots” in the
table. For their mathematics teacher, however, the three dots in Figure 5.1 were signs which
invited the reader to enter a world of generalized mathematical objects. Seen from that
perspective, the composite sign in Figure 5.1 could signify to the seventh-grade students the
mathematical object which the teacher would like it to signify only if the intended meaning
of the sign had been learned by the students. And, even if that meaning had been learned by a
student, there was no guarantee that that student would always interpret the sign according to
what he or she had learned.
Thus, the issue becomes a potentially complex educational one. An exactly similar
argument can be constructed with respect to the composite sign, discussed earlier: “Without
1
using a calculator, find the value of 4 × ( × 128).” An individual seventh-grade student’s
4
cognitive structures must have a large influence on how that student links signs and symbols
in early algebra to mathematical objects which are specified in the curriculum. An important
role for the teacher is to provide experiences for students which will extend and coordinate
their cognitive structures so that they will learn to recognize, comprehend, and use standard
signs and conventions which underlie knowledge, concepts, skills, and principles which are
represented in intended curricula (Herbart, 1898).
Literature Which Helped Frame the Design of the Teaching Intervention

In preparation for the main study, two large pilot studies were carried out (Kanbir,
2014, 2016). The second of those studies involved three classes of seventh-grade students
and one teacher—who was the day-to-day mathematics teacher for each of the classes
(Kanbir, 2016). The following three important lessons were learned as a result of analyzing
data from the pilot study:
1. It should not be assumed that seventh-grade teachers and students are ready to use
sixth and seventh-grade CCSSM content standards with respect to algebra.
2. The fact that the student participants in the pilot study were not randomly allocated
to groups meant that the three groups differed in their initial knowledge and
understandings of algebraic notations and concepts, and this could have had an
impact on the students’ reactions to the intervention tasks. This realization pointed to
the need for random allocation to intervention groups in the main study.
3. The method of professional development for the teacher in the second pilot study
was one-on-one instruction based on notes that had been especially prepared for the
occasion. As it turned out, the volume and the complexity of the material tended to
overwhelm the teacher.
We have also come to believe, somewhat counter-intuitively, that the difference
between arithmetic and algebra cannot be specified entirely in terms of notations, and that
algebraic reasoning is not confined to activities which make use of letters as pronumerals. In
the main study, the first author examined two contexts in which, it seemed to him, algebraic
reasoning occurred. In some of the tasks in the first context, algebraic thinking took place
without the presence of pronumerals; and in the second context, pronumerals were used to
denote variables representing independent and dependent variables. Even though the timing
of the intervention was relatively short in comparison with many longitudinal studies—the
total data-collection period occupied just 24 weeks—it was interesting to investigate the
growth of algebraic thinking toward generalization during that period. In particular, we set
out to find evidence on the issue of whether structural thinking supports functional thinking,
or vice-versa (or neither).
Components of Cognitive Structures 107
For the main study, which was not conducted at the same school as the pilot studies, it
was decided that a “modeling” approach to professional development, along the lines
suggested by Joyce and Showers (2002), would be adopted. The two participating teachers
(Mr. X and Mr. Y) agreed to observe the authors of this book (all three are experienced
mathematics teachers) teaching Grade 8 students at School W, and they would then model
their teaching within their seventh-grade “workshops” on what they had observed.
The teaching method which was adopted by the three authors, when preparing the
model workshops, was based on the “Modes of Communication” approach (see Table 5.1)
developed by Del Campo and Clements (1987, 1990), and reported in Ellerton and Clements
(1991). Receptive language involves the “processing of someone else’s communication” and
expressive language the use of one’s “own language” (Clements & Del Campo, 1987; Ellerton
& Clements, 1991).
Table 5.1
Receptive and Expressive Modes of Communication (Del Campo & Clements, 1987, p. 12)
Language Mode Receptive Language Expressive Language
Spoken Listening Speaking
Written Reading Writing
Pictorial Interpreting diagrams, pictures Drawing
Active Interpreting others’ actions Performing, Demonstrating,

Problem Solving, Gesturing
Imagined — Imagining (Creating), Problem

Posing
For the intervention, the aim was for the model workshops, with seventh- and eighth-grade
students at School W, to show how expressive language forms could become central in middle-
school algebra classrooms which dealt with the structure and functional-thinking approaches
to middle-school algebra. Then, Mr. X and Mr. Y would aim at assisting all participating
students not only to learn to use appropriate expressive language in the course of
developing early-algebra concepts and principles related to the associative properties and the
distributive property for rational numbers, but also to learn to use algebra in expressive ways in
order to model real-life contexts. By developing a sense of ownership of their learning, it was
hoped that the students would move significantly toward the reification of key early-algebra
ideas (Sfard, 1991).
The main purpose of the model workshops would be to show how beginning algebra
students could become engaged in expressive activities which would assist them to learn to
interpret and recognize and use, appropriately, the main “signs” of early algebra, especially
those relating to “structure” and “modeling.” Learners would be expected to become actively,
meaningfully, and creatively, engaged in the construction of appropriate concepts,
relationships, and principles, to the point where they would feel some ownership of the
mathematics that they were studying. The students would regularly be given the opportunity to
discuss ideas and concepts in small groups, to construct answers, to pose problems, and then to
report their findings to others in their class. They would be invited to own the mathematics, and
it was hoped that they would feel that it was a natural thing that they should attempt to
convince others of the sense and value of what they were learning.
With the semiotic view, an ultimate form of algebraic knowledge is to be associated
with objectification. According to Radford (2008), this process of mathematizing involves
various levels of awareness and is manifested by signs such as words, gestures, pictures,
graphs and symbols. In the study described in this book, the research team created and
implemented algebraic tasks and observed seventh-grade students using variables and quasi-
variables (see Warren & Pierce, 2003) in expressive classroom environments. In the process,
the students were expected to go through different layers of semiotic awareness (Radford,
2011).
This view of teaching is consistent with Anna Sfard’s (2008) theory of commognition
in which explanations, interpretations, definitions, etc., become part of an attending/intending
flow with at least three aspects.
1. Learners are engaged in observing, interpreting, experiencing and explaining;
2. Teachers point learners towards an object, the acquisition of which might be
thought of as the aim of the teaching/learning exercise; and
3. The processes of interpreting, observing, explaining constantly involve constructive
mental activity for learners. (Sriraman, 2009)
The workshop environment provided in the research study was also consistent with the
five dimensions that Alan Schoenfeld (2013) claimed mark the robustness of a classroom
learning environment—(a) its mathematical focus, (b) its cognitive demand, (c) the support it
provides to the diverse range of students in the class to be actively engaged, (d) its agency
(the extent to which it provides opportunity for students to make conjectures, give
explanations and arguments, and to develop “voice”), and (e) its use of assessment which
challenges students to reason.
Before the current study took place the research team did not assume that all, or even
most, of the seventh-grade participants would make great progress with generalizing. The
sparse literature on student understandings of the structural properties of real numbers had
not convinced the team that students would make much progress in the relatively brief
number of workshop sessions—over a period of six weeks—that would be devoted to
structure. So far as the effects of the modeling workshops were concerned, Lannin, Barker,
and Townsend (2006) reported that some students found it difficult to move from the
successful use of recursive rules to explicit rules, and that students often preferred to develop,
and continue to use, recursive rules. Would that description also apply for students in the
current study?
Research Questions
Having completed the pilot study, having prepared relevant pencil-and-paper materials,
and artifacts, and having been part of the preliminary teaching in which those materials were
used with eighth-grade students at School W, the research team framed the following six
questions as the main questions to be addressed in the study:
1. What did the participating Grade 7 students know about each of the associative
property for addition, the associative property for multiplication, and the
distributive property, before the intervention workshops took place?
Research Questions 109
2. To what extent were the participating Grade 7 students able to recognize patterns
and to model relationships by using variables before the intervention workshops
took place?
3. What changes in the knowledge and understanding of participating students with
respect to structure and modeling were evident around the middle of the
intervention period (when each student had participated in either the structure or the
modeling workshops, but not both)?
4. Immediately after the two groups had participated in both the structure and
modeling workshops, were there statistically significant differences between their
mean gain scores in their understandings of structure and modeling? Also, what
were the Cohen’s d effect sizes (Cohen, 1988) for the two groups for the first half
and for the second half of the intervention?
5. Immediately after both the structure and modeling workshops were completed,
were there educationally noticeable differences between the concept images of the
students, with respect to the concept of a variable, in comparison with the concept
images that the students had displayed before the intervention began?
6. Twelve weeks after both the structure and modeling workshops were completed,
were there statistically significant differences between the two groups’ mean gain
scores with respect to the retention of what had been learned in regard to
understanding of (a) structure, and (b) modeling?
Further discussion of these research questions will occur in later chapters of this book, and in
Chapter 9 summary answers will be given to each of the questions.
Of interest, is the fact that our identification of the six research questions came after we
had gained approval for the study to be conducted at School W, after we had completed our
historical framework (see Chapter 2), and after we had reviewed pertinent literatures.
Obviously, before we gained final approval for the research to be conducted at School W,
and before we carried out a full review of what we believed to be the most pertinent
literature, we knew that we wanted to conduct an intervention study aimed at improving
seventh-grade students’ understandings of school algebra. However, we wanted the final
forms of the research questions to emerge not only from what we found out about School W
and the seventh-grade participating students from Mr. X and Mr. Y before the classroom
interventions would take place, but also from what we learned as a result of carrying out the
historical analysis, the literature review, and the pre-intervention assessments. Despite the
expectation of many traditional researchers that research questions should be finalized soon
after the details of a study are conceptualized, we believe that the decision to delay the final
formulation of the research questions until relevant preliminary data were known, and a
tentative understanding of the history and pertinent literature was achieved, was totally in
line with the principles of design research.
References
Ausubel, D. P. (1968). Educational psychology: A cognitive view. New York, NY: Holt,
Rinehart and Winston.
Bedney, G., & Meister, D. (1997). The Russian theory of activity: Current applications to
design and learning. New York, NY: Psychology Press.
Blanton, M. L., Brizuela, B. M., Gardiner, A. M., Sawrey, K., & Newman-Owens, A. (2015).
A learning trajectory in 6-year-olds’ thinking about generalizing functional
relationships. Journal for Research in Mathematics Education, 46(5), 511–558.
Blanton, M. L., & Kaput, J. J. (2011). In J. Cai & E. Knuth (Eds.), Early algebraization: A
global dialogue from multiple perspectives (pp. 5–23). Heidelberg, Germany: Springer.
Blanton, M. L., Stephens, A., Knuth, E., Gardiner, A. M. Isler, I., & Kim, J-S (2015). The
46(1), 39–87.
Britt, M. S., & Irwin, K. C. (2011). Algebraic thinking with and without algebraic
representation: A pathway for learning. In J. Cai & E. Knuth (Eds.), Early
Brown, A. L. (1992). Design experiments: Theoretical and methodological challenges in
creating complex interventions. Journal of the Learning Sciences, 2, 141–178.
Bruner, J. (1963). The process of education. Cambridge, MA: Harvard University Press.
Cai, J., & Knuth, E. (Eds.). (2011). Early algebraization: A global dialogue from multiple
Cai, J., Moyer, J. C., Wang, N., & Nie, B. (2011). Examining students’ algebraic thinking in
a curricular context: A longitudinal study. In J. Cai & E. Knuth (Eds.), Early
Campos, D. G. (2010). Peirce’s philosophy of mathematical education: Fostering reasoning
abilities for mathematical inquiry. Studies in Philosophy and Education, 29, 421–429.
Cañadas, M. C., Brizuela, B. M., & Blanton, M. (2016). Second graders articulating ideas
about co-variation with linear functions. Journal of Mathematical Behavior, 41, 87–103.
Carraher, D. W., Schliemann, A. D., & Schwartz, J. L. (2008). Early algebra is not the same
as algebra early. In J. J. Kaput, D. W. Carraher, & M. Blanton (Eds.), Algebra in the
early grades (pp. 235–272). New York, NY: Lawrence Erlbaum Associates.
CCSSM. (2010). Common Core State Standards for Mathematics. Washington, DC: Authors.
Clements, M. A. (2011). Design research: Its potential for meeting Brunei Darussalam’s
educational needs. In I. Cheong (Ed.), Design research in education: Empowering
teachers, learners and researchers (pp. 22–40). Gadong, Brunei Darussalam: Universiti
Brunei Darussalam.
Clements, M. A. (2014). A historical overview of visualization and visualizing in
mathematics. In M. Fried & T. Dreyfus (Eds.), Mathematics and mathematics
education: Searching for common ground (pp. 177–192). New York, NY: Springer.
in children’s understanding of fractional concepts across embodiments, Grades 2
through 5. In J. Novak (Ed.), Proceedings of the Second International Seminar on
Misconceptions and Educational Strategies in Science and Mathematics (Vol. 3, pp.
98–110). Ithaca, NY: Cornell University.
Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Hillsdale,
NJ: Lawrence Erlbaum Associates.
Common Core Writing Team. (2011, May 29). Progression for the Common Core State
Standards in Mathematics (draft). Comment at commoncoretools.wordpress.com. 1.
DeGarmo, C. (1889). Essentials of method. Boston, MA: Heath.
DeGarmo, C. (Ed.). (1900). Apperception: A monograph on psychology and pedagogy.
Boston, MA: D. C. Heath & Co.
de Saussure, F. (1959). Course in general linguistics. New York, NY: The Philosophical
Library.
de Saussure, F. (1974). Cours de linguistique générale (C. Bally & A. Sechehaye, Eds.).
Paris, France: Payot.
Dreyfus, T., &, Eisenberg, T. (1982). Intuitive functional concepts: A baseline study on
intuitions. Journal for Research in Mathematics Education, 13, 360–380.
Ellerton, N. F., & Clements, M. A. (1991). Mathematics in language: A review of language
factors in mathematics learning. Geelong, Australia: Deakin University.
Fried, M. (2008). History of mathematics in mathematics education: A Saussurean
perspective. The Montana Mathematics Enthusiast, 5(2–3), 185–198.
Gagné, R. (1985). The conditions of learning (4th ed.). New York, NY: Holt, Rinehart &
Winston.
Gagné, R. M., & Merrill, M. D. (1990). Integrative goals for instructional design.
Educational Technology Research and Development, 38(1), 23–30.
Gagné, R. M., & White, R. T. (1978). Memory structures and learning outcomes. Review of
Educational Research, 48(2), 187–222.
Herbart, J. F. (1898). Letters and lectures on education (H. M. Felkin & E. Felkin, Trans.).
London, UK: Swan Sonnenschein & Co.
Hewitt, D. (1998). Approaching arithmetic algebraically. Mathematics Teaching, 163, 19–29.
Jeon, K. (2012). Reflecting on PEMDAS. Teaching Children Mathematics, 18(6), 370–377.
Joyce, B., & Showers, B. (2002). Student achievement through staff development (3rd ed.).
Alexandria, VA: Association for Supervision and Curriculum Development.
Kaput, J. (2008). What is algebra? What is algebraic reasoning? In J. Kaput, D. Carraher, &
M. Blanton (Eds.), Algebra in the early grades (pp. 1–18). Mahwah, NJ: Lawrence
Erlbaum/Taylor & Francis Group, & National Council of Teachers of Mathematics.
Kieran, C. (2006). Research on the learning and teaching of algebra. In A. Gutiérrez & P.
Boero (Eds.), Handbook of research on the psychology of mathematics education: Past,
present and future (pp. 11–49). Rotterdam, The Netherlands: Sense Publishers.
Kieran, C. (2011). Overall commentary on early algebraization: Perspectives for research and
teaching. In J. Cai & E. Knuth (Eds.), Early algebraization: A global dialogue from
multiple perspectives (pp. 579–593). Heidelberg, Germany: Springer.
Lannin, J. K., Barker, D. D., & Townsend, B. E. (2006). Recursive and explicit rules: How
can we build student algebraic understanding? The Journal of Mathematical Behavior,
25, 299−317.
Mackenzie, A. A., & White, R. T. (1982). Field-work in geography and long-term memory
structures. American Educational Research Journal, 19(4), 623– 632.
Moss, J., & McNab, S. L. (2011). An approach to geometric and numeric patterning that
fosters second grade students’ reasoning and generalizing about functions and co-
variation. In J. Cai & E. Knuth (Eds.), Early algebraization: A global dialogue from
Otte, M. (1998). Limits of constructivism: Kant, Piaget and Peirce. Science and Education,
7(5), 425–450.
Otte, M. F. (2011). Evolution, learning and semiotics from a Peircean point of view.
Press.
Presmeg, N. (2014). Mathematics education research embracing arts and sciences. In M.
Fried & T. Dreyfus (Eds.), Mathematics and mathematics education: Searching for
common ground (pp. 361–378). New York, NY: Springer.
Radford, L. (2003). Gestures, speech, and the sprouting of signs. Mathematical Thinking and
Learning, 5(1), 37–70.
Radford, L. (2004). Syntax and meaning. In M. J. Høines & A. B. Fuglestad (Eds.),
Proceedings of the 26th Conference of the International Group for the Psychology of
Mathematics Education (Vol. 1, pp. 161–166). Bergen, Norway: International Group
for the Psychology of Mathematics Education.
Radford, L. (2006). Algebraic thinking and the generalization of patterns: A semiotic
perspective. In S. Alatorre, J. L. Cortina, M. Sáiz, & A. Méndez (Eds.), Proceedings of
the 28th Annual Meeting of the North American Chapter of the International Group for
the Psychology of Mathematics Education (Vol. 1, pp. 2–21). Mérida, México:
International Group for the Psychology of Mathematics Education.
Radford, L. (2008). Connecting theories in mathematics education: Challenges and

possibilities. ZDM—The International Journal on Mathematics Education, 40, 317–
327.
Radford, L. (2011). Grade 2 students’ non-symbolic algebraic thinking. In J. Cai & E. Knuth
(Eds.), Early algebraization: A global dialogue from multiple perspectives (pp. 303–
322). Heidelberg, Germany: Springer.
Radford, L. (2015). The epistemological foundations of the theory of objectification.
Isonomia (online journal http://isonomia.uniurb.it/epistemologica), 127–149.
Schoenfeld, A. H. (2013). On forests, trees, elephants and classrooms: A brief for the study
of learning ecologies. ZDM—The International Journal on Mathematics Education,
45(3), 491–495.
Sfard, A. (1991). On the dual nature of mathematical conceptions: Reflections on processes
and objects as different sides of the same coin. Educational Studies in Mathematics,
22(1), 1–36.
Sfard, A. (2008). Thinking as communicating. Cambridge, UK: Cambridge University Press.
Sriraman, B. (2009). What’s all the commotion about commognition? The Montana
Mathematics Enthusiast, 6(3), 541–544.
Sriraman, B., & Lee, K-H. (2011). Commentary on Part II. In J. Cai & E. Knuth (Eds.), Early
Tall, D., & Vinner, S. (1981). Concept image and concept definition in mathematics with
particular reference to limits and continuity. Educational Studies in Mathematics, 12(2),
151–169.
Taylor-Cox, J. (2003). Algebra in the early years? Yes! Young Children, 58(1), 14–21.
Vinner, S., & Dreyfus, T. (1989). Images and definitions for the concept of function. Journal
for Research in Mathematics Education, 20(4), 356–366.
Vinner S., & Hershkowitz, R. (1980). Concept images and some common cognitive paths in
the development of some simple geometric concepts. In R. Karplus (Ed.), Proceedings
of the Fourth International Conference for the Psychology of Mathematics Education
(pp. 177–184). Berkeley, CA: International Group for the Psychology of Mathematics
Education.
Warren, E., & Pierce, R. (2003). Learning and teaching algebra. In B. Perry, G. Anthony &
C. Diezmann (Eds.), Research in mathematics education in Australasia, 2000–2003
(pp. 291–312). Flaxton, Australia: Post Pressed.
Wartofsky, M. (1979). Models, representation and the scientific understanding. Dordrecht,
The Netherlands: D. Reidel.
Wu, H. (2001). How to prepare students for algebra. American Educator, 25(2), 10–17.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015a). Enriching student concept images:
Teaching and learning fractions through a multiple-embodiment approach. Mathematics
Education Research Journal, 27(2), 201–231.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015b). Differences in student visual images
of fractions before and after a teaching intervention: Does a multiple-embodiment
approach to instruction make a difference? Mathematics Education Research Journal,
27(2), 233–261.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015c). Engaging students with multiple
models of fractions. Teaching Children Mathematics, 22(3), 138–147.
Chapter 6
Research Design and Methodology
Abstract: The main study featured a mixed-method design, with complementary quantitative
and qualitative data being gathered and analyzed. Since random allocation of students to two
groups occurred, it was legitimate for null and research hypotheses to be formulated for the
quantitative analyses, and those hypotheses are carefully defined in this chapter. One of the
important challenges was to identify the population to which inferences would be made.
Details relating to the development of appropriate pencil-and-paper tests and an interview
protocol are also given, as are details relating to the calculation of Cohen’s d effect sizes.
Keywords: Campbell and Stanley design, Effect size, Expressive understanding, Joyce and
Showers, Newman interviews, Receptive understanding
As Lesh and Sriraman (2005) have asserted, design research has the potential to
develop useful knowledge about mathematics education and to provide support for policy
makers and curriculum and instructional designers. The research team for the main study
identified a problem—“Why do so many school students find it difficult to learn algebra
well?”—and solving that problem was clearly important for all involved in the study at
School W. School administrators and teachers wanted to improve the quality of their
students’ knowledge and understandings of algebra, and the research team was keen to
design and conduct a study which would not only bring about such an improvement, but
would also make a lasting contribution to the improvement of school algebra. As it turned
out, a sequence of actions was taken in order to illuminate, and ultimately to solve, the
problem, at least as the problem related to seventh-graders at School W. The research team
was kept busy throughout the study designing, developing, implementing, and appropriately
modifying the actual intervention program, and an objective scheme for evaluating the
effects of the intervention was devised.
The bundle of theories which framed the research investigation, and the design-research
approach adopted for the main study, were not chosen without reflection for, as Cobb,
Confrey, diSessa, Lehrer, and Schauble (2003) have emphasized, design research has a
“highly interventionist nature” (p. 10). According to Stylianides and Stylianides (2013), the
fact that there is “a high correlation between classroom-based interventions and studies
following design experiment methodology should not come as surprise” (p. 336).
The preparation phases for the main study included the conception and development of a
program, and associated materials, aimed at facilitating the ultimate goal of addressing difficult
issues relating to the teaching and learning of algebra. Mr. X and Mr. Y were involved as
workshop leaders, for it was recognized that it was desirable that they have important roles in
the production of knowledge by the participating students. Borko (2004) insisted that the design
of interventions should bring researchers and other stakeholders together, through multitiered
designs, so that they will be best placed to adapt and refine existing professional-development
programs and be able to study the impacts of, and on, individual teachers and student outcomes.

116 Ch. 6: Research Design and Methodology
Ann Brown (1992), of the University of California at Berkeley, was one of the pioneers
of design research studies in education. Her Theoretical and Methodological Challenges in
Creating Complex Interventions in Classroom Settings focused on those types of education
research studies in which real instructional environments (contexts and materials) are
deliberately modified. According to Cobb and Gravemeijer (2008), the primary goal of design
research is not to demonstrate or assess whether the envisioned learning hypothesis works—
rather, they argued, “the purpose is to improve the envisioned trajectory developed while
preparing for the experiment by testing and revising conjectures about both the prospective
learning process and the specific means of supporting it” (p. 73). With the study described in
this book—that is to say, “the main study”—the goal was not to compare two different
approaches to middle-school algebra, which, in Slavin’s (2004) terms, would be merely be an
“x versus y comparison” (p. 27). Rather, the aim was to improve teachers’ and researchers’
knowledge about the teaching and learning of middle-school algebra. In order to do that,
relevant quantitative and qualitative data would be gathered, analyzed, and interpreted.
In the main study, members of the research team were not only involved in the selection
of the most appropriate theories which would be utilized, but also in deciding how those
theories should influence the planning, intervention, and evaluation phases of the study. The
theoretical frames chosen for the study were outlined in Chapter 5.
Members of the research team needed to gather data on the “starting points” of the
participating students in order that they would be well-placed to investigate the students’
development of increasingly sophisticated forms of algebraic reasoning (Cobb &
Gravemeijer, 2008). Team members needed to be in a position to document the shifts in the
seventh-grade students’ reasoning, and to do that they knew that they would need multiple
data sources (e.g., from interviews, classroom observations, questionnaires, and pre-teaching,
mid-intervention, post-teaching, and retention performance data). In order to achieve all of
that, it would be necessary to gather and analyze both quantitative and qualitative data. That
approach was in line with the position that the development of sound methods for algebra
teaching and learning should especially look to the findings of mixed-method research.
Setting up the Intervention
Two Pilot Studies

As part of the preparation for the main study the first author (Kanbir) conducted two
pilot studies with seventh- and eighth-grade students, and their mathematics teacher, in a
middle school in the midwest of the United States of America. The school at which the pilot
studies were conducted was not School W, where the main study would be conducted. These
pilot studies took place during the fall of 2014 and the spring of 2015.
The first pilot study: Two different approaches to the concept of a variable. In the
first pilot study (Kanbir, 2014), the first author of this book created two instruments—one
made use of visual-number patterns, and the other of structural properties of rational numbers.
Task-based interviews were carried out for the purpose of identifying the ways the participating
eighth-grade algebra students were thinking about important algebra concepts and principles—
such as a variable, generalization, and algebraic structure. Data were collected from 76 eighth-
grade students enrolled in algebra and pre-algebra courses. The main pencil-and-paper research
instrument comprised 25 items: 7 were concerned with the concept of a variable, 9 with
generalizing from visual-number patterns, and 9 with the structural properties of rational
The Pilot Studies 117
numbers. One of the research aims was to find the Pearson product-moment correlations
between scores on the subtests comprising the three types of items. In addition, data from task-
based interviews were analyzed for the purpose of examining, qualitatively, the ways the
students thought about key aspects of elementary algebra.
Analysis of data from the first pilot study revealed that correlations between the scores
on the subtests comprising the three types of items were very low. That finding suggested that
middle-school algebra programs based entirely on numerical or geometrical patterns, or
entirely on structural aspects of rational numbers, would be inadequate. Analysis of data from
six 30-minute interviews showed that the eighth-grade interviewees rarely viewed or used
algebraic expressions as mathematical entities in their own right. Most of the students
maintained operational views and had great difficulty generalizing. In modeling contexts,
they struggled to identify and describe even apparently simple recursive rules.
As part of the first pilot study, the researcher also created an Algebra Readiness Test
(ART) which was intended to measure eighth-grade students’ readiness to begin studying
secondary-school algebra. The Cronbach alpha reliability index for this ART instrument was
calculated to be 0.84. But, ART was not used in the main study, so did the first pilot study
have any influence on the main study? The answer to that question is “Yes.”
The first pilot study was useful insofar as it alerted the writer to the fact that some
eighth-grade students are not in the least aware of structural properties of rational numbers.
Hung-Hsi Wu (2007), a University of California at Berkeley mathematician, may well have
overstated his case when he claimed that “by the sixth grade most students already know
about the associative and commutative laws of addition and multiplication” (p. 4).
The first pilot study also provided the first author of this book with experience in
interviewing middle-school students, and during the interviews it became clear that some
eighth-graders had not developed notations for representing sequences—and that without
prompting, they might struggle to identify even simple recursive rules that one might have
expected they would be able to identify easily when examining given tables of values.
Furthermore, none of the six interviewees—two of whom were regarded as top students—
was able to identify explicit rules which generalized patterns implicit in given tables of
values.
The second pilot study: Three different approaches to developing seventh-grade
students’ algebraic reasoning. The second pilot study (Kanbir, 2016), which was held at the
same school as the first, lasted 10 weeks. The study applied three different intervention
approaches with students in three seventh-grade classes at a midwestern middle-school. The
seventh-grade mathematics teacher at the school was also an active participant. During the
period of research for the second pilot study, each of the classes followed a different approach
to algebra—these approaches were termed “visual-number,” “modeling,” and “structural.”
The study explored how instruction incorporating the three different approaches to
elementary algebra affected student thinking about algebra. Pre-teaching and post-teaching
pencil-and-paper data were collected, the instruments being the Algebra Readiness Test (the
pencil-and-paper test instrument developed in the first pilot study), a Visual-Number test, a
Modeling test, and a Structure test. Eighteen students (six from each of the three classes) were
interviewed on a 1-1 basis, on two occasions, with the interview protocol recommended by
Newman being used (see Clements, 1980). The goals of the interviews were to ascertain how
the students were thinking about task-based, pencil-and-paper algebra tasks, and to investigate
the students’ functional and structural thinking before and after the classroom interventions.
Analyses of data from the second pilot study indicated that whereas the mean gain
scores for the modeling and visual-number groups were significantly different from zero, the
mean gain score for the structure group increased only slightly. Analysis of data from this
pilot study suggested that for each of the three groups there was no statistically significant
difference between pre- and post-intervention scores for “algebra readiness.” Such a
statement is made with a degree of caution, however, because this second pilot study did not
feature random allocation to groups, and the “modeling” class was regarded, within the
school, as having the best seventh-grade algebra students in the school.
Analysis of interview data generated by the second pilot study indicated that as a result
of participation in the intervention lessons there were changes in students’ concept images of
a variable and in their understanding related to the meaning of expressions and equations
(Kanbir, 2016; Vinner & Dreyfus, 1989).
For the second pilot study, students worked in whole-class environments and the
teacher’s instruction was mostly direct and not dialogical. All the lessons were observed by
one or more of the three authors of this book, who agreed that the seventh-grade students
would probably have learned more if there had been more challenging interactions between
the students and the teacher.
Although both pilot studies generated informative data, it became clear that a different
teaching approach, and a new approach to professional development, were needed for the
main study. The three authors of this book participated in the second pilot study, and that
proved to be important for it enabled them to practice different data-collection and interview
techniques. A total of 36 interviews were conducted, and the authors of this book led
intensive professional development sessions, observed classrooms, analyzed student artifacts,
and took many notes as the teacher taught the three classes. This allowed the three authors to
become more confident in relevant qualitative data-collection techniques and to evaluate, and
modify, instruments. It also helped them plan appropriate lesson sequences and professional
development sessions, and improve the interview protocol for the main study (Kanbir, 2016).
One of the most important general findings arising from the pilot studies was that
analysis revealed that it could not be assumed that seventh-grade students would know much
about structural aspects of rational numbers or about the concept of a variable. When solving
linear equations, the pilot-study students tended to adopt a “find the missing number”
approach which, incidentally, was advocated in the common-core mathematics sequence. By
such an approach, an equation like 3x + 1 = 13 will be interpreted as “what number should I
put in the place of x.” A more productive approach, the research team believed, was to regard
the x in the equation 3x + 1 = 13 as a variable, and the task of solving the equation was not
only to find the value (or values) of x which would make the open sentence a true statement,
but also those values of x which would make it false (Ellerton & Clements, 2011).
Another finding from the pilot studies was that there did not seem to be much
commonality in the ways students thought about structure tasks and about modeling tasks in
which they were asked to identify rules for given patterns. In other words, although both came
under a “beginning algebra” umbrella, the two types of tasks seemed to require different kinds
of thinking. Recognition of that state of affairs seemed to be important, because there is only so
much time allowed for algebra in middle-school mathematics programs. It seemed likely that
students following mainly an algebra sequence which focused on structure would not learn the
same things as students who followed a modeling sequence which focused on patterns and
relationships. That raised the question whether it would be possible to achieve a balance
Preparing for the Main Study 119
whereby students would develop their knowledge and understandings for both kinds of algebra.
Might there be an approach whereby the two types of algebra would be linked? After much
reflection, the research team decided that the concept of variable was equally important within
both approaches, and that recognition seemed to provide a key for bringing the two
components of middle-school algebra together.
Preparing for the Main Study

In an attempt to be well placed to understand interactions between classroom practice
and student learning in the main study, it was decided that an experimental group design
should be used. It is well recognized that randomly assigning students to study groups is a
signature characteristic of true experimental intervention studies in education (Campbell &
Stanley, 1963; Gersten, Baker, & Lloyd, 2000; Mosteller & Boruch, 2002), and therefore it
was decided that if School W would agree to it, random allocation of students to
experimental groups would be a feature of the study design.
In a typical school setting it is not easy to achieve random allocation to groups for a
research study—for that is usually at odds with standard organizational requirements of the
school. More than half a century ago, Campbell and Stanley (1963) argued that the
education research community was struggling with ways to negotiate a balance between
rigorous research design and satisfactory research in school classroom environments. The
problem identified by Campbell and Stanley has never gone away, and in 2002 the Institute
of Education Sciences established the “What Works Clearinghouse” (WWC) with the
specific goal of assessing and improving the quality of education research. The WWC
made a strong commitment to the randomized trial, which was the only method which met
its standards without reservation (Schoenfeld, 2006). Indeed, Grover J. Whitehurst (2003),
a former director of the Institute of Education Sciences, maintained that “randomized trials
are the only sure method for determining the effectiveness of education programs and
practices” (p. 6).
Conducting a high-quality teaching and learning study using an experimental design is
a complex matter. Any number of factors other than the intervention could cause changes in
outcome measures. Campbell and Stanley (1963) listed some threats to internal validity (e.g.,
events can happen between the first and second measurements which can affect measurement
of the experimental variable). Other threats include maturation, instrumentation, statistical
regression effects, effects arising from the novelty of an intervention (the “Hawthorne
effect”), experimental mortality, and selection-maturation interaction. As Trochim and Land
(1982) argued, good research designs should minimize probable alternative explanations for
any hypothesized cause-effect relationships. Trochim and Land called these types of
methodological challenges “threats,” and one of the biggest threats in a random-allocation
study is that the groups may have initial (pre-intervention) performance differences.
At the beginning of the current study the 32 participating seventh-grade students at
School W were randomly allocated to two groups, and the research team agreed to adopt
Campbell and Stanley’s (1963) criterion that the difference between the initial mean scores
of the two groups on a relevant pre-teaching measurement should not exceed one-half of a
standard deviation. Table 6.1 summarizes pre-teaching algebra data for the main study (the
maximum possible score was 20). Clearly, Campbell and Stanley’s criterion requirement
was met.
Table 6.1
Pre-Teaching Algebra Test Mean Scores, Standard Deviations (SD), and Difference of SDs
Group Number of Pre-Teaching Mean Standard Difference

students Score on the Algebra Deviation Between Groups’
(Overall, Test (Maximum (SD) Standard
n = 32) Possible Score was 20.) Deviations
Group 1 16 1.60 0.81

0.21
Group 2 16 1.63 1.02
Based on these data, it would be reasonable to assume that the two groups of seventh-
grade students, which were formed by random allocation, were not only equal in size in
terms of numbers but also with respect to their initial algebraic thinking skills and
understanding. One might add that students in both groups seemed to know very little about
the algebra represented on the initial pre-teaching version of the Algebra Test (which is
reproduced in Appendix B to this book).
Defining Desirable Learning Outcomes

Even though there is a lack of consensus in the literature about what “structure” and
“modeling” lesson sequences should be, in this study the research team clarified the
intervention’s instructional goals in terms of two distinguishable approaches. One of the most
important steps in Herbart’s (1904a, 1904b) analysis of how lessons should be planned was
that intended learning outcomes should be clearly defined (Dunkel, 1970). The main study’s
planned lesson sequences, together with intended, learning outcomes, can be found in
Appendix D, Appendix E and Appendix F to this book.
Professional Development in a Theoretical Context

Liping Ma (1999), after conducting her study of Chinese and U.S. elementary school
teachers, claimed that most elementary-school mathematics teachers in China have a deeper
knowledge and understanding of the structural side of elementary-school mathematics than
persons teaching at the same level in the United States of America. She further argued that
unless an elementary or middle-school teacher has a profound understanding of the structural
side of elementary mathematics it is unlikely that that teacher will be an effective K–8
mathematics teacher. If Ma’s claim is correct then in order to improve elementary- and middle-
school students’ algebra learning it will be important that U.S. middle-school teachers’
knowledge of the structural aspects of elementary algebra is raised to the highest possible level.
The Research Team for the Main Study

As stated in earlier chapters, two well-qualified and experienced middle-school
mathematics teachers, Mr. X and Mr. Y, who had both been teaching in School W for a number
of years, agreed to be actively involved in the study as part of the research team—working
Model Lessons 121
with the three authors of this book. Each of the five members of the research team had had years
of experience in teaching algebra at the middle-school and lower-secondary levels.
A “model-lessons” component of the study was led by the authors of this book—who
have had much experience in teaching actual seventh- and eighth-grade algebra classes. It is
relevant to comment, here, that School W would not have agreed that the study could go
ahead with its middle-school classes unless the Principal of the school and the two
participating teachers (Mr. X and Mr. Y) had believed that the authors were capable of
“producing the goods” so far as teaching middle-school students was concerned.
Model Lessons
Since the teaching intervention would be a major part of the present study, the research
team reflected at length on what would be most likely to assist Mr. X and Mr. Y to be well
prepared for the classes they would take. It was decided that an approach recommended by
Bruce Joyce and Beverley Showers (2002) should be adopted.
Joyce and Showers argued that teachers are most likely to acquire new knowledge and
skills if they are able to model their teaching on their observations of high-quality lessons
which had been conducted in their own classrooms. Toward that end, Mr. X and Mr. Y
agreed to observe the three authors of this book teaching a Grade 8 class at School W over a
period of two weeks, and they would then model their teaching of their seventh-grade classes
on what they had observed. It was expected that as a result of their observations, Mr. X’s and
Mr. Y’s awareness of “structure” and “modeling” approaches would be enhanced to the point
where they would be confident to engage their seventh-grade students in expressive activities
which would enable those students to learn to interpret and use constructively the main
“signs” of middle-school algebra relating to “structure” and “modeling.”
The “model-lessons” component of the study was duly led by the three authors of this
book. They prepared the model lessons with the conscious aim of engaging all students
actively in discussion and reflection. These “workshops” were not based on what can be
found in any existing textbook. The notes for the workshops are reproduced as Appendices
D, E, and F to this book. They were originally prepared for use with eighth-grade students at
School W—but, with suitable modifications, they were also used with seventh-grade students
at School W. For every lesson the main goal was for the students to gain relational
understandings of key algebraic concepts through active group discussion, by solving set
problems, by posing associated problems, and by reporting their findings to the rest of the
class. Each student was invited to explore, with classmates, how algebraic notations could be
used advantageously in curriculum-related structure and modeling situations.
The aim of the workshops was to help middle-school students develop their concepts of
a variable in ways consistent with the CCSSM (2010) common-core sequence, and to help
them identify and generalize number patterns. Mr. X and Mr. Y were pleased to be able to
observe the model lessons given to the eighth-grade students and, after doing so, each
indicated that he would be happy to teach similar lessons to a Grade 7 class.
Intervention Setting and Theoretical Base

This study’s classroom-based intervention was intended to enhance classroom practice
related to seventh-grade algebra. In particular, it was regarded as important that the seventh-
grade students’ algebraic thinking with respect to structure and functional thinking (or
modeling) should improve as a result of their participation in the study. Another important
goal of the intervention was to prepare high-quality workshop notes suitable for use by other
researchers who might want to study middle-school students’ algebraic reasoning.
The Teaching Approach

The approach to teaching which was adopted in the model lessons was based on the
“Expanding Modes of Communication” theory developed by Gina Del Campo and Ken
Clements (1987, 1990), and reported in Ellerton and Clements (1991). Previous research
reported by Clements and Del Campo (1987) showed that often students are happy merely to
“receive” what they are taught to the point where they can reproduce it when answering low-
level, often multiple-choice, questions on written tests. A higher educational goal was to get
students to the point where they were not only able to express curriculum-relevant ideas
using their own words, but were also able to move beyond that to posing and solving
problems associated with what they have learned. It is the latter kind of learning that Del
Campo and Clements (1987) described as “expressive learning” (p. 12), and it was that kind
of learning that the research team wanted to facilitate among the students participating in the
main study.
An important aim of the main study was for all of the seventh-grade students to learn to
use appropriate expressive language while developing concepts and principles related to the
associative properties and the distributive property in middle-school algebra, and to use
algebra to model real-life contexts. Before the study began it was not known whether the
students who would participate in the study knew much about the associative and distributive
properties, or about using algebra to model real-life problems.
The main purpose of the model lessons was to show how beginning algebra students
could engage in expressive activities which would assist them to learn to interpret and use the
main “signs” of early algebra, especially those relating to “structure” and “modeling.” It was
planned that students would regularly be given the opportunity to discuss ideas and concepts
in small groups, to construct answers, to pose problems, and then to report their findings to
others in their class. Through this approach it was hoped that they would not only develop a
feeling that they owned the mathematics that they were studying, but they would also want to
attempt to convince others of its worth.
Another aspect of the planning process was the desire to prepare curricular materials
which could be easily adopted by practicing teachers who were not involved in the study. Of
course, those teachers would not be able to observe the model lessons in which the materials
were used, but the aim was for the curricular materials to be largely self-explanatory.
This view of teaching just outlined is consistent with the idea that students need to go
beyond merely memorizing verbal knowledge and intellectual skills. They need to develop
appropriate imageries, and be able to apply the knowledge, skills, and imageries to new, but
related, situations. Del Campo and Clements (1987) argued that if students are given the
opportunity to engage in conversations with fellow students, if each student learns to describe
his or her verbal knowledge, skills, and imageries to other students, then in the future they
will have a better chance of creating, consolidating, and enhancing cognitive structures which
will enable them to pose and solve problems. It is also likely that they will have developed
attitudes which will make them want to do that.
Summary of the Design 123
P
Prooced
durres
T
Thee reeseaarchh teeamm deevelopped andd addopptedd ann exxpeerim menttal dessignn whic
w ch pperm mittted pree-
teaachiing,, mid-iinteerveentiion, annd ppost-teeachhingg teest meeasuuress to bee coomppareed llegiitim mateely. Thhe
inteervventtionns ccom mprrisedd ccom mbinnatiionns oof llesssonn seequuencces whhichh aaltoogettherr coveeredd 228
sesssioons of reggulaar m mathhem matiics claassrroom m ttimee sppreaad oveer a peeriood oof ssix w weekss. A paaper-
andd-peenccil A Alggebrra Tesst, devveloopeed bby thee auuthoors annd bbassed onn finndiingss frrom m thhe piloot
wass addmiinisstratted, ass a ppree-teaachhingg teest, to 332 sevventth-ggradderrs atttenndinng Schhoool W
stuudiees, w W,
andd stricctlyy paaralllel veersionss of thhis instruumeent weere suubseequuenttly adm minnisttereed aat mmidd-
Thee retenntionn teest, whhichh w
inteervventtionn, ppostt-teaachhingg annd rreteentiion staagess. T was addminnisttereed
to tthe stuudennts 122 weeekks aafterr thhe innterrveentioon worksshopps hhadd beeen com mppleteed, waas iddennticaal
to tthe pree-teeachhingg foormm off thee A
Algeebraa Test.
Te
T
Thee depeenddentt vaariabblees fofor tthe stuudyy were thee teestss sccorees ffor twoo aaspeectss off scchoool
alggebrra——strructuree annd m moodellingg. T
Therre w werre twwo grooupps— —Grroupp 1 annd G Grouup 2— —eaach witth
16 stuudennts,, whhich w weree foormmed byy ranndoom allocaatioon ffrom m thhe 332 sevventth-ggradde sstuddennts aat
Schhoool W w whoo haad aagreeedd to paarticcipaate in tthe stuudyy. Ther
T re aare norrmaallyy tw wo sseveenth-ggradde
maatheemaatics cllassses at Schhoool W W, aandd thhe rranddom m allloccatiion proocedurre w wass strratiifiedd inn thhe
sennse thaat eachh exxisttingg claass prooviidedd eiightt stuudeentss too eaach of thee exxperrim menttal grooups foor
thee maain stuudy. N No ootheer foorm
m off saampple sstraatifiicattionn was eempployedd.
O
One oof thhe grooupss, w w be refferred to as “G
whiich herreaffterr will Grouup 11,” firsst eenggageed iin
struuctuure claassees w witth MMr. X,, annd tthenn inn m moddelinng claassees w withh MMr. Y. “G Grouup 2,”” onn thhe
othher hannd, weere firsst engaaged inn m moddelinng claassees w withh Mr.
M Y Y, aandd thhen in struuctuure claassees
witth MMr. X.
So farr ass thhe sstruuctuure claassees w weree conccernnedd, thhe rreseearch teaam deccideed tthaat inn thhe
tim
me aavaiilabble theere w wouuld be a ffocuus oon jjustt thrree im mporrtannt fi fieldd prropertiies——thhe aassoociaativve
prooperrty forr addditionn, thhe aassoociaativve pproppertty fo
for m
mulltipplicaatioon, andd thhe ddistrribuutivve pproppertty
forr muultiipliccatiionn ovver adddition (annd ssubtracctioon).. Thhe m modellingg classes, whhichh were w e leed bby
Mrr. Y Y, intrrodduceed stuudennts too thhe connceeptss off rrecuursiive annd exppliccit rules forr liineaar
seqquenncees, andd inntrooducedd annd maade usee oof thhe subbscrriptt nootattionn foor ddesccribbingg teerm ms oof
seqquenncees.
A
Aftter thee ppreliimiinarry proofesssioonall ddeveeloppmeent seessioonss—tthatt iss too ssay,, affterr thhe
“mmodeel llesssons” ledd byy thhe threee autthorrs, andd oobseerveed by Mrr. X aandd Mr. M Y Y——haad bbeeen
com mplleteed, aft fter the ppre-teaachhingg teestss hhad beeen addmiinissterred,, annd aft fter prre-teeacchinng
ws hhadd beeen connduucteed, Mrr. X duuly ledd a serriess off woorkkshoops onn strructturee w
inteervview with haalf
of thee seevennth--graadee stuudeentss (GGrouup 1), annd M Mr. Y ledd a serriess off woorkkshoops onn mode m elinng
witth tthe othherr haalf of thee seeveenthh-grradee sttuddentts ((Grooupp 2). T Theen, half-w wayy thhrouughh thhe
inteervventtionn peeriood, thee tw wo grooupps sswaappped teaachhers annd ttoppics. Figuure 6.11 suumm marrizees
stuudy com mpooneentss annd aassoociaatedd tim me inttervvalss forr thhe reeseearcch innveestiggatiionn.
Model Pre-Teaching Test Mid-Intervention Post-Teaching Retention

Lessons Interviews Test Test Test
(PD)
Intervention 1 Intervention 2 Interviews
3 Weeks 4 Weeks 3 Weeks 2 Weeks 12 Weeks
Figgurre 66.1. Sum

mmmaryy of thhe ddesiign andd timee inttervvalss foor thhe sstuddy.
Intervention Fidelity and Implementation

In the main study, issues associated with fidelity of implementation for the two
curricular themes did not present any methodological difficulties. Fidelity of implementation
was not regarded as an issue because all members of the research team believed that the
content to be covered in the workshops, and the methods by which that content would be
assessed, were entirely consistent with CCSSM’s and School W’s expectations.
Table 6.2 summarizes the study components. Mr. X and Mr. Y observed the model-
lessons given by the three authors, and used the printed materials in workshops with their
seventh-grade classes. Both teachers were happy to follow the teaching approaches that they
had observed in the model lessons, and the research team agreed that at any stage either of
Mr. X or Mr. Y could vary the approach if he felt the need to do so. Each of the intervention
lessons was observed by at least one member of the research team other than Mr. X or Mr. Y.
Any variations from what had taken place in the eighth-grade model lessons were noted.
Table 6.2
Summary of the Research Design and Timing in the Planned Research Program
Weeks Weeks Weeks Week Weeks Weeks Week

Group 1–3 4-5 6–7 8 9–10 11–12 24
Pre-
PD for Teaching Mid- Post-Teaching
1 Teachers Test, Workshops Intervention Workshops Test, Retention
(Model Interviews (Structure) Test (Modeling) Interviews Test
Lessons
with an Pre-
8th-Grade Teaching Mid- Post-Teaching
2 Class) Test, Workshops Intervention Workshops Test, Retention
Interviews (Modeling) Test (Structure) Interviews Test
Since the seventh-grade intervention workshops for the two groups occurred at
identical times, no single person was able to observe all of the intervention workshops fully.
Nevertheless, each lesson was observed by at least one of the authors, and during each lesson
the observer(s) made handwritten notes on a specially-prepared “lesson observation
schedule” (reproduced in Appendix G to this book). The notes were concerned with patterns
of classroom interaction between students and students, and between teachers and students.
There was a section in the schedule in which any variations from the model lessons
introduced by Mr. X or Mr. Y were to be described.
Instrumentation
The research team was aware that the weakest part of a teaching intervention in a
research study can be the use of invalid or unreliable instruments for evaluating the extent and
quality of student learning as a result of the intervention (Gersten et al., 2000). After agreement
had been reached by members of the research team with respect to the aims of the intervention,
it was also agreed that the research team should measure students’ initial (“pre-teaching”)
knowledge and understanding with respect to the associative properties for addition and
Instruments: The Algebra Test 125
multiplication, the distributive property and, in addition, the quality of their modeling,
especially in relation to patterns set out as sequences. It was decided that once a pre-teaching
instrument was established then parallel mid-intervention, post-teaching, and retention
instruments would also be created. It was decided that, in fact, the retention instrument should
be identical to the pre-teaching instrument—the two would be administered 21 weeks apart and
therefore problems arising from students “learning” from the first administration were not
expected to arise.
This section provides background information on the measurement instruments
developed and used in the study.
The Main Test Instrument: The Algebra Test

Based on results on the two pilot studies and on considerations of the nature of the
current study’s intervention, the three authors developed and revised a composite pencil-and-
paper Algebra Test which would be suitable for administration to a seventh-grade class in a
normal 45-minute period of time. That instrument, and two parallel versions of it, are
reproduced in Appendix B to this report. Each version of the Algebra Test would be scored
out of 20, with 10 of the 20 marks being allocated to structure responses, and 10 to modeling
responses. The parallel forms of the instrument were used at the mid-intervention and post-
teaching stages of the study. As stated above, it was agreed that the original pre-teaching
version of the Algebra Test would be re-used at the retention stage of the study.
This Algebra Test was prepared with seventh- and eighth-grade students in mind.
Although a few of the questions were adapted from those used in other studies—such as
TIMSS, PISA, and NAEP—no question on any version of the Algebra Test was identical to
any question on any existing test instrument. Many of the questions had been originally
developed for use in the first and second pilot studies (Kanbir, 2014, 2016), and some
questions were modifications of questions used in those studies. The Cronbach-alpha reliability
of the Algebra Test (pre-teaching and retention version) was calculated to be .82, and those of
the two parallel tests were .83 (mid-intervention version) and .85 (post-teaching version).
Figure 6.2 shows Questions 8, and 13, two typical structure questions on the Algebra
Test. With both of these questions a response would be given a mark of 1 if structure was
used to obtain a correct answer—otherwise a mark of 0 was allocated. Thus, for example,
with Question 8, a student who obtained 720 by first multiplying the 72 by 5 to obtain 360
and then multiplying 360 by 2 to obtain 720 was not given a mark. To obtain the mark, a
student needed to associate the 5 with the 2, and then multiply 72 by 10 to obtain 720.
Similarly, to get the mark for Question 13 a student needed to recognize that the distributive
property would be maximally useful—and then correctly multiply 7 by 100 to obtain 700. In
order to obtain the allocated 1 mark it would not be necessary for a student to use the
terminology “associative property” or “distributive property.”
Question 8. Without using a calculator find the value of (72 × 5) × 2, and explain how you
got your answer.
Question 13. What would be a quick method of finding the value of 7 × 97 + 7 × 3 without
using a calculator? What is the property which allows you to use that quick method?
Figure 6.2. Two structure questions (pre-teaching version of the Algebra Test).
Question 14 (see Figure 6.3) was typical of questions concerned with modeling. Four
marks were allocated to Question 14, one each for responses to Parts A and B, and 2 marks
for Part C. The main interest in relation to the task was whether students would be able to
generalize and, if they could do that, to observe the reasons that they would offer for their
generalizations. Would the reasons be mainly inductive, based purely on numerical patterns,
or would students be able to give “reasons for the rule”?
To obtain the mark for Part B an algebraic expression such as 3 + n + 3 + n or 2n + 6 or
6 + 2.n or 2(3 + n), or anything else equal to 2n + 6, needed to be given. To obtain 2 marks
for Question C a response had to explain clearly the thinking behind how the algebraic
expression, given as the response to Question B, was obtained. One mark would be given for
what was deemed to be a partially correct explanation.
Question 14. You have been hired by the Southwestern Fence Company to
make pens for holding cows. A cow pen is a wall of blocks that completely
surrounds the cow. You must leave at least 1 unit block in the middle of each pen
where a cow would go. A cow needs 1 unit block of space.
The first cow pen that you can build looks like this. It holds just one cow,
and there are 8 surrounding blocks altogether:
The second cow pen that you can build looks like this. It holds 2 cows.
The third cow pen that you can build looks like this. It holds 3 cows.
Note that the cow pens must always be in a straight line, left to right.
A. How many surrounding blocks would you need to hold 25 cows?
B. If Sn represents the number of surrounding blocks you would need for a pen
which would hold n cows, what is the rule giving Sn in terms of n?
C. Explain how you got your rule for Part B.
Figure 6.3. A modeling question (from the pre-teaching version of the Algebra Test).
Preparing for the Interviews 127
The Interview Protocol

The research team recognized that the interview data would prove to be of crucial
importance in making qualitative decisions regarding student development with respect to
knowledge of signifiers, and with respect to key aspects for cognitive structure and modeling.
As a result, much effort was put into developing an appropriate interview protocol. The
protocol below emerged from experience in the two pilot studies (described earlier).
The same protocol was used by all three authors—who, between them, conducted all of
the 28 pre-teaching and 28 post-teaching interviews. All interviews were tape-recorded.
During an interview, the interviewer made the following “equipment” available to the
interviewee at appropriate times:
• A sheet of paper with 482 + (18 + 300) on it. [See Question (2) in the Protocol.]
• A sheet of paper with 4 × (1/4 × 128) on it. [See (3) below.]
• A sheet of paper with Tn = 2n + 3 on it. [See (4) below.]
• A sheet of paper with Tn = 5n – 2 on it. [See (5) below.]
• A sheet of paper with 15 – (5 – x) = (15 – 5) – x on it. [See (6) below.]
• A sheet of paper showing the Table in Question 7 below.
• A sheet of paper showing the illustration in Question 8 below.
During interviews, the Newman interview technique (see Clements, 1980) was used by
all three interviewers. Following the interviews, the first author transcribed all of the text, for
all 56 interviews, from the audio-recordings.
Questions Asked in Interviews, and Procedures for Conducting Interviews

1. “I am going to say two words and, as soon as I say them, I want you to say
something, or draw something, or do something—do the first thing that comes into
your head after I say the words. The words are … “distributive property.” Here
are the words again: “distributive property.”
2. Without using a calculator, find the value of 482 + (18 + 300).
[Once an answer is given, ask for the explanation of where that answer came from.]
3. Without using a calculator, find the value of 4 × (1/4 × 128).
[Once an answer is given, ask for the explanation of where that answer came from.]
4. If we write Tn = 2n + 3, then we can say T5 equals 13, because 2 times 5 plus 3
equals 13. What would T11 equal?
When the pupil gives an answer, ask her or him to write down how she or he
obtained that answer. Also, ask the student to explain what she or he thought, in
words.
5. Give the pupil a piece of paper with Tn = 5n – 2 on it and, then ask her or him to
say which values of n would make Tn greater than 20.
When the pupil gives an answer, ask her or him to write down how she or he obtained
that answer. Also, ask the student to explain what she or he thought, in words.
6. Give the pupil a piece of paper with the equation 15 – (5 – x) = (15 – 5) – x on it,
and then, pointing to the x, say: “Which numbers could x equal so that you would
get a true statement?”
When the pupil gives an answer, ask her or him to write down how she or he
obtained that answer. Also, ask the student to explain what she or he had thought,
in words.
7. Give the pupil a piece of paper with the following table on it:
First
1 2 3 4 5 … n
Value
Second …
3 5 7 9 ? ?
Value
Then ask (pointing): What number should we place under the 5 in the table?
Then ask (pointing): What do you think we should we put under the n?
8. The diagram below shows how tables and chairs are arranged in a school cafeteria.
One table can seat 4 people, and tables can be pushed together (but always in a
straight line). When two tables are pushed together, 6 people can sit around the
table (as shown), etc.
A. If 10 tables were pushed together (in a straight line), how many people could sit
around them (assuming the pattern shown above)?
B. If Pn represents the number of people who can sit when n tables are pushed together
(in a straight line), what is the rule giving Pn in terms of n?
Research Hypotheses, and Issues Related to the Quantitative Analyses

Samples from this study’s hypothetical populations were students in either Group 1 or
Group 2, or, sometimes, students in both Group 1 and Group 2. This is summarized in Table
6.3, which relates to four stages of the study—which are referred to as pre-teaching, mid-
intervention, post-teaching, and retention. Table 6.3 introduces subscript notations for class
mean scores on the structure questions on the Algebra Test (each mean score having a
maximum possible value of 10) and on the modeling questions (also with a maximum
possible value of 10), and also for the means of the sum of scores for the structure and
modeling questions (with a maximum of 20).
Summary of Proposed Quantitative Analyses 129
Table 6.3
Subscript Notations for Describing Class Means at Various Points in the Study
Mean Score on Mean Score on Total Mean Score

Structure Modeling (Structure +
Questions Questions Modeling)
(/10) (/10) (/20)
Pre-Teaching
Group 1 X 11 X 12 Y 11
Group 2 X 21 X 22 Y 21
Mid-
Intervention
Group 1 X 13 X 14 Y 13
Group 2 X 23 X 24 Y 23
Post-Teaching
Group 1 X 15 X 16 Y 15
Group 2 X 25 X 26 Y 25
Retention
Group 1 X 17 X 18 Y 17
Group 2 X 27 X 28 Y 27
If the random allocation to groups worked efficiently then each pair of pre-teaching
group means, X 11 and X 21, X 12 and X 22, and Y 11 and Y 21—would be expected to be
approximately equal, although a corresponding statement was not expected to be true for the
mid-intervention means.
Because the design of the study involved random allocation to classes, it was possible
to use, legitimately, inferential statistics to make decisions, from sample statistics, about
population parameters. However, before that could be done there was the difficult task of
defining the relevant populations. That task was difficult because Group 1 and Group 2 were
both groups of seventh-grade students from one school, School W. Obviously, it would be
wrong to assume that the use of inferential statistics based on the two sample groups could
enable legitimate inferences to be made about the algebra knowledge of seventh-grade
classes outside of School W. So the question arises: Why use inferential statistics at all in this
study?
The issue of defining relevant populations was made all the more difficult for the
present study because at some stages of the study the two groups of students had studied
different types of algebra. At the mid-intervention stage for example, Group 1 students had
taken classes in structure but not in modeling, but the reverse was true for Group 2 students.
By using inferential statistical procedures, it should be possible to decide whether mean
differences are sufficiently large for one to be able to make confident decisions about
whether they occurred purely by chance. If that can be done, then the challenge will be for
others to replicate the study at other places to see if similar differences persist elsewhere.
Each of the hypothetical populations would comprise second-semester seventh-grade
students who were attending School W and who were taught in heterogeneous algebra
classes in which there was a focus on the associative and distributive properties (taught by
Mr. X), and on modeling (taught by Mr. Y). However, there could be differences between the
hypothetical populations in relation to whether or not they had participated in either or both
of the structure and modeling interventions.
These considerations led to null and research hypotheses being formulated. The first
eight comparisons related to group mean scores on either the structure or the modeling
component of the Algebra Test—we shall assume that all the parallel forms of the structure
test can reasonably be regarded as equivalent, and that all the parallel forms of the modeling
test can reasonably be regarded as equivalent.
Comparisons of Scores on Tests Administered at the Same Time

Hypotheses relating to pre-teaching differences. It is important to consider whether
the random allocation to the two groups which was employed in the study actually produced
two “equal” groups so far as structure and modeling were concerned. The first and second
hypothetical populations would comprise students who had taken the same pre-teaching test,
but had yet been interviewed, and had not participated in any of the intervention lessons.
Applying the subscript notation introduced in Table 6.3 to hypothetical populations, we
denoted the mean (out of a possible 10) and standard deviation on the structure component of
the initial test, for one of these populations, by µ11 and σ11, respectively. Similarly, we
denoted the mean (out of a possible 10) and standard deviation on the modeling component
of the initial test, for the second of these populations, by µ21 and σ21, respectively.
Then, the following null and research hypotheses could be stated.
First null and research hypotheses (relating to pre-teaching structure). The
difference between Hypothetical Population 1’s mean score on the pre-teaching structure test
and Hypothetical Population 2’s mean score on the same structure test would be zero. If we
denote the null and research hypotheses by H0 and H1, then the hypotheses would be:
H0 : µ11 − µ21 = 0, with distribution from zero being according to the t-distribution, df =
30, and α = .05 (two-tailed).
H1: µ11 − µ21 ≠ 0.
Second null and research hypotheses (relating to pre-teaching modeling). The
difference between Hypothetical Population 1’s mean score on the pre-teaching modeling
test and Hypothetical Population 2’s mean score on the same modeling test would be zero. If,
once again, we denote the null and research hypotheses by H0 and H1, then:
30, and α = .05 (two-tailed).
H1: µ12 − µ22 ≠ 0.
A similar pair of hypotheses could be stated for the sum of the structure and modeling scores
(with a maximum possible score of 20)—but those hypotheses will not be stated here.
Hypotheses relating to mid-intervention differences. We wanted to check whether,
as would be expected, student participation in structure lessons would result in higher
Summary of Proposed Quantitative Analyses 131
performance on the structure test than student participation in modeling lessons. Similarly,
we wanted to check whether student participation in modeling lessons would result in higher
performance on the modeling test than student participation in structure lessons.
Third null and research hypotheses (relating to mid-intervention structure). The
difference between Hypothetical Population 1’s mean score on the structure test and
Hypothetical Population 2’s mean score on the same structure test, after Population 1 had
participated in structure lessons but not modeling lessons, and Population 2 in modeling
lessons, but not structure lessons, would be zero. Thus:
30, and α = .05 (one-tailed).
H1: µ13 − µ23 > 0.
Fourth null and research hypotheses (relating to mid-intervention modeling). The
difference between Hypothetical Population 2’s mean score on the modeling test and
Hypothetical Population 1’s mean score on the same modeling test, after Population 2 students
had participated in modeling workshops, but not structure workshops, and Population 1
students in structure workshops but not modeling workshops, would be zero. Thus:
30, and α = .05 (one-tailed).
H1: µ24 − µ14 > 0.
Hypotheses relating to post-teaching differences. We wanted to check whether, as
would be expected, student participation in structure lessons then modeling lessons would
result in the same performance on the structure test as student participation in modeling lessons
then structure lessons. Similarly, we wanted to check whether, as would be expected, student
participation in structure lessons then modeling lessons would result in the same performance
on the modeling test as would student participation in modeling lessons then structure lessons.
Fifth null and research hypotheses (relating to post-teaching structure). The
difference between Hypothetical Population 1’s mean score on the structure test and
Hypothetical Population 2’s mean score on the same structure test, after Population 1 had just
participated in modeling lessons after having participated in structure lessons, and Population
2 had just participated in structure lessons after having participated in modeling lessons,
would be zero. Thus:
H0 : µ15 − µ25 = 0, with the distribution from zero being according to the t-distribution,
df = 30, and α = .05 (two-tailed).
H1: µ15 − µ25 ≠ 0.
Sixth null and research hypotheses (relating to post-teaching modeling). The
difference between Hypothetical Population 1’s mean score on the modeling test and
Hypothetical Population 2’s mean score on the same modeling test, with Population 1 having
just participated in modeling lessons after having participated in structure lessons and
Population 2 having just participated in structure lessons after having participated in
modeling lessons, would be zero. Thus:
H1: µ16 − µ26 ≠ 0.
Hypotheses relating to retention differences. We wanted to check whether, as would

be expected, student participation in structure lessons then modeling lessons, and then having
a “break” of 12 weeks, would generate the same performance on the structure test as student
participation in modeling lessons then structure lessons followed by a 12-week break.
Seventh null and research hypotheses (relating to retention structure). The difference
between Hypothetical Population 1’s mean score on the structure test and Hypothetical
Population 2’s mean score on the same structure test, after Population 1 had participated in
structure lessons, then modeling lessons, and a period of 12 weeks had then elapsed, and
Population 2 had participated in modeling lessons, then structure lessons, and the same
period of 12 weeks had then elapsed, would be zero. Thus:
df =30, and α = .05 (two-tailed).
H1: µ17 − µ27 ≠ 0.
Eighth null and research hypotheses (relating to retention modeling). The difference
between Hypothetical Population 1’s mean score on the modeling test and Hypothetical
Population 2’s mean score on the same modeling test, after Population 1 had participated in
structure lessons, then modeling lessons and a period of 12 weeks had then elapsed, and
Population 2 had participated in modeling lessons, then structure lessons and the same period
of 12 weeks had then elapsed, would be zero. Thus:
H0 : µ18 − µ28 = 0, with the distribution from zero according to t-distribution, df = 30,
and α = .05 (two-tailed).
H1: µ18 − µ28 ≠ 0.
Comparisons Between Gain Scores at Different Times

From a mathematics education research perspective, it was recognized that it would be
of some interest to measure and compare group gains on the structure and modeling
components of the Algebra Test at various stages of the investigation. It was thought best to
regard the pre-teaching scores as representing baseline scores, and then to compare mean
gains, for the two groups, measured from the baseline scores to scores at the mid-
intervention, post-teaching, and retention stages.
Suppose a student in Group 1 obtained a score of x11 on the pre-teaching structure test
(obviously, 0 ≤ x11 ≤ 10), and the same student obtained a score of x13 on the parallel mid-
intervention structure test (again, obviously, 0 ≤ x13 ≤ 10). Then the gain score will be the
value of x13 – x11, which can be positive, zero, or negative depending on whether x13 is greater
than, or equal to, or less than, x11. The mean gain score and standard deviation of the gain
scores for Group 1, at the mid-intervention stage, could be found by calculating the mean
gain and standard deviation of the gain scores for the 16 students. This idea is easily
generalizable across the study, to cover mean gain scores, etc., at the post-teaching and
retention stages, and for both the structure and modeling tests. Unless otherwise stated, the
baseline from which gain-scores will be calculated will be the pre-teaching scores for the
group being considered.
A comparison of gain scores so that null and research hypotheses with respect to
hypothetical populations could be formulated in rigorous statistical terminology, but detailed
discussion of the technicalities involved would be distracting here. Further comment on
Gain-Score Hypotheses 133
issues involved will be made in Chapter 9 of this dissertation. Here it will suffice to present
formal statements of pairs of null and research hypotheses.
Comparing mid-intervention/pre-teaching mean gain scores. Although the
following statements are not set out in fully-formalized, statistical language, the type of
comparisons to be carried out should be clear in each case.
First mean-gain comparison (structure, mid-intervention versus pre-teaching).
H0 : The difference between the mid-intervention versus pre-teaching mean gain scores
for the two hypothetical populations on the structure test would equal zero, with
the distribution from zero being according to the t-distribution, df = 30, and α = .05
(one-tailed).
H1: The difference between the mid-intervention versus pre-teaching mean gain scores
for the two hypothetical populations on the structure test would be greater than
zero (with Population 1’s mean gain score being greater than Population 2’s).
Second mean-gain comparison (modeling, mid-intervention versus pre-teaching).
H0 : The difference between the mid-intervention versus pre-teaching mean gain scores
for the two hypothetical populations on the modeling test would equal zero, the
distribution from zero being according to the t-distribution, df = 30, and α = .05
(one-tailed).
H1: The difference between the mid-intervention versus pre-teaching mean gain scores
for the two hypothetical populations on the modeling test would be greater than 0
(with Population 2’s mean gain score being greater than Population 1’s).
Third mean-gain comparison (structure, post-teaching versus pre-teaching).
H0: The difference between the post-teaching versus pre-teaching mean gain scores for
the two hypothetical populations on the structure test would equal zero, the
(two-tailed).
the two hypothetical populations on the structure test would not equal zero.
Fourth mean-gain comparison (modeling, post-teaching versus pre-teaching).
the two hypothetical populations on the modeling test would equal zero, the
(two-tailed).
the two hypothetical populations on the modeling test would not equal zero.
Fifth mean-gain comparison (structure, retention versus pre-teaching).
H0: The difference between the retention versus pre-teaching mean gain scores for the
two hypothetical populations on the structure test will equal zero, the distribution
from zero being according to the t-distribution, df = 30, and α = .05 (two-tailed).
two hypothetical populations on the structure test would not equal zero.
Sixth mean-gain comparison (modeling, retention versus pre-teaching).

two hypothetical populations on the modeling test will equal zero, the distribution
two hypothetical populations on the modeling test would not equal zero.
Seventh mean-gain comparison (structure, retention versus post-teaching).
H0: The difference between the retention versus post-teaching mean gain scores for the
two hypothetical populations on the structure test would equal zero, the distribution
two hypothetical populations on the structure test would not equal zero.
Eighth mean-gain comparison (modeling, retention versus post-teaching).
two hypothetical populations on the modeling test would equal zero, the distribution
two hypothetical populations on the modeling test would not equal zero.
Effect Sizes
It was expected that the extent of differential learning occurring during the first half of
the main period of intervention, after the pre-teaching data had been gathered and
immediately before the mid-intervention data were gathered, would be of special interest to
scholars researching middle-school algebra. During that period, 32 students, almost all of the
seventh-graders at School W, would be allocated to two classes (16 in Group 1 and 16 in
Group 2) on a random-allocation basis. Group 1 students would then participate in a series of
45-minute workshops relating especially to the associative and distributive properties for
rational numbers—those workshops would be led by Mr. X.
Simultaneously, Group 2 students would participate in a series of 45-minute workshops
relating especially to how elementary algebra arises in modeling—those lessons would be led
by Mr. Y. Mr. X and Mr. Y were the normal seventh-grade mathematics teachers at School
W, and during the school year all of the participating students had had either Mr. X or Mr. Y
as their regular mathematics teacher.
Six questions would be considered. The first four were “effect-size” questions:
1. What would be the Cohen’s d effect size (Cohen, 1988) generated by the pre-
teaching to mid-intervention sessions on structure with Group 1? It would be
assumed that the control group for this period was Group 2.
2. What would be the Cohen’s d effect size generated by the mid-intervention to post-
teaching sessions on structure with Group 2? It would be assumed that the control
group for this period was Group 1.
3. What would be the Cohen’s d effect size generated by the pre-teaching to mid-
intervention sessions on modeling with Group 2? It would be assumed that the
control group for this period was Group 1.
Planning for Effect-Size Calculations 135
4. What would be the Cohen’s d effect size generated by the mid-intervention to post-
teaching sessions on modeling with Group 1? It would be assumed that the control
group for this period was Group 2.
The other two questions to be considered would be:
5. Would the Group 1 students, who participated in structure workshops in which
there was no obvious emphasis on modeling, improve their scores on the modeling
component of the Algebra Test at the mid-intervention stage?
6. Would the Group 2 students, who participated in modeling workshops in which
there was no obvious emphasis on structure, improve their scores on the structure
component of the Algebra Test at the mid-intervention stage?
It might be argued that it would be invalid to use Group 2 as a control group for Group
1 for pre-teaching to mid-intervention structure effect-size calculations, because the Group 2
students would also have been engaged in algebraic activities. Similarly, it might be argued
that it is invalid to use Group 1 as a control group for Group 2 for pre-teaching to mid-
intervention modeling effect-size calculations, because the Group 1 students would also have
been engaged in algebraic activities. Similarly, it might be argued that it would be invalid to
use Group 1 as a control group for Group 2 for mid-intervention to post-teaching structure
effect-size calculations, because the Group 1 students would also have been engaged in
algebraic activities. And, it might be argued that it is invalid to use Group 2 as a control
group for Group 1 for mid-intervention to post-teaching modeling effect-size calculations,
because the Group 2 students would also have been engaged in algebraic activities. These
possible objections were rejected for the following two reasons:
1. The workshops for both groups would take place during times that the students
would normally have mathematics classes, and control group students would be
“doing” mathematics—as they normally would be doing—at those times.
2. Although structure and modeling can both be regarded as seventh-grade algebra
themes they are, in fact, quite different in content. During observations of modeling
classes, research-team observers never heard the teacher (Mr. Y) or any of his
students mention any of the words “structure,” “associative” or “distributive.”
Similarly, during observations of structure classes, research-team observers never
heard the teacher (Mr. X) or any of his students mention any of the words
“modeling,” “recursive,” “explicit,” “sequence,” or “subscript.”
Although the Group 1 and Group 2 workshops would appropriately be regarded as being part
of a middle-school’s implemented mathematics curriculum, any intersection of the
mathematical content was not at all obvious.
The Cohen’s d effect-size formula (Cohen, 1988) which would be used to measure the
effects of the workshops on student knowledge and understanding was:
d = (M1 – M2)/(Pooled standard deviation),
where M1 and M2 are the mean gain scores for the treatment group and control group,
respectively, and the pooled standard deviation is the square root of the average of the
squared standard deviations (Rosnow & Rosenthal, 1996).
Most education researchers who use Cohen’s (1988) d statistic have accepted Cohen’s
guidelines for interpreting effect sizes. According to Cohen, effect sizes around 0.2 are
regarded as “small,” those around 0.5 as “medium,” and those from about 0.8 onwards as
“high.” A Cohen’s d effect size of zero indicates that the mean of the treated group is the
same as the mean of the untreated group; An effect size of 0.8 would indicate that the mean
of the treated group is at just less than the 80th percentile for the untreated group; and an
effect size of 1.7 would indicate that the mean of the treated group is between the 95th and
96th percentiles for the untreated group.
Post-Hoc Statistical Analyses

Earlier in this chapter formal statements of null and research hypotheses which would
be checked in the study were presented, and effect-size calculations to be determined were
also identified. Certainly, though, other post-hoc calculations could be made, to examine
issues deemed to be relevant by the research team. Thus, for example, it might be of interest
to examine and compare Group 1 and Group 2’s total score (out of a possible 20) when the
structure and modeling sub-scores are added at the four main stages (pre-teaching, mid-
intervention, post-teaching, and retention). Bar graphs and Cartesian graphs might be drawn
to illustrate patterns of effect, for either or both groups, at different stages. Decisions on
which post-hoc calculations might be made, if any, were left until after data had been
gathered and the research team had decided whether it might be profitable to explore patterns
which, it seemed, had emerged.
Issues Related to the Qualitative Analyses
Semiotic Aspects of the Qualitative Analysis

In our review of pertinent literature (see Chapter 5 of this book), emphasis was given to
possible semiotic interpretations of data generated by this study. In our qualitative analyses,
attention was given to how knowledge, properties, concepts, images, principles, values and
attitudes should best be represented by signifiers in workshop notes, in verbal statements
made by the teachers during workshops, and in the Algebra Test. Attention would also be
given to the extent to which the interpretants of the mathematical signifiers corresponded to
the mathematical objects (signifieds) that Mr. X and Mr. Y were hoping the students would
come to appreciate. It was expected that the students’ involvements in workshops, and their
written responses to homework tasks, would mold their language and behaviors so that they
would not only be enabled to understand the signs being used, receptively, but they would
also acquire and learn to use the signs themselves, expressively, as they communicated with
their fellow students and with their teachers.
Evidence for the growth in appropriate receptive and expressive understandings would
be sought by studying classroom discourse patterns during the workshops, especially during
group workshop discussions and public presentations by students to the whole class. Toward
that end, most of the workshops were audiotaped, in order that continuing analyses of
discourse patterns would be possible. It was decided that workshops would not be videotaped
because of the researchers’ conviction that the presence of cameras and videotapers distracts
teachers and students, and affects discourse patterns (Singh & Ellerton, 2013).
Potentially important evidence would also be gleaned by studying, from qualitative
(and especially semiotic) perspectives, students’ handwritten responses to questions on the
tests, and also their written responses to homework tasks (which were regularly examined by
the three authors of this book). Photocopies of all of the students’ handwritten responses
Post-Hoc Statistical Analysis 137
were made in order to facilitate beyond-the-event analyses. The authors made extensive
handwritten comments on the homework scripts in the hope that these would be taken note of
by the students once the scripts had been returned to them.
It was recognized from the outset that one of the richest forms of qualitative data would
be provided by audiotapes of pre- and post-teaching interviews with the students. The
research team was given permission to interview 28 of the participating 32 seventh-grade
students on two different occasions. Each interview was conducted by one of the three
authors and occupied between 20 and 40 minutes. Interviews were carried out on a one-to-
one basis, in quiet, private, areas in School W’s library. As mentioned previously, the
interviews were audiotaped and, in addition, handwritten notes were taken by the
interviewer. The interview protocol remained the same—it was reproduced earlier in this
chapter—for the pre- and post-teaching interviews, and that enabled changes in the
interviewees’ receptive and expressive languages to be identified. Once the 56 interviews
were completed the first author (Kanbir) typed complete transcripts for all of the interviews.
The Qualitative Analyses and Cognitive Structure

The review of literature emphasized the Herbartian idea of apperception, and also
views on long-term memory, the extent of cognitive consciousness, and student learning, by
scholars like Charles DeGarmo, Lev Vygotsky, Jean Piaget, Jerome Bruner, David Ausubel,
Robert Gagné, and Shlomo Vinner. The main idea was that, with respect to a given concept,
a human’s long-term memory comprises a unique configuration of links between potentially
relevant verbal information, imagery, intellectual skills, memories of episodes, and various
kinds of attitudes. A stimulus from a teacher or fellow student can stimulate mental activity
in a student, the form and quality of which will be influenced by the unique configuration of
that student’s long-term memory as it relates to the input.
Seen from a cognitive-structure (or concept-image) perspective, the teacher’s role
becomes one of providing the most appropriate inputs which, in the language of Herbart, will
be “welcomed into the soul.” That is to say, from a Herbartian’s vantage point, a learner’s
long-term memory comprises a unique, largely subconscious, combination of fundamental
verbal aspects of knowledge, skills, images, memories, and attitudes, and it is the educator’s
task to present new ideas so that they will most easily be assimilated and accommodated by a
learner. The challenge for the seventh-grade mathematics teachers, of course, was that there
was not one, but as many different cognitive structures with respect to any new curricular
theme covered as there were students in their classes.
One of the unique contributions of the present study is the attempt to link semiotic
theory with neo-Herbartian apperception theory. As far as we are aware, that has never been
attempted before in algebra education research. The main task for the teacher was to provide
classroom activities and opportunities which would enable learners to enrich their
understandings of the meanings of signs, and their abilities to construct and apply associated
concepts and principles. In order to do this, the teachers needed to recognize the importance,
with any curricular theme, of enriching each student’s ability to acquire and apply pertinent
vocabulary, imagery, and skills, and to do that in such a way that the student would be keen
to work toward further mental development so far as that curricular theme was concerned.
And, hence, the workshops were planned so that semiotic and cognitive aspects of middle-
school algebra would be brought together through a healthy combination of the subconscious
and conscious.
Making use of the large volume of interview transcripts, and of handwritten pencil-and-
paper texts and homework texts, collected at different stages of the study for almost all of the
participating students, it was deemed to be possible, by examining changes in each
participating student’s verbal knowledge, imageries, skills, memory of episodes, and
attitudes, to document changes in that student’s knowledge and application of key signifiers,
and to trace growth towards deeper understandings of mathematical objects. Twenty-five
years ago, Anna Sfard (1991) called that process of change, reification.
Each interview transcript and each handwritten response by a seventh-grade student
was examined for evidence relating to the use of verbal knowledge, imagery, skills, memory
of episodes, and attitudes. The attempt was made to track changes over time, and thereby to
account for cognitive growth, especially with respect to meanings which were given, by the
students, to key signifiers.
Concluding Comments
This chapter has provided details of the design for the main study, and of how
principles taken from the literatures on design research, semiotics, cognitive structure, and on
receptive and expressive involvements in learning, would be combined and applied. In
addition, summary plans for implementation of the study with seventh-grade mathematics
teachers and students at School W were given. Details of the planning for the structure and
modeling workshops were provided as were details of the interview protocol.
Summaries were presented of the quantitative methods which would be used to
investigate the stated research hypotheses, as were methods for analyzing interview data.
In Chapter 7, data generated from the parallel pencil-and-paper tests on structure and
modeling, administered at the pre-teaching, mid-intervention, post-teaching and retention
stages of the study, will be presented and analyzed using quantitative methods (including
inferential statistical methods). Then, in Chapter 8, qualitative analyses—based on interviews
transcripts, classroom observations, handwritten pencil-and-paper test data, and examination
of homework scripts—will be presented. Chapter 9 will draw together conclusions suggested
by the quantitative and qualitative analyses; and the six research questions will be answered,
question by question. Limitations of the study will also be discussed, and possibilities for
future related research outlined. In Chapter 10, reflections on the educational and
mathematics education significance of results obtained in the study will be discussed.
References
Borko, H. (2004). Professional development and teacher learning: Mapping the terrain.
Educational Researcher, 33, 3–15.
Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for
research on teaching. In N. L. Gage (Ed.), Handbook of research on teaching (pp. 171–
246). Washington, DC: American Educational Research Association.
CCSSM. (2010). National Governors Association Center for Best Practices, & Council of
Chief State School Officers. (2010). Common core state standards for mathematics.
Washington, DC: Authors. [Also cited in this book under National Governors
Association Center for Best Practices, & Council of Chief State School Officers. (2010).]
Clements, M. A. (1980). Analyzing children’s errors on written mathematical tasks.

Misconceptions and Educational Strategies in Science and Mathematics (Vol. 3, 98–
Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in
educational research. Educational Researcher, 32(1), 9–13.
Cobb, P., & Gravemeijer, K. (2008). Experimenting to support and understand learning
processes. In A. E. Kelly, R. Lesh, & J. Baek (Eds.), Handbook of design research
methods in education: Innovations in science, technology, engineering, and
mathematics learning and teaching (pp. 68–95). New York, NY: Routledge.
MI: Lawrence Erlbaum Associates.
algebraization (pp. 379–408). New York, NY: Springer.
Gersten, R., Baker, S., & Lloyd, J. W. (2000). Designing high quality research in special
education: Group experimental design. Journal of Special Education, 34, 2–18.
Lesh, R., & Sriraman, B. (2005). Mathematics education as a design science. ZDM—The
International Journal on Mathematics Education, 37(6), 490–505.
Mosteller, F., & Boruch, R. (Eds.). (2002). Evidence matters: Randomized trials in education
research. Washington, DC: Brookings Institution Press.
Rosnow, R. L., & Rosenthal, R. (1996). Computing contrasts, effect sizes, and counternulls
on other people’s published data: General procedures for research consumers.
Psychological Methods, 1, 331–340.
Schoenfeld, A. H. (2006). What doesn’t work: The challenge and failure of the What Works
Clearinghouse to conduct meaningful reviews of studies of mathematics curricula.
22(1), 1–36.
Singh, P., & Ellerton, N. F. (2013). International collaborative studies in mathematics
education. In M. A. Clements, A. Bishop, C. Keitel, J. Kilpatrick, & F. Leung (Eds.),
Third international handbook of mathematics education (pp. 806–839). New York, NY:
Springer.
Slavin, R. E. (2004). Educational research can and must address “what works” questions.
Educational Researcher, 33(1), 27–28.
Stylianides, A. J., & Stylianides, G. J. (2013). Seeking research grounded solutions to
problems of practice: Classroom-based interventions in mathematics education. ZDM—
The International Journal on Mathematics Education, 45(3), 333–342.
Trochim, W., & Land, D. (1982). Designing designs for research. The Researcher, 1, 1–6.
Whitehurst, G. J. (2003). Research on mathematics education. Washington, DC: U.S.
Department of Education.
Chapter 7
Quantitative Analyses of Data
Abstract: Quantitative data from the main study are summarized and analyzed. Both the
structure and modeling workshops generated statistically significant performance gains. Thus,
after students had participated in both the workshops, their performances on both parts of the
Algebra Test—that is to say, on the questions concerning structure and on the questions
concerning modeling—were much improved. Cohen’s d effect sizes for each set of workshops
(the structure workshops and the modeling workshops) were large. The chapter concludes by
introducing two questions. First, although the performance gains were highly statistically
significant, and the effect sizes large, were they educationally significant? And, second, “What
was there about the interventions which generated such apparently impressive results?”
Keywords: Campbell and Stanley design, Effect size, Random allocation to treatments,
Statistical significance
This chapter provides quantitative analyses of pre-teaching, mid-intervention, post-

teaching, and retention data generated by parallel versions of the Algebra Test. Each version of
the Test was designed so that it would be suitable for administration to seventh-grade students
within a 45-minute class period. Each version comprised two kinds of questions—those
concerned with the structural aspects of rational numbers (hereafter referred to as “structure”
questions) and those concerned with modeling tasks, mostly involving linear sequences
(hereafter referred to as “modeling” questions). The retention Test, which was administered to
all participating students 12 weeks after the post-teaching Test, was identical to the pre-
teaching Test. The time when the pre-teaching version of the Test was administered will be
referred to as the “pre-teaching” stage of the study, and the terms “mid-intervention stage,”
“post-teaching stage,” and “retention stage” will be used in a similar way.
Each of the pre-teaching, mid-intervention, post-teaching, and retention Tests was scored
out of 20, with 10 marks being awarded for structure questions and 10 for modeling questions.
The Cronbach-alpha reliabilities of the three versions of the Test were between 0.80 and 0.85.
The three parallel versions of the Algebra Test are reproduced in Appendix B to this
book. It will be recalled that the retention version of the Test was identical to the pre-teaching
version. Question 1, on the pre-teaching and retention versions of the Test was worded: “If
Tn = 13 – 3n, where n can represent various positive counting numbers, which values of n would
make the values of Tn positive?” Question 1 on the mid-intervention version was worded in
exactly the same way, only “Tn = 13 – 3n” was replaced by “T n = 14 – 4n”; and on the post-
teaching version, “Tn = 13 – 3n” was replaced by “T n = 17 – 4n.”
Overview of Quantitative Data
Table 7.1 and Table 7.2 summarize the pre-teaching, mid-intervention, post-teaching and
retention scores for all 32 seventh-grade student participants (16 in Group 1, 16 in Group 2).

142 Ch. 7: Quantitative Analysis of Data
Table 7.1
Group 1 Students’ Pre-Teaching, Mid-Intervention, Post-Teaching, and Retention Scores
for the Structure and Modeling Questions (Out of 10, in Each Case)
Number Pre-T Pre-T Mid-Int. Mid-Int. Post-T Post-T Retention Retention
and Sex of Structure Modeling Structure Modeling Structure Modeling Structure Modeling
Student (/10) (/10) (/10) (/10) (/10) (/10) (/10) (/10)
Student 0 1 4.5 2 3 3 3.5 3

1.1 (F)
Student 0 2 6.5 2 8.5 3.5 5 5
1.2 (F)
Student 0 2 10 4.5 8 8 9 7
1.3 (F)
Student 0.5 2 6 2.5 9 9 5 4
1.4 (F)
Student 0 1 2 2.5 5 2.5 6.5 4
1.5 (F)
Student 2 1.5 6 2 6 6 4.5 7.5
1.6 (F)
Student 0 2 7 2 8 5 7 5
1.7 (F)
Student 0 1 8 2 8 3 7 5
1.8 (M)
Student 0 1 1 1 2.5 2.5 3 2
1.9 (M)
Student 0 1 1.5 1 2 2 2.5 1
1.10 (M)
Student 0 1 4 3 8 8 6.5 7.5
1.11 (M)
Student 0 1 3 1 3.5 2 3 2
1.12 (F)
Student 0 1 7 1 9 6.5 6 2
1.13 (F)
Student 0 2 3.5 1 3 5 3.5 4.5
1.14 (F)
Student 0 1 1 1 4 1 2 1
1.15 (F)
Student 0 2 7 1 6 1.5 7 2
1.16 (F)
Mean 0.16 1.41 4.88 1.84 5.84 4.28 5.06 3.91
Score/10
Standard 0.51 0.49 2.72 0.98 2.55 2.56 2.02 2.18
Deviation
Overview of Quantitative Data 143
Table 7.2
Group 2 Students’ Pre-Teaching, Mid-Intervention, Post-Teaching, and Retention Scores
for the Structure and Modeling Questions (Out of 10, in Each Case)
Number Pre-T. Pre-T Mid-Int. Mid-Int. Post-T Post-T Retention Retention
and Sex of Structure Modeling Structure Modeling Structure Modeling Structure Modeling
Student (/10) (/10) (/10) (/10) (/10) (/10) (/10) (/10)
Student 0 2 1 5.5 4 6 7 5
2.1 (M)
Student 0 1 1 6 7 4.5 5.5 5.5
2.2 (M)
Student 0 1 1 3.5 8 6 7 6.5
2.3 (F)
Student 0.5 1 2 3 2 1 4.5 1
2.4 (F)
Student 0.5 2 2.5 4 4.5 3.5 5 6
2.5 (M)
Student 1.5 2 4 5.5 7 5.5 8 4
2.6 (M)
Student 2.5 2 5 5.5 8 8 8 6.5
2.7 (F)
Student 1 2 4 5 7 7.5 8 6
2.8 (F)
Student 0.5 1 3 6.5 8 7.5 6 7.5
2.9 (F)
Student 0 1 0 1 2.5 0.5 1 1.5
2.10 (M)
Student 0 1 0.5 1.5 5 1.5 5 2
2.11 (M)
Student 0 1 0 1 6 2 6 1.5
2.12 (M)
Student 0 1 0.5 0 4.5 1 4 2
2.13 (F)
Student 0 1 0 3.5 5.5 3.5 3 2.5
2.14 (F)
Student 0 1 1 4 8 6.5 6 7
2.15 (F)
Student 1 1 0.5 5.5 5 4.5 3 7
2.16 (M)
Mean 0.47 1.31 1.63 3.81 5.75 4.31 5.44 4.47
Score/10
Standard 0.72 0.48 1.61 2.03 1.95 2.54 2.08 2.34
Deviation
In Figure 7.1, the overall trends of mean scores at the different stages of the study can
be seen. The mean scores for both groups improved for each of the Tests at the mid-
intervention and post-teaching stages, with the greater increases at the mid-intervention stage
being associated with the fact that the group with the greater mean score, on a topic, had just
received instruction on that topic (and those in the other group had not). Thus, for example,
at the mid-intervention stage, Group 1 students who had just participated in workshops which
focused on structure had a much greater mean gain on the structure questions than did Group
2, whose students had just participated in workshops which focused on modeling. The
reverse was the case for Group 2 students, with Group 2 students having a greater mean gain
on modeling questions than Group 1 students. Note that at the retention stage both groups
gained slightly lower mean scores on the structure subtest, and one of the groups obtained a
lower mean score on the modeling subtest than it had at the post-teaching stage.
The next three sections of this chapter will provide statistical and other analyses related
to the hypotheses and conjectures set out in Chapter 6.
OV ERALL T REN D O F M EAN S CO RES

Strcuture G1 Structure G2 Modeling G1
5.84
Modeling G2
5.75
5.44
5.06
4.88
4.47
4.31
4.28
MEAN SCORE (/10)
3.91
3.81
1.84
1.63
1.41
1.31
0.47
0.16
P R E-T E A C H I N G MID-INTERVENTION POST-TEACHING RETENTION
TESTING STAGES
Figure 7.1. Bar graphs, showing mean scores of the two groups at different stages, on the
structure and modeling subtests.
Analyses of Pre-Teaching Data
Because random-sampling procedures were adopted in this study, when Group 1 and
Group 2 were being formed, the first and second hypothetical populations should have
comprised students who were similar in their characteristics. The relevant samples from these
populations comprised seventh-grade students at School W who, at the beginning of the study,
had not taken the pre-teaching version of the Algebra Test, had not been interviewed, and had
not participated in any of the intervention lessons. Entries in Table 7.1 and Table 7.2 show,
among other things, the mean scores of Group 1 and Group 2 students on the pre-teaching tests
for structure and modeling (the maximum possible score, for each test, was 10).
Quantitative Analysis of Pre-Teaching Data 145
Pre-Teaching Group Differences with Respect to Structure

The first null hypothesis to be considered was that the difference between Hypothetical
Population 1’s mean score on the pre-teaching structure subtest and Hypothetical Population
2’s mean score on the same structure subtest would be zero. If we denote the null and
research hypotheses by H0 and H1, then:
H0: µ11 − µ21 = 0, with the distribution from zero being according to the
(independent samples) t-distribution, df = 30, and α = .05 (two-tailed).
H1: µ11 − µ21 ≠ 0.
Entries in Tables 7.1 and 7.2 indicate that the relevant sample means and standard
deviations for structure were, respectively, 0.16 and 0.51 (for Group 1) and 0.47 and 0.72
(for Group 2). An independent samples t-test analysis gave t = –1.42, and with 30 degrees of
freedom, the null hypothesis was accepted. At the pre-teaching stage there was no
statistically significant difference between the two groups with respect to knowledge of
structure.
Obviously, at the pre-teaching stage both groups knew very little about those questions
on the Algebra Test concerned with the “structure of rational numbers.”
Pre-Teaching Group Differences with Respect to Modeling

The second null hypothesis to be considered was that the difference between
Hypothetical Population 1’s mean score on the pre-teaching modeling subtest and
Hypothetical Population 2’s mean score on the same modeling subtest would be zero. If we
denote the null and research hypotheses by H0 and H1, then:
H0: µ12 − µ22 = 0, with the distribution from zero being according to the t-
distribution, df = 30, and α = .05 (two-tailed).
H1: µ12 − µ22 ≠ 0.
Entries in Table 7.1 and Table 7.2 indicate that the relevant sample means and standard
deviations for modeling were, respectively, 1.41 and 0.49 (for Group 1) and 1.31 and 0.48
(for Group 2). An independent samples t-test analysis gave t = .58, and with 30 degrees of
freedom, the null hypothesis was accepted. There was no statistically significant difference
between the two groups with respect to knowledge of modeling at the pre-teaching stage.
Obviously, at the pre-teaching stage both groups knew very little about the kind of
modeling represented by the questions on the Test.
Mid-Intervention Group Differences with Respect to Structure

The third null hypothesis to be considered was that the difference between Hypothetical
Population 1’s mean score on the mid-intervention structure subtest and Hypothetical
Population 2’s mean score on the same structure subtest would be zero. If we denote the null
and research hypotheses by H0 and H1, then:
H0: µ13 − µ23 = 0, with the distribution from zero being according to the t-distribution,
df = 30, and α = .05 (one-tailed).
H1: µ13 − µ23 > 0.
deviations for structure were, respectively, 4.88 and 2.72 (for Group 1) and 1.63 and 1.61 (for
Group 2). A t-test analysis gave t = 4.19, and with 30 degrees of freedom, and assuming a 1-
tailed test, the null hypothesis was rejected. It is reasonable to assume, therefore, that from the
point of view of structure, there was a statistically significant difference between the two
groups with respect to knowledge of structure at the mid-intervention stage, and that Group 1
students (who had only recently participated in structure workshops) had come to know more
about the structure of rational numbers than had the Group 2 students (who had participated in
modeling workshops). That said, the Group 1 students still had much to learn with respect to
structure.
Mid-Intervention Group Differences with Respect to Modeling

The fourth null hypothesis to be considered was that the difference between
Hypothetical Population 1’s mean score on the mid-intervention modeling subtest and
Hypothetical Population 2’s mean score on the same modeling subtest would be zero. If we
denote the null and research hypotheses by H0 and H1, then:
H1: µ24 − µ14 > 0.
(for Group 2). A t-test analysis gave t = 3.48, and with 30 degrees of freedom and assuming a
1-tailed test, the null hypothesis was rejected. There was a statistically significant difference
between the two groups with respect to knowledge of modeling at the mid-intervention stage,
and the Group 2 students (who had only recently participated in modeling workshops) had
come to know more about modeling than had the Group 1 students (who had been involved
in structure workshops). That said, the Group 2 students still had much to learn with respect
to modeling.
Post-Teaching Group Differences with Respect to Structure

The fifth null hypothesis to be considered was that the difference between Hypothetical
Population 1’s mean score on the post-teaching structure subtest and Hypothetical Population
2’s mean score on the same structure subtest would be zero. If we denote the null and
research hypotheses by H0 and H1, then:
H1: µ15 − µ25 = 0.
deviations for structure were, respectively, 5.84 and 2.55 (for Group 1) and 5.75 and 1.95
(for Group 2). A t-test analysis gave t = .12, and with 30 degrees of freedom and, assuming a
2-tailed test, the null hypothesis was accepted. There was no statistically significant
difference between the two groups with respect to knowledge of structure at the post-
teaching stage. Clearly, too, both Group 1 and the Group 2 students still had a lot to learn
with respect to structure.
Quantitative Analysis of Post-Teaching and Retention Data 147
Post-Teaching Group Differences with Respect to Modeling

The sixth null hypothesis to be considered was that the difference between Hypothetical
Population 1’s mean score on the post-teaching modeling subtest and Hypothetical
Population 2’s mean score on the same modeling subtest would be zero. If we denote the null
H1: µ16 − µ26 ≠ 0.
(for Group 2). A t-test analysis gave t = –.04, and with 30 degrees of freedom and, assuming
a 2-tailed test, the null hypothesis was accepted. There was no statistically significant
difference between the two groups with respect to knowledge of modeling at the post-
teaching stage. That said, both Group 1 and Group 2 students still had much to learn with
respect to modeling.
Retention Group Differences with Respect to Structure

The seventh null hypothesis to be considered was that the difference between
Hypothetical Population 1’s mean score on the retention structure subtest and Hypothetical
Population 2’s mean score on the same structure subtest would be zero. If we denote the null
H1: µ17 − µ27 = 0.
deviations for structure were, respectively, 5.06 and 2.02 (for Group 1) and 5.44 and 2.08 (for
Group 2). A t-test analysis gave t = –.53, and with 30 degrees of freedom and assuming a 2-
tailed test, the null hypothesis was accepted. There was no statistically significant difference
between the two groups with respect to knowledge of structure at the retention stage. That said,
both Group 1 and the Group 2 students still had a lot to learn with respect to structure.
Retention Group Differences with Respect to Modeling

The eighth null hypothesis to be considered was that there would be no difference
between Hypothetical Population 1’s mean score on the retention modeling subtest and
Hypothetical Population 2’s mean score on the same modeling subtest. If we denote the null
H0: µ18 − µ28 = 0, with the distribution from zero being according to the t-
distribution, df = 30, and α = .05 (two-tailed).
H1: µ18 − µ28 = 0.
deviations for modeling were, respectively, 3.91 and 2.18 (for Group 1) and 4.47 and 2.34 (for
Group 2). A t-test analysis gives t = –.70, and with 30 degrees of freedom and assuming a 2-
tailed test, the null hypothesis was accepted. There was no statistically significant difference
between the two groups with respect to knowledge of modeling at the retention stage. That
said, both Group 1 and the Group 2 students still had a lot to learn with respect to modeling.
Comparisons Between Mean Gain Scores at Different Times

Eight comparisons of group gain scores were made. For the first six of these, Group 1
students’ mean gain scores were compared with Group 2’s mean gain scores, with the pre-
teaching scores being regarded as baseline scores; then mean gains from the baseline scores
were compared for the two groups at the mid-intervention, post-teaching, and retention stages.
First Mean-Gain Comparison (Structure, Mid-Intervention Versus Pre-Teaching)

The null and research hypotheses were based on the assumption that since students in
Group 1 had participated in structure workshops but students in Group 2 had not, the mean
gain for Group 1 should be greater than the mean gain for Group 2. The null and research
hypotheses were:
H0: The difference between the mid-intervention versus pre-teaching mean gain
scores for the two hypothetical populations on the structure subtest would
equal zero, with the distribution from zero being according to the t-distribution,
scores for the two hypothetical populations on the structure subtest would be
greater than zero (with Group 1’s mean gain score being greater than Group 2’s).
Gain scores for individual students could be calculated from the entries in Table 7.1
and Table 7.2. For this first mean-gain comparison, which compared the mid-intervention
versus pre-teaching mean gains for Group 1 and Group 2 on the structure subtest, the mean
gain for Group 1 was 4.72 (standard deviation 2.70), and the mean gain for Group 2 was 1.16
(standard deviation 1.08). A t-test analysis yielded t = 4.91, and with 30 degrees of freedom
and assuming a 1-tailed test, the null hypothesis was rejected. The research hypothesis, that
the mean mid-intervention versus pre-teaching gain score for Group 1 on structure was
greater than the mean gain score for Group 2 on structure, was accepted.
Second Mean-Gain Comparison (Modeling, Mid-Intervention Versus Pre-Teaching)

The null and research hypotheses were based on the assumption that because students
in Group 2 had recently participated in modeling workshops but students in Group 1 had not,
the mean gain for Group 2 on the modeling subtest should be greater than the mean gain for
Group 1. The null and research hypotheses were:
scores for the two hypothetical populations on the modeling test would equal
zero, with the distribution from zero being according to the t-distribution, df =
30, and α =.05 (one-tailed).
scores for the two hypothetical populations on the modeling test would be
greater than zero (with Group 2’s mean gain score being greater than Group 1’s).
Gain scores for individual students can be calculated from the entries in Table 7.1 and
Table 7.2. For this second mean-gain comparison, which compared the mid-intervention versus
Quantitative Analysis of Gain-Score Data 149
pre-teaching mean gains for Group 2 and Group 1 on the modeling test, the mean gain for
Group 2 was 2.49 (standard deviation 1.87), and the mean gain for Group 1 was .44 (standard
deviation .96). A t-test analysis yielded t = 3.91, and with 30 degrees of freedom and assuming
a 1-tailed test, the null hypothesis was rejected and the research hypothesis, that the mean mid-
intervention versus pre-teaching gain score for Group 2 on modeling was greater than the mean
gain score for Group 1 on modeling, was accepted.
Third Mean-Gain Comparison (Structure, Post-Teaching Versus Pre-Teaching)

both Group 1 and Group 2 had recently participated in structure workshops led by the same
teacher, the difference between the mean gains for Group 1 and Group 2 on the structure
subtest should, theoretically, be zero. The null and research hypotheses were:
H0: The difference between the post-teaching versus pre-teaching mean gain scores
for the two hypothetical populations on the structure subtest would equal zero,
with the distribution from zero being according to the t-distribution, df = 30,
for the two hypothetical populations on the structure subtest would not equal
zero.
Table 7.2. For this third mean-gain comparison, which compared the post-teaching versus pre-
teaching mean gains for Group 1 and Group 2 on the structure subtest, the mean gain for Group
1 was 5.69 (standard deviation 2.55), and the mean gain for Group 2 was 5.28 (standard
deviation 1.85). A t-test analysis yielded t = .52, and with 30 degrees of freedom and assuming
a 2-tailed test, the null hypothesis was accepted. The mean post-teaching versus pre-teaching
gain scores for Group 1 and Group 2 on structure were not statistically significantly different.
Fourth Mean-Gain Comparison (Modeling, Post-Teaching Versus Pre-Teaching)

both Group 1 and Group 2 had recently participated in modeling workshops led by the same
teacher the difference between the mean gains for Group 1 and Group 2 should, theoretically,
be zero. The null and research hypotheses were:
for the two hypothetical populations on the modeling subtest would equal zero,
for the two hypothetical populations on the modeling subtest would not equal
zero.
Table 7.2. For this fourth mean-gain comparison, which compares the post-teaching versus pre-
teaching mean gains for Group 1 and Group 2 on the modeling subtest, the mean gain for
Group 1 was 2.88 (standard deviation 2.41), and the mean gain for Group 2 was 3.00 (standard
deviation 2.35). A t-test analysis yielded t = –.15, and with 30 degrees of freedom and
assuming a 2-tailed test, the null hypothesis was accepted. The mean post-teaching versus pre-
teaching gain scores for Group 1 and Group 2 on modeling were not statistically significantly
different.
Fifth Mean-Gain Comparison (Structure, Retention Versus Pre-Teaching)

The null and research hypotheses were based on the assumption that since the students
in both Group 1 and Group 2 had participated in structure workshops led by the same
teacher, the difference between the retention versus pre-teaching mean gains for Group 1 and
Group 2 on the structure subtest should, theoretically, be zero. The null and research
hypotheses were:
H0: The difference between the retention versus pre-teaching mean gain scores for
the two hypothetical populations on the structure subtest would equal zero,
the two hypothetical populations on the structure subtest would not be equal.
Table 7.2. For this fifth mean-gain comparison, which compares the retention versus pre-
teaching mean gains for Group 1 and Group 2 on the structure subtest, the mean gain for
Group 1 was 4.91 (standard deviation 2.12), and the mean gain for Group 2 was 4.97
(standard deviation 1.79). A t-test analysis yielded t = – .09, and with 30 degrees of freedom
and assuming a 2-tailed test, the null hypothesis was accepted. The mean retention versus
pre-teaching gain scores for Group 1 and Group 2 on structure were not statistically
significantly different.
Sixth Mean-Gain Comparison (Modeling, Retention Versus Pre-Teaching)

The null and research hypotheses were based on the assumption that both Group 1 and
Group 2 had participated in modeling workshops led by the same teacher and therefore the
retention versus pre-teaching mean gains for Group 1 and Group 2 on the modeling subtest
should, theoretically, be equal. The null and research hypotheses were:
the two hypothetical populations on the modeling subtest would equal zero,
the two hypothetical populations on the modeling subtest would not equal zero.
Table 7.2. For this sixth mean-gain comparison, which compares the retention versus pre-
teaching mean gains for Group 1 and Group 2 on the modeling subtest, the mean gain for
Group 1 was 2.50 (standard deviation 2.06), and the mean gain for Group 2 was 3.16
(standard deviation 2.24). A t-test analysis yielded t = –.86, and with 30 degrees of freedom
and assuming a 2-tailed test, the null hypothesis was accepted. The difference between the
mean retention versus pre-teaching gain scores for Group 1 and Group 2 on modeling was
not statistically significantly different from zero.
Quantitative Analysis of Gain-Score Data 151
Seventh Mean-Gain Comparison (Structure, Retention Versus Post-Teaching)

There was a 12-week period between the times when the post-teaching and parallel
retention Tests were administered and it was a matter of interest whether Group 1 students
would retain whatever they had learned during the structure workshops over that period more
than would Group 2 students. During the 12 intervening weeks the participating students did
not focus on either structure or modeling in their mathematics classes.
This seventh mean-gain comparison checks whether the difference between mean gain
scores for Group 1 and Group 2 on the retention versus post-teaching structure subtests
differed from zero.
H0: The difference between the retention versus post-teaching mean gain scores for
the two hypothetical populations on the structure subtest would equal zero,
the two hypothetical populations on the structure subtest would not equal zero.
Table 7.2. For this seventh mean-gain comparison, which compares the retention versus post-
teaching mean gains for Group 1 and Group 2 on the structure subtest, the mean gain for
Group 1 was –0.78 (standard deviation 1.70), and the mean gain for Group 2 was –0.31
(standard deviation 1.63). A t-test analysis yielded t = .80, and with 30 degrees of freedom
mean retention versus pre-teaching gain scores for Group 1 and Group 2 on structure was not
statistically significantly different from zero.
Eighth Mean-Gain Comparison (Modeling, Retention Versus Post-Teaching)

This eighth mean-gain comparison checks whether the difference between mean gain
scores for Group 1 and Group 2 on the retention versus post-teaching modeling subtests
equaled zero.
the two hypothetical populations on the modeling subtests would equal zero,
the two hypothetical populations on the modeling tests would not equal zero.
Table 7.2. For this eighth mean-gain comparison, which compares the retention versus post-
teaching mean gains for Group 1 and Group 2 on the modeling subtests, the mean gain for
Group 1 was –0.38 (standard deviation 1.95), and the mean gain for Group 2 was 0.16
(standard deviation 1.28). A t-test analysis yielded t = –.91, and with 30 degrees of freedom
mean retention versus pre-teaching gain scores for Group 1 and Group 2 on modeling was
not statistically significantly different from zero.
Calculation of Effect Sizes

The following four Cohen’s (1988) d effect sizes were computed:
1. Effect of pre-teaching to mid-intervention workshop sessions on structure with
Group 1 (it was assumed that the control group for this period was Group 2);
2. Effect of mid-intervention to post-teaching workshop sessions on structure with
3. Effect of pre-teaching to mid-intervention workshop sessions on modeling with
4. Effect of mid-intervention to post-teaching workshop sessions on modeling with
Group 1 (it was assumed that the control group for this period was Group 2).
Summary results of the calculations are shown in Table 7.3.
Table 7.3
Effect Sizes for Four Intervention Workshops
Type of Period Group Involved in Control Cohen’s d
Workshop (Pre-T to Mid-I, Relevant Workshops Group Effect
(Structure or or Mid-I to (Group 1 or Group 2) Size
Modeling) Post-T)
Structure Pre-T to Mid-I Group 1 Group 2 1.74
Structure Mid-I to Post-T Group 2 Group 1 1.70
Modeling Pre-T to Mid-I Group 2 Group 1 1.38
Modeling Mid-I to Post-T Group 1 Group 2 1.07
Adopting Cohen’s (1988) criteria, whereby effect sizes around 0.2 are “small,” those
around 0.5 are “medium,” and those more than 0.8 onwards are “large,” it can be seen, from
entries in Table 7.3, that the effects on performance on the structure subtest of the structure
workshops were very large, for both Group 1 and Group 2. The effects for performance on
the modeling subtest of the modeling workshops were also large, especially for the Group 2
students. This conclusion is amplified by Figure 7.2 and Figure 7.3 in which the mean subtest
scores at four stages are depicted and corresponding mean scores for each group are joined
by line intervals.
In Figure 7.2 and Figure 7.3 the “fall-off” in mean scores for three of the four
comparisons between the administrations of the post-teaching Test and the retention Test is
exaggerated with respect to time. That is because whereas the time between the pre-teaching
and mid-intervention administrations of the test was about only three weeks, and the same
was true for the period between the mid-intervention and post-teaching administrations of the
tests, the time between the post-teaching and retention administrations of the Test was about
12 weeks. Nevertheless, the fall-offs do raise important issues which could usefully be taken
up by future researchers.
Structure and Modeling Mean Score Trends 153
Group 1 and Group 2 Structure Mean Score Trends

Mean score(/10)
Pre-T (Week 1) Mid-Int (Week 3) Post=T (Week 6) Ret-T (Week 18)

Timeline
Figure 7.2. Group 1 and Group 2 students’ pre, mid-, post-teaching, and retention mean
scores on the structure subtest (maximum possible score was 10).
Group 1 and Group 2 Modeling Mean Score Trends

5
Group 1 Group 2 4.31 4.47
4.5
3.81
4 4.28
3.91
3.5
Mean Score(/10)
3
2.5
2
1.5 1.41
1.84
1 1.31
0.5
0
Pre-T (Week 1) Mid-Int(Week 3) Post-T (Week 6) Ret-T (Week 18)
Timeline
Figure 7.3. Group 1 and Group 2 students’ pre-, mid-, post-teaching, and retention mean
scores on the modeling subtest (maximum possible score was 10).
Summary and Concluding Comments on the Quantitative Analyses

The random allocation to groups was successful in that the initial, pre-teaching mean
scores of the two groups so formed were approximately equal on both the structure and
modeling subtests within the Algebra Test.
The structure intervention workshops were highly effective in that they generated large
mean gains on the structure tests (for Group 1 at the mid-intervention stage, and for Group 2
at the post-teaching stage). The modeling intervention workshops were also very effective in
that they generated impressive mean gains on the modeling tests for both groups. The gains
for modeling were not as large as the gains for structure. Analyses of student responses to
questions on the retention tests (which were administered 12 weeks after the post-teaching
tests) suggested that for both groups there had been a slight (statistically non-significant) fall-
off in student understanding, between the post-teaching and retention stages for structure, but
Group 2 had actually seemed to improve with respect to modeling.
The pattern for the results was very similar to that reported by Zhang, Clements and
Ellerton (2015), for a study involving fifth-grade students which had a similar design to the
main study reported in this book. In that study, which also took place at School W, fifth-
grade students were randomly allocated to two groups. One of the groups took part in
multiple-embodiment workshops while the other group did not take classes in mathematics;
then, the second group participated in the same type of multiple-embodiment workshops
while the first group did not take classes in mathematics. The same pattern of results found in
the current study was obtained. Immediately after the multiple-embodiment workshops the
first group had a large mean gain on a fractions test but the other group (which had not yet
participated in the workshop) did not have a statistically significant mean gain. However,
after the second group had participated in the workshops, it had a large mean gain on the
fractions test, and the two groups obtained almost identical mean scores on the test. Then,
after a 12-week break, a parallel retention test was administered to both groups and it was
found that although the mean score for one of the groups had fallen a little from its post-
teaching high, there were still very large pre-teaching to retention mean gains. None of the
“fall-offs” for the groups was statistically significant.
Two additional comments should be made. First, Vaiyavutjamai (2003) has shown that
highly statistically significant gains and large effect sizes do not necessarily correspond to
educationally significant results. And second, in the current study we were challenged by the
same question that Zhang et al. (2015) faced: “What was there about the teaching
interventions which generated such apparently impressive results?” Answers to that question
with respect to the present study will be offered in Chapter 9 of this book. However, some
light will be shed on the question, and on possible answers to it, in the next chapter—which
provides a summary and analyses of qualitative data.
References
Vaiyavutjamai, P. (2003). Dilemmas associated with quantitative analyses of effects of
traditional mathematics teaching. In H. S. Dhindsa, S. B. Lee, P. Achleitner, & M. A.
Clements (Eds.), Studies in science, mathematics and technical education (pp. 259–
268). Gadong, Brunei Darussalam: Universiti Brunei Darussalam.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015). Conceptual (mis)understandings of
fractions: From area models to multiple embodiments. Mathematics Education
Research Journal, 27(2), 233–261.
Chapter 8
Qualitative Analyses of Data
Abstract: Qualitative data from the main study are summarized, analyzed, and interpreted
from the perspective of Herbart’s theory of apperception and Del Campo and Clements’s
theory of receptive-expression modes of communication. For many of the students, there was
evidence of “significant growth,” but for some, there was “no evidence.” Findings from these
analyses complemented and supported findings from the quantitative analyses in Chapter 7.
Qualitative analyses of pre-teaching data suggested that the students remembered very little,
if anything, about structures and modeling that they had previously studied—despite the fact
that common-core expectations would be that they should have had a strong grasp.
Keywords: Apperception, Attitudes to algebra, Expressive modes of communication,

Imagery in algebra, Johann Friedrich Herbart, Receptive modes of communication
Intended, Implemented and Received Curricula
Consistent with a design-research approach, the research team decided to frame the
investigation within a composite theoretical base which, it was expected, would be the most
helpful in solving the problem that had been identified. Interviews with the seventh-graders in
the pilot study (Kanbir, 2014, 2016), as well as analysis of pre-teaching data for the main
study, revealed that if, for example, seventh- or eighth-grade interviewees were shown a sign
1
like “Find the value of 64 × ( × 120),” most of them would not link, in their minds, the 64
32
1 1
and the . Rather, they would proceed by trying to find the product of and 120 and,
32 32
having done that, they would then multiply their result by 64. This suggested that the students
had not really received the intended message of the person(s) who created the sign.
1
The reader might protest that the request “Find the value of 64 × ( × 120)” is a
32
sentence, and is not really a “sign.” But, from a Peircean perspective it is a sign: whoever
framed the request had a mathematical object in mind—that object was the associative
property of multiplication of rational (or real) numbers. Students who responded by initially
1
trying to multiply and 120 had seen the sign, but had not responded to it in a way which
32
showed that they recognized the mathematical object that the sign represented. When the
same students were asked to say whether (a × b) × c was always, or sometimes, or never,
equal to a × (b × c), where a, b, and c represented numbers, they did not grasp the meaning of
the question, and none of them used the words “associative property of multiplication.”
Similarly, when shown a sign like “describe a quick method for finding the value of
7 × 97 + 7 × 3,” none of the pilot-study students, or the students at the pre-teaching stage of the

156 Ch. 8: Qualitative Analyses of Data
main study, immediately recognized that 7 × 97 + 7 × 3 was equal to 7 times (97 + 3), or 700.
None of them gave any indication that the task might be related to what the CCSSM (2010)
sequence for elementary and middle schools referred to as the “distributive property.”
Instead, the interviewees proceeded to try to use, in a rote way, the PEMDAS (“Please
Excuse My Dear Aunt Sally”) order-of-operations mnemonic.
The quality of the expressive understandings of the structural properties of numbers of
the seventh-grade pilot-study students and the students in the main study at the pre-teaching
stage was no better than the quality of their receptive understandings. When specifically
asked to give verbal descriptions of the associative property for addition, or the associative
property for multiplication, or the distributive property of multiplication over addition, none
of them knew the meanings of those terms (although a few of them recalled that they had
heard their teachers use such expressions earlier, in elementary or middle-school mathematics
classes).
A similar conclusion was also reached with respect to the students’ pre-teaching
responses to modeling tasks. Whereas, they were capable of identifying recursive rules, such
as “add 3,” when shown, in tables of values, successive terms in sequences, they were not
able to identify explicit rules by writing statements such as “the nth term is equal to 3n – 1.”
More basically, their comprehension of the inter-connectedness of terms in tables of values
was extremely limited.
At the pre-teaching stage, then, pilot-study students and students in the main study had
neither receptive nor expressive understandings of key algebraic symbols which the authors
of the CCSSM (2010) document, of the NCTM Standards ( National Council of Teachers of
Mathematics, 2000), and of middle-school textbooks (e.g., Charles, Branch-Boyd,
Illingworth, Mills, & Reeves, 2004), presumed that they should know. It is one thing to argue
that “when teachers ask students to represent relationships that already make sense to them,
the transition from words to variables actually is not as difficult as might be expected”
(Knuth, Stephens, Blanton, & Gardiner, 2016, p. 66, original emphasis retained), but it is
another thing to work out what the reality of the situation is with middle-school students in
schools which have not been part of large early-algebra intervention projects such as those
described by Maria Blanton and her co-researchers (see, e.g., Blanton, Brizuela, Gardiner,
Sawrey, & Newman-Owens, 2015; Blanton & Kaput, 2011; Blanton, Stephens, Knuth,
Gardiner, Isler & Kim, 2015). There can be large differences between intended, implemented
and received curricula (Westbury, 1980).
The question arose: what can be done in ordinary schools to assist middle-school
students and their teachers to make sense of the relationships? The main study described in
this book represented an attempt to answer that question.
Peirce’s Triadic Semiotic Position and Herbart’s Theory of Apperception

Since the pilot-study students did not recognize or know the meanings of important
signs, it was impossible for them to reach out and grasp the associated “mathematical objects”
prescribed in the curriculum until they had acquired receptive and expressive knowledge and
understandings of the meanings of those signs. With Peirce’s triadic theory, appropriate
interpretants might be regarded as kinds of bridges between the “signs” and the “mathematical
objects” which were being signified. Before the students could learn the meanings of the signs
intended by those who framed the CCSSM sequence they would need to be involved in
relevant educative situations (Campos, 2010). It was the first author’s recognition of that state
of affairs which led him to choose Herbart’s theory of apperception as part of the bundle of
Peirce’s and Herbart’s Theories 157
theories which defined a composite theoretical base for the study. The theory had direct
application in the lesson planning for the workshop interventions.
Research team members decided to design an intervention which would include the
collection and analysis of pre- and post-teaching interview data as well as pre-teaching, mid-
intervention, post-teaching, and retention pencil-and-paper test data. By that means, the
research team would gather data relating to the students’ receptive and expressive
understandings of the key signs before, during, and after the interventions.
Johann Friedrich Herbart (1904a, 1904b), a German philosopher, educator, and
psychologist, emphasized the need for teachers to take account of what modern scholars have
called students’ “cognitive structures” (Gagné & White, 1978) or “concept images” (Tall &
Vinner, 1981) with respect to the mathematical object which it is intended the students will
learn. During the second-half of the nineteenth century, Herbartian thinking on this matter
became a cause célèbre among educators in many parts of the world (see, e.g., Adams 1898;
Cole, 1912; Ellerton & Clements, 2005; Hayward, 1904) but, suddenly, early in the twentieth
century, Herbartianism lost favor. Harold B. Dunkel’s (1970) scholarly book, Herbart and
Herbartianism: An Educational Ghost Story, in tracing the influence of Herbartianism in the
United States, sought to explain why Herbartianism’s “fame blazed up like a meteor and
meteor-like was extinguished” (p. 4). A discussion of why that occurred is beyond the scope
of the present study (but interested persons might consult Ellerton and Clements, 2005).
Richard Selleck (1968), a distinguished education historian, stated that whatever
reservations commentators might have of Herbart’s views, “his work has a complexity,
subtlety and coherence which make it more impressive than the writings of comparative
amateurs such as Froebel or Pestalozzi” (p. 227). Such an assessment is hardly an
exaggeration, for Herbart was a philosopher good enough to hold the Chair in Philosophy at
Königsberg University not long after it had been held by Immanuel Kant. Herbart’s theory of
apperception was chosen to complement Peirce’s triadic theory for the main study because, it
could be argued, the emphasis of some modern education researchers on the need to take
account, when planning instruction, of what is already “in the students’ heads,” could be
traced to Herbart’s ideas.
Although, probably, twentieth-century scholars such as Lev Vygotsky, Jean Piaget,
Jerome Bruner, David Ausubel, Robert Gagné, and Shlomo Vinner, would not have
considered themselves to be neo-Herbartianists, each of them was nevertheless concerned
with planning instruction so that it would take maximum advantage of what prospective
learners already knew, and of how they would be likely to think, about a topic which was
about to be taught. Gagné and White (1978) argued that cognitive structure could be regarded
as being made up of four separable components—verbal knowledge, intellectual skills,
imageries, and episodes. Later, Gagné added “attitudes” and “motor skills” to the list (see
Gagné, 1985; Gagné & Merrill, 1990). Although these components were conceptually
separable, Gagné argued that it was the idiosyncratic cognitive links between them, already
existing in a learner’s long-term memory, which most influenced what and how that learner
would learn from an instructional sequence.
Ian Westbury’s (1980) distinction between intended, implemented and received
curriculum is particularly important for interpreting the qualitative analyses carried out in this
chapter. The received curriculum for an individual learner might be thought of as that
individual’s cognitive structure, with respect to a desired mathematical “object,” which is the
result of the student having participated in an instructional intervention aimed at helping
students to achieve that mathematical object. Although the received curriculum is not the
scores which the student obtains on post-teaching or retention tests, nevertheless those scores
are generated by a student’s received curriculum and the student’s responses to the signs
which appeared on the pencil-and-paper tests. The received curriculum for an individual
learner is not fixed in time. From that perspective, it was a matter of interest, in the main
study, to investigate whether, and how, a participating student’s cognitive structure
immediately before the intervention differed from her or his cognitive structure immediately
after the intervention, and also 12 weeks after the intervention.
In this chapter the qualitative analyses will be mainly, but not solely, based on data
gathered with respect to pre- and post-intervention cognitive structures which characterized
the thinking of the participating seventh-grade students, especially in relation to the
associative properties for addition and multiplication, the distributive property, and the
elementary notions of modeling which required some knowledge of the concept of a variable.
Data will be analyzed in an attempt to describe the participating students’ pre- and post-
intervention cognitive structures in terms of verbal knowledge, intellectual skills, imagery,
episodes, and attitudes—with Gagné’s “motor skills” not being considered relevant to this
study. The aim will be to show how students’ understandings of the mathematical objects
associated with key middle-school algebra concepts were enhanced as they developed their
verbal knowledge, intellectual skills, imagery, episodes, and attitudes. This kind of analysis
not only required the identification of changes in the separable components of cognitive
structures, but also some discussion of how the links in cognitive structure between those
components changed.
Analyses of Qualitative Data Generated by the “Structure” Intervention
One focus of the main study was to describe the extent to which the participating
seventh-grade students learned to make associative and distributive transformations, with
addition and multiplication of real numbers (which, in the context of the study, was of
rational numbers only, for irrational numbers were not considered). This required the
gathering of data from students both before and after they participated in lessons emphasizing
associative and distributive properties of rational numbers. As part of this focus, the
investigation gathered data on whether students responded confidently and accurately, both
receptively and expressively, to signs for which the mathematical objects under consideration
were the associative and distributive properties of rational numbers.
Of course, it is not possible to “videotape” what is going on in a student’s mind—what a
learner knows, and how he or she thinks, has to be inferred from what is said or done in
situations involving (or possibly involving) the mathematical objects under consideration.
Table 8.1 summarizes types of data likely to be generated by certain activities and tasks
related to students’ responses to associative and distributive properties.
Careful examination of summary statements in Table 8.1 will suggest how students’
responses might be interpreted. In the table, a distinction is made between receptive and
expressive responses by the students. Herbartian apperception theory was deemed to be
especially useful for thinking in terms of bridging the educational gap between interpreting
signs used to signify structures and the actual “signifieds”—that is to say, the mathematical
objects being signified.
Components of Cognitive Structure 159
Table 8.1
Evidences for Qualities of Components of Students’ Concept Images with Respect to the
Associative and Distributive Properties
Component Examples of Receptive Examples of Expressive Evidences for Aspects of
of Working Outcomes for Outcomes for Concept Image for
Memory “Structure” “Structure” “Structure”
Verbal Can say, or write down, Can explain the meanings Verbal knowledge can be
Knowledge accurately and from of the associative and tested through pencil-and-
memory, what has been distributive properties paper tests, or through task-
learned about the accurately, and can apply based interviews.
associative and distrib- them in relevant but
utive properties. There is unrehearsed situations.
no need for this to be
verbatim (Gagné &
White, 1978).
Intellectual Can follow a demon- Can express why PEMDAS Willingness to employ
Skills stration for a task—e.g., rules for order of operations directly an associative or
17 + 84 = 17 + (83 + 1) should not be applied before distributive transformation
= possible implications of the with tasks like, “Find the cost
(17 + 83) + 1 = 100 + 1 associative and distributive of 7 apples at 99 cents each,”
= 101, and appreciates properties are considered— and “Find the value of
the sense of proceeding and generate examples 2 × (50 × 7.13).”
in that way. showing this.
Imageries Recognizes visual pat- Can generate visual patterns The middle-school CCSSM
terns in arithmetic when of calculations for which (2010) sequence refers to
these are pointed out— the associative and diagrams illustrating the
e.g., recognizes after distributive properties sense of m(a + b) = ma + mb,
someone has pointed it would be appropriate—e.g., and a(bc) = (ab)c. However,
out, that both 97 × 5 + 398 + 403. many seventh-graders find it
3 × 5 and 17 × 5.01 + 83 Can draw diagrams difficult to link the diagrams
× 5.01 have “common” illustrating the sense of the to the algebraic statements,
factors and therefore can number properties. and also to realize that the m,
be dealt with using the a, b and c are being used as
distributive property. variables.
Episodes Remembers episodes in Recalls episodes in Recalls working with other
mathematics classes mathematics classes when students in group work, and
when someone explains someone explained how planning for and making
how useful it could be to useful it can be to use the associated group presen-
use associative and associative and distributive tations on number properties.
distributive transfor- transformations in mental Recalls details of the number
mations in mental calculations, and can create topics which were the subject
calculations. tasks in which such of group discussion and
transformations would be presentation.
useful.
Attitudes Makes observable Makes explicit comments Of special interest is whether
positive or negative on the (ir)relevance and a student’s positive or
responses when number value of associative and negative responses affect the
properties are being distributive transformations. quality of his or her learning
studied. of number properties.
The most comprehensive set of qualitative data was generated by responses by students in
the 56 one-on-one interviews (28 at the pre-teaching stage, and 28 at the post-teaching stage).
Those interviews were audiotaped, with each interview lasting approximately 30 minutes.
Although the interview data would provide the foundation for the analysis which will now be
presented, not all of it will be summarized in this chapter. After all of the pencil-and-paper data
had been collected, and the pencil-and-paper test and homework data analyzed, the five
members of the research team (Mr. X, Mr. Y, and the three authors) met to discuss aspects of
the study, and that discussion was audiotaped. Statements made in that discussion were
regarded as qualitative data, and will be summarized toward the end of Chapter 9.
In order to simplify the analyses, qualitative data for the “structure” aspect of the study
will be dealt with before data from the “modeling” aspect of the study are considered.
Analyzing Samples of Interview Data Relating to “Structure”

Each pre-teaching and post-teaching interview began with the interviewer making the
following statement:
I am going to say two words to you and, as soon as I say them, I want you to say
your head after I say the words. Here are the words … “distributive property.”
Here are the words again: “distributive property.” What comes into your mind?
The excerpts in Figure 8.1 come from pre- and post-teaching interview transcripts for
Student 2.5—whose pre-teaching response was the only answer at the pre-teaching stage in
which a student gave evidence of knowing something about the distributive property
and of being able to express it verbally. However, the student confused the associative property
for multiplication with the distributive property for multiplication over addition, and did not
give a correct instance illustrating either the distributive property or the associative property for
multiplication. Student 2.5’s post-teaching response to the same request—to say or do
something in response to the words “distributive property”—is also shown in Figure 8.1.
Student 2.5, Pre-Teaching Interview: I remember from my lower grades. I don’t

remember exactly, but it is something about when you take something and you put
something somewhere else. You do one thing at a time.
Interviewer: That is something you will be learning about. Did you have anything or any
problem come in your mind? Can you describe what you are saying or illustrate what you
are saying?
Student 2.5: Okay, I have written something. In parentheses 3 times 7 and times 4
outside of the parentheses. And I switched around and it becomes 4 times in parentheses
3 times 7.
Student 2.5, POST-Teaching Interview: You will be multiplying 37 times 4 and you
would break it down as 30 times 4 and you add with 7 times 4, you and you get 148.
Figure 8.1. Student 2.5’s responses to the words “distributive property” in the pre- and post-
teaching interviews.
Other interview data indicated that not only did Student 2.5’s concept image of the
distributive property change with respect to verbal knowledge, but also with respect to
intellectual skills, imagery, and episodes (see Appendix H).
Interview Data—For Structure 161
The concept image of another student, Student 2.6, also obviously changed between the
pre- and post-teaching stages. Data in Figure 8.2 show how, at the two stages, he responded
to the initial request (regarding the distributive property) in qualitatively different ways. It
should be noted that at the post-teaaching stage both Student 2.5 and Student 2.6 were able to
generate their own examples to illustrate distributive transformations.
Student 2.6’s concept image with respect to structure also changed for the associative
property of multiplication. One of the interview questions was “Without using a calculator,
find the value of 4 × (1/4 × 128),” and Figure 8.3 shows his responses in the pre- and post-
teaching interviews.
Student 2.6, Pre-Teaching Interview: Distributing numbers from multiple numbers [He
could not provide any examples or any extra information].
Student 2.6, POST-Teaching Intervview: Here is an example if you have 2 times 15 and 2
times 85 you would do this as 2 times in parenthesis 15 plus 85 and get 2 times 100
which is 200
Figure 8.2. Student 2.6’s written pre- and post-teaching interview responses to the request
regarding the “distributive property.”
1
Task: Without using a calculator find the value of 4 × ( × 128) without using a
4
calculatorr.
Student 2.6, Pre-Teaching Interview: I would first make ¼ into a decimal. Then I would
make ¼ have a 10,000 or 100 denominator. I could get to 100 by multiplying 4 times 25.
Then you multiply 0.25 and 128.
Student 2.6, POST-Teaching Interview: I switched this parenthesis around these first
two numbers. From there, 4 times one fourth and I got 1, and I multiplied it by 128 and
got 128.
Figure 8.3. Student 2.6’s written pre- and post-teaching interview responses to a question
concerned with the associative prroperty for multiplication.
For the task “Find the value of 98 × 6 + 98 × 4,” Student 1.7 knew, at the post-teaching
stage, what to do, and was able to articulate the correct idea for what she did—but she did not
give the correct name for the property (see Figure 8.4). Her responses illustrate the fact that
someone’s concept image with respect to some topic is not made up solely of separate
components (verbal, skills, imagery, episodes, attitudes). Relationships between the
components are also important in defining the concept image. The quality of a student’s
response to a task is likely to be determined by the cognitive “push” of the dominant
component(s) of the concept image that she or he associates with the task.
Student 1.7, POST-Teaching Interview: 980. I did 6 plus 4 is 10 and I multiplied 98

times 10 so I got 980.
Interviewer: What is the name of the property you just used?
Student 1.7: Associative.
Figure 8.4. A student, in responding to “Find the value of 98 × 6 + 98 × 4,” demonstrated a
correct expressive understanding of the distributive property, without knowing its name.
For the “98 × 6 + 98 × 4” task, knowledge of the actual name of the property was not
really important in the context of the interview. However, it is easy to imagine a classroom
situation in which it could be important. Consider what might happen, for example, if the
teacher were to be talking about the associative property for multiplication, or the distributive
property, but some of the students mixed the labels. The end result might be that the students
would not understand what the teacher was saying.
In order to collect evidence relating to a student’s memory of episodes and their
attitudes to the mathematics with which they had been concerned in the intervention
workshops, students were asked directly what they remembered most about the workshops,
what they liked most and what they liked least about them. Although it is recognized that an
interviewee might give guarded, less-than-frank, answers to such questions, the responses
could nevertheless be revealing. In the main study, the responses helped the first author to
map students’ cognitive structures (see Appendix H). Figure 8.5 shows the response by
Student 1.1 to a post-teaching interview question.
Interviewer: Think about the special lessons you’ve just had with Mr. X and Mr. Y. What
is the thing that you remember most?
Student 1.1, POST-Teaching Interview: I remember most about pattern and visualizing
the crossing-the-river problem. Mr. X. taught us the distributive and the associative
properties and how to use parentheses.
Interviewer: What did you like least about the lessons?
Student 1.1: I was okay mostly but with the associative property sometimes I forget how
to use a parenthesis and how to work with the problems.
Figure 8.5. Student 1.1 indicated that although she really liked the “crossing-the-river”
modeling task, the associative properties sometimes confused her.
The student indicated that she remembered, most of all, her participation in the
“crossing-the-river” task—which is interesting because of all the tasks in which the students
engaged, for both the structure and modeling interventions, the “crossing-the-river” task was
the only one for which students were expected to make hands-on manipulations of special
equipment.
Cognitive Growth—For Structure 163
Cognitive Growth in Structure-Related Content Domains

Immediately before the first intervention workshops the 32 participating students were
asked to respond in writing to the written question: “Do you have any idea what the
distributive property for multiplication over addition for real numbers states?” Of the 32
responses, 4 replied “Yes,” and 28 replied “No.” When asked to explain in writing what the
distributive property was all about, none of the four who answered “Yes,” gave a
mathematically adequate reply—but two of the responses seemed to point in the right
direction. The four statements were:
• “When you distribute something or divide it.”
• “It is when you do multiplication over addition first.”
• “When you write out numbers in word form and solve the problem.”
• “You distribute the numbers to multiply.”
Although there were no elaborations of these statements it seems to have been the case that at
the pre-teaching stage not one of the 32 students could link the words “distributive property”
with what the distributive property is—despite the fact that that property is much emphasized
in CCSSM’s (2010) statement of the middle-school intended curriculum.
Qualitative analysis of responses to questions on the pre-teaching version of the pencil-
and-paper Algebra Test revealed that the lack of knowledge concerning structure was not
confined to the distributive property. Among questions on structure on the Algebra Test were:
2. A really important property for numbers and for algebra is called the associative
property for multiplication. Describe this property in your own words?
3. Suppose you were asked to calculate the value of 940 + (60 + 403) in your head
(without writing anything down, or using a calculator). How would you do it, and
which property would you be using?
8. Without using a calculator find the value of (72 × 5) × 2, and explain how you got
your answer.
11. If 20 × (10 + 5) = (20 × 10) + (20 × y), what must y equal? Explain how you got
your answer.
13. What would be a quick method of finding the value of 7 × 97 + 7 × 3 without using
a calculator? What is the property which allows you to use that quick method?
1
15. What would be a quick method of finding the value of 64 × ( × 120), without
32
using a calculator?
A careful reading of these questions will reveal that the students were encouraged to write
what they knew about the associative properties for addition (Question 3), the associative
property for multiplication (Questions 2, 8, and 15) and the distributive property (Questions
11 and 13).
During the test administration at the pre-teaching stage, students were asked to provide
answers to every question—and, not surprisingly, therefore, written responses to all questions
were provided by all 32 students. In fact, hardly any of the 32 students gave a correct response
to any of the above questions. Most of them did not give correct verbal definitions, and they
obviously did not know what the associative properties for addition and multiplication, or the
distributive property, were. Four of the 32 students indicated that they had vague memories of
having, at some time in the past, dealt with the properties in mathematics classes, but none of
them offered a mathematically-adequate description of what the properties were. For
Question 3, only one student combined the 940 and 60 before adding 403; for Question 8, no
student multiplied 5 by 2 before multiplying by 72. And so on. There could be no other
conclusion than this: at the pre-teaching stage, the 32 students did not know the relevant
structures, and could not apply them.
At the post-teaching stage, however, we shall see that more than half of the participating
students were able to demonstrate a relatively strong knowledge of the properties.
Changes in Concept Images for the Associative Property of Addition

The question on the associative property of addition was such that none (or perhaps 1)
of the 32 students gave a correct answer at the pre-teaching stage, but 26 of them did so at the
post-teaching stage. For a question on the associative property of multiplication, the numbers
of students correct at pre- and post-teaching stages were 1 and 27, respectively. And, for a
question on the distributive property for multiplication over addition, the numbers correct at
the pre- and post-teaching stages were 1 and 24, respectively. The gains are so large that it is
worth reflecting on what seems to have happened in the students’ minds.
Table 8.2 summarizes pre- and post-teaching qualitative data for all 28 interviewees in
terms of their reactions to, and knowledge of, the associative property of addition of rational
numbers. The data summary is not just based on interview responses, but also on responses to
pencil-and-paper questions at the pre- and post-teaching stages. Note that the concept images
of 23 of the 28 students (i.e., about 82%) changed in educationally significant ways as a result
of the intervention lessons (see, also, data reported in Appendix H to this book).
Table 8.2
Summary of Data from 28 Interviewees in Relation to Concept Images
for the Associative Property for Addition
Evidence for Correct Pre- Evidence for Correct Post-
Teaching Component Teaching Component
in Long-term Memory in Long-term Memory
Attitudes
Attitudes
Episodes
Episodes
Imagery
Imagery
Verbal
Verbal
Skills
Skills
Strong
0 0 0 0 0 18 21 20 21 21
Evidence
Some
Evidence 0 0 0 1 1 5 2 3 2 2
No
28 28 28 27 27 5 5 5 5 5
Evidence
In Table 8.2, the five “cognitive structure” components—verbal knowledge, intellectual

skills, imagery, episodes, and attitudes—are mentioned for the pre- and post-teaching stages.
Changes in Concept Images—For Structure 165
The term “strong evidence” was used to indicate that there was definite evidence that a
student had developed a sound knowledge of the concept, or had developed relevant skills, or
could evoke appropriate imagery, or could recall relevant episodes, or had developed positive
or otherwise appropriate attitudes. “Some evidence” implied that although there was
evidence, it was not strong; “no evidence” indicated that no evidence existed in relation to
that component of long-term memory. The movement from “no evidence” toward “strong
evidence” is striking—it occurred with respect to each of the five cognitive structure
components.
Changes in Concept Images for the Associative Property of Multiplication

Qualitative analysis of pre- and post-teaching student responses to pencil-and-paper and
interview questions revealed that at the pre-teaching stage none of the 32 participating
students had a well-formed knowledge of the associative property for multiplication. They
lacked knowledge of definitions (verbal knowledge), did not have well-developed and
appropriate intellectual skills, and could not evoke appropriate images. A few of them
remembered having heard the expression “associative property for multiplication” being used
by the teacher at some stage in their schooling, but they did not remember details, and they
had no well-developed attitudes because it was not something they had ever known or
thought about.
Entries in Table 8.3 show the strength of evidence for the presence of, and connections
between, each the five “cognitive structure” components—verbal knowledge, intellectual
skills, imagery, episodes, and attitudes—of the participating students at the pre- and post-
teaching stages. There can be no doubt that as a result of the intervention major changes
occurred for most (but not all) of the students with respect to their thinking about the
associative property for multiplication. In fact, with both the associative property for addition
and the associative property for multiplication, about 75% of the participating students
displayed an obvious movement from “no evidence” toward “strong evidence” for all five
categories of their long-term memories.
Table 8.3
Summary of Data from 28 Interviewees in Relation to Concept Images for the
Associative Property for Multiplication
Evidence for Pre-Teaching Evidence for Post-
Component in Long-term Teaching Component in Long-
Memory term Memory
Attitudes
Episodes
Attitudes
Imagery
Imagery
Episodes
Verbal
Verbal
Skills
Skills
Strong
1 0 0 1 1 16 15 18 18 18
Evidence
Some
0 1 1 0 0 10 7 3 3 3
Evidence
No
Evidence 27 27 27 27 27 2 6 7 7 7
Changes in Concept Images for the Distributive Property for Multiplication Over
Addition
Entries in Table 8.4 show that for each of the five “cognitive structure” components—
verbal knowledge, intellectual skills, imagery, episodes, and attitudes—major changes
occurred, as a result of the intervention, in the thinking of most of the participating students
with respect to the distributive property for multiplication over addition. As with changes for
the associative properties for addition and multiplication, about three-fourth of the students
displayed an obvious movement from “no evidence” toward “strong evidence,” and the
changes related to each of the five cognitive-structure components. More details related to
changes in students’ cognitive structures with respect to the associative properties for addition
and multiplication, and for the distributive property for multiplication over addition, will be
given later in this chapter when interview data for individual participating students will be
examined.
Table 8.4
Summary of Data from 28 Interviewees in Relation to Concept Images
for the Distributive Property for Multiplication Over Addition
Evidence for Pre- Evidence for Post-

Teaching Component in Teaching Component in
Long-term Memory Long-term Memory
Attitudes
Attitudes
Episodes
Episodes
Imagery
Imagery
Verbal
Verbal
Skills
Skills
Strong
0 0 0 0 0 15 19 20 20 20
Evidence
Some
0 0 0 3 0 6 3 3 3 3
Evidence
No
Evidence 28 28 28 25 28 7 6 5 5 5
Analyses of Qualitative Data Generated by the Modeling Intervention
Table 8.5 summarizes types of data generated when students engaged in tasks
associated with the algebra of modeling. Actually, almost all of the relationships which the
students were asked to consider were those arising from linear sequences—that is to say,
functions whose domain was either the set of natural numbers or the set of natural numbers
and zero. Entries in Table 8.5, for which a distinction is made between receptive and
expressive responses by the students, suggest how students’ responses were interpreted.
Components of Students’ Concept Images—For Modeling 167
Table 8.5
Evidences for Qualities of Components of Students’ Concept Images with Respect to
Modeling Relationships
Component Examples of Receptive Examples of Expressive Evidences for Aspects of
of Long-term Outcomes for Outcomes for Cognitive Structure for
Memory Modeling Modeling Modeling
Verbal Can remember the Can explain the meanings Knows, and can express
Knowledge meanings of terms such as of terms such as “first verbally, the difference
“table of values,” “first term, “second term,” and between recursive and
term,” “second term.” “nth term” accurately, and explicit descriptions of
can apply them in sequences.
unrehearsed situations.
Intellectual Can follow discussion Can discuss aspects of a Is aware that tables
skills related to linear given table of values for a of values can express
sequences—e.g., if the linear sequence. Learns to the values of terms in a
teacher states that “because use the subscript notation linear sequence, and of
the nth term equals 5n then for describing sequences. conventions (e.g., often—
the sixth term equals 30” but not always—three
then the student will be able dots (“...”) indicate the
to state the values of the need to leap to the nth
first and second, terms, etc. term).
Imagery Recognizes that with linear Can generate visual With situations relevant
sequences there is a descriptions of relation- to real-life, students can
constant difference between ships between successive give a “reason for the
successive terms. Can terms of a linear sequence rule,” and accompany this
follow someone’s (e.g., “It starts at 1, and by describing or
explanation of relationships goes up by 3 each time”). illustrating appropriate
between terms and the real imagery.
situation.
Episodes Remembers when a teacher Is able to identify and Recalls working with
explained how to interpret describe, verbally and in other students in group
tables of values and states writing, recursive and work, and planning for
that with recursive specifi- explicit specifications and making associated
cations one needs to give from tables of values, and group presentations on
the first term and the rule remembers having doing modeling tasks.
for “going to the next this on previous occasions.
term.”
Attitudes Makes positive or negative Makes positive or negative Declines to participate
responses when working in comments to others on the actively in group
groups or when giving value of studying linear discussion; or displays
presentations to others. sequences. leadership when organiz-
ing a group.
Interview Data: Student Responses to Modeling Tasks

The first pieces of data relate to Student 1.16’s responses, in the pre- and post-teaching
interviews, to the task described at the top of Figure 8.6.
Task [The interviewee is shown a piece of paper on which “Tn = 2n + 3” is written, in

large print. Then the interviewer states “If you write Tn = 2n + 3, then Tn equals 2 times n
and plus 3. So if we write that, then we can say T5 equals 13, because 2 times 5 plus 3
equals 13. So what do you think T11 would equal?”]
Student 1.16’s response in the pre-teaching interview: 143
Interviewer: Is that your final answer?
Student 1.16: Yeah.
Interviewer: Could you explain why you think the answer is 143?
Student 1.16: I got 13 from here and times 11 and I got 143.
The above response suggested that Student 1.16 had not comprehended the question—in
other words, she had not understood the meaning of the sign which comprised the
question.
Student 1.16’s response in the POST-teaching interview: 25.
Interviewer: How did you get that so quickly?
Student 1.16: So you do 2 times 11 plus 3.
Figure 8.6. Learning to comprehend subscript notation for linear sequences.

Student 2.12’s pre- and post-teaching responses to the interview question shown in
Figure 8.7 revealed his improvement with respect to the concept of a variable.
Task [The interviewee is shown a piece of paper on which “Tn = 5n – 2” is written, in

large print. Then the interviewer states ““If Tn = 5n – 2, tell me which values of n would
make Tn greater than 20.”]
Student 2.12’s response in the pre-teaching interview: First I tried 2 for n. 5 times 2 is
10 and minus 2 is 8. Later I multiplied 8 by 2 is 16. That is my answer.
Student 2.12’s response in the POST-teaching interview: I am thinking about “above 5.”
Because 5 times 5 is 25 minus 2 is 23. But 4 makes it 18.
Figure 8.7. Evidence for having learned to comprehend subscript notation.

The captions for Figure 8.6 and Figure 8.7 include the verb “comprehend,” which
implies “knowing the meaning of the sign which has been used”—and, therefore the
discussion will benefit if we enter the world of Peircean semiotics. The mathematical object
being presented in each of the tasks, in Figure 8.6 and Figure 8.7, is a linear sequence (or
arithmetic progression), and there is a large amount of learning needed to comprehend fully
the meaning of the signs. In Figure 8.7, for example, Student 2.12 only partly knew the meaning
of the sign at the pre-teaching stage, and his interpretation of the sign’s meaning was not
appropriate. But, at the post-teaching stage, he knew that n was being used as a variable
which could take natural-number values only, and that for some of natural-number values the
statement would be true and for some it would be false. By a process of trial-and-error, he
determined that the answer was “above 5,” which was almost correct—correct was “greater
than or equal to 5.”
Student 2.12’s cognitive development between the pre-teaching and the post-teaching
stages was impressive. He was developing appropriate verbal knowledge, intellectual skills,
and imagery. At the post-teaching stage, he was also able to refer to relevant episodes that he
Conventions with Tables of Values 169
recalled from the workshops. In the post-teaching interview, it was clear that he had gained
an interest in the mathematics of modeling.
The next example to be discussed is in relation to Student 1.13’s pre-and post-teaching
responses to a table-of-values task. During the interview with Student 1.13 a sheet of paper
displaying the following table of values was placed in front of her.
First 1 2 3 4 5 ... n
Value
Second 3 5 7 9 ? ... ?
Value
The pre-teaching interview proceeded in this way:
Interviewer [pointing]: What number should we place under the 5 in the table?
Student 1.13: Eleven.
Interviewer: Tell me how you got your answer.
Student 1.13: It looks like the second row of numbers has odd numbers. The numbers are
already there, three, five, seven, and nine. So, I was thinking that those are odd
numbers. So I just added to eleven.
Interviewer: What do you think we should put under the n?
Student 1.13: Fifteen.
Interviewer: How did you get that?
Student 1.13: Because if you continue the same odd number order, the next one would be
thirteen and the next question mark [pointing under the n] would be fifteen.
In this excerpt, Student 1.13 failed to recognize that there was a relationship between
the two rows of numerals. The number below the “5” was to be “11,” she said, because the
numbers along the bottom row were odd numbers, and 11 was the next odd number. There
appeared to be no recognition that each number below was 1 more than twice the number
above. Also, Student 1.13 did not know what the three dots, …, were intended to convey. She
did not recognize that the dots were a signal to make a cognitive leap from the particular to
the general.
But, after all, why should she have known that? Probably, she had never been
introduced to the meaning of the “…” convention by which the reader is expected to
formulate an explicit, nth-term rule. Sometimes, “…” is merely used to indicate that “some of
the following terms are not going to be shown”—see, for example, the use of “…” in Task 4
for Group 1, in Appendix F to this book. And, because Student 1.13 was not aware of
possible meanings of “…”, and because she did not see a need to relate entries in the first row
with entries in the second, it was only to expected that she would say that 15 should be below
the “n.” There was no knowledge of different overall “taken-as-shared” meanings of the “…”
sign, and therefore she had no way of doing what, cognitively, the task “wanted” her to do.
Analysis of data from the post-teaching interview with Student 1.13, however, revealed
an altogether different story (see Figure 8.8). Evidence for impressive pre- to post-teaching
cognitive growth is strong. Figure 8.8 points to Student 1.13 having developed appropriate
verbal language and symbols. She had learned to identify and use appropriate intellectual skills,
and had responded appropriately to a table of values whose orientation differed from those she
had been used to seeing in class. Most importantly, she was now able to demonstrate a newly-
found ability to express the explicit and recursive specifications of linear sequences in formally
correct ways. All of this was done by Student 1.13 with great confidence and enthusiasm.
Student 1.13, POST-teaching interview

Interviewer: W Whhat about under the n?
Student 1.13: If you find a recursive rule … you cannoot do times because three times
three is nine … so you have to do it a different way [she is thinking out loud]. 2 times 3 is
6 and minus 1 does not work [she is guessing and checking]. Is it okay if I make a tabble
here?
Interviewer: Yes.
Student 1.13: You could do a recursive rule w

whhich pluses 2 each time, and write Tn+1 =
Tn + 2, T1 = 3. You could also write an explicit rule whicch is 2 times n plus one, Tn =
2n +1.
Figure 8.8. Student 1.13’s post-teaching interview response to the “table-of-values” task.
Cognitive Growth in Modeling Related to the Subscript Notation for Sequences

Each of the parallel versions of the pencil-and-paper Algebra Tesst included four
questions in which the subscript notation for specifying the nth teerm of a linear sequence was
specifically used in the writtten form of the question. At the pre-teaching stage, none of the
participating students had been taught that notation and, therefore, it was only to be expected
that the students would not know the “taken-as-shared” meaning of the signs. On the pre-
teaching version of the Algebra Tesst, 2 of the 32 participating students gave correct responses
to these questions; but, at the post-teaching stage, 17 students gave correct answers.
The pre-teaching entries in Table 8.6 suggest that a few of the students had a strong
intuitive idea of what the subscript notation meant, and others had some idea. The level of
understanding of the meaning of the notation obviously improved as a result of the students’
participation in the modeling workshops.
Analysis of classroom-observation data showed that although most of the seventh-grade
students developed a receptive understanding of the subscript notation, many experienced
difficulty learning to use the notation in an expressive way. Sometimes, students did not
know whether the variable n should appear as a subscript or at the “normal” level. Whether it
would be wise to introduce the subscript notation to middle-school students is something
which might be determined by further research— —ccertainly, the issue is an important one
because it would be foolish to allow difficulties with notation to stand in the way of greater
positive involvement in linear-sequence tasks inviting generalizations.
Cognitive Growth in Modeling 171
Table 8.6
Summary of Data from 28 Interviewees in Relation to the Use of the Subscript
Notation in Modeling Tasks (Associated with Linear Sequences)
Evidence for Pre-Teaching Evidence for Post-Teaching
Component in Long-Term Component in Long-Term
Memory Memory
Imagery
Imagery
Attitude
Episode
Episode
Attitude
Verbal
Verbal
Skills
Skills
Strong
3 2 2 2 2 20 21 21 22 21
Evidence
Some
Evidence 0 12 10 2 12 1 3 3 0 3
No
Evidence 25 14 16 24 14 7 4 4 6 4
Cognitive Growth in Modeling Related to Generalizing for the nth Term

Each of the parallel versions of the pencil-and-paper Algebra Test included four
questions in which respondents, having been given a pattern or set of circumstances which
would enable the nth term to be determined, were asked to specify the nth term of a linear
sequence. None of the participating students could do that at the pre-teaching stage, but 17 of
them (53%) consistently did it correctly at the post-teaching stage. Classroom-observation
data indicated that many students found it difficult to generalize linear patterns, and some of
the difficulty was related to the students’ lack of confidence in using the subscript notation.
Entries in Table 8.7 reveal that although considerable progress was made so far as
generalizing to the nth case was concerned, 11 of the 28 interviewees continued to struggle to
make generalizations for linear sequence patterns, even at the post-teaching stage.
Table 8.7
Summary of Data from 28 Interviewees in Relation to Generalizing for the nth Term
Evidence for Pre-Teaching Evidence for Post-Teaching
Component in Long-Term Component in Long-Term
Memory Memory
Imagery
Imagery
Attitude
Episode
Episode
Attitude
Verbal
Verbal
Skills
Skills
Evidence
Found 0 0 0 0 0 17 17 17 17 17
Some
Evidence 2 2 2 2 2 6 6 6 6 6
No
Evidence 26 26 26 26 26 5 5 5 5 5
All writers on middle-school algebra, and the CCSSM standards, place great importance
on developing middle-school students’ abilities to generalize and, clearly, for both the
structure and modeling workshop interventions in the current study, one of the most
important aims was to assist the students to “grow” so far as their appreciation and
understanding of the mathematical object of generalization was concerned.
With the structure sessions, one important aim was for students to gain receptive and
expressive understandings of the idea that if a, b, c, represent any rational numbers then the
value of a(b + c) must be equal to ab + ac—with the caveat that whatever value a is given in
the expression a(b + c) that must also be the value which a is given in ab + ac, etc. For
mathematicians, and for most school teachers of mathematics, that caveat is not a matter of
concern, but for many seventh-grade students the generality of the number properties is
something which is difficult to understand. The achievement of an understanding requires a
cognitive leap so far as the concept of a variable is concerned. That is why one of the most
important aims of the structure workshops was for students to come to realize that the
associative and distributive properties are true for all rational numbers. In the workshops the
students considered whether they were true for natural numbers, integers (positive and
negative, and zero) and fractions. Although there has been some recent attention given to
aspects of teaching middle-school students to develop more “abstract representations,” so far
as the number properties are concerned (see, e.g., Ding & Li, 2014), there is a dearth of
reported research on the matter.
The main mathematical object to be associated with the modeling workshops was to
assist the seventh-grade students to generalize by finding, describing, and applying the nth
term of a linear sequence. Radford (2006) defined three levels of generalization—factual,
contextual, and symbolic—which might be applied to the thinking of middle-school students
as they develop their algebraic thinking with respect to linear functions. In Table 8.8, the first
author (Kanbir) included Radford’s three “levels” and, having recognized the difficulty that
many of the participating seventh-grade students experienced in regard to the subscript
notation for linear sequences, he added a fourth level, which he called “post-symbolic
generalization.”
In classroom observations of the workshops it appeared to be the case that although the
structure workshops and the modeling workshops were both regarded, by the teachers and
students alike, as being concerned with legitimate forms of school “algebra,” the signs being
used, the forms of speech, and the kinds of generalizations aimed for in the structure
workshops were different from the notations and generalizations sought in the modeling
workshops. Although student generalization has received much attention in the algebra-
education research literature over the past two decades (see, e.g., Blanton, Brizuela, Gardiner,
Sawrey, & Newman-Owens, 2015; Branton, Stephens, Knuth, Gardiner, Isler, & Kim, 2015;
Kieran, 2011), not much attention has been paid to generalization in the context of numerical
structures.
Students’ Generalizations 173
Table 8.8
Examples of Students’ Generalizations on Various Written Test Items
Level of Description of Thinking at the Example of Seventh-Grade Students’
Generalization Nominated Level of Generalization Work at the Level
Factual A process of generalization has begun, “There will be 10 triangles, and since
Generalization but thinking is still “in the realm of there are 3 matches for each triangle,
arithmetic” (Radford, 2006, p. 10), and a there will be 30 matches altogether.”
mathematical object has not been
recognized and described.
Contextual Such a description includes a mixture of “How many rooms the queen has,
Generalization. mathematical symbols and natural times 2 plus 2?”
language. According to Radford (2006), “It has to be times 2 plus 2. Whatever
with contextual generalization a tables we have you multiply by 2 and
sequence is identified verbally and add two.”
relationships between a figure and the
next figure are identified.
Symbolic A student’s thinking can shift between Builds an expression like “n + n +
Generalization recursive thinking and explicit thinking, 3,” irrespective of whether the person
and it is at this stage that generalizations can transform that into “n × 2 + 3,”
step into the realm of algebra. Radford or “2n + 3.”
(2006), after calling this process of
noticing, “objectification,” argued that it
corresponds to an attempt to identify and
describe a mathematical object.
Post-Symbolic Successfully uses one of two different
Generalization types of algebraic syntax. The first asks
students to identify and use a recursive
formula in order to generalize; the
second asks them to identify, to notate,
and to use an explicit formula with a
subscript notation.
In Figure 8.9 an analysis of student responses at the pre- and post-teaching stages for
the two “horizontal table-of-values” interview tasks, discussed earlier in this chapter, is given.
Levels of generalization achieved by the 28 interviewees at the two stages are shown. The
cognitive growth away from “factual generalization” strategies, which were very common at
the pre-teaching stage, toward higher levels of generalization, is evident.
A similar phenomenon is illustrated, for two different tasks, in Figure 8.10. It can be
seen that at the pre-teaching stage, most students used inappropriate counting-on strategies,
and did not understand what was needed to answer the question about what should be
associated, in the sequence, with the value n. At the post-teaching stage, most of the students
had made some “progress,” but it was still the case that about one-fourth of them were using
the inappropriate factual generalization approach, despite their having participated in the
workshop sessions on modeling.
Student Thinking on a Modeling Task Student Thinking on a Modeling Task

Related to the nth Term and Three (…) Related to the nth Term and Three (…)
Dots on a Horizontal Table Dots on a Horizontal Table
Pre-Teaching Post-Teaching
28 0 0 0 0 9 2 3 3 11
Counting/Arithmetic
Counting/Arithmetic
Post-Symbolic
Post-Symbolic
Contextual
Contextual
Symbolic
Symbolic
Factual
Factual
26 1 1 0 0 5 4 3 6 10
Pre-Teaching Post-Teaching
Student Thinking on a Modeling Task Student Thinking on a Modeling Task
Related to nth-Term Generalization. Related to nth-Term Generalization.
Figure 8.9. Student generalizations in the pre-and post-teaching interviews.
Two Modeling Interview Tasks and Students' Semiotic

Development
30 28 26
25
20
15
11 10
9
10
5 6
3 3 4 3
5 2
0 0 0 0 1 1 0 0
0
Pre-T Interview_Task 12 Post-T Interview _Task 12 Pre-T Interview_Task 14 Post-T Interview _Task 14
Counting/Arithmetic Factual Contextual Symbolic Post-Symbolic
Figure 8.10. Seventh-grade students’ cognitive growth on two different tasks inviting
students to state the rule for the nth term.
Analyses of Student Responses to Six Interview Tasks

This section will summarize responses given in pre- and post-teaching interviews by six
students (one low-, one average-. and one high-achiever from Group 1, and one low-, one
Student Responses to Interview Tasks—Structure 175
average- and one high-achiever from Group 2) to each of six interview questions. The basis
for classification and selection of the students (as low, average, high) was the score obtained
on the pre-teaching version of the Algebra Test.
There are two aims for this section. The first is to show the effect of the intervention
workshops on students’ cognitive structures. When pre-teaching and post-teaching responses
are compared, effects of instruction are suggested. The second aim is to enable the reader to
assess differences in cognitive growth patterns between low, average and high achievers.
The method of presentation will be as follows: first the task will be stated, and then
responses of the two low achievers, the two average achievers, and the two high achievers, at
pre- and post-teaching stages, will be shown. The symbol “S” will be used to denote
“Student.”
Occasionally, a student gave an appropriate answer at the pre-teaching stage, but
answers at the post-teaching stage tended to be offered much more confidently, and were
indicative of cognitive growth toward understanding of the mathematical objects being
considered. One only has to examine CCSSM’s (2010) middle-school document carefully to
come to realize that generalization is regarded as an important objective, but the following
evidence would suggest that learning to generalize is something much more difficult than is
commonly recognized. Readers are invited to keep that thought in mind as they take account
of the data.
Task: “I am going to say two words to you and, as soon as I say them, I want you to
say something, or draw something, or do something—do the first thing that comes
into your head after I say the words. The words are … “distributive property.” Here
are the words again: “distributive property.” What comes to your mind?
Pre-Teaching Interview Post-Teaching Interview

Student S: Breaking the problem down S: Breaking down a problem and
1.1 Interviewer: Can you give me an moving parentheses.
(Low example what do you mean by that? Interviewer: Can you write an
Score) S: 2 plus 2 times 6 minus 4 equals 4 example?
times 2 which equals 8. S:
Student S: I remember from my lower grades. I S: You will be multiplying 37 times

2.5 don’t remember exactly, but it is 4 and you would break it down as 30
(Low something about … You take something times 4 and you add 7 times 4, and
Score) and you put something else. You do one you get 148.
thing at a time.
Interviewer: Can you describe or
illustrate what you are saying?
S: Okay. I have written something. In
parentheses, 3 times 7 and times 4 outside
of the parentheses. And I switched it
around and it becomes 4 times in the

parentheses 3 times 7.
Student S: I think of “parentheses.” S: Parentheses. Breaking down to

1.6 Interviewer: Could you write something numbers.
(Average down or draw something? What else? Interviewer: Can you write an
Score) S: An addition sign. example?
S: 94 times 4 can be written as 90
times 4 plus 4 times 4.
Student S: “Parentheses,” “exponent,” S: If you want to calculate 9 times

2.2 “addition” and “subtraction.” 110 you could do first 9 times 100
(Average Interviewer: Can you give an example? plus 9 times 10. It would make this
Score) S: I would do seven plus 2 in calculation easier.
parentheses first. I got 9, and then 9
minus 4 gives 5.
Student S: I have “method” and “subtraction.” S: I wrote “easier and faster”

1.11 Interviewer: Why did you write those? because the properties are used to
(High S: Those are of two things that first make harder math equations go by
Score) came to my mind. Because I remember quicker.
something about the distribution (sic.) Interviewer: What does the
property. distributive property actually mean?
Interviewer: Thanks for that. S: I don’t remember if it is
distributive or associative, but if
there is an equation that has two of
the numbers say 44 and you can just
combine those together as one, but
then the remaining numbers …
Student S: “Math.” S: “Multiplication.”
2.7 Interviewer: Anything else? What is the Interviewer: Why did you write
(High distributive property? multiplication?
Score) S: I don’t know. S: Because when I think of that, it’s
easier to do multiplication than
addition and subtraction.
Interviewer: Give me an example.

Why do you think it is easier?
Task: “Without using a calculator, find the value of 482 + (18 + 300).”

Student 1.1 S: I did 300 plus 18 which is 318. S: 800. I did 482 plus 18 and got
(Low Score) Then 482 plus 318 which equals 500. And I added 300 and got 800.
800.
Student 2.5 S: First I added 18 plus 300 in the S: 800.

(Low Score) parentheses. I got 318 by adding the Interviewer: How did you get that?
two numbers, and then I added 482 S: I added first 482 plus 300 which
with 318 and got 800. is 782 and plus 18.
Interviewer: What is the name of
the property?
S: The associative property.
Student 1.6 S: First I did 482 plus 18 makes it S: You could do 482 plus 18 which
(Average 490; and 300 more--I got 790. would be 500 and plus 300 and you
Score) would get 800.
Student 2.2 S: First I added 18 plus 300 in the S: I did 482 plus 18 which is 500,
(Average parentheses. Then I got 318 adding and I added 300 and got 800.
Score) two numbers; and then I added 482 Interviewer: Do you know the
with 318 and got 800. name of the property?
S: I used the associative property
and moved the parentheses.
Student 1.11 S: First, using the order of S: So I would move the parentheses
(High Score) operation, parentheses comes first. I around 482 plus 18. And then add
did, first, 300 plus 18 and later 482 those two together which would get
plus 318. 500. Then 500 plus 300 which is
S: I am still working on … My 800.
final answer is 800. Interviewer: Do you know the
name of the property you just used?
S: Distributive? Or associative? I
think it is associative.
Student 2.7 S: I did 300 plus 18. Then, I did 482 S: 800. Because I put the
(High Score) plus 318. I did first 300 plus 400 parentheses around 482 plus 18,
which is 700 and then 82 plus 18 and plus 300, which is 800.
which is 100 and I added 700 and Interviewer: Why are you
100. allowed to do that?
S: The associative property.
Task: “Without using a calculator, find the value of 4 × (1/4 × 128).”
Student S: So far I have done 128 times 25 and I S: I got 2048. I did 4 times 4 which
1.1 got 3200. is 16 and I did 16 times 128.
(Low Interviewer: What are you going to do
Score) now?
S: Times 4 and I got 12800.
Student S: First I turned 128 to 128 ones and I S: First I would multiply one-fourth
2.5 cross simplified and got 32. Then I times 128 because the parentheses
(Low multiply 32 over 1 to 4 and got 128. come first; and I would get 128 over
Score) 4 which is 32. Finally, 4 times 32
gets 128.
Student S: I did 4 times 128 and I got 512. I did S: You could do 4 times one-fourth
1.6 512 over 1 times one over four. I made and times 128. I did four over one
(Average this a fraction to make multiplication times one over four which is one and
Score) easy. And then I did 512 over 4 and I got times 128. I got 128.
1210. Interviewer: What is the name of the
Interviewer: Why did you start with 4 property you just used?
times 128? S: The associative property of
S: I wanted to get rid of all the whole addition.
numbers.
Student S: I would first multiply one-fourth times S: I changed the parentheses around
2.2 128 and make that into an improper 4 times one-fourth which is 1; and I
(Average fraction. multiplied it by 128 and got 128.
Score) Interviewer: Okay. You divided 128 by 4 Interviewer: Do you know the name
and you got 32. of the property?
S: Yes. I think 32 times 4. I would then S: This would be the distributive
multiply it by four and get 128. property.
Student S: First, one-fourth times 128. It would be S: I would move the parentheses
1.11 128 over 4. That is an improper fraction. around 4 and multiply 4 times one-
(High The method we used is called “tip over” fourth first. What I would do, 4 as a
Score) and multiply it. I would do 128 over 4 and fraction is 4 over 1 so 4 over 1 times
it is 32. But, still I have to multiply by 4. 1 over 4 you do cross simplify and
Which means 128. the problem becomes one over one
and problem becomes 128 times 1.
So the answer is 128.
Task: “If Tn = 5n – 2, tell me which values of n would make Tn greater than 20.”
Student S: 10. S: n equals 6. So 5 times 6 is 30 and

1.1 Interviewer: Why did you choose minus 2 is 28.
(Low 10? Interviewer: Do you think this is the
Score) S: If you 5 times 10 you get 50 which only value which makes it greater than
is greater than 20. I was just thinking 20?
20 and how you do 2 times 10 to get S: You can also do it with 5. 5 also
20 and 10 times 5 which is more than makes it greater than 20.
20
Student S: A value could be 5. S: It would probably be 5 and up.
2.5 Interviewer: Okay. What made you Because if we put 5 times 4 and subtract
(Low decide that? two that would be 18 [T4]. But you can
Score) S: Because 5 times 5 is 25 and minus do T5 which is 5 times 5 minus 2, which
2 which is 23. It is still higher than is 23.
20.
Interviewer: Is that the only number
that could make Tn greater than 20?
S: Any number higher than 5 could
work.
Student S: 5 would work. Because 5 times 5 S: It could be 5 because T5 equals 5
1.6 is 25 and minus 2 is 23 which is times 5 minus 2 which is more than 20.
(Average greater than 20. So T6 also works as well. Any number
Score) Interviewer: What values did you bigger than 5 makes it greater than 20.
find?
S: Five.
Student Responses to Interview Tasks—Modeling 181
Student S: 6 would be one of them and 7. S: Any number greater than 5. If you
2.2 Would it be anything over 5, maybe? take 5 times 5 and minus 2 is more than
(Average Interviewer: Anything over 5? 20. You could do the same thing for 6,
Score) S: Yes. Anything over 5. Because 5 7, 8, and so on.
times 5 is 25 and minus 2 is 23 which
is more than 20.
Student S: Well! I did math in my head. First S: I just decided to pick a number, I
1.11 number would be 5. Because 5 times starter with a 4 and 5 times 4 is 20 and
(High 5 is 25 and subtract 2 which gives me minus 2 is 18. I decided to go up one
Score) 23. Any number above 5 would be an number and put it in because 4
answer. becomes less than 20. My answer is 5
and above.
Student S: Six times five minus 2 equals 28 S: 5, 6, 7, 8, and like more …

2.7 Interviewer: What else? Interviewer: What about 4?
(High S: 6, 7, and then 8. S: No. 5 times 4 minus 2 is 18 which is
Score) less than 20.
Task: Give the pupil a piece of paper with the following table on it:
First 1 2 3 4 5 ... n
Value
Second 3 5 7 9 ? ... ?
Value
Student S: I think it would be 5 plus 3 S: It would be 10. Because 4 plus 5 is

1.1 Interviewer: Why did you think that 9 and 9 plus 1 is 10 [pointed to the
(Low way? question mark under the 5].
Score) S: Oh no actually it would be 5 plus 2
you could get 7. Interviewer: What do you think we
Interviewer: Okay. You think that this should put under the n?
question mark should be 7 because you S: 12. Because I was thinking it is
add 2. going the same way which is 10, 11,
Interviewer: What do you think we and 12 [pointed to the question mark
should put under the n? under the n.]
S: It would be 9. Because this [pointing
the empty top cell] is 6, and the n is 7.
Then on the bottom, this would be
[pointing to the question mark] be 7 plus
2 which is 9.
Student S: I think it would be 11 maybe. S: Well under 5 would be 11. Because
2.5 Interviewer: Why did you say “maybe”? you’re going up by 2 each time.
(Low S: Because each time you add 2 to the
Score) bottom numbers. And 9 plus 2 is 11. Interviewer: What should go below
Interviewer: What do you think we the n?
should put under the n? S: It could be n plus 2. Every time
S: It could be 13, maybe. you’re adding 2, the same sum, which
Interviewer: How did you get 13? means n plus two.
S: Because this question mark on the top
would be 7 and on the bottom would be
13.
Student S: I think it would be 11. S: 11. It is plus three, plus four, plus
1.6 five, and plus 6 which is 5 plus 6.
(Average Interviewer: What do you think we Interviewer: What do you think we
Score) should put under the n? should put under the n?
S: I think it is 14. I think that it’s going S: Tn equals 2n plus 1
up by like 2, 3, 4, 5, and 6. If there [Tn = 2n + 1]
would 6 it goes up by 7 and then this Interviewer: What do you call this
would go up by 8. Wait … that would rule?
be 15. S: It is an explicit formula.
Student S: 11. Would be 11. Because it goes up S: This would be 11. Because every
2.2 by two [pointing out the second values] time this goes up by 2. So 9 plus 2 is
(Average 3 plus 2 is 5, 5 plus 2 is 7, and this one 11.
Score) [pointing beneath the 5] 9 plus 2 is 11. Interviewer: What do you think we
Interviewer: What do you think we should put under the n?
should put under the n? S: Would it be Tn?
S: 15. Because. This one [pointing the Interviewer: Can you try to find Tn?
blank part of the second row] would be S: 2n plus 1. Tn = 2n + 1.
13 and two more would be 15.
Student S: First, I would find a pattern. I could S: The difference between each row
1.11 subtract second row minus first row goes up each time by 1. 4 then 9
(High numbers, because, the second row which is the difference is 5 and then
Score) numbers are always bigger. Each time the next difference will be 6. So, the
the difference between the numbers number under the 5 is 11.
grows by 1. Last set up number Interviewer: What is your final
difference before the comma is 5. I answer?
would add 6 to 5 which gives me 11. S: 11.
Interviewer: What is your final answer? Interviewer: What should go below
S: I would say the question mark is equal the n?
to 11. S: So the pattern is the difference
Interviewer: What should go below the between the second row and the first
n? row is 1, and goes up each time by 1.
S: To get to that, I would look at the next I have to go up by 6 from 5 so the
two empty boxes. The top one would 7 first question mark would be 11. Then
because the first row goes up by 1 and for the question mark below n … I
the bottom row goes up by 2 which is 7 believe it would be Tn.
to 13. Then, the bottom number right Interviewer: So what do you think Tn
below the n will be 15. would be equal?
S: In order to find that we have to
find the explicit rule. But I am not
sure how to write this. The difference
goes up by 1 each time. Tn equals n
times 2 plus 1.
Student S: That is 11. S: 11.

2.7 Interviewer: How did you get your Interviewer: Why is it 11?
(High answer? S: To get 3 here you do 1 times 2 plus
Score) S: 3 plus 2 is 5, 5 plus 2 is 7, 7 plus 2 is 1 and 5 times 2 and plus 1 is 11.
9, and 9 plus 2 is 11.
Interviewer: What should go below the n? Interviewer: What should go below
S: 15. the n?
Interviewer: Why do you think the S: This is Tn equals 2n plus one [writes
answer is 15? down Tn = 2n + 1, T1 = 3], and T1
S: Because, you are adding 2 to each equals 3.
number. Interviewer: Okay. Well done!
Interviewer: Thank you. Write down
your final answer.
Task:
B. If Pn represents the number of people who can sit when n tables are pushed
together (in a straight line), what is the rule giving Pn in terms of n?
Student S: You could have 30 people. S: Each time they are adding 2
1.1 Interviewer: How did you do it? more people. I counted and got 22.
(Low S: You have to take 2 people off the Interviewer: Okay. What about the
Score) ends. First I took one person from n-table case?
each table. 40 minus 10 which is 30 S: You want me to write an explicit
people. something?
Interviewer: Yes. Can you come up
Interviewer: What about the n-table with the explicit formula?
case? S: Pn +1 equals n + 22.
S: You would do 5 tables. You will
have 15 people.
Interviewer: How did you get 15?
S: I did 5 times 3. I thought 15 people
could sit around 5 tables.
Interviewer: Because you look at 3
people around the table and multiply
by three?
S: Yes.
Student S: I got 22. S: There will be one person on each

2.5 Interviewer: Did you count them all? side you can sit on. Three people get
(Low S: Yes. These two first and I kept one table each time. There will be 10
Score) adding each one. people on top side and 10 people on
Interviewer: What about the n-table bottom side and 2 on each ends.
case? Total, 22 people.
S: n tables …We can find what the n Interviewer: What about the n-table
is and count to see how many people case?
there would be around the n tables. S: n would equal three plus n
because it depends on how many
there are and then subtract one if you
are between two tables.
Student S: Total 22 people. 10 people could sit S: 22 people. If there is 10 tables 10

1.6 on each side and 2 people at the ends. people for each sides and 2 people at
(Average the ends. So 10 times 2 plus 2.
Score) Interviewer: What about the n-table
case? Interviewer: What about the n-table
S: Would still be add the sides and 2 case? What if you have n tables?
ends. S: Pn equals P2 plus 2.
Interviewer: What if you had 50
tables?
S: 102.
Interviewer: What if you have n
tables? You can create your general
rule about this design.
S: Add the sides and ends.
Interviewer: Ends means plus 2 right?
S: Yes.
Student S: 22 people can sit around 10 tables. S: 22. Because 10 people on each
2.2 Interviewer: Tell me more about it. side and 2 people on the ends which
(Average S: I have 10 people here and one more is 10 plus 10 plus 2.
Score) which is 11 and I have another 10
there (bottom or top of ten tables) and Interviewer: What about the n-table
then one more—the total would be 22. case? You wrote down a formula
Interviewer: Okay. What about the n- there. What type of rule is that?
table case? S: Explicit rule.
Interviewer: How did you get that?
S: Take two sides which is S: You add 2 each time it increases,
multiplying by two. You would do the based on a pattern. Pn equals 2n plus
number of tables times 2 and plus two. 2
Because two people on each end of the
table.
Student S: 3 tables makes, 3 bottom and 3 top S: 3 tables makes, 3 bottom and 3
1.11 and 2 ends which is 8 people. If you top and always 2 ends which is 8
(High think 10 tables, the answer would be people. If you think 10 tables, the
Score) 22. answer would be 22.
Interviewer: How did you get that? Interviewer: How did you get that?
S: Because on each sides [pointing to S: By multiplying 10 times 2 and 2
the top and bottom] the numbers are extra ends.
equal—10 plus 10. It would be 22.
Interviewer: 10 plus 10 plus one plus
one which is 22 people.
Interviewer: What about the n-table Interviewer: What about the n-table
case? case?
S: It is difficult. S: I believe it will have to be a
Interviewer: If you have 100 tables? variable. For example, Tn.
S: It would be 202 people. Interviewer: What does Tn equal?
Interviewer: What about 50 tables? S: Tn would equal n times 2 plus 2.
S: It would be 102 people. Interviewer: How would you write
Interviewer: Can you verbalize your it?
formula based on n tables? S: I wrote Tn = 2n + 2.
S: I don’t know. Interviewer: How would you get 22
from that, if you have 10 tables?
S: You would follow how the rule
goes. If n were 10 and it would be
10 times 2 plus 2 equals 22.
Student S: Twenty-two people. Because it S: 22. Because there are 3 [people]

2.7 depends on how many tables, so if you on each end and 3 times 2 plus the
(High have 10 tables, 10 people on one side middle tables (on top and bottom) 8
Score) and 10 people on the other side. And plus 8 and total would be 22.
one for each side of the tables
[Pointed to the ends of the first and Interviewer: What about the n-table
the last table]. case?
Interviewer: 10 tables how many S: Tn equals 2n plus 2 and T1 equals
people again? 4 [Tn = 2n + 2, T1 = 4].
S: 10 plus 10 plus one plus one which Interviewer: How did you get the
is 22 people. 2n + 2?
S: Because for the first table there is
Interviewer: What about the n-table 4 and 1 times 2 is 2 plus is 4; for the
case? two tables 6, which is 2 times 2 and
S: Whatever tables you are, you plus 2; the three tables, 3 times 2 is 6
multiply it by 2 and add two. plus 2 is 8.
Interviewer: How many tables?
S: 9, because n can be any number.
Interviewer: How did you get the 9?
S: I just put a regular number.
Qualitative Analyses of the Implemented Curriculum 187
Qualitative Analyses of the Implemented Curriculum

Details of the workshop procedures were well documented (see Appendices D, E, F),
the aim being to provide sufficient documentation to enable the study to be replicated—either
at School W or at any other middle school. The quantitative analyses in Chapter 7 and the
qualitative analyses in this chapter generated data which, when analyzed, had a sufficiently
clear interpretation to make it worthwhile for the study to be replicated.
Details related to the instruments used in the main study are given in Appendices A, B,
C, D, E, F and G, which appear toward the end of this book. These appendices are titled:
Appendix A: Protocol for Algebra Interviews with Seventh-Graders
Appendix B: Algebra Tests (Three Parallel Versions are Reproduced)
Appendix C: “Questionnaire” Completed by Seventh-Grade Students at School W at the
Beginning of the Algebra Workshops on “Structure”
Appendix D: Statement of Instructional Aims for the Structure Workshops with the
Seventh-Grade Students at School W
Appendix E: Detailed Lesson Plans for Four Workshops on “Structure” for Seventh-
Grade Students at School W, Including Homework Challenges for Each
Workshop
Appendix F: Detailed Plans for Group Tasks in the Modeling Workshops: Finding
Recursive and Explicit Rules for Patterns
Appendix G: Classroom Observation Schedule
Appendix H: Pre-Teaching to Post-Teaching “Growth” with Respect to the Five Basic
Cognitive Structure Components
Appendix I: Generalization Categories (after Radford, 2006)
It will not be possible, here, to discuss, extensively, what actually transpired in the
intervention workshops in which the students in Group 1 and Group 2, and Mr. X and Mr. Y,
were active participants. However, since each workshop was observed by at least one of the
three authors of this book, and since an observation schedule was completed for each
workshop, it was possible to make some reasonably objective comments about both the
structure and intervention workshops.
Fidelity of Implementation of Workshops

Both Mr. X and Mr. Y had observed the three authors of this book lead very similar
workshops with an eighth-grade class at School W, and when they themselves led workshops
with seventh-grade students at School W they distributed almost the same workshop notes to
the seventh-graders as the three authors had distributed to the eighth-graders. The seventh-
grade classes proceeded in almost exactly the same way as had the model eighth-grade
classes, and the same homework “challenges” were used. Those observing the lessons believe
that both Mr. X and Mr. Y faithfully based their workshops on what they had observed in the
model lessons. That statement should not be read to imply that the two teachers never
exercised their own initiatives. They often added their own interpretations of the mathematics
under consideration, and linked activities to the special interests of the participating
students—all of whom were well known to both teachers.
During all of the workshops there were absolutely no discipline problems—all students
seemed to be highly motivated to learn the algebra under consideration. Class sizes were
small (there were usually 15 or 16 students present in workshop sessions) and, because of the
random sampling, each class could be described as “heterogeneous,” covering a wide range
of mathematical performance. School W is a smallish, rural public school, and mathematics
performance data gathered in previous years would suggest that although the school is by no
means “selective” so far as its intakes are concerned, the school maintains good academic
standards—probably slightly better than “average.” The three observers agreed that almost all
the participating students, and both the participating teachers, seemed to enjoy responding to
the challenges presented by and in the structure and modeling workshops.
The Receptive-Expressive Balance

The workshop notes which were issued to participating students had been especially
prepared for the intervention classes. Before the main study took place the workshop notes
did not exist. These notes became the basis for how the workshops proceeded. The notes are
reproduced in Appendix E (for structure workshops) and Appendix F (for modeling
workshops).
A careful examination of the notes will reveal that the desirability of implementing an
educationally sound and balanced receptive-expressive approach to teaching and learning
(Clements & Del Campo, 1987; Del Campo & Clements, 1987; 1990) was taken very seriously
in this intervention study. In all workshop sessions, the students spent much time (usually
about half of the allocated lesson time) discussing the mathematics under consideration with
fellow students in small groups. Group members knew that they would be expected to give
detailed reports on their thoughts and findings to the whole class, and that each student would
be expected to participate actively when their groups were reporting to the whole class. The
classroom observers (at least one of the three authors of this book observed any particular
workshop “lesson”) believe that the above description could be applied to each workshop
session in the intervention.
Of course, Mr. X and Mr. Y also participated actively in the workshop sessions,
introducing topics, monitoring group discussions, chairing student-presentation sessions,
commenting on aspects of the topics which many of the groups were finding difficult, and
adding interesting “extras.” Homework was set for each session (see Appendix E), and students
submitted written responses to homework tasks to their teachers. The last session for each of
Concluding Comments—Qualitative Analyses 189
the structure and modeling workshops took place in the form of a whole-class overview, led
by either Mr. X or Mr. Y, but with inputs from at least one of the three authors of this book.
It should be possible for replications of the main study described in this book to be
carried out at schools other than School W. The workshop notes have already been trialled
with prospective middle-school teachers taking an “Algebra for Middle-School Teachers”
course, and they proved to be very successful and helpful for the students. Hopefully, middle-
school teachers and teacher educators will attempt to replicate the study, and it will be
interesting to note variations which will be needed, given different students, teachers and
circumstances.
Concluding Comments with Respect to the Qualitative Analyses

The qualitative analyses complemented the quantitative analyses, both suggesting that
the interventions had been successful. The quantitative analyses revealed that gains were
statistically significant, and that effect sizes were large. The qualitative analyses revealed that
most of the participating students’ concepts images were considerably improved as a result of
their involvement in the workshops. In both the structure and modeling workshops, verbal
knowledge of relevant concepts was enhanced, intellectual skills improved, imagery evoked,
episodes committed to memory, and positive attitudes engendered. Linking Peirce’s triadic
semiotic theory to Herbart’s theory of apperception provided an ideal theoretical base, with
the emphasis on concept-image development providing the needed bridge from the sign-rich
environments of structure and modeling. Most of the students made worthwhile progress
toward the pertinent mathematical objects which were expressed in the workshops’ aims.
It is important though, to draw attention to some of the difficulties experienced. At the
retention stage, there were slight decreases from the post-teaching highs in performance that
had occurred for Groups 1 and 2 for structure, and for Group 1 for modeling. That finding
should provide the challenge for all interested persons at School W to take the next step by
providing additional workshops on structure and modeling during the next year or so, at least.
It should be noted, too, that there was evidence indicating that about one-fourth of the
participating students did not always cope with the mathematical complexities which
confronted them in the workshops.
Structure and modeling are parts of the middle-school mathematics syllabus at School
W, but the mathematical objects that they address are more sophisticated than is often
realized. Teachers should not attempt to “teach” them in a few lessons. No matter what has
been written elsewhere, the authors’ experiences in the two pilot studies and in this
intervention study at School W suggest that many middle-school students begin their seventh-
grade studies knowing hardly anything about the associative and distributive properties for
real numbers, and many struggle to identify and state explicit rules for even simple linear
sequences. Whether the word “many” (used twice in the last sentence) should be replaced by
“most” is something that should be determined by research.
Structure and modeling were much emphasized at different levels in the common-core
mathematics sequence (CCSSM, 2010). They are themes which should be consciously built
into curricula for all grade levels in the middle-school. The success of the receptive-
expressive approach in the structure and modeling workshops in the main study suggests that
attempts to replicate and extend the study should be encouraged.
In order to make sense of the qualitative data, particularly the pre-and post-teaching
interview data, the first author created ordered pairs which were indicative of the strength of
the presence of a component in cognitive structure. For a particular component, the extent of
evidence was assessed on a three-point scale, 0 (corresponding to no evidence), 1 (some

evidence), and 2 (strong evidence). The term “strong evidence” was used to indicate that
there was definite evidence that a student had a sound verbal comprehension of the concept,
or was able to identify and use relevant intellectual skills, or could evoke appropriate
imagery, or could recall relevant episodes, or had developed positive or otherwise appropriate
attitudes. “Some evidence” implied that although there was evidence, it was not strong; and
“no evidence” indicated that there was no evidence that that component was present in long-
term memory (see Appendix H). The first coordinate of an ordered pair indicates the extent of
the evidence for the presence of the component at the pre-teaching stage, and the second
indicates the extent of evidence at the post-teaching stage.
The first author also created a coding for “overall growth,” based on the sum of five
memory components with a maximum score of 10 at the pre-teaching stage, and also a
maximum of 10 at the post-teaching stage. For example, if, for the five components, the
ordered pair (6, 10) summarized the pre-teaching/post-teaching measurements then a growth
measurement of 4 (i.e., 10 – 6) would be assumed.
By definition, “significant growth” corresponded to an overall growth—for the five
components—of at least 7, “modest growth” corresponded to a growth ranging from 3
through 6, and “no growth” corresponded to an overall growth of at most 2 (see Appendix I).
For many of the students, there was evidence of “significant growth,” but for some, there was
“no evidence.”
Ultimately, though, analysis of cognitive growth using this approach is flawed, because
if someone gained a score of 10 at the pre-teaching stage then any growth would not be
measured—yet it is possible that there could have been important conceptual growth for that
person during the intervention. The idea of measuring cognitive growth is, nevertheless,
important, and hopefully future researchers will devise better methods of doing this.
The emphasis in this chapter has been on how the classroom interventions affected the
thinking and attitudes of the seventh-grade students. In attempting to find evidence of the
extent to which an intervention links intended, implemented and received aspects of curricula,
it is not enough to assess and compare pre-intervention and post-intervention scores on tests
(Cai, Morris, Hohensee, Hwang, Robison, & Hiebert, 2017). One needs to attempt to assess
the ways which the intervention affects students’ verbal knowledge, skills, imageries,
episodic memories, and attitudes. This study has suggested how that might be achieved.
References
Adams, J. A. (1898). The Herbartian psychology applied to education. Boston, MA: D. C.
Heath.
Blanton, M. L., & Kaput, J. J. (2011). In J. Cai & E. Knuth (Eds.), Early algebraization: A
global dialogue from multiple perspectives (pp. 5–23). Heidelberg, Germany: Springer.
Blanton, M. L., Stephens, A., Knuth, E., Gardiner, A. M. Isler, I., & Kim, J.-S. (2015). The
algebra intervention in third grade, Journal for Research in Mathematics Education,
46(1), 39–87.

Cai, J., Morris, A., Hohensee, C., Hwang, S., Robison, V., & Hiebert, J. (2017). Making
Cole, P. R. (1912). Johann Friedrich Herbart. In P. Monroe (Ed.), A cyclopaedia of education
(Vol. 3, pp. 250–254). New York, NY: Macmillan.
87, 103–121.
theory and practice (pp. 313–320). Sydney, Australia: Mathematics Education Research
Group of Australasia.
Winston.
Gagné, R. M., & Merrill, M. D. (1990). Integrative goals for instructional design. Educational
Technology Research and Development, 38(1), 23–30.
Hayward, F. H. (1904). The secret of Herbart. London, UK: Sonnenschein.
Knuth, E. Stephens, A., Blanton, M., & Gardiner, A. (2016, March). Build an early
foundation for algebra success. Kappanmagazine.org, 97(6), 65–68.
Authors. [Also cited under CCSSM. (2010).]
Selleck, R. J. W. (1968). The new education 1870–1914. London, UK: Sir Isaac Pitman and
Sons.
151–169.
Chapter 9
Answers to Research Questions, and Discussion
Abstract: Answers to the six main research questions are given, and issues arising from the
answers are discussed. Both the quantitative and qualitative analyses have pointed to the
success of both the structure and the modeling workshops. Initially, the seventh-grade
participants had very little knowledge of the associative and distributive properties—they
did not know the definitions, and could not apply the properties to numerical calculations. A
similar situation was true so far as modeling was concerned—whereas, initially, some
students could identify recursive rules for simple linear sequences, none could identify
explicit rules. Relevant algebraic conventions and language were not known. As a result of
the students’ active engagement in workshops in which the students learned appropriate
language and conventions, and made generalizations in terms of variables, most of the
participating students—but not all of them—showed strong improvement in relation to
structure and modeling. Students’ knowledge of definitions and skills improved, they
developed appropriate imagery, and their self-confidence when asked to answer questions
relating to structure and modeling improved. The results are linked to the theories of Charles
Sanders Peirce, Johann Friedrich Herbart, and Gina Del Campo and Ken Clements.
Keywords: Attitudes to school algebra, Expressive modes of communication, Imagery in

school algebra, Johann Friedrich Herbart, Modeling in school mathematics, Receptive modes
of communication, Structure in school algebra
The intervention study described in this book took the form of a design-research
investigation. The research team defined the problems to be investigated, crafted the precise
wording of six research questions, and made all organizational decisions for the
investigation, including the choice of methods which would be used to obtain answers to the
questions. In this penultimate chapter, answers to the six main research questions are given,
and issues which arose as the study progressed are discussed. Then, limitations of the study
are considered, implications for the teaching and learning of middle-school algebra are
discussed, and recommendations for future research are stated. Finally, comments on the
study by Mr. X and Mr. Y are presented.
Answer to Research Question 1

Question 1: What did the participating Grade 7 students know about each of the
associative property for addition, the associative property for multiplication, and
the distributive property, before the intervention workshops took place?
The research team decided that it was important to identify and describe pre-intervention
knowledge and understandings of the seventh-grade participating students with respect to (a) the
associative properties (for addition and multiplication of rational numbers), and the distributive

194 Ch. 9: Answers to Research Questions and Discussion
property for multiplication over addition for rational numbers, and (b) recursive and explicit
rules for linear sequences.
The document analysis, carried out in Chapter 4, drew attention to statements in the
CCSSM (2010) standards document which elaborated upon what might reasonably have been
expected of the participating seventh-grade students so far as structure and modeling were
concerned, before they took part in the structure and workshop interventions. The analysis
also elaborated upon the expectations with respect to structure and modeling implied by the
content with respect to those themes in the mathematics textbook used by the seventh-graders
at School W—specifically, the book by Charles, Branch-Boyd, Illingworth, Mills, and
Reeves (2004).
The document analysis revealed that it was expected that students would know, and be
able to apply, all of the field properties, including the associative and distributive properties,
before they began their seventh-grade studies. Hung-Hsi Wu, an author of a common-core-
inspired textbook (Wu, 2011), has stated that by the sixth grade most students should already
know the associative and commutative laws of addition and multiplication. The common-
core mathematics sequence even assumes that first-grade students will get to know the
associative property for addition. However, the pilot studies and the main study generated
data which raised serious doubt on whether those assumptions were correct.
Quantitative analyses of students’ responses to structure questions on the pre-teaching
version of the Algebra Test indicated that the means for both Group 1 and Group 2 on the
structure subtest were extremely low. Out of a maximum score of 10, the sample mean
scores, and corresponding standard deviations, were, 0.16 and 0.51 (for Group 1) and 0.47
and 0.72 (for Group 2). In other words, it seems that the students knew virtually nothing
about the associative properties for addition and multiplication, or about the distributive
property. That finding was entirely consistent with analyses carried out by the first author in
his two pilot studies (Kanbir, 2014, 2016) involving seventh- and eighth-grade students at
another midwestern middle school (located about 50 miles away from School W).
Qualitative analyses of pre-teaching interview data, and of data generated by student
responses to an initial “structural knowledge” questionnaire, complemented the results from
the pre-teaching quantitative analyses. Immediately before the first intervention workshops,
the 32 participating students were asked to respond in writing to written questions seeking
information on whether they knew the meanings of the terms “associative property for
addition,” “associative property for multiplication,” and “distributive property” (see
Appendix C). Analysis of the student responses indicated that none of the 32 participating
students had a firm knowledge of any of the properties. When, in pre-teaching interviews, the
participating students were specifically asked to give verbal descriptions of the distributive
property, only one of them could give anything like an accurate definition. In the pre-
teaching interviews, none of the students recognized signs pointing to the mathematical
“objects” which, according to CCSSM’s (2010) specifications, they ought to have known.
To sum up, then, at the pre-teaching stage most of the students did not have appropriate
receptive or expressive knowledge or understandings of any of the three structural properties
under consideration. Some had vague memories of having heard the names of the properties
before, but they did not know what the properties actually stated, and had no relevant imagery
with respect to them. In Herbartian apperception terms, their cognitive structures, or concept
images, were deficient—the students did not know how to define the properties, or when they
could be useful. Yet, textbook authors (e.g., Charles et al., 2004), and at least some of those
Answers to Research Question 1 and Research Question 2 195
who developed the common-core middle-school sequence (e.g., Hung-Hsi Wu), seemed to
think that it was reasonable to expect that the students would have known the properties.
For example, when the 28 pre-teaching interviewees were shown the sign “Find the
1 1
value of 4 × ( × 128),” none of them linked, in their minds, the 4 and the . Most of them
4 4
1
proceeded by trying to find the value of by 128 and, having done that, then to multiply that
4
result by 4. They proceeded in that way because the first letter in the PEMDAS mnemonic
(“Please Excuse My Dear Aunt Sally”) was “P,” and “M” came later: therefore, they
reasoned, they had to attend to what was inside the parentheses first. The students had not
1
recognized the mathematical object signified by the sign “Find the value of 4 × ( × 128).”
4
Similarly, when shown a sign like “describe a quick method for finding the value of
6 × 97 + 4 × 97,” none of the students in the pre-teaching interviews recognized that that sign
pointed toward the distributive property, and that 6 × 97 + 4 × 97 is equal to 97 times (6 + 4).
Instead, most of the interviewees proceeded according to PEMDAS and, because “M” came
before “A” in that mnemonic, they multiplied separate terms and then added.
We concluded that pre-teaching student responses to pencil-and-paper and interview
questions revealed that none of the participating students had either a well-formed
knowledge, or appreciation, of the power of any of the associative property for addition, the
associative property for multiplication, or the distributive property for multiplication over
addition. They lacked knowledge of definitions (verbal knowledge), did not have well-
developed and appropriate intellectual skills, and could not evoke appropriate images. A few
of them remembered having heard expressions like “associative property” and “distributive
property” being used by teachers, but they did not remember details, and they had no well-
developed attitudes regarding the structural properties of rational numbers because those
properties were not something that they had ever known or thought about.
From the perspective of the concept of a variable, at the pre-teaching stage none of the
participating students seemed to be aware that with a statement like “if a, b and c represent
any rational numbers then a + (b + c) = (a + b) + c,” the letters a, b and c are being used as
variables. The concept of a variable was not something that the students knew much about.

Question 2: To what extent were the participating Grade 7 students able to
recognize patterns and to model relationships by using variables before the
intervention workshops took place?
The common-core mathematics document (CCSSM, 2010) indicated that seventh-grade
students should have been sufficiently familiar with the concept of a variable that they would
be able to engage in elementary aspects of functional thinking—that is to say, they would be
able to use algebraic syntax to express a relationship between a well-defined pair of
variables. For example, fifth-grade students were expected to have experience with, and have
learned, the concepts of explicit and recursive rules for sequences. Middle-school teachers
are encouraged to introduce the signs of algebra (including letters for variables) to their
students well before asking them to manipulate algebraic symbols formally. Thus, if
common-core expectations were correct, the participating students would have become
acquainted with the semantics of algebraic formulations—with how the signs relate to
corresponding mathematical objects and real-life situations—before they were asked to pay
much attention to the syntax of such formulations.
At the pre-teaching stage, before the actual classroom intervention began, analyses of
paper-and-pencil test data and interview data indicated that the students’ thinking with respect
to modeling was not well advanced. For example, the students did not know the meanings of
the terms “recursive rule,” “explicit rule,” or “sequence.” The pencil-and-paper Algebra Test
and the interview protocol included questions and tasks involving modeling which were closely
aligned to CCSSM’s (2010) expectations for seventh-graders, and the research team was
interested in determining the extent to which the students were able, at the pre-teaching stage,
to use recursive and explicit modeling language and concepts to describe sequence patterns.
The team was also interested to explore whether the participants had any idea of the meaning
of the most pertinent modeling signs. At the outset, for example, could they make sense of the
sign “Suppose Tn = 3n – 1, where n can represent any natural number.” And, were they able to
associate that statement with the ordered set of numbers 2, 5, 8, 11, …? Did they know what
the term “natural number” meant? Since there was an intention to use the subscript notation for
sequences in the current study, answers to these questions were of special interest.
In relation to the subscript notation, two issues were identified: first, could individual
students interpret statements in which someone else used the subscript notation? And,
second, were individual students fluent in using the notation to describe sequences?
Quantitative analysis of pre-teaching responses to questions on the modeling subtest of
the Algebra Test indicated that before they participated in the study the Group 1 and the Group
2 students had virtually no receptive or expressive understanding of functional relationships in
which algebraic notations were employed. With a maximum possible score of 10, the recorded
sample pre-teaching mean scores and standard deviations on the modeling subtest were,
respectively, 1.41 and 0.49 (for Group 1) and 1.31 and 0.48 (for Group 2). Overall, only 2 of
the 32 students displayed some idea of how to handle the subscript notation or to specify the
nth term of a linear sequence. Only one interviewee seemed to know that the use of three dots
(…) in a “horizontal table of values” was an invitation to give the general rule for the nth term.
The terms “recursive rule” and “explicit rule” were not known. As stated previously, that
finding was consistent with the results of the two pilot studies which the first author had
conducted at the school where the pilot studies took place (Kanbir, 2014, 2016).
Analyses of qualitative data from the pre-teaching interviews which related to modeling
aspects of the study pointed to conclusions identical to those reached from the pencil-and-
paper data. Before the intervention lessons took place, students had neither receptive nor
expressive understandings of key functional-thinking concepts which the authors of the
CCSSM (2010) sequence presumed that seventh-grade students should know.
For example, consider some of the pre-teaching data with respect to the three dots
(“…”) in the following table of values:
First 1 2 3 4 5 ... n
Value
Second 3 5 7 9 ? ... ??
Value
During pre-teaching interviews conducted for the second pilot study, and for the pre-teaching
interviews in the main study, the seventh-graders did not know an appropriate convention
represented by the three dots, and thought that the n (in the last upper cell) represented 7
Answer to Research Question 3 197
(“because 5 + 1 + 1 equals 7”). In a similar way, they thought that the “??” represented 15
(“because 9 + 2 + 2 + 2 equals 15”). Analysis of student responses to this “horizontal-table-
of-values” task revealed that seventh-grade students who have never been exposed to tables
in which they are expected to make a cognitive leap to the “nth term” are unlikely to
understand the meaning of certain sign conventions and, unless these are explained to them,
will find it almost impossible to grasp the corresponding mathematical object—which is
concerned with generalizing. Whereas, a few interviewees (2 out of 28) identified simple
recursive rules, such as “add 2,” when asked to summarize successive terms in sequences
which were expressed in horizontal tables of values, they were not able to identify and
communicate explicit rules by saying, or writing, statements such as “the nth term is equal to
2n + 1” or “Sn = 2n +1.”
Basically, analyses of both quantitative and qualitative pre-teaching data showed that at
the pre-teaching stage about 95 percent of the participating students did not have a receptive
understanding of curriculum-relevant meanings of important signs in algebra, and did not
grasp what Radford (2006) called “symbolic mathematical” objects. In particular, they did
not know how to apply the concept of a variable using the subscript (sequence) notation—for
at the pre-teaching stage the notation was meaningless to them.
Thus, for example, pre-teaching student responses to pencil-and-paper and interview
questions revealed that none of the participating students had a well-formed knowledge of, or
appreciation of the power of, the concept of an “nth term” of a linear sequence. They could
not make explicit generalizations, largely because the intended “mathematical object” was
unfamiliar to them. Furthermore, they did not recall ever having seen the symbol “Tn” used,
and had neither receptive nor expressive understanding of it. They did not know how to
introduce and use a variable to represent an important aspect of a given real-life situation.
They lacked knowledge of definitions (verbal knowledge), had not developed appropriate
intellectual skills, and could not evoke appropriate images. They were unaware of the
appropriate meaning of the “three dots (…)” convention. They had not developed attitudes
towards modeling because modeling was not something they had ever known or thought
about. Although some of them remembered seeing tables of values which were presented in
vertical form, tables of values with a different orientation were confusing.

Question 3: What changes in the knowledge and understanding of participating
students with respect to structure and modeling were evident around the middle of
the intervention period (when students had participated in either the structure or
the modeling workshops, but not both)?
On the mid-intervention structure subtest of the Algebra Test, the sample means (with a
maximum possible mean score of 10) and the standard deviations were, respectively, 4.88
and 2.72 (for Group 1) and 1.63 and 1.61 (for Group 2). Mid-intervention mean scores and
standard deviations for the modeling subtest were, respectively, 1.84 and 0.98 (for Group 1)
and 3.59 and 2.30 (for Group 2) (see Figures 9.1 and 9.2).
At the mid-intervention stage, each group’s mean score was considerably higher, on one
of the subtests—but not on both subtests—than had been the case at the pre-teaching stage. The
increases at the mid-intervention stage were associated with the fact that the groups had just
participated in workshops in which they had dealt with matters related to questions on half of
the test. Thus, for example, at the mid-intervention stage, Group 1 had just participated in
workshops which focused on structure, and Group 1 showed a much greater mean gain on the
structure questions than did Group 2, whose students had just participated in workshops which
focused on modeling. The reverse was the case for Group 2, with Group 2 students having a
greater mean gain on modeling than Group 1 students.
S T RU CT U RE ( P RE -M ID )
6
Group 1 Group 2
4.9
5
MEAN SCORE(/10)
2
1.6
1
0.5
0 0.2
WEEK 1 TIMING WEEK 3
Figure 9.1. Comparison of pre-teaching and mid-intervention structure means,

Group 1 and Group 2.
M O D ELIN G ( P RE -M ID )
Group 1 Group 2
4 3.6
3.5
3
MEAN SCORE (/10)
2.5
2
1.5 1.4
1.8
1 1.3
0.5
0
WEEK 1 WEEK 3
TIMING
Figure 9.2. Comparison of pre-teaching and mid-intervention modeling means,

Group 1 and Group 2.
The mean gains for the two groups suggest that the structure intervention led to an
educationally significant increase in the performance of Group 1 students on the structure
subtest, and the modeling intervention led to an educationally significant increase in the
performance of Group 2 students on the modeling subtest. The Group 1 students also showed
a small (but statistically non-significant) gain on the modeling subtest, and Group 2 students
showed a small (but statistically non-significant) gain on the structure subtest. It seemed that
the structure workshops did not have an important effect on Group 1 students’ learning with
respect to modeling, and the modeling workshops did not have an important effect on Group
2 students’ learning with respect to structure.
Values calculated for Cohen’s (1988) d effect sizes for the interventions were
consistent with the trends depicted in Figure 9.1 and Figure 9.2. The effect size of the pre-
teaching to mid-intervention workshop sessions on structure was calculated for Group 1 (it
was assumed that the control group for this period was Group 2). Similarly, the effect size for
Group 2 was calculated for the pre-teaching to mid-intervention workshop sessions on
modeling (it was assumed that the control group for this period was Group 1). A summary of
the results of the calculations is given in Table 9.1.
Table 9.1
Effect Sizes for Intervention Workshops, Pre-Teaching to Mid-Intervention
Type of Group Involved
Workshop in Relevant
Period Control Cohen’s d
(Structure Workshops
(Pre-T to Mid-I) Group Effect Size
or (Group 1 or
Modeling) Group 2)
According to Cohen (1988), Cohen’s d effect sizes above 0.8 can be regarded as
“large.” It can be seen, from entries in Table 9.1, that the effect of the structure workshops on
Group 1’s performance on the structure subtest was very large. The effect of the modeling
workshops on Groups 2’s performance on the modeling subtest was also very large.

Question 4: Immediately after the two groups had participated in both the
structure and modeling workshops, were there statistically significant differences
between their mean gain scores in structural understanding and modeling? Also,
what were the Cohen’s d effect sizes for the two groups for the first half and for
the second half of the intervention?
For the structure subtest of the post-teaching version of the Algebra Test, 14 of the 32
participating students scored at least 70%, with 11 getting at least 80% and 23 of the 32
students getting at least 50%. For the modeling subtest questions of the Algebra Test a similar,
but slightly lower result, was obtained by the students. Of 32 participants, 8 of them scored at
least 70%, with 6 of them around 80%, and 16 of them obtained at least 50%. A post-teaching
versus pre-teaching mean gain-score analysis for Group 1 and Group 2 on the structure subtest,
showed that the mean gain for Group 1 was 5.69 (standard deviation 2.55), and the mean gain
for Group 2 was 5.28 (standard deviation 1.85). The post-teaching versus pre-teaching mean
gains for Group 1 and Group 2 on the modeling subtest were 2.88 (standard deviation 2.41), for
Group 1, and 3.00 (standard deviation 2.35) for Group 2. The mean post-teaching versus pre-
teaching gain scores for Group 1 and Group 2 on structure were not statistically significantly
different. Similarly, the mean post-teaching versus pre-teaching mean gain scores for Group 1
and Group 2 on modeling were not statistically significantly different.
In Figure 9.3, the overall trends of mean scores at the pre- and post-teaching stages of
the study can be seen. The mean scores for both groups improved for each of the subtests at
the post-teaching stage.
OV ERALL T REND S O F M EAN S CO RES

Strcuture G1 Structure G2 Modeling G1 Modeling G2
5.84
5.75
MEAN SCORE (/10)
4.31
4.28
1.41
1.31
0.47
0.16
PRE-TEACHING POST-TEACHING
TESTING STAGES
Figure 9.3. Bar graphs, showing mean scores of the two groups at pre-teaching and post-
teaching stages, on the Structure and Modeling subtests.
It will be recalled that pre-teaching to mid-intervention effect sizes were reported in

Table 9.1. It was also possible to calculate mid-intervention to post-teaching effect sizes (see
Table 9.2). The effect sizes for both structure and modeling were large, and it seemed to be
the case that the structure intervention workshops were particularly effective.
Table 9.2
Effect Sizes for Four Intervention Workshop Periods*
Type of Period Group Involved in Control Cohen’s d
Workshop (Pre-T to Mid-I, or Relevant Workshops Group Effect Size
(Structure or Mid-I to Post-T) (Group 1 or Group 2)
Modeling
Structure Mid-I to Post-T Group 2 Group 1 1.70
Modeling Mid-I to Post-T Group 1 Group 2 1.07

* Note that this table is, basically, identical to Table 7.3. It is repeated here because of our intention
to make each chapter stand alone.

Question 5: Immediately after both the structure and modeling workshops were
completed, were there educationally noticeable differences between the concept
images of the students, with respect to the concept of a variable, in comparison
with the concept images that the students had displayed before the intervention
began?
The qualitative analyses of data in Chapter 5 took into consideration Westbury’s (1980)
distinction between intended, implemented, and received curricula. It was important for the
research team to recognize, and take account of, the fact that the received curriculum for an
individual student was something much more educationally significant than mere
improvement in his or her scores on two subtests. Rather, the received curriculum for an
individual was seen to be an ongoing alignment of that individual’s concept images with
important mathematical objects. This process of objectification was consistent with Charles
Sanders Peirce’s triadic theory by which curriculum-appropriate interpretants could be
thought of as bridges between the “signs” and the “mathematical objects” being signified.
The need to bridge signs and mathematical objects was closely linked to Johann Friedrich
Herbart’s theory of apperception, which emphasized the need to take account of the
apperceptive process—although each learner would construct new knowledge, the
construction was to be achieved by encouraging the learner to link what they already knew to
what they were expected to learn. Although there is a sense in which this process was
rediscovered by Lev Vygotsky early in the twentieth century, it should be recognized that it
had been often emphasized by neo-Herbartians in different parts of the world in the second half
of the nineteenth century. Vygotsky (1978/1930) called on teachers to provide “scaffolding” to
enable learners to progress from what they already knew to what they needed to learn.
The detailed analyses of quantitative data presented in Chapter 7 revealed that about 80
percent of the participating students’ concept images with respect to structure changed in
educationally significant ways. The structure workshops helped individual students to
connect idiosyncratic “cognitive structure” components—verbal knowledge, intellectual
skills, imagery, episodes, and attitudes—in ways which enabled at least some of them to
begin to visualize relationships between concepts and principles in new ways. And, because
many of them took the opportunity to construct their own concept images, they were likely to
remember much of what they had learned. However, some of the students failed to “see” the
mathematical objects which were on the agenda. Furthermore, even those who benefited the
most still had much to learn so far as the desired mathematical objects were concerned.
The effect sizes for the modeling workshop were large, but not as large as those for the
structure workshops. It appeared to be the case that about half of the students’ concept
images for modeling were constructed and reconstructed in educationally significant ways.
Most participants recognized that they were being given the opportunity to generalize, but
about half of them struggled as they attempted to grasp the mathematical objects that they
could only partially see ahead of them.
Changing Attitudes: “I was Really Good at All that Stuff”

At the post-teaching stage almost all of the participating students indicated that they
believed that they had benefited greatly from being involved in the workshops. In post-
teaching interviews, many of them spontaneously expressed very positive feelings about
the workshops. Among their comments were:
Student 1.4, Post-Teaching Interview: I remember the distributive and associative
problems, because I was really good at all that stuff. But with the recursive and
explicit ideas, I found that it was very hard to write a rule.
Student 1.5, Post-Teaching Interview: The Tn one was kind of unusual because I
had never seen a subscription (sic.) before. That was kind of interesting.
Student 1.6, Post-Teaching Interview: I liked the charts with the explicit and
recursive rules.
Student 1.7, Post-Teaching Interview: I liked getting to learn a lot of new things
… Even my Mom, who is an advanced math teacher, said she does not teach
recursive and explicit rules, and my sister—she is a junior—does not know that
stuff. I like it, and remember at all.
Student 1.8, Post-Teaching Interview: I really liked the distributive property with
Mr. X. I will mostly remember changing the problem so that I made it an easier
problem. In the beginning I did not know what I was doing [referring to the
modeling lessons]. Later I got to know how to do some of the problems.
Student 1.14, Post-Teaching Interview: When we switched over lessons I thought
the patterns were really fun. They can be hard but eventually you will find them
out. It just takes a lot of processes and thinking. I remember just doing all the
patterns. When we did the crossing-the-river problem my group figured out the
formula very fast. When you do the recursive rule you lower the n and when you
do the second part you put a number up, and n down.
Student 2.2, Post-Teaching Interview: I learned more about the associative and the
distributive properties. I did not know much about them before. Sequences, also, I
didn’t know anything about them. I thought it was very cool.
Student 2.3, Post-Teaching Interview: I remember most the Tn + 1, because I really
got that. It was pretty easy. But, I did not like switching the teachers. Because it
was hard for me to schedule and remember Mr. X or Mr. Y’s class. Not so much
about the mathematics.
Student 2.4, Post-Teaching Interview: I learned how to get a recursive and an
explicit formula. Also, how to deal with an equation that involves the distributive
or the associative property.
Student 2.5, Post-Teaching Interview: I liked how we learned about the
distributive and associative properties. Because the distributive property would be
very useful with some giant problems.
Student 2.8, Post-Teaching Interview: I liked the distributive and associative
properties. I liked the worksheets and the homework questions.
Student 2.9, Post-Teaching Interview: I liked the modeling lessons because we had
not learned to do that before. I liked it, even though it was not easy to write things
[referring to the subscript notation and the recursive representation] and to find
some of the formulas. I also liked Mr. X’s class [on structure]. If I had to do some
real-life problems, it would be easy, like, to break it down and make it easier
[referring to the distributive property].
Student 2.13, Post-Teaching Interview: I remember mostly some of the recursive
rules and I also remember the associative property.
Student 2.15, Post-Teaching Interview: If you have a separate multiplication
problem and two different numbers and you are multiplying the same numbers you
just add the different numbers in the parentheses and multiply by the same number.
As the five participating researchers read these comments, all of which were made by real
seventh-graders who had participated in the study, we began to say to each other: “Wow—
Those workshops really made more of a difference than we had imagined.” The last of the
above quotations, from Student 2.15, was made by a girl who made impressive gains on the
Algebra Test, for both the structure and modeling subtests. Although we found it difficult to
follow what she was trying to say, in the passage quoted, it was clear that she knew what she
meant, and that she felt very positive about what she had experienced in the workshops. She
had constructed new knowledge, and she knew that that knowledge was important for her.
Question 6: Twelve weeks after both the structure and modeling workshops were
completed, were there statistically significant differences between the two groups’
mean gain scores with respect to the retention of what had been learned in regard
to understanding of (a) structure, and (b) modeling?
As stated before, the received curriculum for an individual learner is not fixed in time.
From that perspective, it was a matter of interest, in the current study, to investigate whether,
and how, a participating student’s cognitive structure immediately after the intervention
(which might be thought of as that student’s “initial received curriculum”) differed from her
or his cognitive structure 12 weeks after the intervention (which might be thought of as her,
or his, “retained curriculum”).
Analyses of retention data (gathered 12 weeks after the post-teaching tests and 20
weeks after the pre-teaching tests) enabled the research team to evaluate how well
participating seventh-grade students had retained the structural and functional knowledge and
understandings that they had developed during the intervention lessons. Like Gersten, Baker
and Lloyd (2000), we asked an important question about the effects of our teaching
intervention study: “Do the effects last beyond a very brief period over time?” (p. 3).
Differences between mean retention scores and mean pre-teaching scores for Group 1,
for both structure and modeling, were highly statistically significantly different from zero.
The same was true for Group 2. Similarly, the mean gains for Group 1 and Group 2 on the
modeling test, when retention scores were compared with pre-teaching scores on modeling,
were highly statistically significantly different from zero. That said, it was noted that for
three of the four comparisons there were declines during the retention period. The declines
were not statistically significant but from a mathematics education perspective, they were
interesting (see Figure 9.4).
The declines in mean scores during the retention period—for all but Group 2 for the
modeling subtest, for which there was a small gain—served as a timely reminder. This result
indicates that it is not enough simply to teach a topic and, then, after having assessed the
students’ learning, proceed to the next topic and “forget” what has been done before.
OVERA L L TREND S OF M EA N S C ORES

Strcuture G1
5.84 Structure G2 Modeling G1 Modeling G2
5.75
5.44
5.06
4.47
4.31
MEAN SCORE (/10)
4.28
3.91
POST-TEACHING RETENTION
TESTING STAGES
Figure 9.4. Bar graphs, showing mean scores of the two groups at post-teaching and
retention stages, on the structure and modeling subtests.
Limitations of the Study
Despite CCSSM’s (2010) emphasis on the importance of structural properties with

respect to elementary and middle-school algebra, other than the commendable research by
Melvia Ding and Xiaobao Li (2010, 2014), there has not been much research reported in
scholarly outlets on the teaching and learning of such properties in middle schools. For the
main study, the decision was made to limit the investigation of the participating seventh-
grade students’ developing knowledge of structural properties to the associative properties
for addition and multiplication and to the distributive property. Time constraints—for all
three authors of this book and also for Mr. X and Mr. Y at School W—meant that it was not
feasible to conduct a larger study which would have taken into account closure properties,
commutative properties, identity elements, inverse properties, and also order properties, and
with a wider range of middle-school students. Future researchers might profitably extend the
scope of the study reported in this book by covering other structural aspects of elementary
and middle-school algebra, and by involving students from a wider range of grade levels.
So far as the modeling aspect of the study was concerned, there has been much recent
research on aspects of modeling with respect to the teaching and learning of school algebra.
However, analysis of data generated by the two pilot studies conducted by the first author
(Kanbir, 2014, 2016) raised the possibility that many seventh- and eighth-grade students are
unaware of mathematical conventions with respect to descriptions of generalizations and, in
particular, to the subscript notation for specifying recursive and explicit rules for sequences.
The fact that the students struggled to learn the subscript notation is not surprising, given that
that notation was a long time coming among mathematicians (Cajori, 1928). So, one of the
important aspects of the analysis of the effects of the modeling lessons in the main study was
to explore whether most of the seventh-grade students were capable of learning how to use
Possibilities for Future Related Research 205
the subscript notation effectively and with understanding. There are many other aspects of
modeling covered in the recent literature which were not addressed in the study—despite the
already large volume of related research, more research is needed on this theme.
There is a sense in which the study, with only 32 participating students and two
participating teachers, was not large. However, almost the complete cohort of the seventh-
graders students attending a public school was involved in the study, and random allocation
to treatment groups for the intervention was achieved. Extensive, and hopefully rich,
interview data arising from a carefully-developed interview protocol, were generated.
Clearly, the large volume of data which was generated was both a strength and a weakness of
the study—the volume was too large to do justice to it in this book. The study was unfunded,
and all five members of the research team were very busy doing what they normally do.
The design of the study was limited because it allowed for the possibility of a teacher
effect which could possibly have biased the data. That was because one of the teachers (Mr.
X) taught all the seventh-grade “structure” workshop sessions, and the other teacher (Mr. Y)
taught all the “modeling” sessions. The three authors, who observed the workshop sessions,
fully appreciated the hard work and enthusiasm of both teachers, who were experienced, well
qualified, and diligent.
Possibilities for Future Related Research

Despite the above limitations, the results of the study are, we believe, important. This is
the first major study on algebra education to combine the theories of Peirce, Herbart, and Del
Campo and Clements, and the results are sufficiently interesting, and potentially important,
to warrant replications of the study.
There are other possibilities for related research. We need to investigate further, for
example, when the subscript notation should be introduced to middle-school or secondary-
school students. We need to know whether mathematics teacher-education graduates are
beginning their teaching careers without strong understandings of the structural and modeling
approaches to middle-school algebra. We need to investigate whether the structure and
modeling approaches are mutually independent of each other and, if so, what the implications
of that are for curriculum design. We need to know whether middle-school students, and their
teachers, realize that PEMDAS, the order-of-operations mnemonic which is so widely used
in U.S. schools, has, from a mathematical perspective, severe limitations. In fact, often
students and their teachers are not aware that students doing order-of-operations calculations
should attend to any implications of the associative and distributive properties before
applying the PEMDAS mnemonic.
There are many other issues, and in that sense the analyses presented in this book can be
regarded as the end of the beginning. Appendix H and Appendix I present tables that include
data which suggest further possibilities for creative analysis and design. In Appendix H there
are five tables which are intended to show the strength of evidence for the presence of the
five fundamental components of cognitive structure—verbal knowledge, intellectual skills,
imagery, episodes and attitudes—with respect to well-defined mathematical objects (such as
the distributive property). A method for “measuring” growth is described. The analyses and
the method are tentative, but the results arising from the analyses are interesting and
consistent with the results from the quantitative and qualitative analyses presented in Chapter
7 and Chapter 8, respectively. There would appear to be considerable scope for tightening
and developing the ideas implicit in the analyses given in Appendix H.
Appendix I provides another tentative analysis of data. It assumes Radford’s (2006)

theory, in which he outlined three levels of generalization with modeling tasks. The first
author added a fourth level—post-symbolic generalization. This approach has not been
further developed in this book, because it seemed to be less useful for classifying students’
movement toward understanding the “structure of rational numbers” than it was for the
modeling component of the study. However, there could be much scope for further
development of the ideas.
Although CCSSM (2010) emphasized the importance of structure throughout its
elementary and middle-school sequences, there seems to have been an attitude among
researchers, curriculum developers and textbook writers that students will, somehow,
automatically grow into an acceptable knowledge of the main structural properties. Analyses
of data in the current study, and in the pilot studies, indicated that it is unlikely that that
expectation is realistic. Much more research is needed on the issue.
Final Comments on the Workshop “Lessons”

When we re-examined the notes which were handed to the students for the workshops,
the following six points became clear:
1. In the structure component of the study, the mathematics in the notes started “where
the students were at.” The students had spent much time in recent years studying
operations with counting numbers, common fractions, decimal fractions, and, more
recently, negative numbers. In the structure workshops they were able,
immediately, to discuss the mathematics to which they were being introduced in the
notes.
2. Both the structure and modeling notes uncompromisingly took a mathematical
approach—with the associative and distributive properties, and elementary aspects
of the mathematics of linear sequences being addressed in mathematically honest,
but unadorned ways. Properties were clearly illustrated, and it became clear to the
students that they would be expected not only to remember the rather long names
(associative, distributive, recursive, etc.) of the properties, but also to get to know
what the properties actually meant.
3. The notes, for both the structure and modeling workshops, assumed that the
students would be placed in groups, and that they would spend much of the lesson
time discussing the mathematics that they were expected to learn. Also, the notes
made it clear that each small group of students would be expected to make
presentations to the whole class about what they “discovered.”
4. The students knew that they would be tested on what they were learning, and for
most of them it became important that they would come to “understand” the
mathematics that they were studying.
5. For each lesson, homework exercises were set, and written responses by the
students were submitted to the research team for examination. Written comments
were made on each homework submission by members of the research team—and
most of the students liked that.
6. With the modeling workshops, the students particularly liked the “crossing-the-
river” task, which involved them in manipulating simple equipment (colored blocks
and rulers). They also responded very positively to a challenge to pose their own
sequence problems—which they loved reporting to the whole class.
Comments on the Study by the Two Participating Teachers 207
The notes were consciously developed with the aim of getting students to communicate
“expressively” (Del Campo & Clements, 1987, 1990). Anyone wishing to attempt to replicate
the investigation can examine the workshop notes, which have been reproduced as Appendix E
and Appendix F to this book. Those workshop notes were successful in that they obviously
helped Mr. X and Mr. Y to engage all 32 participating students actively and expressively as
they attempted to learn structure and modeling aspects of middle-school algebra. Future studies
could involve analysis of students’ argumentation during group discussions and presentations
to other students (Toulmin, 1969), or the gestures they make to others, and to themselves
(Zurina & Williams, 2011), or how they use Twitter, or other forms of technology, to
communicate with each other in relation to homework tasks. There are many possibilities.
Comments on the Study by the Two Participating Teachers

It will be appropriate to end this chapter with comments made by the two participating
teachers on their impressions of the study. Mr. X and Mr. Y met with the three authors of this
book after the retention data had been gathered and analyzed, and the discussion which ensued
was audiotaped. Both teachers indicated that they had enjoyed being involved in the study, and
that they felt that they had profited, personally and professionally, from their participation.
They said that they intended to use the workshop notes with students in future years.
Mr. X and Mr. Y also commented that despite their overall very positive feelings about
the study, their involvement and their students’ involvement in it had placed pressure on
them. They were expected to follow the school’s syllabus, and the time spent by the students
on tests added to an already sizeable amount of class time that the students had had to spend
on tests which were outside of the normal schedule.
Some of the actual questions that the three authors asked of Mr. X and Mr. Y, and their
responses, are reproduced below:
Ellerton: From your perspective, was the study worth doing? Think about
the amount of time it took, disruptions to normal schedule, new content, the
amount of effort you had to expend, etc.
Mr. X: It fitted the seventh-grade curriculum.
Mr. Y: The time could have been tighter. There were some breaks and interruptions.
But, I enjoyed seeing the model lessons and then presenting them to the seventh-
graders. The quality of what was there was good, and the students’ interaction
with each other was also good.
Clements: How did you randomly allocate the students to the two groups for
this study?
Mr. X: I put all the names in a hat and then got Group 1 and Group 2 by
withdrawing names from the hat.
Kanbir: What was your overall impression about the students’ progress?
Mr. X: They are considerably better as a result of having participated in this study.
Some of the students obtained detailed understandings of most of the concepts, but
some concepts weren’t understood fully. As for retention, they remembered the
skills and applications, but didn’t always remember the names of the properties.
Mr. Y: I enjoyed having kids working in groups on the different tasks. The groups
then shared what they’d done. I liked it, and I also think that the kids liked it.
Ellerton: Is there anything that you will do differently, in the future, as a

result of your involvement in this study?
Mr. X: I will definitely incorporate the ideas on the structure properties when we
do operations at the beginning of seventh-year. That would fit in nicely. I have
also thought about Mr. Y’s lessons on modeling. I think there is a strong
connection between linear functions and the structure of the modeling tasks.
Mr. Y: I will incorporate the modeling content and graphing. We can look at the
patterns and at their graphs. The tasks provide good representations of linear
progressions and of how math can predict the future. Graphing would be a kind of
extension on what we were doing.
Kanbir: What was the highlight of the study, for you?

Mr. X: I remember the day before a presentation when the kids knew that they
would have to talk to the rest of the class. They were summarizing the structures
that they had been working on. Everybody was working together in small groups
on assigned tasks. That was very beneficial and convenient—very impressive.
Mr. Y: When they came together in a group and they had to come up with
examples, they made up some very creative and unique examples. They created
great questions and it worked perfectly. Some of the students really went well
beyond what I expected. It was really good for the kids.
A Call to Action
The teachers’ comments drew attention to the fact that planning and conducting
effective education research in classroom settings is a difficult thing to do in these times
when mandated tests and curricula have increasingly become part of what is normally done
in schools. Nevertheless, we believe the answers which have been provided in this chapter to
the six main research questions are important from a mathematics education perspective. The
study will have been worthwhile if those answers are noticed and acted upon by teachers,
curriculum developers, and mathematics education researchers.
References
Publishing Co.
87, 103–121.
Authors. [Also cited under CCSSM. (2010).]
Toulmin, S. (1969). The uses of argument. Cambridge. UK: Cambridge University Press.
Vygotsky, L. S. (1978/1930). Mind in society: The development of higher psychological
processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Trans.). Cambridge,
MA: Harvard University Press.
Wu, H. (2011). Understanding numbers in elementary school mathematics. Providence,
RI: American Mathematical Society.
Zurina, H., & Williams, J. (2011). Gesturing for oneself. Educational Studies in Mathematics
77(2–3), 1–14.
Chapter 10
Postscript: Framing Research Aimed at Improving School Algebra
Abstract: This final chapter is written as a guide to persons wishing to carry out research
which aims to improve middle-school students’ understanding of school algebra to the point
where not only will the students be able to generalize freely, but will also be able to apply the
algebra that they learn. The first point made in the chapter is that mathematics education
researchers need to take the history of school mathematics more seriously, because the six
purposes of school algebra identified in the historical analysis presented in Chapter 2 of this
book were important not only in helping the research team identify the importance of
language factors in school algebra, but also in designing the study which would be carried
out. The second point made is that in a design-research study the theoretical frame is likely to
be not one single theory but a composite theory arising from a bundle of part-theories that
are suggested by needs revealed in the historical analysis. The third, and final point is the
need for mathematics education researchers to remember that, ultimately, the aim of school
mathematics is to help students learn mathematics well, so that the students will be competent
and confident to use it whenever they might need it in the future. Research designs should be
such that tight assessments can be made with respect to whether the results of the studies will
help educators improve the teaching and learning of algebra in schools. Suggestions for
organization and teaching methods which will generate appropriate discourse patterns in
algebra classrooms are made, and an invitation to replicate the main study is extended.
Keywords: Algebra education, Apperception, Charles Sanders Peirce, Design research in

mathematics education, Expressive modes of communication, Imagery in school algebra,
Johann Friedrich Herbart, Modeling in school mathematics, Receptive modes of
communication, Semiotics, Structure in school algebra
Sometimes the Most Appropriate Theories for Mathematics Education Research Will
Come from the Past
Most U.S. doctoral dissertations in mathematics education come in a fairly standard
form. The first chapter provides an introduction to the problem, or problems, to be
investigated. A statement of the “theoretical lens” for the study is often given in the first
chapter, and research questions are identified. Then, in the second chapter there is a literature
review which takes special account of the writings of scholars whose ideas have been
considered to be the most relevant to the research. Then follows three chapters on design,
methodology (including instrumentation), and results. The final chapter interprets the results,
offers answers to the research questions, states conclusions, comments on perceived
limitations of the study, and makes suggestions for further research.
The first author’s dissertation (Kanbir, 2016) did not depart too much from the pattern
just described. This book, however, has a different structure. The first obvious difference is
that there is a lengthy chapter—Chapter 2—which provides a historical framework in which

212 Ch. 10: Improving School Algebra
the motive for conducting the main study emerges from our investigations into the history of
school algebra in different nations. At various parts of that chapter we argued that the
development of historical frameworks should routinely be part of most research studies in
mathematics education.
Although, there was no published history of school algebra written from an
international perspective, Chapter 2 tentatively identified, from historical considerations, six
intersecting but, nonetheless, separable purposes of school algebra. Our analysis drew
attention to a disconnect between many secondary-school algebra texts—found in school-
algebra curricular statements, in school-algebra textbooks, and in school-algebra
classrooms—and the cognitive structures of middle-school students who have used those
texts. We argued that the disconnect was largely brought about by semiotic factors—many
students did not learn to understand the signs used in algebra texts, or in algebra classrooms,
and their idiosyncratic interpretations of those signs often prevented them from learning the
mathematical “objects” that the signs were intended to represent. The students’ receptive
understandings and expressive interpretations have often been so inadequate that even after
having studied algebra, formally, for several years, some do not comprehend even apparently
simple statements, like “y equals 8 times z.” To make matters worse, often this state of affairs
is not recognized by most mathematicians, curriculum designers, external test constructors,
and even by some teachers. The situation becomes particularly serious when students think
they understand what they do not.
Giving Precedence to Peirce’s, Herbart’s, and Del Campo and Clements’s Theories
The reader might have wondered why we prioritized the theoretical positions of Charles
Sanders Peirce, Johann Friedrich Herbart, and Gina Del Campo and Ken Clements when
contemplating the theories which would be most relevant as we reflected on the fundamental
problem—“Why do so many school students find it difficult to learn algebra well?” Semiotics
is a thriving field of intellectual endeavor today, so why choose Peirce’s theory when there
are so many competing, and more “modern,” semiotic theories? Similarly, many cognitive
psychologists have investigated the role of consciousness in problem solving, so why look
back to Herbart’s theory of apperception? Our short answer to those questions would be that
although mathematicians have always been interested in the history of mathematics, and in
tracing the lineage of a mathematician from earlier mathematicians, neither mathematicians
nor mathematics education researchers have shown much interest in the history of school
mathematics, including the historical development of theories related to issues associated
with the learning of school mathematics. We believe that many of the main ideas relating to
how children think about, and learn, algebra were laid out, admittedly using different forms
of language, by Peirce, more than a century ago, and by Herbart, two centuries ago.
Peirce’s triadic emphasis on “signifiers,” “interpretants,” and “signifieds” has had a
considerable influence on philosophy, semiotics, and mathematics education, and the number
of researchers applying Peirce’s ideas to school mathematics has increased in recent years (see,
e.g., Presmeg, 2014; Radford, 2008). The first author’s (Kanbir’s) identification of Peirce’s
writings as providing an important piece of the jigsaw puzzle which made up the theoretical
frame for the main study, raises another problem associated with traditions relating to U.S.
doctoral dissertations—there is an expectation, among many experienced mathematics
education researchers, that theoretical frames should arise from “modern” theorists whose
ideas are often regarded as, somehow, being “more advanced.” Peirce died over a century ago
and surely, a critic might argue, there are more modern theorists whose writings would offer a
Discussion of the Composite Theoretical Base 213
more compelling contribution to the theoretical frame for the study. In response, we would
say we looked at numerous other modern theoretical expositions by mathematics educators
whose writings have been influenced by semiotics, but we did not find any which expressed
the signifier-interpretant-signified triad as clearly and as elegantly as Peirce. Therefore, it
was his triadic theoretical position that we chose as our guide as we set about planning the
main study and creating workshop notes for the structure and modeling interventions.
We also chose to take seriously the educational ideas of Johann Friedrich Herbart,
despite the fact that relatively few modern mathematics education researchers have displayed
much interest in Herbart’s views, or in the modifications to those views put forward by the
Herbartians in the second half of the nineteenth century. Herbart was not the first notable
scholar to formulate a view on apperception—intellectual giants like Gottfried Leibniz (see
Dascal, 1987), John Locke (Coventry & Kriegel, 1989), and Immanuel Kant (Castañeda,
1990), for instance, put forward views on what resides in a person’s “consciousness,” and on
how that influences a person’s learning (Lange, 1896). But Herbart’s writings probably had
the greatest impact on theory concerned with how knowledge of students’ long-term memory
structures ought to affect teaching practices, and on how students learn (Grinder, 1989).
Like modern-day constructivists, the Herbartians argued that learners constructed their
own knowledge (Adams, 1898). Herbart’s (1904a) theory of apperception implied that when
attempting to learn mathematics every learner identifies “messages” from stimuli which he or
she codes in a unique way. Mathematical text (like, for example, a written passage in a
textbook, or a spoken text in a classroom) is interpreted by different students in different
ways, because each learner has a unique cognitive structure which generates a unique
interaction with the text. Therefore, although a teacher of mathematics can play a vital role in
facilitating the mathematical learning of a student, he or she can never be completely sure of
what his or her students have learned, are learning, or will learn. This message, from
Herbart’s writings on apperception, was popularized by the Herbartians in many parts of the
world during the period 1850–1900. Later writers who influenced thinking among
mathematics educators made essentially the same point, without giving much credit to
Herbart. In Russia around 1930, for example, Lev Vygotsky (1962) asked teachers to assist
individual learners to embrace new material by “scaffolding” the old information to the
new—which should be within their “zones of proximal development.” Then came Jean
Piaget’s theory of equilibration, by which humans were said to learn through the twin
processes of assimilation and accommodation, which could be stimulated by cognitive
dissonance (Piaget, 1970; Sinclair, 1990). The Herbartian emphasis on the need for the
teacher to engineer maximally rich learning environments for students anticipated much of
the more radical and social constructivist and interactionist positions of Vygotsky (1962),
Jean Lave and Etienne Wenger (1991), and Heinrich Bauersfeld (1995).
The Herbartians stressed the need for teachers to assist students to “chunk” new
knowledge so that, via the process of apperception, new knowledge entities would be
welcomed into, and integrated with, the learner’s “soul” (DeGarmo, 1889, 1895; Hayward,
1904; Henderson, 1911). Thus, for example, Charles McMurry and Frank McMurry (1897)
provided illustrated lessons showing how students could be assisted to construct general truths
through individually-constructed notions. In the 1950s, Jerome Bruner, in breaking away from
a dominant behaviorist psychology which had gripped U.S. educators during the first half of
the twentieth century, emphasized the idea that whatever happened in students’ minds as they
were studying mathematics (and other subjects) was important and worthy of much deeper
study than had been recognized by the behaviorists (Bruner, 1963). In the last quarter of the
twentieth century, Ernst von Glasersfeld (1990), a “radical constructivist,” wrote: “Good
teachers and perceptive cognitive psychologists have always been aware of the fact that what
we call knowledge does not enter the uninitiated head in large, complex wholes, but must be
built up from components that, all too often, have to be very small elementary pieces” (p. 30).
Von Glasersfeld’s collaborator, Les Steffe (1990), asserted that “in any communication
between two human beings, signals can be transmitted between the communicators, but not
the intended or received meanings” (p. 7). That was precisely Herbart’s (1904b) position. As
John Adams (1898) argued, the Herbartians had “no reverence for hard facts” but recognized
that “every man is his own fact-maker, whether he will or no” (p. 67).
The Herbartians questioned the effectiveness of so-called concrete aids to learning, and
emphasized that materials, even live materials, did not assist learning if those materials did not
fit the apperceptive needs of the learner (Adams, 1898; Rein, 1893). Throughout the twentieth
century, a sequence of mathematics education researchers (see, e.g., Hiebert & Carpenter,
1992) would, on that issue, reach the same conclusion as had the Herbartians in the 1890s.
During the twentieth century many cognitive psychologists stressed the importance of
prior knowledge so far as the potential to learn is concerned. For example, David Ausubel
(1968) asserted that if he had to reduce all educational psychology to just one principle, he
would say this: “the most important single factor influencing learning is what the learner
already knows” (p. vi). In the nineteenth century, Herbart and the Herbartians emphasized
exactly that same thing (Adams, 1898). The Herbartians also emphasized, in their five steps,
the need to prepare students mentally for the main ideas that a lesson would introduce. In the
1960s, Ausubel (1968) would refer to a similar process as “providing an advance organizer.”
During the late 1970s, and in the 1980s, Robert Gagné moved away from his earlier
emphasis on learning hierarchies to embrace a view which drew attention to the importance of
“cognitive structure.” Gagné and White (1978) described a learner’s cognitive structure for a
particular topic of learning as having four main components—verbal knowledge, intellectual
skills, imagery, and episodes (that is to say, recollections of relevant incidents). Later, Gagné
would add “attitudes” to the list (Gagné & Merrill, 1990). However, for Gagné, a learner’s
cognitive structure for a topic was more than what was included in those five separate
components of long-term memory—of fundamental importance was how the components
were idiosyncratically linked. According to Gagné and White (1978), a learner’s response to
what is presented in a mathematics lesson must be a function of the learner’s cognitive
structure for the topic under consideration. Such a position was not very different from the
influence that Herbart and the Herbartians attributed to apperception. The same could be said
concerning the idea of “concept image” put forward by Shlomo Vinner and his co-workers
during the 1980s (see, e.g., Vinner & Dreyfus, 1989; Vinner & Hershkowitz, 1980).
For Herbart and the Herbartians, teaching had an important role to play in the process
of educating moral individuals. The curriculum therefore needed to be carefully planned, and
lessons carefully structured and taught so that appropriate learning would occur. To achieve
that end, the curriculum and the lessons needed to take into account the present social, moral
and cognitive states of the learners. That has always been a hard message for mathematicians
and for many school teachers of mathematics to grasp, because school algebra, geometry,
trigonometry, calculus, etc., have, for so long, been regarded as culture-free, fixed forms of
truth to be passed on faithfully to students. It was not until well into the twentieth century
that highly regarded mathematics educators would be prepared to argue that the core part of
mathematical enculturation in schools comes into effect at the meta-level and is learned
Modern Theories Emerging from Past Theories 215
indirectly (Bauersfeld, 1995; Clements, Keitel, Bishop, Kilpatrick, & Leung, 2013), and that
children’s mathematical constructions are profoundly influenced by social and cultural
conditions (Bishop, 1988; Cobb, 1989). Around 1820, Herbart regarded as axiomatic that
aspect of his theory, as did the Herbartians 70 to 80 years later.
Erich Wittmann (1998) argued that “mathematics education requires the crossing of
boundaries and depends on results and methods of considerably diverse fields, including
mathematics, general didactics, pedagogy, sociology, psychology, history of science, and
others” (pp. 87–88). According to Wittmann, scientific knowledge about the teaching of
mathematics is not gained simply by combining results from these fields, but “presupposes a
specific didactic approach that integrates different aspects into a coherent and comprehensive
picture of mathematics teaching and learning” (p. 88). Thus, “theoretical studies in the related
areas become significant only insofar as they are linked to the core and thus receive a specific
meaning” (p. 90). Emphasizing the core does not diminish the importance of the related areas.
This emphasis on the need for multiple-perspective studies that combine multiple
research methodologies is in line with the Herbartians’ view that school curricula should be
more integrated or, as Charles DeGarmo (1895) expressed it, feature “a close correlation of
studies” (p. 217). DeGarmo urged a thorough reconceptualization of the school curriculum,
in order “to develop the apperceiving power of the mind” (p. 217). He asked for serious
consideration to be given to unifying “all the studies of the elementary school,” including
arithmetic and any other branches of mathematics. This, DeGarmo maintained, could
“prevent duplication, eliminate non-essentials, and save time and effort” (p. 217).
Three decades ago, Gina Del Campo and Ken Clements (1987, 1990) distinguished
between “receptive” and “expressive” understandings of mathematical concepts. According
to Del Campo and Clements, it is one thing to be able to give meaning, in one’s own mind, to
the mathematics that others express (e.g., in passages in textbooks, or in explanations by
teachers or fellow students), and another thing to understand, and therefore be able to think
about, talk about, write about, and illustrate, the meanings and implications of that
mathematics. Research indicated that the latter was harder for learners to achieve than the
former (Clements & Del Campo, 1987). The research team for the study described in this
book wanted the participating students to be able to demonstrate both receptive and
expressive understandings, and that desire was built into the materials prepared for the
structure and modeling workshops. In those workshops, participating students were expected
to discuss ideas in small groups and to make reports to the rest of the class on what they were
learning. Nevertheless, the teachers introduced lessons, and provided review sessions in
which important concepts, skills, imageries, and relationships which the students were
expected to learn, were identified and emphasized.
The first main point, then, in this concluding chapter, is that when mathematics
education researchers are seeking to establish appropriate theoretical frameworks for research
investigations in the field of mathematics education they should not restrict themselves to
considering “modern” theories only—sometimes the most appropriate theory, or theories,
will come from the past. Furthermore, even if modern theories are chosen, it is important to
pay attention to the lineage of those theories—what were their origins?
In the study described in this book the overall theoretical framework comprised a
unique composition of theories, and not just one theory. The main investigation was such that
a solution arising from a design-research approach to research was sought. Once the main
problem had been identified, and various components of that problem identified, it occurred to
the research team that no single theoretical position from the past, by itself, would be sufficient to
illuminate adequately a path which might be followed in order to solve the problem.
Did the Intervention Improve the Participating Students’ Knowledge of, and Ability to
Generalize, and Apply, School Algebra?
Given that our fundamental task was to begin to answer the question “Why do so many
school students find it difficult to learn the subject well?”, it was important that our
intervention would assist us to make realistic statements about what might be done to help
middle-school school student learn algebra well.
We determined that the intervention should have three essential goals:
1. The design of the study would need to be such that our analyses of data would
enable us to make definite statements on whether students’ concepts, principles and
relationships, for curriculum-relevant school algebra, improved as a result of
participation in the intervention, and whether improvements were retained after the
intervention was completed.
2. In particular, we wanted to know whether participating students improved their
knowledge of algebraic language—symbols, syntax, semantics, and relationships—
as a result of their participation in the intervention; and, if that proved to be the case,
whether the improvement was still evident 12 weeks after the study was completed.
3. From the point of view of algebraic structure and modeling, was there evidence
pointing to the likelihood that five components of the participating students’
cognitive structures—verbal knowledge, skills, imagery, episodes, and attitudes—
had been enriched?
We believe that, to a certain extent, we achieved those goals—but not as fully as we would
have liked.
When the first author (Kanbir) was, formally, defending his doctoral dissertation there
were about 25 persons present. Generally speaking, their reactions to the study, as it was
summarized in the presentation, were very positive. Toward the end of the session, questions
were invited from those present, and it became clear that the overall feeling was that the
study had been successful. However, one of the questioners correctly noted that after the
participating students had been involved in all of the workshop sessions the mean post-
teaching scores fell well short of the maximum possible means, both for structure and for
modeling. Despite the fact that gain scores were highly statistically significantly different
from zero, that the effect sizes for both the structure and modeling interventions were large,
that 12 weeks after the sessions the students had retained most of what they had learned, and
that interviews had revealed obvious and strong qualitative gains for most students with
respect to the five cognitive structure components, the questioner’s observation was
important. Some of the students had not improved their knowledge of structure and modeling
as much as we would have liked.
The absence of a perfect result was not a major concern to us, however. We recalled that,
according to the common-core CCSSM (2010) sequence, before any of the workshops occurred
all of the participating seventh-grade students should have been fully familiar with the algebraic
structural properties which constituted a major focus in the study—specifically, the associative
and distributive properties for rational numbers. They should also have been able, also, to
identify recursive and explicit specifications applying to real situations for which linear
Moving Forward 217
sequences were appropriate. Pre-intervention testing and interviewing revealed that that was
far from being the case, and pilot-study data gathered by the first author at another school
indicated that it had not been the case, for seventh- and eighth-graders at that school, either.
On reflection, we believe that our intervention, and our answers to the six research
questions, represent the end of a beginning. Obviously much more needs to be done if the
teaching and learning of algebra in middle schools is to generate higher quality learning than
has been the case in the past, and indeed was the case during this intervention study. That
said, we believe that this book contains the seeds of what can be done to improve the situation.
The historical analysis of the purposes of school algebra, presented in Chapter 2, is, as far as
we know, a first, and we hope that current scholars in mathematics education will not only
read it, but will react to it, critique it, and hopefully attempt similar historical analyses.
Moving Forward
Twenty years ago Kaye Stacey and Mollie MacGregor summarized their interpretations
of data with respect to why lower-secondary students experienced difficulties in learning
algebra by drawing attention to the following four factors:
1. Students' interpretations of algebraic symbolism are based on other
experiences that are not helpful;
2. The use of letters in algebra is not the same as their use in other contexts;
3. The grammatical rules of algebra are not the same as ordinary language rules;
4. Algebra cannot say a lot of the things that students want it to say.
(Stacey & MacGregor, 1997, p. 110)
Notably, each of these factors had a semiotic aspect. Stacey and MacGregor had obviously
found that students were not interpreting algebraic “signifiers” in ways that mathematics
teachers and mathematicians expected. The British research project, Concepts in Secondary
Mathematics and Science (CSMS), conducted in the late 1970s and early 1980s, had also
shown that many lower-secondary students did not learn the meanings of algebraic
symbolisms in ways that their teachers desired (Hart, 1981). Around the same time, Anne
Newman’s (1977) research, in Australia, showed that language issues were of central
importance so far as the teaching and learning of school mathematics, in general, was
concerned—they were not confined to school algebra. Lean, Clements and Del Campo
(1990), in reporting data generated by elementary school pupils in Australia and in Papua
New Guinea, found that the pupils often misinterpreted the syntax and semantics of
arithmetic word problems which could be solved by addition and subtraction. In fact, both
Newman (1977) and Lean et al. (1990) argued, unambiguously, that difficulties with
semantic structures were much more responsible for school students’ mathematical errors
than faulty algorithms or knowledge of the meanings of important mathematical words. The
way was being prepared for greater attention to be given to the need to help school children
to learn to connect appropriate signifiers to mathematically appropriate signifieds.
The historical analysis in Chapter 2 of this book led us to combine three theories—from
Peirce, Herbart, and Del Campo and Clements—which, taken together, seemed to have the
potential to throw new light on the issues involved. Peirce’s triadic theory drew attention to
the need to develop students’ understandings of the syntax and semantics of school algebra;
Herbart’s apperception drew attention to the need to take account of individual differences in
cognitive structures; and Del Campo and Clements’s distinctions between receptive and
expressive understandings motivated us to design workshop sessions which would result in

students becoming actively engaged in the learning process.
From our perspective, the package of part-theories, consciously linked together as part
of the design-research process, illuminated an otherwise obscure pathway that school
mathematics teachers and mathematics educators might need to take if they are serious about
improving the quality of mathematics learning, and about getting students to value positively
and to apply what they learn. There are many theories on how discourse patterns in
mathematics classrooms are linked to the quality of learning (see, e.g., Herbel-Eisenmann,
Choppin, Wagner, & Pimm, 2012), but the difficulty has been to work out how to apply the
theories in ways which will generate better student understanding of key principles in
algebra.
A decision to adopt a single modern theory for a mathematics education research
investigation would provide no guarantee that results of the study would generate improved
student learning. There can be little doubt that if the demand for “algebra for everyone”
(Edwards, 1990) is to have to any chance of becoming something more than a slogan then
teachers, teacher educators, and algebra education researchers, will need to become
increasingly aware of the importance of language factors in algebra learning. That is an
important lesson from history. In particular, the need to link “signifiers” with appropriate
“signifieds”—that is to say, signifieds of interest to, and within the cognitive reach of,
learners—by developing appropriate discourse patterns which will engage students in
expressive forms of learning, needs to be heeded by researchers and practitioners alike.
Hopefully, the research planned, implemented, and evaluated by the research team of
teachers and mathematics educators, for the main study described in this book, offers a model
for what can be done to improve middle-school students’ learning of algebra. The study was
by no means perfect and, although its results were impressive, they could have been better.
That said, the study was carried out in realistic circumstances—none of the participants in the
intervention received any grant money toward the project, and there was no time release for
anyone. The materials used in the intervention—workshop notes, assessment instruments,
interview protocols—were all prepared by the research team. The pencil-and-paper test
instruments were administered during normal class times, and all interviews took place
during the students’ normal mathematics sessions. The test materials, and the interview
protocols—as they appear in the appendices to this book—have not been modified, and
probably can be improved.
It is important to recognize the clarity, with respect to the issue of the purposes of
algebra in school mathematics, generated by the historical analysis presented in Chapter 2 of
this book. In the 1920s and 1930s, Lao Genevra Simons (1924, 1936) reported high-quality
research on the history of school algebra in the early North American colonies and in the
United States of America up to 1850. But, there has been nothing comparable since, and the
absence of such historical work has resulted in modern mathematics educators often lacking
perspective on how issues which arose in earlier times were dealt with during those times.
We believe that there is a serious need for a detailed study of the history of school algebra,
and that such a study would be particularly helpful if it were to be based on comparative
interpretations of analyses of data from many nations—and not just on data from North
America, or Europe, or Asia, etc.
In any case, we look forward to others—especially persons with strong pedagogical
content knowledge with respect to school algebra (Hill, Ball, & Schilling, 2008)—attempting
not only to replicate the main study summarized in this book, but also to improve upon it,
and extend it to other components of school algebra, at various levels of schooling. In any
replication, the emphasis should not only be on the type of algebra to be taught, but also
on how the algebra needs to be presented in order that effects on the learners’ verbal
knowledge, skills, imageries, memories of episodes and attitudes will be as educationally
worthwhile as possible (Fullan & Pomfret, 1977).
References
Adams, J. (1898). Herbartian psychology applied to education: Being a series of essays
applying the psychology of J. F. Herbart. Boston, MA: D. C. Heath.
Ausubel, D. P. (1968). Educational psychology: A cognitive view. New York, NY: Holt,
Rinehart & Winston.
Bauersfeld, H. (1995). Language games in the mathematics classroom: Their function and
their effects. In P. Cobb & H. Bauersfeld (Eds.), The emergence of mathematical
meaning: Interaction in classroom cultures (pp. 271–291). Hillsdale, NJ: Lawrence
Erlbaum.
Bishop, A. J. (1988). Mathematical enculturation. Dordrecht, The Netherlands: Kluwer.
Castañeda, H.-N. (1990). The role of apperception in Kant’s transcendental deduction of the
categories. Nous, 24(1), 147–157.
Misconceptions and Educational Strategies in Science and Mathematics (Vol. 3, pp.
98–110). Ithaca, NY: Cornell University.
Clements, M. A., Keitel, C., Bishop, A. J., Kilpatrick, J., & Leung, F. (2013). From the few
to the many: Historical perspectives on who should learn mathematics. In M. A.
Clements, A. J. Bishop, C. Keitel, J. Kilpatrick, & F. Leung (Eds.), Third international
handbook of mathematics education (pp. 7–40). New York, NY: Springer.
Cobb, P. (1989). Experiential, cognitive and anthropological perspectives in mathematics
education. For the Learning of Mathematics, 9(2), 32–43.
Coventry, A., & Kriegel, U. (1989). Locke on consciousness. History of Philosophy
Quarterly, 25(3), 221–242.
Dascal, M. (1987). Leibniz: Language, signs, and thought. Philadelphia, PA: John Benjamins.
DeGarmo, C. (1895). Herbart and the Herbartians. New York, NY: Charles Scribner’s Sons.
Edwards, E. A. (Ed.). (1990). Algebra for everyone. Reston, VA: National Council of
Teachers of Mathematics.
Grinder, R. E. (1989). Educational psychology: The master science. In M. C. Wittrock & F.
Farley (Eds.), The future of educational psychology (pp. 3–18). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Murray.
Henderson, E. N. (1911). Apperception. In P. Monroe (Ed.), A cyclopaedia of education
Herbel-Eisenmann, B., Choppin, J., Wagner, D., & Pimm, D. (Eds.). (2012). Equity in
discourse for mathematics education: Theories, practices, and policies. Dordrecht, The
Netherlands: Springer.
Hiebert, J., & Carpenter, T. P. (1992). Learning and teaching with understanding. In D. A.
Grouws (Ed.), Handbook of research on mathematics teaching and learning (pp. 65–
97). Reston, VA: National Council of Teachers of Mathematics.
Hill, H. C., Ball, D. L., & Schilling, S. G. (2008). Unpacking pedagogical content
knowledge: Conceptualizing and measuring teachers’ topic-specific knowledge of
students. Journal for Research in Mathematics Education, 39(4), 372–400.
Kanbir, S. (2016). An intervention study aimed at enhancing seventh-grade students’
Lange, K. (1896). Apperception: A monograph on psychology and pedagogy. Boston, MA:
D. C. Heath & Co.
Lave, J., & Wenger, E. (1991). Situated learning. Legitimate peripheral participation.
Cambridge, UK: Cambridge University Press.
affecting children’s understanding of arithmetic word problems. Educational Studies in
Mathematics, 21(2), 165–191.
McMurry, C. A., & McMurry, F. (1897). The method of recitation. New York, NY:
Macmillan.
Authors. [Also cited under CCSSM (2010).]
Newman, M. A. (1977). An analysis of sixth-grade pupils’ errors on written mathematical
tasks. Victorian Institute for Educational Research Bulletin, 39, 31–43.
Piaget, J. (1970). Science of education and the psychology of the child. New York, NY:
Viking Press.

possibilities. ZDM—The International Journal on Mathematics Education, 40, 317–
327.
Rein, W. (1893). Outlines of pedagogics. London, UK: Sonnenschein.
Simons, L. G. (1924). Introduction of algebra into American schools in the eighteenth
century. Washington, DC: Government Printing Office.
Simons, L. G. (1936). Bibliography of early American textbooks on algebra published in the
colonies and the United States through 1850, together with a characterization of the
first edition of each work. New York, NY: Scripta Mathematica.
Sinclair, H. (1990). Learning: The interactive recreation of knowledge. In L. P. Steffe & T.
Wood (Eds.), Transforming children’s mathematics education: International
perspectives (pp. 19–29). Hillsdale, NJ: Lawrence Erlbaum.
Stacey, K., & MacGregor, M. (1997). Ideas about symbolism that students bring to algebra.
The Mathematics Teacher, 90(2), 110–113.
Steffe, L. P. (1990). Early childhood years. In L. P. Steffe & T. Wood (Eds.), Transforming
children’s mathematics education: International perspectives (pp. 3–15). Hillsdale, NJ:
Lawrence Erlbaum.
Education.
von Glasersfeld, E. (1990). Environment and communication. In L. P. Steffe & T. Wood
(Eds.), Transforming children’s mathematics education: International perspectives (pp.
30–38). Hillsdale, NJ: Lawrence Erlbaum.
Vygotsky, L. (1962). Thought and language (E. Hanfman & G. Vakar, Trans.). Cambridge,
MA: MIT Press. (Original work published, 1934)
Wittmann, E. (1998). Mathematics education as a “design science.” In A. Sierpinska & J.
Author Biographies
Sinan Kanbir is an Assistant Professor of Mathematics

Education, University of Wisconsin—Stevens Point. He
received his Ph.D. in mathematics education from Illinois
State University, a masters degree in computer and
educational technology from Marmara University, and a
bachelors degree in mathematics education from Gazi
University. Prior to his appointment at UW-Stevens Point,
he had taught mathematics at the middle- and high-school
levels, and university-level courses for prospective middle-
and elementary-school teachers. His main research interests
include the development of students’ and teachers’ algebraic
reasoning, particularly in the elementary and middle
grades, and the development of pre-service and in-service
teachers’ mathematical knowledge for teaching. He currently
teaches elementary and middle-school mathematics pre-
service teacher-education courses. His main research has
been in the areas of algebra education and the history of
school mathematics.
M. A. (Ken) Clements’s masters and doctoral degrees were

from the University of Melbourne, and at various times in
his career he taught, full-time, for a total of 15 years, in
primary and secondary schools. He has taught in six
universities, located in five nations, and is currently
Professor in the Department of Mathematics at Illinois
State University. He has served as a consultant and as a
researcher in Australia, Brunei Darussalam, India,
Malaysia, Papua New Guinea, South Africa, Thailand, the
United Kingdom, the United States of America, and
Vietnam. He served as co-editor of the three International
Handbooks of Mathematics Education—published by
Springer in 1996, 2003 and 2013—and, with Nerida
Ellerton, co-authored a UNESCO book on mathematics
education research. He has authored or edited 33 books
and over 200 articles on mathematics education, and is
honorary life member of both the Mathematical
Association of Victoria and the Mathematics Education
Research Group of Australasia.

History of Mathematics Education, DOI 10.1007/978-3-319-59204-6
224 Author Biographies
Author Biographies (Continued)
Nerida F. Ellerton has been Professor in the

Department of Mathematics at Illinois State University
since 2002. She holds two doctoral degrees—one in
Physical Chemistry and the other in Mathematics
Education. Between 1997 and 2002 she was Dean of
Education at the University of Southern Queensland,
Australia. She has taught in schools and at four
universities, and has also served as consultant in
numerous countries, including Australia, Bangladesh,
Brunei Darussalam, China, Malaysia, the Philippines,
Thailand, the United States of America, and Vietnam.
She has written or edited 18 books and has had more
than 150 articles published in refereed journals or edited
collections. Between 1993 and 1997 she was editor of
the Mathematics Education Research Journal, and
between 2011 and 2014 she was Associate Educator of
the Journal for Research in Mathematics Education. In
2016 she received an award for top researcher at Illinois
State University.
List of Appendices
Appendix Page
A Protocol for Algebra Interviews with Seventh-Graders 227
B Algebra Test (Three Parallel Versions are Reproduced) 229
C “Questionnaire” Completed by Seventh-Grade Students at

School W at the Beginning of the Algebra Workshops on “Structure” 237
D Statement of Instructional Aims for the Structure Workshops with the

Seventh-Grade Students at School W 239
E Detailed Lesson Plans for Four Workshops on “Structure” for Seventh-

Grade Students at School W, Including Homework Challenges for
Each Workshop 241
F Detailed Plans for Group Tasks in the Modeling Workshops:

Finding Recursive and Explicit Rules for Patterns 257
G Classroom Observation Schedule 287
H Pre-Teaching to Post-Teaching “Growth” with Respect to the

Five Basic Cognitive Structure Components 289
I Generalization Categories (After Radford, 2006) 295

Appendix A
Protocol for Algebra Interviews with Seventh-Graders
1. The interviews should be tape-recorded.

2. The interviewer should have the following:
(a) A sheet of paper with 482 + (18 + 300) on it. [See (2) below.]
(b) A sheet of paper with value of 4 × (¼ × 128) on it. [See (3) below.]
(c) A sheet of paper with Tn = 2n + 3 on it. [See (4) below.]
(d) A sheet of paper with Tn = 5n – 2 on it. [See (5) below.]
(e) A sheet of paper with 15 – (5 – x) = (15 – 5) – x on it. [See (6) below.]
(f) A sheet of paper showing the Table in Question 7 below.
(g) A sheet of paper showing the illustration in Question 8 below.
3. Ask each student the questions listed below, and make notes as you go.
1. I am going to say two words and, as soon as I say them, I want you to say
your head after I say the words. The words are … “distributive property.” Here are
the words again: “distributive property.”
2. Without using a calculator, find the value of 482 + (18 + 300).

[Once an answer is given ask for an explanation of where that answer came from.]
3. Without using a calculator, find the value of 4 × (¼ × 128).

[Once an answer is given ask for an explanation of where that answer came from.]
4. If we write Tn = 2n + 3, then we can say T5 equals 13, because 2 times 5 plus 3 equals 13.
What would T11 equal?
[When the pupil gives an answer, ask her or him to write down how she or he obtained
that answer. Also, ask the student to explain what she or he thought, in words.]
227
228 Appendix A: Protocol for Algebra Interviews
5. Give the pupil a piece of paper with Tn = 5n – 2 on it, then ask her or him to say which
values of n would make Tn greater than 20.
6. Give the pupil a piece of paper with the equation 15 – (5 – x) = (15 – 5) – x on it, and
then, pointing to the x, say: “Which numbers could x equal so that you would get a true
statement?”
7. Give the pupil a piece of paper with the following table on it:
First
1 2 3 4 5 … n
Value
Second
3 5 7 9 ? … ?
Value
8. The diagram below shows how tables and chairs are arranged in a school cafeteria. One
table can seat 4 people, and tables can be pushed together (but always in a straight line).
When two tables are pushed together, 6 people can sit around the table (as shown), etc.
B. If Pn represents the number of people who can sit when n tables are pushed together
(in a straight line), what is the rule giving Pn in terms of n?
Appendix B: Algebra Test (Pre-Teaching and Retention Version) 229
Appendix B
Algebra Test (Three Parallel Versions are Reproduced)

Algebra Test (Pre-Teaching Version, and also the Retention Version)
1. If Tn = 13 – 3n, where n can be any whole number, which values of n would make the
values of Tn positive?
2. A really important property for numbers and for algebra is called the associative property
for multiplication. Describe this property in your own words.
3. Suppose you were asked to calculate the value of 940 + (60 + 403) in your head (without
writing anything down, or using a calculator). How would you do it, and which property
would you be using?
4. If Sn = 101 + 50n, where n can be any whole number, what is the value of S4?
5. A student is creating towers out of unit cubes. Each unit cube, by itself, has 6 square
faces, but when two unit cubes are stuck together, one exactly on top of the other, there
are only 10 faces in the tower (including the top and the bottom). The first tower has 1
unit cube and 6 faces. The second tower has 2 unit cubes, one on top of the other, and the
third tower has 3 unit cubes, etc. We say that the surface area of the first tower is 6 units,
of the second tower is 10 units, etc.
n=1 n=2 n=3 n=4
What is the surface area of a tower with 50 cubes?
6. If Tn = 2n + 5, what is value of Tn + 1 – Tn?

230 Appendix B: Algebra Test (Pre-Teaching and Retention Version)
7. Which values of x would make this statement true? 5(x – 3) = 5x – 10.

Explain how you got your answer.
8. Without using a calculator find the value of (72 × 5) × 2, and explain how you got your
answer.
9. What must x equal if 12 – (8 – 4) = (12 – x) – 4? Explain how you got your answer.
10. If 96 = 16y + 32y, what must y equal?

11. If 20 × (10 + 5) = (20 × 10) + (20 × y), what must y equal?

12. (a) What number should replace the question mark below 5 in the second row of the
following table?
Appendix B: Algebra Test (Pre-Teaching and Retention Version) 231
First row of
numbers 1 2 3 4 5 … n
Second row of
numbers 3 6 9 12 ? … ?
(b) In the table above, what should replace the question mark which is below the n?
13. What would be a quick method of finding the value of 7 × 97 + 7 × 3 without using a
calculator? What is the property which allows you to use that quick method?
14. You have been hired by the Southwestern Fence Company to make pens for holding
cows.
A cow pen is a wall of blocks that completely surrounds the cow. You must leave at least
1 unit block in the middle of each pen where a cow would go. A cow needs 1 unit block
of space.
The first cow pen that you can build looks like this. It holds just one cow, and there are 8
surrounding blocks altogether:
232 Appendix B: Algebra Test (Pre-Teaching and Retention Version)
(a) How many surrounding blocks would you need to hold 25 cows?
(b) If Sn represents the number of surrounding blocks you would need for a pen which
would hold n cows, what is the rule giving Sn in terms of n?
Explain how you got your rule for Part (b).
1
15. What would be a quick method of finding the value of 64 × ( × 120), without using a
32
calculator?
Appendix B: Algebra Test (Mid-Intervention Version) 233
Algebra Test (Mid-Intervention Version)

_______________________________________________________________________
_______________________________________________________________________
1. In this question Tn = 14 – 4n, where n can represent various positive counting numbers.
Which values of n would make the values of Tn positive?
for addition. Describe this property in your own words.
would you be using?
5. Mr. Y wants to know how many students can sit around a row of hexagonal desks. If one
desk is by itself then six students can sit around it, but if two desks are pushed together,
then only 10 students can sit around the pattern of tables. If three desks are pushed
together in a row, as shown above, then 14 students can sit around the pattern of tables
n=1 n=2 n=3

How many students could sit around a pattern of 50 hexagonal tables arranged in a long
straight row, like the patterns shown?
6. If Tn = 3n + 2, what is value of Tn + 1 – Tn?

234 Appendix B: Algebra Test (Mid-Intervention Version)
answer.
10. If 72 = 12x + 24x, what must x equal?


following table?
First row of
numbers 1 2 3 4 5 ... n
Second row of
numbers 4 8 12 16 ? ... ?
Appendix B: Algebra Test (Mid-Intervention Version) 235
14. A king is building a house for his queen, but there has to be security rooms all around the
queen’s bedroom, as shown in the first diagram below. You can see that there are 8
surrounding security rooms altogether:
The queen complained that one room was not enough for her, so the king arranged for a
two-room version to be built, like this. There were 10 security rooms needed.
But, after a while, the queen said she needed three rooms, so the king arranged for the
following to be built.
Notice that the queen’s rooms must always be in a straight line, left to right.
(a) How many surrounding security rooms would you need if the queen had 25 rooms?
(b) If Rn represents the number of surrounding security rooms needed if the queen had n
rooms, what is the rule giving Rn in terms of n?
1
24
calculator?
236 Appendix B: Algebra Test (Post-Teaching Version)
Algebra Test (Post-Teaching Version)

___________________________________________________________________________
1. If Tn = 17 – 4n, where n can represent various positive counting numbers, which values of
n would make the values of Tn positive?
for multiplication. Describe this property in your own words.
would you be using?
5. A student is creating towers out of unit cubes. Each unit cube, by itself, has 6 square
faces, but when two unit cubes are stuck together, one exactly on top of the other, there
are only 10 faces in the tower (including the top and the bottom). The first tower has 1
unit cube and 6 faces. The second tower has 2 unit cubes, one on top of the other, and the
third tower has 3 unit cubes, etc. We say that the surface area of the first tower is 6 units,
of the second tower is 10 units, etc.
n=1 n=2 n=3 n=4
What is the surface area of a tower with 50 cubes?
6. If Tn = 3n + 4, what is value of Tn +1 – Tn?

Appendix B: Algebra Test (Post-Teaching Version) 237
answer.
10. If 72 = 12y + 24y, what must y equal?


238 Appendix B: Algebra Test (Post-Teaching Version)
following table?
First row of
numbers 1 2 3 4 5 … n
Second row of
numbers 4 8 12 16 ? … ?
14. You have been hired by the Southwestern Fence Company to make pens for holding
cows.
A cow pen is a wall of blocks that completely surrounds the cow. You must leave at least
one unit square in the middle of each pen where a cow would go. A cow needs 1 square
unit of space.
The first cow pen that you can build looks like this. It holds just one cow, and there are 8
surrounding blocks altogether:
Appendix B: Algebra Test (Post-Teaching Version) 239
(a) How many surrounding blocks would you need to hold 25 cows?
(b) If Sn represents the number of surrounding blocks you would need for a pen which
would hold n cows, what is the rule giving Sn in terms of n?
1
calculator? 24
Appendix C
“Questionnaire” Completed by Seventh-Grade Students at School W
at the Beginning of the Algebra Workshops on “Structure”
1. Do you have any idea what the associative property for addition for real numbers states?
Your Answer: Circle whichever is appropriate for you— Yes
No
If your answer was “Yes,” write down, in your own words what you think the associative
property for addition for real numbers is:
2. Do you have any idea what the associative property for multiplication for real numbers
states?
No
If your answer was “Yes,” write down, in words what you think the associative property
for multiplication for real numbers is:
3. Do you have any idea what the distributive property connecting multiplication and
addition of real numbers states?

No
If your answer was “Yes,” write down, in words what you think the distributive property
connecting multiplication and division of real numbers states:
241
Appendix D
Statement of Instructional Aims for the Structure Workshops
with the Seventh-Grade Students at School W
Aims:
1. To identify how well the students know, before any of the planned lessons take place, the
formal statements associated with the following field properties with respect to real
numbers:
(a) The associative property for addition [i.e., if a, b, c represent any real numbers then
(a + b) + c = a + (b + c)]
(b) The associative property for multiplication [i.e., if a, b, c represent any real
numbers then (a × b) × c = a × (b × c)]
(c) The distributive property combining addition and multiplication [i.e., if a, b, c
represent any real numbers then a × (b + c) = (a × b) + (a × c)].
2. To engage the seventh-grade students in meaningful discussions with fellow students
about how knowledge of the above three properties can facilitate calculations, especially
mental calculations.
3. To identify how well the seventh-grade students learn to apply the associative and
distributive properties with respect to simple, but appropriate, mental calculations.
4. To assist the students to develop confidence as well as competence with respect to
understanding and applying the three properties which are the special focus of the
sessions.
Preparation of Material:
1. Workshop 1 (Group Task 1) and Workshop 1 (Group Task 2) are handouts for
students. It is intended that these handouts be completed by the seventh-grade students, as
they engage in group discussions. The class would be subdivided into groups of 3 or 4,
with half of the groups being assigned Group Task 1 (which is concerned with the
associative property for addition) and the other half of the groups being assigned to
Group 2 (which is concerned with the associative property for multiplication).
2. An initial questionnaire has been prepared. The aim of this is to glean information about
what the students already know about the associative properties for addition and
multiplication of real numbers, and what they already know about the distributive property.
3. Typed homework materials, for each of the first two sessions, have also been prepared.
These materials are to be submitted to the teacher at the beginning of the second and third
sessions. The two homework sheets will be mostly review, but there will also be
“challenging questions” on each sheet.
243
244 Appendix D: Instructional Aims for Structure Workshops
4. This “summary sheet” has also been prepared, the main purpose of which is to indicate
clearly the intended sequence for the first two sessions.
Intended Sequence for the First Two Sessions:

1. Introduction: Students will each have a folder in which to keep all of their worksheets
and homework. This should greatly facilitate opportunities for the researchers to look at
students’ work.
2. Allocation to groups and re-arrangement of tables and administration of questionnaire.
3. Group discussion should focus on the questions asked on printed handouts (for either
Workshop Group Task 1 or Workshop Group Task 2). Half of the groups will have the
Task 1 handout, and the other half will have the Task 2 handout. The students will be told
that in their group they should read the materials, agree on good answers to the questions,
and then decide who will teach the students doing the other task that day when they meet
the next day. On the second day, group members will be responsible for teaching students
in the other group the mathematical ideas that they learned on the first day. Note that on
the second day, all of the students will have, and be able to refer to, both the handouts for
Workshop Group Task 1 and Workshop Group Task 2.
4. At the end of each of the first two sessions, each of the students will be given a
“homework” sheet which he or she will be expected to complete overnight and submit to
the teacher at the beginning of the next session. The quality of the students’ responses on
these homework sheets will be assessed before being returned to the students.
Brief Comments on the Expected Forms of Sessions 3 through 5:
1. Note that during the third and fourth sessions all students will work in their groups, and a
printed handout on two key aspects of the distributive property—“expanding”
parentheses, and factoring—will be made available to the students. It is expected that the
pattern will again be group discussions in the third session, followed by group-teaching
of the “other” groups during the fourth session.
2. Then, in the fifth session, the teacher will provide a review of what has been covered
during the first four sessions. The aim of this review is to bring into focus the main
concepts and principles that the students have been grappling with.
3. This set of five sessions will then be repeated with a second set of seventh-grade students
at School W (beginning soon after the set of five sessions is completed).
Appendix E
Detailed Lesson Plans for Four Workshops on “Structure” for Seventh-Grade Students at
School W, Including Homework Challenges for Each Workshop
Workshop 1: Notes for the First Workshop Sessions on “Structure”—An Important

Number Property: The Associative Property for Addition
INSTRUCTIONS: You will be working with two or three others, and your group’s task is to
discuss the questions, work out what they mean, and then answer them.
Question 1.1: If a = 12, b = 6 and c = 2, what are the values of each of the following?
a + (b + c)
(a + b) + c
a – (b – c)
(a – b) – c
(a – b) + c
Now write down what you’ve found, neatly in the space below:
Question 1.2: Repeat each of the five parts of Question 1.1, only this time let a = 51, b = 26,
and c = 12. Then, write down what you’ve found neatly in the space below:
Question 1.3. Repeat Question 1.1, only this time let a = 1¼, b = ¾, and c = ½.
Question 1.4. Repeat Question 1.1, only this time let a = –3, b = –2, and c = 1.
Question 1.5: Repeat Question 1.1, only this time let a, b and c be any three real numbers
that your group chooses. Then, write down what you’ve found neatly in the space below:
245
246 Appendix E: Lesson Plans (Structure Workshop 2)
Question 1.6: Do you think a + (b + c) will always equal (a + b) + c, no matter which

numerical values you allow a, b and c to be?
Talk about that question, and decide who, from your group, will tell the whole class, during
your next mathematics lesson, about what your group was thinking.
Question 1.7: Do you think a – (b – c) will always equal (a – b) – c, no matter which number
values you allow a, b and c to represent?
Talk about this, and decide who, from your group, will tell the whole class during your next
mathematics class, about what your group was thinking.
Question 1.8: Do you think a – (b – c) will always equal (a – b) + c, no matter which number
values you allow a, b and c to represent? Talk about this, and decide who, from your group,
will tell the whole class, during the next mathematics lesson, about what your group was
thinking.
An Important Number Property Which You Must Remember

Here is an important number property for you to learn and remember.
If a, b, c represent any three real numbers then (a + b) + c always equals a + (b + c). This is
called the associative property for addition. It is always true, no matter which values you
give a, b and c.
But, as we have found, the associative property does NOT hold for subtraction, because, in
most cases, a – (b – c) does not equal (a – b) – c.
Discuss: Under what circumstances will a – (b – c) equal (a – b) – c? Write down your
conclusions.
Appendix E: Lesson Plans (Structure Workshop 2) 247
Workshop 2: Notes for the Second Workshop Sessions on “Structure”—An Important

Number Property: The Associative Property for Multiplication
a × (b × c)
(a × b) × c
a ÷ (b ÷ c)
(a ÷ b) ÷ c
(a ÷ b) × c
Question 1.2: Repeat each of the five parts of Question 1.1, only this time let a = –16, b = 8,
and c = –2. Then, write down what you’ve found neatly in the space below:
Question 1.3. Repeat Question 1.1, only this time let a = 1¼ , b = 2 , and c = ½.
5
Question 1.4. Repeat Question 1.1, only this time let a = –3, b = –2, and c = 1.
Question 1.5: Repeat Question 1.1, only this time let a, b and c be any three real numbers,
other than 0, that your group chooses. Why did we exclude zero from this task?
Write down what you’ve found neatly in the space below:
Question 1.6: Do you think a × (b × c) will always equal (a × b) × c, no matter which

the next mathematics class, about what your group was thinking.
Question 1.7: If we exclude zero from the possible values, do you think a ÷ (b ÷ c) will
always equal (a ÷ b) ÷ c, no matter which number values you allow a, b and c to represent?
Talk about this, and decide who, from your group, will tell the whole class, during the next
mathematics lesson, about what your group was thinking.
Question 1.8: Do you think a ÷ (b ÷ c) will always equal (a ÷ b) × c, no matter which number
values you allow a, b and c to represent? Talk about this, and decide who, from your group,
will tell the whole class, during the next mathematics lesson, about what your group was
thinking
An Important Number Property Which You Must Remember

If a, b, c represent any three real numbers then (a × b) × c always equals a × (b × c). This is
called the associative property for multiplication. It is always true, no matter which real-
number values you give a, b and c.
But, as we have found, the associative property does NOT hold for division, because, in most
cases, a ÷ (b ÷ c) does not equal (a ÷ b) ÷ c. Now discuss: Under what circumstances will
a ÷ (b ÷ c) equal (a ÷ b) ÷ c? Write down your conclusions.
Workshop 3: Notes for the Third Workshop Sessions on “Structure”—An Important

Number Property: The Distributive Property for Multiplication Over Addition, Part I
a × (b + c)
a×b+a×c
a × (b – c)
a×b–a×c
(b + c) × a
Question 1.2: Repeat each of the five parts of Question 1.1, only this time let a = 51, b = 26,
and c = 12. Then, write down what you’ve found neatly in the space below:
Question 1.3: Repeat Question 1.1, only this time let a = 1½, b = ¾, and c = ¼.
Question 1.4: Repeat Question 1.1, only this time let a = –3, b = –2, and c = 1.
Question 1.5: Repeat Question 1.1, only this time let a, b and c be any three real numbers
that your group chooses. Then, write down what you’ve found neatly in the space below:
Question 1.6: Do you think a × (b + c) will always equal a × b + a × c, no matter which

the next mathematics lesson, about what your group was thinking.
Question 1.7: Do you think a × (b – c) will always equal a × b – a × c, no matter which

number values you allow a, b and c to represent?
Talk about this, and decide who, from your group, will tell the whole class, during the next
Question 1.8: What do you think (a – b) × c might equal, no matter which number values you
allow a, b and c to represent? Check your conjecture with some numbers. Decide who, from
your group, will tell the whole class, during the next mathematics lesson, about what your
group was thinking.
An Important Number Property Which You Must Remember (Also, Remember its
Name!)
If a, b, c represent any three real numbers then a × (b + c) is always equal to a × b + a × c.

This is called the distributive property for multiplication over addition. It is always true,
no matter which values you give a, b and c.
Discuss: What are the main differences between the associative property for addition, the
associative property for multiplication, and the distributive property linking multiplication
and addition?
Workshop 4: Notes for the Fourth Workshop Sessions on “Structure”—An Important

Number Property: The Distributive Property for Multiplication Over Addition, Part II
We say “the factors of 12 are 1, 2, 3, 4, 6 and 12.” What do we mean when we say that 4 (for
example) is a factor of 12?
Question 1.1: If a = 11, b = 8 and c = 5, calculate the values of ab + ac. Then answer the
following questions:
Is 11 a factor of the value of ab + ac?
Is 13 a factor of value of ab + ac?
Is it true that a × (b + c) equals (a× b) + (a × c)? Notice that in algebra this is usually written
a(b + c) = ab + ac,
with the multiplication signs on the left and right sides dropped, and the two pairs of
parentheses on the right side also dropped.
If you swap the sides you get
ab + ac = a(b + c),
and that emphasizes that a and (b + c) are both factors of ab + ac.
We say that a is a “common factor” of ab and ac because a is common to ab and ac. The
other factor, (b + c), represents the sum of what is left over after the a is removed.
Question 1.2
If a, b and c represent any three numerical values, must the value of a be a factor of both
ab + ac and ab – ac?
If a, b and c represent any three numerical values, must the value of (b + c) be a factor of
ab + ac?
If a, b and c represent any three numerical values, must the value of (b – c) be a factor of
ab – ac?
Explain why, in algebra, (a × b) + (a × c) must always equal a × (b + c), no matter which real-
number values you allow a, b, c to be.
Explain why, in algebra, (a × b) – (a × c) must always equal a × (b – c), no matter which real-
number values you allow a, b, c to be.
What are factors of: 3x – 15?

What are factors of 2x2 + 4x?
Question 1.3: Repeat each of the parts of Question 1.1, only this time let a = 51, b = 26, and
c = 12. Then, write down what you’ve found neatly in the space below:
Question 1.4. Repeat the parts of Question 1.1, only let a = 1½, b = ¾, and c = ¼.
Question 1.5. Repeat the parts of Question 1.1, only let a = –3, b = –2, and c = 1.
Question 1.6: Do you think ab + ac + ad will always equal a(b + c + d), no matter which
the mathematics class tomorrow morning, about what your group was thinking.
Question 1.7: Do you think 5x + 3x + 2 will always equal 10x, no matter which number
values you allow x to represent?
Talk about this, and decide who, from your group, will tell the whole class during the next
Question 1.8: Do you think 3a + 3b will always equal 6ab. no matter which number values
you allow a and b to represent?
An Important Number Property Which You Must Remember. (Also, Remember its
Name!)
If a, b, c represent any three real numbers then a(b + c) is always equal to ab + ac. This is
called the distributive property for multiplication over addition. It is always true, no matter
which values you give a, b and c. Since a(b + c) is always equal to ab + ac it follows that a
and (b + c) are factors of ab + ac.
Discuss: What are the main differences between the associative property for addition, the
associative property for multiplication, and the distributive property linking multiplication
and addition?
Appendix E: Homework Challenges (Structure Workshop 1) 253
Homework Challenges for Distribution to Seventh-Grade Students Following

Workshop 1
1. Using mental arithmetic only (don’t write down anything), calculate the following:
(a) 102 + 798
(b) 89 + 101 + 10
Write a short paragraph below about how the associative property for addition can be used
to find the answers to (a) and (b) quickly.
2. Write a short paragraph about how the associative property for addition can be used to find
an answer to the following:
(a) 8500 + 601 + 899
(b) 3½ + 16¼ + ¼
Write a short paragraph, at the top of the back of this page, which makes clear how
the associative property for addition can be used to find the answer.
254 Appendix E: Homework Challenges (Structure Workshop 1)
(c) This final homework question is mathematically very challenging. If any seventh-grade
student gets its completely right, with correct reasoning, then he or she will be doing very
well!
Under what circumstance will it be true that, if a, b and c represent real numbers then
(a – b) – c = a – (b – c)?
Remember, too, members of your group should be ready to talk about the first
workshop at the next session (tomorrow).
Homework Challenges for Distribution to Seventh-Grade Students Following

Workshop 2
25 × (4 × 19)
Write a short paragraph below about how the associative property for multiplication can
be used to find the answer quickly.
(–2.5 × 0.93) × 4
Write a short paragraph below about how the associative property for multiplication can
be used to find the answer quickly.
3. How could you use an associative property to find answers to the following?
(a) 48 × 52 × 1
12
(b) –3½ × 16 × – 4
7
Write two short paragraphs below about how the associative property for multiplication
could be used to find the answers to (a) and (b). Do you need to use another property?
A Final, Very Challenging Homework Question
If any seventh-grade student gets its completely right, then he or she will be doing very well!
Under what circumstance will it be true that, if a, b and c represent real numbers then
(a ÷ b) ÷ c = a ÷ (b ÷ c)?
Remember, too, members of your group should be ready to talk about the second
workshop at the next session.
Homework Challenges for Distribution Following Workshop 3
1. Using mental arithmetic only (don’t write down anything, and don’t use a calculator),
calculate the following:
(a) How much would 11 apples cost at $0.99 each?
(b) How much would 9 newspapers cost at 55 cents each?
(c) If one book weighs 1¼ pounds, how much would 12 of the books weigh?
Write a short paragraph below about how the distributive property can be used to find the
answers to (a), (b) and (c) quickly.
2. How could you use the distributive property to find an answer to the following?
(a) 17 × 97 + 17 × 3
(b) 11 × 9.9 + 89 × 9.9

Write a short paragraph which makes clear how the distributive property could be used to
find the answers to (a) and (b) quickly.
3. Explain why the truth of the statement 5x + 3x = 8x could be justified by referring to the
distributive property.
4. Simplify, as much as possible: 5x(1 – x) – 4 + 5x2. When did you use the distributive
property when doing the simplification?
Remember, too, members of your group should be ready to talk about the third
Homework Challenges for Distribution Following Workshop 4
24 × 4 + 76 × 4
Write a short paragraph below about how the distributive property can be used to find the
answer quickly.
(–2.5 × 6) + (4 × –2.5)
Write a short paragraph below about how the distributive property could be used to find
the answer quickly.
3. Simplify as much as possible 5x + 4 – 3(x – 2), justifying each step.
Remember, too, members of your group should be ready to talk about the fourth
Appendix F
Detailed Plans for Group Tasks in the Modeling Workshops:
Finding Recursive and Explicit Rules for Patterns
Workshop Group Tasks: Finding Recursive and Explicit Rules for Patterns
work out what are called the “recursive” and “explicit” rules for patterns which you identify
for your two tasks.
But first we’ll go over an example.
Worked Example: n Sn
(a) See if you can work out
1 2
the pattern, and then
2 7
replace the question marks
3 12
in the table.
4 17
(b) In this question we 5 ??
could summarize the … …
pattern by writing “S1 = ?? 37

2, and Sn + 1 = Sn + 5.”
n ??
That means the first term, S1, is 2, and you get the next term from the previous term by
adding 5. Thus S2 = 2 + 5 = 7; S3 = 7 + 5 =12, S4 = 12 + 5 =17, etc. The expression “Sn + 1 = Sn
+ 5” is said to be the recursive rule for the pattern, and you should also give the first term S1.
If you want to find S20, for example, you start with 2 (which is S1), add 5 to get S2, and then
just keep on adding 5 until you get to S20. But it could take you a long time to get to S20.
Often it is quicker to look for an explicit form of the rule.
In this question the nth term, Sn, is equal to 5n – 3. Can you see where the “5” comes from?
Where did the “– 3” come from?
261
262 Appendix F: Lesson Plans (Modeling Workshop Group 1)
“Sn, = 5n – 3” is called an explicit rule for the pattern.

Now let’s see if you can work out recursive or explicit rules (sometimes just one of them,
sometimes both) for the tasks allocated to your group. Then, your group will also be expected
to create a pattern of its own, and then see if you can get others in the class to understand
what your pattern is.
FIVE TASKS FOR GROUP 1

You should work together to find the recursive and explicit rules for the patterns in the first
two tasks. With Task 3, “Crossing the River,” your teacher will take the lead, and the whole
class will work on it together. Then, Task 4 is harder, and for that task you only need to find
an explicit rule. Finally, for Task 5, your group will be expected to create a pattern of its
own.
Task 1 (for Group 1):
Talk together and work out the pattern between the two variables—the number of days and
the number of completed weeks. Then, find the missing values.
Completed Days
Weeks n Dn
1 7
2 14
3 ?
… …
10 ?
… …
? 364
… …
n ?
(a) What is the recursive rule for the pattern?
(b) What is the explicit rule for the pattern?
(c) What is the value of n if Dn = 2562?
How are you going to explain this pattern to the rest of the class? Who will do the talking?
Appendix F: Lesson Plans (Modeling Workshop Group 1) 263
Figure 1 Figure 2 Figure 3
The first three terms for the pattern of the tiles shown above are T1 =3, T2 = 6, and T3 = 9.
Number of Figure Number of Tiles in

(n) that Figure (Tn)
1 3
2 6
3 9
… …
8 ?
9 ?
… …
? 120
… …
n ?
(c) How many tiles would there be in Figure 80? Why?
Now work out how you are going to explain the pattern to the rest of the class. Who will
do the talking?
Task 3 for Group 1: “Crossing the River”

(This will be done by all students in the class, but you will still work on the task in your
group)
Some armed robbers are chasing 8 adults and 2 very strong and fit children who have found
gold in a long-lost cave out in the desert. The 8 adults and 2 children, who are about 4 miles
ahead of the robbers, come to a river which is infested with very many man-eating
crocodiles. There is a little old boat, and someone has left handwritten instructions:
CAREFUL: This old boat will sink if you try to have 2 adults in it. It will even sink if
you try to have 1 adult and 1 child in it. But it can carry 1 child, or 2 children, or 1
adult.
The problem is: What is the least number of crossings with the boat so that all 8 adults
and the 2 children get to the other side of the river? A crossing is going across to the other
side—so, going across to the other side and then coming back would count as two (2)
crossings.
Once you arrive at your answer for the least number of crossings check that you’re correct
with a teacher. Then complete the following table (which applies to the situation where there
are always 2 very strong children, but n adults needing to cross the river):
Number of adults Smallest possible
needing to cross (n) number of crossings
(Cn)
1 ?
2 ?
3 ?
… …
8 ?
9 ?
… …
? 121
… …
n ?
Now answer these questions.

Task 4 (for Group 1)—This is a harder task!

(a) You don’t have to find the recursive pattern for this one but you do have to find an
explicit pattern. Talk together and work out an explicit pattern connecting n and Sn.
Then, find the missing values, complete the table, and answer the questions in (b), (c) and
(d) below.
Value of n Value of Sn
1 2
2 5
3 10
4 17
… …
10 ?
… …
n ?
(b) What is the value of S100?

(c) If Sn = 1297, what is the value of n?
Now work out how you are going to explain the pattern to the rest of the class. Who will do
the talking?
[Note: There is a recursive pattern for this—but it is VERY hard to find it. If you’re brave
you might try to find it!]

Your group should make up an interesting pattern task—not too easy, not too hard.
During the next session you will be asked to explain your task to the whole class. You may
also be asked to give an explicit rule for your pattern.

own.
(a) Talk together and work out the pattern between the two variables—the number of
cars and the number of tires. Then, find the missing values in the following tables
Cars (n) Tires (Sn)
1 4
2 8
3 ?
? 240
… …
100 ?
… …
n ?
(b) What is the recursive rule for the pattern?
(c) What is the explicit rule for the pattern? What is the value of n if Sn = 172?
the talking?
Appendix F: Lesson Plans (Modeling Group 2) 267

Mr. Y wants to know how many students can sit around a row of hexagonal shaped desks.
The challenge is to be able to answer (a), (b) and (c) and be able explain to the class, the
formula for Sn for the pattern.
If one desk is by itself then six students can sit around it. If two desks are pushed together,
then 10 students can sit around the pattern of desks. If three desks are pushed together in a
row, as shown above, then 14 students can sit around the pattern of desks.
(a) Fill in the following table.
Number of Number of students
hexagonal desks that can sit around
the desks
n Sn
1 6
2 10
3 14
4
5
6
7
8
...
n

(c) What is the explicit rule for the pattern? What is the value of n if Sn = 172?
(d) Imagine that 100 of the hexagonal desks were pushed together in a row. How many
students could sit around that pattern of desks?
the talking?

group)
adult.
crossings.
(Cn)
1 ?
2 ?
3 ?
… …
8 ?
9 ?
… …
? 121
… …
n ?

Figure 1 Figure 2 Figure 3
The first three terms for the pattern of small triangular tiles shown above are 2, 8, and 18.
(a) Complete this table:
Figure Number Numbers of

(n) Small Triangles
(Tn)
1 2
2 8
3 18
4
5
6
...
n Sn =
(b) What is an explicit rule for the pattern?

(c) How many tiles would there be in Figure 80? Why?
the talking?


own.
At a fun park it costs $10 for entry and then $3 for each ride. Talk together and work out the
pattern between the two variables—total cost $Cn and the number of rides you have. Then,
find the missing values.
Rides (n) Cost ($Cn)
0 10
1 13
2 16
3 ?
… …
20 ?
? 103
… …
n ?

(b) What is an explicit rule for the pattern?
(c) What is the value of n if Cn = 79?
the talking?
Based on the same pattern, draw Figure 4, and then complete the following table which
should be based on the number of matches Mn, in Figure n (for different values of n).
Figure Number of
n Matches, Mn,
Making up Figure
n
1 3
2 5
3 ?
4 ?
n Mn = ?

(b) What is the explicit rule for the pattern? [Careful: It’s not Mn = 2n + 3.]
(c) What is the value of n if Mn = 201?
the talking?

group)
adult.
crossings.
(Cn)
1 ?
2 ?
3 ?
…
8 ?
9 ?
…
? 121
…
n ?

Task 4 (for Group 3)—This is a harder task:

Then, find the missing values, and answer the questions in (b), (c) and (d) below.
1 2
2 6
3 12
4 20
10 110
… …
n ?

the talking?
[Note: There is a recursive pattern for this—but it is VERY hard to find it. If you’re brave
you might try to find it! If you think you’ve found it tell a teacher!]


own.
A young salesman agreed to a contract in which he has a starting salary of $1000 per month.
His monthly salary will increase by $120 at the beginning of each new month. Talk together
and work out the pattern between his monthly income $Sn and the number of months that he
has completed at work. Then, find the missing values.
Number of Monthly
Completed Salary
Months (n) ($Sn)
0 1000
1 1120
2 1240
3 ?
… …
? 3400
… …
n ?

(c) What is the value of n if Sn = 4960?
the talking?
(a) Based on the same pattern, draw the fourth figure. Then complete the table. Suppose
the symbol Tn is used to represent the number of matches needed for Figure n.
Number of Figure Number of Matches
n in the Figure
Tn
1 6
2 9
3 12
4 ??
5 ??
? 39
… …
n ??

(c) What is the explicit rule for the pattern?
(d) What is the value of n if Tn = 201?
the talking?

group)
CAREFUL: This old boat will sink if you try to have 2 adults in it. It will even sink if you
try to have 1 adult and 1 child in it. But it can carry 1 child, or 2 children, or 1 adult.
crossings.
with a teacher. Then complete the following table (which applied to the situation where there
(Cn)
1 ?
2 ?
3 ?
… …
8 ?
9 ?
… …
? 121
… …
n ?

Task 4 (for Group 4)—This is a quite a difficult task:

Then, find the missing values, and answer the questions in (b) and (c) below.
Value of n Value of
Sn
1 1
2 3
3 6
4 10
5 15
… …
n ?
the talking?


an explicit rule. Finally, for Task 5, your group will be expected to create a pattern of its own.
A new car salesman has a retaining salary of $2500 per month and is then paid an additional
$700 for every car he sells during the month.
Cars sold in Income
the month (n) ($Wn)
0 2500
1 3200
2 3900
3 ?
… …
? 24200
… …
n ?
(a) Talk together and work out the pattern between his monthly income $Wn and the
number of cars he sells. Then, find the missing values.
(d) What is the value of n if Wn = 10200?
the talking?
A pattern of tiles is shown, and the following table shows the area for the first two Figures.
Replace the question marks in the following table.
Position Area of the shape in that
position, An square units
1 1 square unit
2 4 square units
3 ?
4 ?
n An = ? square units
Suppose the symbol An is used to represent the area of Figure n (in square units)
(c) What is the value of n if An = 441?
the talking?

group)
adult.
crossings.
(Cn)
1 ?
2 ?
3 ?
… …
8 ?
9 ?
… …
? 121
… …
n ?

Task 4 (for Group 5)—This is a quite a difficult task.

You don’t have to find the recursive pattern for this one but you do have to find an explicit
pattern.
(a) Talk together and work out an explicit pattern connecting n and Tn. Then, find the
missing values, and answer the questions in (b) and (c) below.
Value of n Value of Tn
1 0.1
2 0.4
3 0.9
4 1.6
5 2.5
… …
n ?
(b) What is the value of T20?

(c) If Tn = 129.6, what is the value of n?
the talking?


own.
The cost to rent a construction crane is $500 per day, plus $220 per hour that you use the crane.
Complete the following table, showing the cost if the crane was used for 2, 3, 4, or 5 hours
on a day. [Note that an extra part of an hour is counted as a full hour.]
Number of Total Cost

Hours (or Part of Hire, in
Hours) that the Dollars
Crane is Used (Tn)
(n)
0 500
1 720
2
3
4
5
…
n Tn =
(a) How much would it cost on a day when the crane was used for 15 hours?
(b) How much would it cost on a day when the crane was used for 14 hours and 35
minutes?
(c) What is the recursive rule for the pattern?
(d) What is the explicit rule for the pattern?
(e) What is the value of n if Tn = 2920?
How are you going to explain the pattern to the rest of the class? Who will do the talking?
Task 2 for Group 6: Trapezoidal Tables

Suppose you could seat 5 people comfortably at a table shaped like a trapezoid.
If you joined 2 trapezoid tables back-to-back, 8 people can seat at the two tables.
(a) How many people could be comfortably if you joined 3 trapezoid tables (in a straight
line)? Now complete this table, in which Sn represents the number of people who
could be comfortably seated if you joined n trapezoid tables (in a straight line).
Number of Tables Number of People

n Seated Sn
1 5
2 8
3 ?
4 ?
… …
? 41
…
n ?
(d) What is the value of n if Sn =110?
How are you going to explain the pattern to the rest of the class? Who will do the talking?

group)
adult.
crossings.
(Cn)
1 ?
2 ?
3 ?
… …
8 ?
9 ?
…
? 121
…
n ?


Then, find the missing values, and answer question (b) below.
1 0
2 3
3 8
4 15
… …
10 ?
… …
n ?

Appendix G
Classroom Observation Schedule
Teacher:________________________ Date:___________
Observer:_______________________ Topic:__________________
1. Any resources (e.g., handouts) used by the teacher and by the students?
2. Number of students present?
3. Name(s) of any absent student(s) (if known)?
4. Were there any noticeable differences between this lesson and the corresponding model
lesson (with eighth-grade students)? Comment.
5. What were interesting questions asked publicly by the teacher?
287
288 Appendix G: Classroom Observation Schedule
6. What were interesting comments/questions made by a student?
7. Comment on how well groups of students worked together.
8. Did the teacher talk to the whole class about the algebra, and if so what did he say?
9. Were there any especially memorable episodes in the lesson? (Details)
10. Did most students seem to engage and learn the algebra well? (Comments)
11. Observer’s overview of the workshop:

Appendix H
Pre-Teaching to Post-Teaching “Growth” with the Five Basic Cognitive Structure Components
The tables in this Appendix include ordered pairs which are intended to indicate the
“extent of evidence” for the presence of a component in cognitive structure at the pre- and
post-teaching stages, for each of the 28 interviewees (13 from Group 1, 15 from Group 2).
For a particular component, the extent of evidence was assessed on a three-point scale: 0 =
no evidence, 1 = some evidence, and 2 = strong evidence. The first coordinate of an ordered
pair indicates the extent of the evidence for the presence of the component at the pre-teaching
stage, and the second coordinate indicates the extent of evidence at the post-teaching stage.
Thus, for example, (0, 2) would indicate that there was no evidence of presence of that
component at the pre-teaching stage, but strong evidence at the post-teaching stage.
An indicator of pre-post-teaching “overall growth” is also given, for each student, in the
columns on the right of the tables. For example, an ordered pair, (0, 2) is taken to indicate
a “growth” of 2 (because 2 – 0 = 2) for a particular component and the “overall growth,” for
a student, is the sum of the growths for the five separate components.
Note that the use of the superscript “++” indicates “significant growth,” from a
cognitive-structure perspective, during the intervention period; and that the use of the
superscript “+” indicates “modest growth.” The term “significant growth” corresponds to an
overall growth of at least 7, and “modest growth” corresponds to a growth ranging from 3
through 6.
289
290 Appendix H: Pre-/Post-Teaching Qualitative Growth
DISTRIBUTIVE PROPERTY, COGNITIVE STRUCTURE GROWTH,

28 INTERVIEWEES
Table H1
Summary of Pre-Teaching to Post-Teaching “Growth” for the Five Basic Cognitive
Structure Components, with Respect to the Distributive Property for Multiplication
Over Addition, of 13 Group 1 and 15 Group 2 Interviewees
Verbal Skills Imagery Episodes Attitudes “Growth”
Student 1.1 (0, 2) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 6+

Student 1.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.4 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 1.5 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.7 (0, 2) (0, 2) (0, 2) (1, 2) (0, 2) 1 → 10++
Student 1.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.12 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.13 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.15 (0, 0) (0, 0) (0, 0)) (0, 0) (0, 0) 0→0
Student 1.16 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.1 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.4 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.5 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.7 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.8 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 2.9 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.10 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.11 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 2.12 (0, 2) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.13 (0, 2) (0, 0) (0, 0) (0, 0) (0, 0) 0→2
Student 2.14 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.15 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Twenty of the 28 interviewees experienced significant growth, and three (Students

1.1, 2.12 and 2.14) experienced modest growth. For five students (Students 1.12.
1.15, 2.4, 2.10 and 2.13), however, there was no noticeable growth.
Appendix H: Pre-/Post-Teaching Qualitative Growth 291
ASSOCIATIVE PROPERTY FOR ADDITION,

COGNITIVE STRUCTURE GROWTH, 28 INTERVIEWEES
Table H2
Structure Components, with Respect to the Associative Property for Addition, of 13
Group 1 and 15 Group 2 Interviewees
Student 1.1 (0, 2) (0,2) (0, 2) (0, 2) (0, 2) 0 → 10++

Student 1.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.4 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.5 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.7 (0, 1) (0, 1) (0, 1) (0,1) (0, 1) 0 → 5+
Student 1.8 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 1.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.12 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.13 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 1.15 (0, 0) (0, 0) (0, 0)) (0, 0) (0, 0) 0→0
Student 1.16 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.1 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.4 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.5 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.7 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.9 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.10 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.12 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.13 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.14 (0, 1) (0, 1) (0, 1) (0, 2) (0, 2) 0 → 7+
Student 2.15 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Twenty of the 28 interviewees experienced significant growth, and three (Students

1.7, 2.5 and 2.14) experienced modest growth. For five students (Students 1.12.
1.15, 2.4, 2.10 and 2.13), however, there was no noticeable growth.
ASSOCIATIVE PROPERTY FOR MULTIPLICATION,

Table H3
Structure Components, with Respect to the Associative Property for Multiplication,
of 13 Group 1 and 15 Group 2 Interviewees
Student 1.1 (0, 1) (0, 0) (0, 0) (0, 0) (0, 0) 0→1

Student 1.2 (0, 2) (0, 1) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 1.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.4 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.5 (0, 2) (0, 1) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 1.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.7 (0, 1) (0, 1) (0, 2) (0,2) (0, 2) 0 → 8++
Student 1.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.12 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.13 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.15 (0, 1) (0, 0) (0, 0)) (0, 0) (0, 0) 0→1
Student 1.16 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++

Student 2.1 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.2 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 9++
Student 2.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.4 (0, 1) (0, 0) (0, 0) (0, 0) (0, 0) 0→1
Student 2.5 (0, 1) (0, 0) (0, 0) (0, 0) (0, 0) 0→1
Student 2.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.7 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.9 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.10 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.11 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.12 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.13 (0, 1) (0, 0) (0, 0) (0, 0) (0, 0) 0→1
Student 2.14 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.15 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Eighteen of the 28 interviewees experienced significant growth, and three (Students

2.11, 2.12, and 2.14) experienced modest growth. For seven students (Students 1.1,
1.12. 1.15, 2.4, 2.5, 2.10 and 2.13), however, there was no noticeable growth.
Appendix H: Pre-/Post-Teaching Qualitative Growth 293
SUBSCRIPT NOTATION AND VARIABLE,

Table H4
Structure Components, with Respect to Subscript Notation and Variable (for Linear
Sequences), of 13 Group 1 and 15 Group 2 Interviewees
Student 1.1 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++

Student 1.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.4 (0, 2) (1, 2) (0, 2) (1, 2) (1, 2) 3 → 10++
Student 1.5 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 1.6 (0, 2) (1, 2) (0, 2) (1, 2) (1, 2) 3 → 10++
Student 1.7 (0, 2) (1, 2) (1, 2) (0,2) (1, 2) 3 → 10++
Student 1.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.12 (0, 0) (1, 1) (1, 1) (0, 0) (1, 1) 3→3
Student 1.13 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.15 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.16 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++

Student 2.1 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.2 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 10 → 10++
Student 2.3 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.4 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.5 (2, 2) (2, 2) (2, 2) (2, 2) (2, 2) 10 → 10
Student 2.6 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.7 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.8 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.9 (0, 2) (1, 2) (1, 2) (0, 2) (1, 2) 3 → 10++
Student 2.10 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.11 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.12 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.13 (0, 0) (0, 1) (0, 1) (0, 0) (0,1) 0→3
Student 2.14 (0, 0) (1, 1) (1, 1) (0, 0) (1, 1) 3→3
Student 2.15 (2, 2) (2, 2) (2, 2) (2, 2) (2, 2) 10 → 10
Nineteen of the 28 interviewees experienced significant growth. For nine students

(Students 1.12. 1.15, 2.4, 2.5, 2.10, 2.11, 2.13, 2.14, and 2.15), however, there was
no noticeable growth.
RECURSIVE AND EXPLICIT RULES, 3 DOTS (…),

Table H5
Structure Components, with Respect to Recursive and Explicit Rules, and 3 Dots
(…) for a Linear Sequence of 13 Group 1 and 15 Group 2 Interviewees
Student 1.1 (0, 1) (0, 1) (0, 2) (0, 2) (0, 2) 0 → 8++

Student 1.2 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.3 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.4 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.5 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.6 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.7 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0→ 10++
Student 1.8 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.11 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.12 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.13 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 1.15 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 1.16 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++

Student 2.1 (0,1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5++
Student 2.2 (0, 1) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.3 (1, 2) (1, 2) (1, 2) (1, 2) (1, 2) 5 → 10+
Student 2.4 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.5 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.6 (0, 1) (0, 1) (0, 1) (0, 1) (1, 1) 0 → 5+
Student 2.7 (1, 2) (1, 2) (1, 2) (1, 2) (1, 2) 5 → 10++
Student 2.8 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.9 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Student 2.10 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0 → 10++
Student 2.11 (0, 1) (0, 1) (0, 1) (0, 1) (0, 1) 0 → 5+
Student 2.12 (0, 0) (0, 0) (0, 0) (0, 0) (0, 0) 0→0
Student 2.13 (0, 1) (0, 1) (0, 1) (0, 1) (0,1) 0 → 5+
Student 2.14 (0, 1 ) (0, 1) (0, 1) (0, 0) (0, 1) 0 → 5+
Student 2.15 (0, 2) (0, 2) (0, 2) (0, 2) (0, 2) 0 → 10++
Seventeen of the 28 interviewees experienced significant growth, and seven (Students

2.1, 2.3 2.6, 2.10, 2.11, 2.13, and 2.14) experienced modest growth. For four students
(Students 1.8, 1.12. 1.15, and 2.12), however, there was no noticeable growth.
Appendix I
Generalization Categories (after Radford, 2006)
In Chapter 8 of this book, the levels of generalization for elementary algebra put
forward by Luis Radford in 2006 were described. Radford described three levels of
generalization—which he termed “Factual Generalization,” “Contextual Generalization,” and
“Symbolic Generalization” (these are defined in Table 8.8 of Chapter 8). In his dissertation
the first author (Kanbir) added a fourth level, which was termed “Post-Symbolic
Generalization,” because he thought that that was needed for the purposes of the main study.
Although Radford (2006) called his first, and lowest, level of generalization, “Factual
Generalization,” he also included a “Counting/Arithmetic” category which preceded Factual
Generalization. As the name “Counting/Arithmetic” implies, it was to be applied to learners
who insisted on remaining in the realm of arithmetic, and did not try to generalize. In the
current study, almost all of the seventh-grade students at the pre-teaching stage, would have
been in that category. When tackling questions which could be modeled by linear sequences,
the idea of generalizing did not occur to them.
With Question 7 in the interview protocol (see Appendix A), interviewees were given a
sheet of paper with the following table of values on it, and were then asked what should
replace the question marks beneath the 5, and beneath the n.
First
1 2 3 4 5 … n
Value
Second
3 5 7 9 ? … ?
Value
At the pre-teaching stage all of the interviewees saw this task as purely something
concerning arithmetic. No interviewee showed any inclination to want to give an algebraic
expression like 2n + 1 in the box below the n. For them, the top row of the table of values
was increasing by 1’s, and therefore n had to be 7; and the bottom row was increasing by 2’s,
and therefore 15 should be below the n. The students did not know the meaning of the
convention of the three dots (…). But, at the post-teaching stage, most of these same students
saw the table of values in a different light—almost certainly because they had been
generalizing from tables of values and real-life contexts during the modeling workshops
which were part of the intervention.
The second interview question to be discussed here was Question 8. It provided an
image of a possible real situation (students sitting around tables joined together in the school
cafeteria). However, at the pre-teaching stage, most of the students did not seem to recognize
that the question was inviting them to generalize. Most of their responses were of the
Counting/Arithmetic variety. That had changed by the post-teaching stage, with most
students then being capable of offering contextual, symbolic, and even post-symbolic
generalizations.
The following table, which was developed from an analysis of the interview data,
should enable the reader to “see” the extent of the effects of the modeling workshops on
students’ generalizations.
295
Appendix I: Generalization Categories (after Radford) 297
Table 1
Summary of Pre-Teaching to Post-Teaching Responses to Two Interview Tasks (Questions 7 and 8 on
the Interview Protocol) Inviting Generalizations for Linear Sequence Modeling Tasks
nth term and three dots (…) with a Horizontal 10 Tables Pushed Together?
Interviewee Table of Values Explicit Rule for n Tables Pushed Together?
(n =28) Pre-Teaching Post-Teaching Pre-Teaching Post-Teaching
Student 1.1 Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic
Student 1.2 Counting/Arithmetic Contextual Counting/Arithmetic Contextual
Student 1.3 Counting/Arithmetic Symbolic Counting/Arithmetic Symbolic
Student 1.4 Counting/Arithmetic Post-Symbolic Counting/Arithmetic Post-Symbolic
Student 1.5 Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic Factual
Student 1.6 Counting/Arithmetic Post-Symbolic Counting/Arithmetic Symbolic
Student 2.1 Counting/Arithmetic Contextual Counting/Arithmetic Contextual
Student 2.3 Counting/Arithmetic Post-Symbolic Factual Post-Symbolic
Student 2.4 Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic Contextual
Student 2.5 Counting/Arithmetic Symbolic Counting/Arithmetic Symbolic
Student 2.7 Counting/Arithmetic Post-Symbolic Contextual Post-Symbolic
Student 2.8 Counting/Arithmetic Symbolic Counting/Arithmetic Post-Symbolic
Student 2.9 Counting/Arithmetic Post-Symbolic Counting/Arithmetic Symbolic
Student 2.10 Counting/Arithmetic Factual Counting/Arithmetic Factual
Student 2.11 Counting/Arithmetic Contextual Counting/Arithmetic Symbolic
Student 2.12 Counting/Arithmetic Factual Counting/Arithmetic Factual
Student 2.13 Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic Symbolic
Student 2.14 Counting/Arithmetic Counting/Arithmetic Counting/Arithmetic Factual
Composite Reference List
Adams, J. (1898). Herbartian psychology applied to education: Being a series of essays

applying the psychology of J. F. Herbart. Boston, MA: D. C. Heath.
Albree, J. (2002). Nicolas Pike’s Arithmetic (1788) as the American Liber Abbaci, In D. J.
Curtin, D. E. Kullman, & D. E. Otero (Eds.), Proceedings of the Ninth Midwest History
of Mathematics Conference (pp. 53–71). Miami, FL: Miami University.
Alexander, J. (1709). A synopsis of algebra, being the posthumous work of John Alexander,
of Bern ... To which is added an appendix by Humfrey Ditton ... (translated from the
Latin by Samuel Cobb). London, UK: Christ’s Hospital.
Alexandri, J. (1693). Synopsis algebraica, opus posthumum. London, UK: Christ’s Hospital.
Al-Khalili, J. (2011). The house of wisdom: How Arabic science saved ancient knowledge
and gave us the Renaissance. New York, NY: The Penguin Press.
Allen, F. B., Douglas, E. C., Richmond, D. E., Rickart, C. E., Swain, H., & Walker, R. J.
(1965). First course in algebra: Teacher’s commentary, Part 1. Stanford, CA: School
Mathematics Study Group.
Andrews, A. (1813). Cyphering book (handwritten manuscript). Cheshire, CT: Cheshire
Episcopal Academy.
Artigue, M., Assude, T., Grugeon, B., & Lengant, A. (2001). Teaching and learning algebra:
Approaching complexity through complementary perspectives. In H. Chick, K. Stacey,
J. Vincent, & J. Vincent (Eds.), The future of the teaching and learning of algebra
(Proceedings of the 12th ICMI Study) (pp. 21–32). Melbourne, Australia: The
University of Melbourne.
Ausubel, D. P. (1968). Educational psychology: A cognitive view. New York: Holt, Rinehart
& Winston.
Banchoff, T. (2012). Algebraic thinking and geometric thinking. In C. E. Greenes & R.
Rubinstein (Eds.), Algebra and algebraic thinking in school mathematics (pp. 99–110).
Bartell, T., Wager, A., Edwards, A., Battey, D., Foote, M., & Spencer, J. (2017). Toward a
framework for research linking equitable teaching with the Standards for Mathematical
Practice. Journal for Research in Mathematics Education, 48(1), 7–21. doi:10.5951/
jresematheduc.48.1.0007
Battista, M. T. (2002). Shape makers: A computer environment that engenders students’
construction of geometric ideas and reasoning. Computers in Schools, 17(1/2), 105–120.
Bauersfeld, H. (1995). Language games in the mathematics classroom: Their function and
their effects. In P. Cobb & H. Bauersfeld (Eds.), The emergence of mathematical
meaning: Interaction in classroom cultures (pp. 271–291). Hillsdale, NJ: Lawrence
Erlbaum.
Bedney, G., & Meister, D. (1997). The Russian theory of activity: Current applications to
design and learning. New York, NY: Psychology Press.
Beman, W. W., & Smith, D. E. (1902). Academic algebra. Boston, MA: Ginn.
Bestor, A. (1956). The restoration of learning. New York, NY: Alfred A. Knopf.
Bézout, É. (1794). Cours de mathématiques a l’usage des gardes du pavillon et de la maine.
Paris, France: Baudelot & Eberhart.
Bishop, A. J. (1988). Mathematical enculturation. Dordrecht, The Netherlands: Kluwer.

300 Composite Reference List
Blanton, M. L., & Kaput, J. J. (2011). Functional thinking as a route into algebra in the
elementary grades. In J. Cai & E. Knuth (Eds.), Early algebraization: A global
Blanton, M. L., Stephens, A., Knuth, E., Gardiner, A. M., Isler, I., & Kim, J.-S. (2015). The
46(1), 39–87.
Bolton, F. E. (1900). The secondary school system of Germany. London, UK: Edward
Arnold.
Booth, L. R. (1984). Algebra: Children’s strategies and errors. Windsor, Berks, UK: NFER-
Nelson.
Booth, L. R. (1988). Children’s difficulties in beginning algebra. In A. F. Coxford & A. P.
Schulte (Eds.), The ideas of algebra: K–12, 1988 Yearbook. Reston, VA: National
Borel, E. (1904). Les exercices pratiques de mathématiques dans l’enseignement secondaire.
Revue Générale des Sciences Pures et Appliqués, 15, 431–440.
Borko, H. (2004). Professional development and teacher learning: Mapping the terrain.
Bourdon, L. P. M. (1831). Elements of algebra, by Bourdon, translated from the French for
the use of colleges and schools. Boston, MA: Hilliard, Gray, Little and Wilkins.
Bray, M., & Kwo, O. (2014). Regulating private tutoring for public good: Policy options for
supplementary education in Asia. Hong Kong, China: Education Research Centre
(CERC), UNESCO.
Bray, M., & Lykins, C. (2012). Shadow education: Private supplementary tutoring and its
implications for policy makers in Asia. Manila, The Philippines: Asian Development
Bank.
Bressoud, D. M. (2016). Book review: The Math Myth and Other STEM Delusions. Notices
of the American Mathematical Society, 63(10), 1181–1183.
Britt, M. S., & Irwin, K. C. (2011). Algebraic thinking with and without algebraic
representation: A pathway for learning. In J. Cai & E. Knuth (Eds.), Early
Brock, W. H. (1981). The Japanese connection: Engineering in Tokyo, London, and Glasgow
at the end of the nineteenth century. The British Journal of the History of Science,
14(3), 227–243.
Brock, W. H., & Price, M. H. (1980). Squared paper in the nineteenth century: Instrument of
science and engineering, and symbol of reform in mathematical education. Educational
Studies in Mathematics, 11, 365–381.
Cai, J., & Knuth, E. (2011a). A global dialogue about early algebraization from multiple
Composite Reference List 301
Cai, J., & Knuth, E. (2011b). Preface to Part I. In J. Cai & E. Knuth (Eds.), Early
algebraization: A global dialogue from multiple perspectives (pp. 3–4). Heidelberg,
Germany: Springer.
Cai, J., & Knuth, E. (Eds.). (2011c). Early algebraization: A global dialogue from multiple
Cai, J., Morris, A., Hohensee, C., Hwang, S., Robison, V., & Hiebert, J. (2017). Making
Cai, J., & Moyer, J. (2012). Developing algebraic thinking in earlier grades: Some insights
from international comparative studies. In C. E. Greenes & R. Rubinstein (Eds.),
Algebra and algebraic thinking in school mathematics (pp. 169–180). Reston, VA:
Cai, J., Moyer, J. C., Wang, N., & Nie, B. (2011). Examining students’ algebraic thinking in
a curricular context: A longitudinal study. In J. Cai & E. Knuth (Eds.), Early
Publishing Co.
Campbell, D. T., & Stanley, J. C. (1963). Experimental and quasi-experimental designs for
research on teaching. In N. L. Gage (Ed.), Handbook of research on teaching (pp. 171–
246). Washington, DC: American Educational Research Association.
Cañadas, M. C., Brizuela, B. M., & Blanton, M. (2016). Second graders articulating ideas
about co-variation with linear functions. Journal of Mathematical Behavior, 41, 87–103.
Carr, W., & Kemmis, S. (2004). Education, knowledge, and action research. London, UK:
Routledge.
Carraher, D. W., Schliemann, A. D., & Schwartz, J. L. (2008). Early algebra is not the same
as algebra early. In J. J. Kaput, D. W. Carraher, & M. Blanton (Eds.), Algebra in the
early grades (pp. 235–272). New York, NY: Lawrence Erlbaum Associates.
Carslaw, H. S. (1914). The teaching of mathematics in Australia: Report presented to the
International Commission on the Teaching of Mathematics. Sydney, Australia: Angus
& Robertson Ltd.
Castañeda, H-N. (1990). The role of apperception in Kant’s transcendental deduction of the
categories. Nous, 24(1), 147–157.
CCSSM. (2010). National Governors Association Center for Best Practices, & Council of
Chief State School Officers. (2010). Common core state standards for mathematics.
Washington, DC: Authors.
Chambers, C. E. (1964). The Cuisenaire Gattegno method of teaching mathematics. Reading,
UK: Educational Explorers.
Chateauneuf, A. O. (1929). Changes in the content of elementary algebra since the beginning
of the high school movement as revealed by the textbooks of the period (Unpublished
Ph.D. thesis). University of Pennsylvania.
Chorlay, R. (2011). The multiplicity of viewpoints in elementary function theory: Historical
and didactical perspectives. In V. Katz & C. Tzanakis (Eds.), Recent developments on
introducing a historical dimension in mathematics education (pp. 57–66). Washington,
DC: Mathematical Association of America.
Clairaut, A.-C. (1746). Elémens d’algèbre. Paris, France: Guérin.
Clement, J., Lochhead, J., & Monk, G. S. (1981). Translation difficulties in learning
mathematics. American Mathematical Monthly, 88, 286–290.
Clements, M. A. (1979). Relationships between the University of Melbourne and the
secondary schools of Victoria, 1890–1912 (Unpublished PhD thesis). The University of
Melbourne.
Clements, M. A. (1980). Analysing children’s errors on written mathematical tasks.
Clements, M. A. (1992). Mathematics for the minority: Some historical perspectives on
school mathematics in Victoria. Geelong, Australia: Deakin University.
Clements, M. A. (2003). An outsider’s view of North American school mathematics
curriculum trends. In G. M. A. Stanic & J. Kilpatrick (Eds.), A history of school
mathematics (Vol. 2, pp. 1509−1580). Reston, VA: National Council of Teachers of
Mathematics.
Clements, M. A. (2011). Design research: Its potential for meeting Brunei Darussalam’s
educational needs. In I. Cheong (Ed.), Design research in education: Empowering
teachers, learners and researchers (pp. 22–40). Gadong, Brunei Darussalam: Universiti
Brunei Darussalam.
Clements, M. A. (2014). A historical overview of visualization and visualizing in
mathematics. In M. Fried & T. Dreyfus (Eds.), Mathematics and mathematics
education: Searching for common ground (pp. 177–192). New York, NY: Springer.
Clements, M. A., Bishop, A. J., Keitel, C., Kilpatrick, J., & Leung, F. (Eds.). (2013). Third
international handbook of mathematics education. New York, NY: Springer.
Clements, M. A., & Ellerton, N. F. (1996). Mathematics education research: Past, present
and future. Bangkok, Thailand: UNESCO.
Clements, M. A., & Ellerton, N. F. (2015). Thomas Jefferson and his decimals 1775–1810:
Neglected years in the history of U.S. school mathematics. New York, NY: Springer.
Clements, M. A., Evans, K. M., Green, L., Smith, G., & Watterson, G. (1967). Mathematics
for today and tomorrow. Melbourne, Australia: Sun Books.
Clements, M. A., Grimison, L. A., & Ellerton, N. F. (1989). Colonialism and school
mathematics in Australia 1788–1988. In N. F. Ellerton & M. A. Clements (Eds.),
School mathematics: The challenge to change (pp. 50–78). Geelong, Australia: Deakin
University.
Clements, M. A., Keitel, C., Bishop, A. J., Kilpatrick, J., & Leung, F. (2013). From the few
to the many: Historical perspectives on who should learn mathematics. In M. A.
Clements, A. J. Bishop, C. Keitel, J. Kilpatrick, & F. Leung (Eds.), Third international

handbook of mathematics education (pp. 7–40). New York, NY: Springer.
Cobb, P. (1989). Experiential, cognitive and anthropological perspectives in mathematics
education. For the Learning of Mathematics, 9(2), 32–43.
Cobb, P., Confrey, J., diSessa, A., Lehrer, R., & Schauble, L. (2003). Design experiments in
educational research. Educational Researcher, 32(1), 9–13.
Cobb, P., & Gravemeijer, K. (2008). Experimenting to support and understand learning
processes. In A. E. Kelly, R. Lesh, & J. Baek (Eds.), Handbook of design research
methods in education: Innovations in science, technology, engineering, and
mathematics learning and teaching (pp. 68–95). New York, NY: Routledge.
Cocker, E. (1677). Cocker’s arithmetick: Being a plain and familiar method suitable to the
meanest capacity for the full understanding of that incomparable art, as it is now
taught by the ablest school-masters in city and country. London, UK: John Hawkins.
Colburn, W. (1825). An introduction to algebra upon the inductive method of instruction.
Boston, MA: Hilliard, Gray & Co.
Cole, P. R. (1912). Johann Friedrich Herbart. In P. Monroe (Ed.), A cyclopaedia of education
(Vol. 3, pp. 250– 254). New York, NY: Macmillan.
Colebrooke, H. T. (1817/2013). Algebra with arithmetic and mensuration from the Sanscrit
of Brahmegupta and Bhascara (translated by H. T. Colebrooke). New York, NY:
Cambridge University Press.
Common Core Writing Team. (2011, May 29). Progression for the Common Core State
Standards in Mathematics (draft). Comment at commoncoretools.wordpress.com. 1.
Coventry, A., & Kriegel, U. (1989). Locke on consciousness. History of Philosophy
Quarterly, 25(3), 221–242.
Cremin, L. A. (1956). The problem of curriculum-making: An historical perspective. In
Association of Supervision and Curriculum Development (Ed.), What shall high
schools teach? (pp. 6–26). Washington, DC: Association of Supervision and
Curriculum Development.
Cremin, L. A. (1964). The transformation of the school: Progressivism in American
education 1876–1957. New York, NY: Vintage Books.
da Ponte, J. P., & Guimarāes, H. M. (2014). Notes for a history of the teaching of algebra. In
A. Karp & G. Schubring (Eds.), Handbook on the history of mathematics education
(pp. 459–472). New York, NY: Springer.
Dascal, M. (1987). Leibniz: Language, signs, and thought. Philadelphia, PA: John Benjamins.
Davies, C. (1837). Elements of algebra translated from the French of M. Bourdon. New
York, NY: Wiley & Long.
Davies, C. (1846). Elementary algebra: Embracing the first principles of the science. New
York, NY: A. S. Barnes and Co.
Wesley.
Day, J. (1814). An introduction to algebra, being the first part of a course in mathematics,
adopted to the method of instruction in the American colleges. New Haven, CT: Howe
& Deforest.
DeGarmo, C. (1895). Herbart and the Herbartians. New York, NY: Charles Scribner’s Sons.
DeGarmo, C. (Ed.). (1900). Apperception: A monograph on psychology and pedagogy.

Boston, MA: D. C. Heath & Co.
de L’Hôpital, G. F. A. (1696). Analyse des infiniment petits, pour l’intelligence des lignes
courbes. Paris, France: L’Imprimerie Royale.
de Saussure, F. (1959). Course in general linguistics. New York, NY: The Philosophical
Library.
de Saussure, F. (1974). Cours de linguistique générale (C. Bally & A. Sechehaye, Eds.).
Paris, France: Payot.
Didis, M. G., & Erbas, A. K. (2015). Performance and difficulties of students in formulating
and solving quadratic equations with one unknown. Educational Sciences: Theory &
Practice, 15(4), DOI: 10.12738/estp.2015.4.2743.
Dienes, Z. P. (1960). Building up mathematics. London, UK: Hutchinson & Co.
Dienes, Z. P. (1967). Groups and co-ordinates. Paris, France: Hutchinson Educational.
Dienes, Z. P. (1968). Powers, roots and logarithms. New York, NY: Herder and Herder.
Dilworth, T. (1762). The schoolmaster’s assistant. Being a compendium of arithmetic, both
practical and theoretical (11th ed.). London, UK: Henry Kent.
Ding, M. (2016a). Developing preservice elementary teachers’ specialized content
knowledge for teaching fundamental mathematical ideas: The case of the associative
property. International Journal of STEM Education, 3(9), 1–19.
Ding, M. (2016b). Opportunities to learn: Inverse operations in U.S. and Chinese elementary
mathematics textbooks. Mathematical Thinking and Learning, 18(1), 45–68.
87, 103–121.
Ditton, H. (1709). Appendix to the algebra of John Alexander. London, UK: Christ’s
Hospital.
Dreyfus, T., & Eisenberg, T. (1982). Intuitive functional concepts: A baseline study on
intuitions. Journal for Research in Mathematics Education, 13, 360–380.
Durell, F., & Robbins, E. R. (1901). Grammar school algebra. Harrisburg, PA: R. L. Myers
& Co.
Education Department of Victoria and Australian Council for Educational Research. (1966).
Background in mathematics: A guidebook to elementary mathematics for teachers in
primary schools. Melbourne, Australia: Authors.
Edwards, E. A. (Ed.). (1990). Algebra for everyone. Reston, VA: National Council of
Teachers of Mathematics.

theory and practice (pp. 313–320). Sydney, Australia: Mathematics Education
Research Group of Australasia.
Ellerton, N. F., & Clements, M. A. (2012). Rewriting the history of school mathematics in
North America 1607–1861. New York, NY: Springer.
Ellerton, N. F., & Clements, M. A. (2014). Abraham Lincoln’s cyphering book and ten other
extraordinary cyphering books. New York, NY: Springer.
Ellerton, N. F., & Clements, M. A. (2017). Samuel Pepys, Isaac Newton, James Hodgson and
the beginnings of secondary school mathematics: A History of the Royal Mathematical
School within Christ’s Hospital, London 1673–1868. New York, NY: Springer.
Ellis, A. B. (2011). Algebra in the middle school: Developing functional relationships through
quantitative reasoning. In J. Cai & E. Knuth (Eds.), Early algebraization: A global
Exner, R. M., & Rosskopf, M. F. (1970). Proof. In M. Rosskopf (Ed.), The teaching of
secondary school mathematics: Thirty-third yearbook (pp. 196–240). Washington, DC:
Filloy, E., & Sutherland, R. (1996). Designing curricula for teaching and learning algebra. In
A. J. Bishop, M. A. Clements, C. Keitel, J. Kilpatrick, & C. Laborde (Eds.),
International handbook of mathematics education (pp. 139–160). Dordrecht, The
Netherlands: Kluwer Academic Publishers.
Freudenthal, H. (Ed.). (1962a). Report on the relations between arithmetic and algebra in
mathematical education up to the age of 15 years. Groningen, The Netherlands: J. B.
Wolters.
Freudenthal, H. (1962b). Logical analysis and critical survey. In H. Freudenthal (Ed.), Report
on the relations between arithmetic and algebra in mathematical education up to the
age of 15 years (pp. 20–41). Groningen, The Netherlands: J. B. Wolters.
Fried, M. (2008). History of mathematics in mathematics education: A Saussurean
perspective. The Montana Mathematics Enthusiast, 5(2–3), 185–198.
Fullan, M., & Pomfret, A. (1977). Research on curriculum and instruction implementation.
Review of Educational Research, 47(2), 335–397.
Winston.
Gavin, M. K., & Sheffield, L. J. (2015). A balancing act: Making sense of algebra.
Mathematics Teaching in the Middle School, 20(8), 460–466.
Giacardi, L. (2009). The school as “laboratory”: Giovanni Vailati and the project to reform
mathematics teaching in Italy. International Journal for the History of Mathematics
Education, 4(1), 5–28.
Godfrey, C. (1913). The teaching of calculus. Mathematical Gazette, 7, 233–240.
Grimison, L. A. (1988, July). The introduction of calculus into the secondary mathematics
curriculum—How, when and why? Paper presented to the annual conference of the
Mathematics Education Research Group of Australasia, Geelong, Australia.
Grinder, R. E. (1989). Educational psychology: The master science. In M. C. Wittrock & F.
Farley (Eds.), The future of educational psychology (pp. 3–18). Hillsdale, NJ: Lawrence
Erlbaum Associates.
Grinstein, L. S., & Michaels, B. (1977). Calculus: Readings from the Mathematics Teacher.
Hacker, D. M. (2016). The math myth and other STEM delusions. New York, NY: The New
Press.
Murray.
Henderson, E. N. (1911). Apperception. In P. Monroe (Ed.), A cyclopaedia of education
Herbart, J. F. (1898). Letters and lectures on education (H. M. Felkin & E. Felkin, Trans.).
London, UK: Swan Sonnenschein & Co.
Herbel-Eisenmann, B., Choppin, J., Wagner, D., & Pimm, D. (Eds.). (2012). Equity in
discourse for mathematics education: Theories, practices, and policies. Dordrecht, The
Netherlands: Springer.
Hertel, J. T. (2016). Investigating the implemented curriculum of navigation cyphering
books. For the Learning of Mathematics, 36(3), 4–10.
Hewitt, D. (1998). Approaching arithmetic algebraically. Mathematics Teaching, 163, 19–29.
Hiebert, J., & Carpenter, T. P. (1992). Learning and teaching with understanding. In D. A.
Grouws (Ed.), Handbook of research on mathematics teaching and learning (pp. 65–
97). Reston, VA: National Council of Teachers of Mathematics.
Hill, H. C., Ball, D. L., & Schilling, S. G. (2008). Unpacking pedagogical content
knowledge: Conceptualizing and measuring teachers’ topic-specific knowledge of
students. Journal for Research in Mathematics Education, 39(4), 372–400.
education research and design. In A. E. Kelly, R. A. Lesh, & J. Y. Baek (Eds.),
Handbook of design research methods in education: Innovations in science,
technology, engineering and mathematics learning and teaching (pp. 96–110). New
York, NY: Routledge.
Hodgson, J. (1723). A system of the mathematics. London, UK: Thomas Page.
House, P. (1988). Reshaping school algebra: Why and how? In A. F. Coxford (Ed.), The
ideas of algebra, K–12 (pp. 1–7). Reston, VA: National Council of Teachers of
Mathematics.
Husky, M. E. (2011). David Eugene Smith and the algebra curriculum. Unpublished M.Sc.
thesis, Illinois State University.
Jahnke, H. N. (1983). Origins of school mathematics in early nineteenth century Germany.
Journal of Curriculum Studies, 18(1), 85–94.
Jeon, K. (2012). Reflecting on PEMDAS. Teaching Children Mathematics, 18(6), 370–377.
Jones, P. S., & A. F. Coxford (1970). First steps toward “revision,” 1894–1920.In P. S. Jones
& A. F. Coxford (Eds.), A history of mathematics education in the United States and
Canada (pp. 36–45). Washington, DC: National Council of Teachers of Mathematics.
Kanbir, S. (2016a, April 12). Three different approaches to middle-school algebra: Results of
Kanbir, S. (2016b). An intervention study aimed at enhancing seventh-grade students’
Kaput, J. (2008). What is algebra? What is algebraic reasoning? In J. Kaput, D. Carraher, &
M. Blanton (Eds.), Algebra in the early grades (pp. 1–18). Mahwah, NJ: Lawrence
Erlbaum/Taylor & Francis Group, & National Council of Teachers of Mathematics.
Karp, A. (2007). Interview with Henry Pollak. The International Journal for the History of
Karp, A., & Schubring, G. (Eds.). (2014). Handbook on the history of mathematics
education. New York, NY: Springer.
Karpinski, L. C. (1980). Bibliography of mathematical works printed in America through
1850. New York, NY: Arno Press.
Keitel, C. (2006). Perceptions of mathematics and mathematics education in the course of
history—A review of Western perspectives. In F. K. S. Leung, K-D Graf, & F. Lopez-
Real (Eds.), Mathematics education in different cultural traditions—A comparative
study of East Asia and the West: The 13th ICMI Study (pp. 81–94). New York, NY:
Springer.
Kiang, L. Y. (2012). Mathematics K–12: Interview with Wu Hung-Hsi. Mathematical
Medley, 38(1), 2–15.
Kieran, C. (2006). Research on the learning and teaching of algebra. In A. Gutiérrez & P.
Boero (Eds.), Handbook of research on the psychology of mathematics education: Past,
present and future (pp. 11–49). Rotterdam, The Netherlands: Sense Publishers.
Kieran, C. (2007). Learning and teaching algebra at the middle school through college levels:
Building meaning for symbols and their manipulation. In F. K. Lester (Ed.), Second
handbook of research on mathematics teaching and learning (pp. 707–762). Charlotte,
NC: Information Age Publishing; and Reston, VA: National Council of Teachers of
Mathematics.
Klein, F. (1872). Vergleichende betrachtungen über neuere geometrische forschungen
(“comparative review of recent researches in geometry”), which was reprinted in 1893
in Mathematische Annalen, 43, 63–100. An English translation by Mellen Haskell
appeared in 1893 in the Bulletin of the New York Mathematical Society, 2, 215–249.
Klein, F. (1949). Elementary mathematics from an advanced standpoint: Arithmetic, algebra,
analysis (translated from the German by E. R. Hedrick & C. A. Noble). New York, NY:
Dover Publications.
Klein, F. (2004). Elementary mathematics from an advanced standpoint: geometry
(translated from the German by E. R. Hedrick & C. A. Noble). New York, NY: Dover
Publications.
Kleiner, I. (1986), The evolution of group theory: A brief survey. Mathematics Magazine,
59(4), 195–215.
Kline, M. (1973). Why Johnny can’t add: The failure of the new math. New York, NY: St.
Martin’s Press.
Knuth, E. Stephens, A., Blanton, M., & Gardiner, A. (2016, March). Build an early
foundation for algebra success. Kappanmagazine.org, 97(6), 65–68.
Küchemann, D. E. (1981). Algebra. In K. M. Hart (Ed.), Children’s understanding of
mathematics: 11−16 (pp. 102–119). London, UK: John Murray.
Laborde, C. (2001). Integration of technology in the design of geometry with Cabri-
Geometry. International Journal of Computers for Mathematical Learning, 6, 283–317.
Lacroix, S. F. (1815). Complément des elémens d’algébre, à l’usage de l’école centrale des
quatre-nations (3rd ed.). Paris, France: Chez Courcier.
Lange, K. (1896). Apperception: A monograph on psychology and pedagogy. Boston, MA:
D. C. Heath & Co.
Lannin, J. K., Barker, D. D., & Townsend, B. E. (2006). Recursive and explicit rules: How
can we build student algebraic understanding? The Journal of Mathematical Behavior,
25, 299−317.
Laplace, P.-S. (1799). Traité de mécanique céleste. Paris, France: J. B. M. Duprat.
Lave, J., & Wenger, E. (1991). Situated learning. Legitimate peripheral participation.
Cambridge, UK: Cambridge University Press.
Lawson, D. R. (1990). The problem, the issues that speak to change. In E. L. Edwards (Ed.),
Algebra for everyone (pp. 1–6). Reston, VA: National Council of Teachers of
Mathematics.
Layton, E. T. (1932). The persistence of learning in elementary algebra. Journal of
Educational Psychology, 23(1), 46–55.
affecting children’s understanding of arithmetic word problems. Educational Studies in
Mathematics, 21(2), 165–191.
Lesh, R., & Sriraman, B. (2005). Mathematics education as a design science. ZDM—The
International Journal on Mathematics Education, 37(6), 490–505.
Leung, F. K. S., Park, K., Holton, D., & Clarke, D. (Eds.). (2014). Algebra teaching around
the world. Rotterdam, The Netherlands: Sense Publishers.
level Mathematics: A case study (Unpublished M.Ed. dissertation). Universiti Brunei
Darussalam.
Lochhead, J., & Mestre, J. P. (1988). From words to algebra: Mending misconceptions. In A.
F. Coxford & A. P. Schulte (Eds.), The ideas of algebra, K–12, 1988 yearbook (pp.
127–135). Reston, VA: National Council of Teachers of Mathematics.
Loemker, L. (Ed.). (1969). Leibniz: Philosophical papers and letters. Dordrecht, The
Netherlands: D. Reidel.
López, J., Robles, I., & X Martínez-Planel, R. (2015). Students’ understanding of quadratic
equations. International Journal of Mathematical Education in Science and
Technology. doi:10.1080/0020739X.2015.119895
Lüken, M. M. (2012). Young children’s structure sense. Journal für Mathematik-
Didaktik, 33, 263–285.
MacGregor, M. E. (1991). Making sense of algebra: Cognitive processes influencing
MacGregor, M., & Stacey, K. (1997). Students’ understanding of algebraic notation.
Mackenzie, A. A., & White, R. T. (1982). Field-work in geography and long-term memory
structures. American Educational Research Journal, 19(4), 623–632.
May, K. O., & Van Engen, H. (1959). Relations and functions. In H. P. Fawcett, A. M. Hach,
C. W. Junge, H. W. Syer, H. Van Engen, & P. S. Jones (Eds.), The growth of
mathematical ideas, grades K–12: Twenty-fourth yearbook (pp. 65–110). Washington,
DC: National Council of Teachers of Mathematics.
McMurry, C. A., & McMurry, F. (1897). The method of recitation. New York, NY: Macmillan.
McQualter, J. W. (1975). Some influences on mathematics curriculum making: A historical
review. Australian Mathematics Teacher, 30(1), 15–26.
Meserve, B. E., & Sobel, M. A. (1962). Mathematics for secondary school teachers.
Englewood Cliffs, NJ: Prentice-Hall.
Milgram, R. J. (2013). Common Core’s effect on math. Retrieved from
https://www.youtube.com/watch?v=MuexIMWBOQY
Monroe, W. S. (1917). Development of arithmetic as a school subject. Washington, DC:
Government Printing Office.
Moore, E. H. (1903). On the foundations of mathematics. Science, 17(428), 401–416.
Moore, E. H. (1906). The cross-section paper as a mathematical instrument. School Review,
14, 317–338.
Moss, J., & McNab, S. L. (2011). An approach to geometric and numeric patterning that
fosters second grade students’ reasoning and generalizing about functions and co-
variation. In J. Cai & E. Knuth (Eds.), Early algebraization: A global dialogue from
Mosteller, F., & Boruch, R. (Eds.). (2002). Evidence matters: Randomized trials in education
research. Washington, DC: Brookings Institution Press.
Mullis, I. V. S., Martin, M. O., Foy, P., & Arora, A. (2012). TIMSS 2011 international
results in mathematics. Chestnut Hill, MA: TIMSS & PIRLS International Study
Center, Boston College.
National Council of Teachers of Mathematics. (2006). Curriculum focal points for
Prekindergarten through Grade 8 mathematics: A quest for coherence. Reston, VA:
Author.
Authors. (Also cited as CCSSM (2010)
National Mathematics Advisory Panel. (2008). Foundations for success: The final report of
the Mathematics Advisory Panel. Washington, DC: U.S. Department of Education.
Newman, M. A. (1977). An analysis of sixth-grade pupils’ errors on written mathematical
tasks. Victorian Institute for Educational Research Bulletin, 39, 31–43.
Newton, I. (1739). Philosophiæ naturalis principia mathematica (Vol. 1). Geneva,
Switzerland: Barrillot & Filii.
Nordgaard, M. A. (1928). Introductory calculus as a high school subject. In J. R. Clark & W.
D. Reeve (Eds.), Selected topics in the teaching of mathematics (Third Yearbook of the
National Council of Teachers of Mathematics) (pp. 65–101). New York, NY: Teachers
College, Columbia University.
Olney, E. (1885). First principles of algebra: An introduction to the author’s complete, or
university algebra. New York, NY: Sheldon & Company.
Otte, M. (1998). Limits of constructivism: Kant, Piaget and Peirce. Science and Education,
7(5), 425–450.
Otte, M. F. (2011). Evolution, learning and semiotics from a Peircean point of view.
Owens, K. D., Lean, G. A., Paraide, P., & Muke, C. (2017). History of number: Evidence
from Papua New Guinea and Oceania. New York, NY: Springer.
Oxford and Cambridge Schools Examination Board. (1899). Oxford-Cambridge Schools
Agreement, 1899. Cambridge, UK, Author.
Papert, S. (1980). Mindstorms: Children, computers, and powerful ideas. New York, NY:
Basic Books.
Pappano, L. (2012). The algebra problem. Harvard Education Letter, 28(3), 1–2.
Peirce, C. S. (1931/1958). Collected papers (Vol. 1–VIII). In C. Hartshorne, P. Weiss, & A.
Burks (Eds.), Cambridge, MA: Harvard University Press.
Press.
Perel, W. M., & Vairo, P. D. (1967). New mathematics and old teachers. The Educational
Forum, 31(3), 345–348.
Perry, J. (1899). Practical mathematics. London, UK: Department of Science and Art of the
Committee of the Council on Education.
Perry, J. (Ed.). (1902). Discussion on the teaching of mathematics. London, UK: Macmillan.
Perry, J. (1912). Practical mathematics. Nature, 90, 34–35.
Phillips, C. J. (2014). The new math: A political history. Chicago, IL: The University of
Chicago Press.
Piaget, J. (1970). Science of education and the psychology of the child. New York, NY:
Viking Press.
Pike, N. (1788). The new complete system of arithmetic, composed for the use of the citizens
of the United States. Newburyport, MA: John Mycall.
Pike, N. (1793). The new and complete system of arithmetic composed for the use of the
citizens of the United States (abridged for the use of schools). Newbury-Port, MA: John
Mycall, Isaiah Thomas.
Price, M. H. (1986). The Perry movement in school mathematics. In M. H. Price (Ed.), The
development of the secondary curriculum (pp. 103–155). London, UK: Croom Helm.
Price, M. H. (1994). Mathematics for the multitude? Leicester, UK: Mathematical Association.
Radford, L. (2003). Gestures, speech, and the sprouting of signs. Mathematical Thinking and
Learning, 5(1), 37–70.
Radford, L. (2004). Syntax and meaning. In M. J. Høines & A. B. Fuglestad (Eds.),
Proceedings of the 26th Conference of the International Group for the Psychology of
Mathematics Education (Vol. 1, pp. 161–166). Bergen, Norway: International Group
for the Psychology of Mathematics Education.
possibilities. ZDM—The International Journal on Mathematics Education, 40, 317–327.
Radford, L. (2011). Grade 2 students’ non-symbolic algebraic thinking. In J. Cai & E. Knuth
(Eds.), Early algebraization: A global dialogue from multiple perspectives (pp. 303–
322). Heidelberg, Germany: Springer.
Radford, L. (2015). The epistemological foundations of the theory of objectification.
Isonomia (online journal http://isonomia.uniurb.it/epistemologica), 127–149.
Raub, A. N. (1888). Methods of teaching: Including the nature, object, and laws of education,
methods of instruction, and methods of culture. Philadelphia, PA: E. L. Raub & Co.
Ray, J. (1848). Ray’s algebra, Part First: On the analytic and inductive methods of
instruction. Cincinnati, OH: Winthrop B. Smith & Co.
Rech, J. F., & Harrington, J. (2000). Algebra as a gatekeeper: A descriptive study at an urban
university. Journal of African American Studies, 4(4), 63–71.
Rein, W. (1893). Outlines of pedagogics. London, UK: Sonnenschein.
Rittle-Johnson, B., Loehr, A. M., & Durkin, K. (2017). Promoting self-explanation to improve
mathematics learning: A meta-analysis and instructional design principles. ZDM
Mathematics Education doi:10.1007/s11858-017-0834-z.
Rivera, F. D. (2013). Teaching and learning patterns in school mathematics: Psychological
and pedagogical considerations. Dordrecht, The Netherlands: Springer.
Roberts, D. L. (2012). American mathematicians as educators, 1893–1923. Boston, MA:

Docent Press.
Rosnow, R. L., & Rosenthal, R. (1996). Computing contrasts, effect sizes, and counternulls
on other people’s published data: General procedures for research consumers.
Psychological Methods, 1, 331–340.
Rosskopf, M. F. (1970). Mathematics education: Historical perspectives. In M. Rosskopf
(Ed.), The teaching of secondary school mathematics: Thirty-third yearbook (pp. 3–29).
Washington, DC: National Council of Teachers of Mathematics.
Ruthven, K. (2008). Mathematical technologies as a vehicle for intuition and experiment: A
foundational theme of the International Commission on Mathematical Instruction, and a
continuing preoccupation. International Journal for the History of Mathematics
Education, 3(2), 91–102.
Schoenfeld, A. H. (2006). What doesn’t work: The challenge and failure of the What Works
Clearinghouse to conduct meaningful reviews of studies of mathematics curricula.
Schoenfeld, A. H. (2013). On forests, trees, elephants and classrooms: A brief for the study
of learning ecologies. ZDM—The International Journal on Mathematics Education,
45(3), 491–495.
Schorling, R. (1926). Suggestions for the solution of an important problem that has arisen in
the last quarter of a century. In C. M. Austin, H. English, W. Betz, W. C. Eells, & F. C.
(Republished, 1995).
Schubring, G. (2011). On how n and i turned out to become indices in mathematical
sequences and formulae. Educational Studies in Mathematics, 77(2–3), 331–350.
Schultze, A. (1909). Graphic algebra. New York, NY: The Macmillan Company.
Selleck, R. J. W. (1968). The new education 1870–1914. London, UK: Sir Isaac Pitman and
Sons.
22(1), 1–36.
Sfard, A. (2008). Thinking as communicating. Cambridge, UK: Cambridge University Press.
Shaughnessy, J. M. (2010). Linking research and practice: The research agenda project.
Journal for Research in Mathematics Education, 41(3), 212–215.
Simons, L. G. (1924). Introduction of algebra into American schools in the eighteenth
century. Washington, DC: Government Printing Office.
Simons, L. G. (1936). Bibliography of early American textbooks on algebra published in the
colonies and the United States through 1850, together with a characterization of the
first edition of each work. New York, NY: Scripta Mathematica.
Simpson, T. (1755). A treatise of algebra (2nd ed.). London, UK: John Nourse.
Sinclair, H. (1990). Learning: The interactive recreation of knowledge. In L. P. Steffe & T.
Wood (Eds.), Transforming children’s mathematics education: International
perspectives (pp. 19–29). Hillsdale, NJ: Lawrence Erlbaum.
Singh, P., & Ellerton, N. F. (2013). International collaborative studies in mathematics
education. In M. A. Clements, A. Bishop, C. Keitel, J. Kilpatrick, & F. Leung (Eds.),
Third international handbook of mathematics education (pp. 806–839). New York, NY:
Springer.
Slavin, R. E. (2004). Educational research can and must address “what works” questions.
Educational Researcher, 33(1), 27–28.
Smith, D. E. (1905). Movements in mathematical teaching. School Science and Mathematics,
5(3), 205–209.
Smith, D. E. (1913). Laboratory methods in mathematics. In P. Monroe (Ed.), A cyclopaedia
of education (Vol. 4, pp. 159–160). New York, NY: Macmillan.
Smith, D. E. (1926). A general survey of the progress of mathematics in our high schools in
the last twenty-five years. In C. M. Austin, H. English, W. Betz, W. C. Eells, & F. C.
(Republished, 1995)
Snyder, T. D. (1993). 120 years of American education: A statistical portrait. National
Center for Educational Statistics, Washington, DC: U.S. Government Printing Office.
Snyder, T. D., & Dillow, S. A. (2011). Digest of education statistics 2010. Washington, DC:
National Center for Educational Statistics, U.S. Government Printing Office.
Spielhagen, F. R. (2011). The algebra solution to mathematics reform (completing the
equation). New York, NY: Teachers College Press.
Sriraman, B. (2009). What’s all the commotion about commognition? The Montana
Mathematics Enthusiast, 6(3), 541–544.
Sriraman, B., & Lee, K.-H. (2011). Commentary on Part II. In J. Cai & E. Knuth (Eds.),
Early algebraization: A global dialogue from multiple perspectives (pp. 367–373).
Stacey, K., & MacGregor, M. (1997). Ideas about symbolism that students bring to algebra.
The Mathematics Teacher, 90(2), 110–113.
Steen, L. A. (1992). Does everybody need to study algebra? Basic Education, 37(4), 9–13.
Steffe, L. P. (1990). Early childhood years. In L. P. Steffe & T. Wood (Eds.), Transforming
children’s mathematics education: International perspectives (pp. 3–15). Hillsdale, NJ:
Lawrence Erlbaum.
Sterry, C., & Sterry, J. (1790). The American youth: Being a new and complete course of
introductory mathematics, designed for the use of private students. Providence, RI:
Authors.
Stewart, S. (2017). And the rest is just algebra. Basel, Switzerland: Springer International.
Stigler, J. W., & Hiebert, J. (1999). The teaching gap. New York, NY: The Free Press.
Struik, D. J. (1948). A concise history of school mathematics (2 vols.). New York, NY:
Dover.
Stylianides, A. J., & Stylianides, G. J. (2013). Seeking research grounded solutions to
problems of practice: Classroom-based interventions in mathematics education. ZDM—
The International Journal on Mathematics Education, 45(3), 333–342.
Suzuki, S. (2012). The Kyoto School and J. F. Herbart. In P. Standish & N. Saito (Eds.),
Education and the Kyoto School of philosophy: Pedagogy for human transformation
(pp. 41–54). Dordrecht, The Netherlands: Springer.
Swetz, F. (1992). Fifteenth and sixteenth century arithmetic texts: What can we learn from
them? Science & Education, 1, 365–378.
Tahta, D. (1990). Gratifying usefulness. Mathematics Teaching, 132, 57–58.
151–169.
Taylor-Cox, J. (2003). Algebra in the early years? Yes! Young Children, 58(1), 14–21.
Thomas, G. (1997). What´s the use of theory? Harvard Educational Review, 67(1), 75–104.
Thomas, I., & Andrews, E. J. (1809). Preface to the third octavo edition of A new and
complete edition of arithmetick, composed for the use of citizens of the United States:
Seventh edition for the use of schools. Boston, MA: Author.
Thorndike, E. L., Cobb, M. V., Orleans, J. S., Symonds, P. M., Wald, E., & Woodyard, E.
(1923). The psychology of algebra. New York, NY: Macmillan.
Toh, M. (2008, June 17). Tuition Nation. Asiaone News, 17 June.
http://www.asiaone.com/News/Education/Story/A1Story20080616-71121.html
Toulmin, S. (1969). The uses of argument. Cambridge, UK: Cambridge University Press.
Trochim, W., & Land, D. (1982). Designing designs for research. The Researcher, 1, 1–6.
Turnbull, H. W. (Ed.). (1961). The correspondence of Isaac Newton, 1688–1694 (Vol. 3).
Cambridge, UK: Royal Society/Cambridge University Press.
Twidle, J. (2006). Is the concept of conservation of volume in solids really more difficult
than for liquids, or is the way we test giving us an unfair comparison? Educational
Research, 48(1), 93–109.
Tzanakis, C. (1991). Is it possible to teach abstract algebraic structures in high school? A
historical approach. Euclides, 11(2), 24–34.
Ueno, K. (2006). From Wasan to Yozan: Comparison between mathematical education in the
Edo period and the one after the Meiji restoration. In F. K. S. Leung, K.-D. Graf, & F.
Lopez-Real (Eds.), Mathematics education in different cultural traditions—A
comparative study of East Asia and the West: The 13th ICMI Study (pp. 65–79). New
York, NY: Springer.
UNESCO. (1973). New trends in teaching mathematics (Vol. 3). Paris, France: Author.
University of Melbourne (1855, March 24). Registered correspondence, W. E. Hearn to E. G.
Mayne (held in the University archives).
University of Melbourne (1900). Matriculation examination regulations (held in the
University archives).
Usiskin, Z. (1987). Why elementary algebra can, should, and must be an eighth-grade course
for average students. Mathematics Teacher, 80(6), 428–438.
Usiskin, Z. (1988). Conceptions of school algebra and uses of variables. In A. F. Coxford &
A. P. Schulte (Eds.), The ideas of algebra, K–12, 1988 yearbook (pp. 8–19). Reston,
VA: National Council of Teachers of Mathematics.
Vaiyavutjamai, P. (2003). Dilemmas associated with quantitative analyses of effects of
traditional mathematics teaching. In H. S. Dhindsa, S. B. Lee, P. Achleitner, & M. A.
Clements (Eds.), Studies in science, mathematics and technical education (pp. 259–
268). Gadong, Brunei Darussalam: Universiti Brunei Darussalam.

Vaiyavutjamai, P., Ellerton, N. F., & Clements, M. A. (2005). Students’ attempts to solve
two elementary quadratic equations: A study in three nations. In P. Clarkson, A.
Downton, D. Gronn, M. Horne, A. McDonagh, R. Pierce, & A. Roche (Eds.), Building
connections: Research, theory and practice (pp. 735–742). Sydney, Australia:
Mathematics Education Research Group of Australasia.
Venema, P. (1714). Een kort en klare onderwysinge in de beginselen van de algebra ofte
stel-konst: Waar in niet alleen de voornaamste quetien Van Willem Bartjens en Martin
Wilkens see sinlijk Verhandeld, maar ook t’effens byna alle de stelkonstige question uit
de algebra Van Abraham de Graaf: ales in een seer nette ordre en geschiktheid na
malkanderen zijn gesteld en opgelost. Amsterdam, The Netherlands: Author.
Education.
von Glasersfeld, E. (1990). Environment and communication. In L. P. Steffe & T. Wood
(Eds.), Transforming children’s mathematics education: International perspectives (pp.
30–38). Hillsdale, NJ: Lawrence Erlbaum.
Vredenduin, P. G. J. (1962). Functions and relations. In H. Freudenthal (Ed.), Report on the
relations between arithmetic and algebra in mathematical education up to the age of 15
years (pp. 99–110). Groningen, The Netherlands: J. B. Wolters.
Vygotsky, L. (1962). Thought and language (E. Hanfman & G. Vakar, Trans.). Cambridge,
MA: MIT Press (Original work published, 1934).
Vygotsky, L. S. (1978/1930). Mind in society: The development of higher psychological
processes (M. Cole, V. John-Steiner, S. Scribner, & E. Souberman, Trans.). Cambridge,
MA: Harvard University Press.
Warren, E., & Pierce, R. (2003). Learning and teaching algebra. In B. Perry, G. Anthony, &
C. Diezmann (Eds.), Research in mathematics education in Australasia, 2000–2003
(pp. 291–312). Flaxton, Australia: Post Pressed.
Wartofsky, M. (1979). Models, representation and the scientific understanding. Dordrecht,
The Netherlands: D. Reidel.
Webster, S. (1808). Mathematics, compiled from the best authors, and intended to be the
text-book of the course of private lectures on these sciences in the University of
Cambridge (2nd ed.). Cambridge, MA: William Hilliard.
Wentworth, G. A. (1881). Elements of algebra. Boston, MA: Ginn, Heath, & Co.
Whitacre, I., & Nickerson, S. D. (2016). Prospective elementary teachers making sense of
multidigit multiplication: Leveraging resources. Journal for Research in Mathematics
Education, 47(3), 270–307.
Whitehurst, G. J. (2003). Research on mathematics education. Washington, DC: U.S.

Department of Education.
Wittman, E. (1998). Mathematics education as a “design science.” In A. Sierpinska & J.
Wolfe, C. (1739). A treatise of algebra, with the application of it to a variety of problems in
arithmetic, to geometry, trigonometry and conic sections. London, UK: A. Bettesworth
and C. Hitch.
Wu, H. (2001). How to prepare students for algebra. American Educator, 25(2), 10–17.
Wu, H. (2011a). Understanding numbers in elementary school mathematics. Providence,
RI: American Mathematical Society.
Wu, H. (2011b). The mis-education of mathematics teachers. Notices of the American
Mathematical Society, 58(3), 372–384.
Young, J. W. A. (1914). The teaching of mathematics in the elementary and the secondary
school. New York, NY: Longmans, Green, and Co.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015a). Conceptual (mis)understandings of
fractions: From area models to multiple embodiments. Mathematics Education
Research Journal, 27(2), 233–261.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015b). Enriching student concept images:
Teaching and learning fractions through a multiple-embodiment approach. Mathematics
Education Research Journal, 27(2), 201–231.
Zhang, X., Clements, M. A., & Ellerton, N. F. (2015c). Engaging students with multiple
models of fractions. Teaching Children Mathematics, 22(3), 138–147.
Zurina, H., & Williams, J. (2011). Gesturing for oneself. Educational Studies in Mathematics,
77(2–3), 1–14.
Author Index
A Brock, William H., 35

Adams, John A., 157, 213, 214 Brown, Ann L., 88, 116
Albree, Joseph, 27 Bruner, Jerome, 103
Alexander, John, 18
Alexandri, Johannis, 18, 21 C
Al-Khalili, James, 15 Cai, Jinfa, 3, 12, 31, 45, 49, 71, 80, 88,
Allen, Frank B., 43 95, 101, 190
Andrews, Alfred, 29 Cajori, Florian, 2, 8, 14, 15, 26, 27, 29,
Andrews, E. J., 27 35, 204
Arora, Alka, 3 Campbell, Donald T., 119
Artigue, Michèle, 12 Campos, Daniel G., 102, 156
Assude, Teresa, 12 Cañadas, M. C., 87
Australian Council for Educational Capraro, Mary M., 41–42, 83
Research, 41 Carpenter, Thomas P., 214
Ausubel, David P., 102, 103, 214 Carr, Wilfred, 62
Carraher, David W., 101
B Carslaw, Horatio Scott, 32, 37
Baek, John Y., 6, 60, 88 Castañeda, Hector-Neri, 213
Baker, Scott, 119, 203 CCSSM (common core, National
Ball, Deborah Loewenberg, 218 Governors Association Center for Best
Banchoff, Thomas Francis, 44 Practices, & Council of Chief State
Barker, David, 108 School Officers), 31, 41, 72, 73, 74,
Bartell, Tonya, v 75, 76, 77, 78, 79, 83, 87, 88, 93, 94,
Battey, Dan, v 97, 102, 156, 189, 194, 195, 196, 204,
Battista, Michael, 43 206
Bauersfeld, Heinrich, 213 Chambers, C.E., 12
Bedney, Gregory, 103 Charles, Randall I., 80, 81, 82, 83, 156,
Beman, Wooster Woodruff, 36 194
Bestor, Arthur, 34 Chateauneuf, Amy Olive, 31
Bézout, Étienne, 23 Chazan, Daniel, 1, 12
Bishop, Alan, J., 11, 14, 17, 215 Choppin, Jeffrey, 218
Blanton, Maria L., 45, 84, 87, 88, 95, Chorlay, Renaud, 12
101, 156, 172 Clairaut, Alexis-Claude, 23
Bolton, Frederick E., 23 Clarke, David J., 7
Bonnycastle, John, 27 Clement, John J., 65
Booth, Leslie, 2 Clements, McKenzie A. (Ken), 3, 4, 5, 7,
Borel, Émile, 35, 36 11, 13, 14, 15, 17, 18, 21, 23, 24, 29,
Boruch, Robert, 119 31, 32, 33, 34, 35, 36, 37, 39, 43, 44,
Branch-Boyd, Judith C., 80, 156, 194 45, 47, 48, 49, 65, 66, 67, 68, 85, 88,
Bray, Mark, 4 104, 107, 117, 118, 122, 127, 154,
Bressoud, David M., xxi 157, 188, 207, 214, 215, 217
Britt, Murray, 95, 96 Cobb, Margaret V., v
Brizuela, Barbara M., 45, 87, 88, 101, 156 Cobb, Paul, 115, 116, 215

318 Author Index
Cocker, Edward, 27 Erbas, Ayhan Kursat, 44

Cohen, Jacob, 109, 134, 135, 152, 199 Euler, Leonard, 26
Colburn, Warren, 29, 34 Evans, Ken McR., 43
Cole, Percival R., 157 Exner, Robert M., 39
Colebrooke, Henry Thomas, 15
Confrey, Jere, 115 F
Coventry, Angela, 213 Filloy, Yagüe Eugenio, 17
Coxford, Arthur F., 14 Foote, Mary, v
Cremin, Lawrence Arthur, 34 Foy, Pierre, 3
Freudenthal, Hans, 40–41
D Fried, Michael A., 96
da Ponte, João Pedro, 12, 13, 15, 25, 27 Fujii, Toshiakira, 2, 96
Dascal, Marcello, 213 Fullan, Michael, 219
Davies, Charles, 26, 27, 33
Davis, Robert B., 38, 39, 42, 82 G
Day, Jeremiah, 37 Gagné, Robert Mills, 102–104, 105, 157,
de L’Hôpital, Guillaume F. A., 49 214
De Saussure, Ferdinand, 91, 93, 96, 97 Gardiner, Angela Murphy, 45, 84, 88,
DeGarmo, Charles, 102, 103, 213, 215 101, 156, 172
Del Campo, Gina, 7, 66, 68, 85, 107, 122, Gavin, M. Katherine, 73
188, 207, 215, 217 Gersten, Russell, 119, 124, 203
Didis, Makbule Gozde, 44 Giacardi, Livia, 35, 36
Dienes, Zoltan Paul, 12, 39, 40 Godfrey, Charles, 37
Dieudonné, Jean, 38–39 Gravemeijer, Koeno, 116
Dillow, Sally, 34 Green, Lyn, 43
Dilworth, Thomas, 27 Grimison, Lindsay A., 13, 37
Ding, Meixia, 41–42, 83, 204 Grinder, Robert E., 213
DiSessa, Andy, 115 Grinstein, Louise S., 48
Ditton, Humfrey, 18–21 Grugeon, Brigitte, 12
Douglas, Edwin C., 43 Guimarāes, Henrique Manuel, 12, 13, 15,
Dreyfus, Tommy, 102, 104, 214 25, 27
Dunkel, Harold B., 66, 67, 103, 157
Durell, Fletcher, 37 H
Durkin, Kelley, v Hacker, Andrew, xxi
Harrington, Judy, 32
E Hart, Kath, 3, 24, 217
Education Department of Victoria, 41 Hayward, Frank Herbart, 157, 213
Edwards, Ann, 5 Henderson, Ernest N., 213
Edwards, Edgar L., 218 Herbart, Johann Friedrich, 7, 66, 67, 102,
Eisenberg, Ted, 109 103, 104, 106, 120, 157, 213, 214
Ellerton, Nerida F., 3, 4, 5, 13, 15, 18, 21, Herbel-Eisenmann, Beth, 218
23, 24, 29, 31, 33, 34, 44, 45, 47, 49, Hershkowitz, Rina, 102, 104, 214
67, 104, 107, 118, 122, 136, 154, 157 Hertel, Joshua T., 48
Ellis, Amy, 84 Hewitt, Dave, 96
England and Wales in the Oxford and Hiebert, James, 49, 190, 214
Cambridge Examination Board, 23 Hill, Heather, 218
Author Index 319
Hjalmarson, Margret, 62, vi Land, Douglas A., 119

Hodgson, James, 21, 22, 34 Lange, Karl, 213
Hohensee, Charles, 49, 190 Lannin, John K., 108
Holton, Derek, 7 Laplace, Pierre-Simon, 33
House, Peggy A., 2 Lave, Jean, 213
Husky, Matthew E., 14 Lawson, Dene R., 13
Hwang, Stephen, 49, 190 Layton, Edwin T., vi
Lean, Glendon A., 48, 65, 217
I Lee, Kyeong-Hwa, 96
Illingworth, Mark, 80, 156, 194 Lehrer, Richard, 115
Irwin, Kay, 95, 96 Lenfant, Agnès, 12
Isler, Isil, 84, 88 Lesh, Richard A., 6, 60, 62
Izsák, Isaac, 1, 3, 14, 34 Leung, Frederick, 7, 11, 14, 215
Li, Xiaobao, 41–42, 83, 204
J Lim, Ting Hing, 2, 44
Jahnke, Hans N., 33 Lloyd, John Wills, 119, 203
Jeon, Kyungsoon, 93 Lochhead, Jack, 65
Jones, Phillip S., 14 Loehr, Abbey M., vi
Joyce, Bruce, 107, 121 Loemker, Leroy E., 14
López, Jonathan, 44
K Lüken, Miriam, 84
Kanbir, Sinan, 6, 42, 61, 71, 72, 78, 82, 91, Lykins, Chad, 4
106, 116, 118, 125, 155, 194, 204, 211
Kaput, James J., 95, 156 M
Karp, Alexander, 42, 48 Ma, Liping, 5, 41, 120
Karpinski, Louis C., 27 MacGregor, Mollie, 2, 31, 65, 217
Keitel, Christine, 11, 14, 33, 36, 39, 45, Mackenzie, Andrew A., 104
48, 215 Martin, Michael O., 3
Kelly, Anthony. Eamonn, 6, 60, 88, v Martínez-Planel, Rafael, 44
Kemmis, Stephen, 62, v May, Kenneth O., 37
Kiang, Leong Yu, 3, 4 McNab, Susan London, 101
Kieran, Carolyn, 3, 71, 84, 87, 88, 172 McQualter, Jock W., 37
Kilpatrick, Jeremy, 1, 3, 11, 14, 17, 33, Meister, David, 103
34, 36, 39, 45, 48, 215 Merrill, M. David, 102, 103, 157, 214
Kim, Jeeseok, 46, 84, 88, 172 Meserve, Bruce, E., ii
Klein, Felix, 43, 44 Mestre, José, 65
Kleiner, Israel, 37 Michaels, Brenda, 48
Knuth, Eric, 3, 12, 31, 45, 71, 80, 84, 88, Milgram, R. James, 31
156, 172 Mills, Darwin, 80, 156, 194
Kriegel, Uriah, 213 Monk, George S., 65
Küchemann, Dietmar, 3 Monroe, Walter Scott, 27
Kwo, Ora, 4 Moon, Bob, 38, 39
Moore, Eliakim Hastings, 35
L Morris, Anne, 49, 190
Laborde, Colette, 43 Moss, Joan, 101
Lacroix, Sylvestre François, 26 Mosteller, Frederick, 119
320 Author Index
Moyer, John C., 12, 95 R

Muke, Charlie, 48 Radford, Luis, 91, 95–97, 99–101, 108,
Mullis, Ina V.S., 3 172, 173, 187, 206, 212, 295
Raub, Albert, N., 37
N Ray, Joseph, 37
National Governors Association Center Rech, Janice F., 32
for Best Practices, & Council of Chief Reeves, Andy, 80, 156, 194
State School Officers. see under Rein, Wilhelm, 214
“CCSSM” Richmond, Donald E., 43
National Mathematics Advisory Panel, Rickart, Charles E., 43
24, 25, 31 Rittle-Johnson, Bethany, v
NCTM (National Council of Teachers of Rivera, Ferdinand, 84
Mathematics), 12, 41, 156 Robbins, Edward, 37
Newman, M. Anne, 217 Roberts, David Lindsay, 14, 15, 35, 36
Newman-Owens, Ashley, 45, 88, 156 Robison, Victoria, 49, 190
Newton, Isaac, 33 Robles, Izraim, 44
Nickerson, Susan D., 77 Rosenthal, Robert, 135
Nie, Bikai, 95, 101 Rosnow, Ralph L., 135
Nordgaard, Martin A., 35 Rosskopf, Myron F., 35, 39
Ruthven, Kenneth, 35, 36
O
Olney, Edward, 33 S
Orleans, Jacob S., v Sawrey, Katie, 45, 88, 156
Otte, Michael Friedrich, 93 Schauble, Leona, 115
Owens, Kay Dianne, 48 Schilling, Stephen G., 218
Oxford and Cambridge Examination Schliemann, Annalucia, 101
Board, 23 Schloring, Raleigh, 1
Schoenfeld, Alan, 108, 119
P Schubring, Gert, 12, 48
Papert, Seymour, 43 Schultze, Arthur, 36
Pappano, Laura, 32 Schwartz, J. L., 101
Paraide, Patricia, 48 Selleck, Richard J. W., 157
Park, Kyungmee, 7 Sfard, Anna, 3, 8, 12, 37, 43, 107, 108,
Peirce, Charles Sanders, 7, 66, 91–93, 97, 138
102 Sheffield, Linda Jensen, 73
Perel, William M., 4, 39 Showers, Beverley, 107, 121
Perry, John, 35 Simons, Lao Genevra, 218
Phillips, Christopher J., 39 Simpson, Thomas, 23
Piaget, Jean, 213 Sinclair, Hermine, 213
Pierce, Robyn, 108 Singh, Parmjit, 136
Pike, Nicolas, 27–29 Slavin, Robert E., 116
Pimm, David, 218 Smith, David Eugene, 1, 33, 36
Pomfret, Alan, 219 Smith, Gordon, 43
Porro, F. D., 3, 14, 29 Snyder, Thomas D., 34
Presmeg, Norma, 91, 212 Sobel, Max A, ii
Price, Michael H., 35 Spencer, Joi, v
Author Index 321
Spielhagen, Frances R., 12, 33 V

Sriraman, Bharath, 96, 108 Vairo, Philip D., 4, 39
Stacey, Kaye, 31, 217 Vaiyavutjamai, Pongchawee, 2, 44, 49,
Stanley, Julian, 119 65, 154
Steen, Lynn A., 13 Van Engen, Henry, 37
Steffe, Les, 214 Venema, Pieter, 27
Stephens, Ana, 84, 88, 101, 156, 172 Vinner, Shlomo, 102, 104, 118, 157, 214
Stephens, W. Max, 2, 96 von Glasersfeld, Ernst, 214
Sterry, Consider, 29 Vredenduin, Pieter Gaele Johannes, 37, 38
Sterry, John, 29 Vygotsky, Lev, 201, 213
Stewart, Sepideh, v
Stylianides, Andreas J., 115 W
Stylianides, Gabriel J., 115 Wager, Anita, v
Sutherland, Rosamund, 17 Wagner, David, 218
Suzuki, Shoko, 67 Wald, Elva, v
Swain, Henry, 42 Walker, Robert J., 43
Swetz, Frank, 33 Wang, Ning, 95
Symonds, Percival M., v Warren, Elizabeth, 108
Wartofsky, Marx, 105
T Watterson, Geoff, 43
Tahta, Dikran, ii Webster, Samuel, 26, 29
Tall, David, 102, 104, 157 Wenger, Etienne, 213
Taylor-Cox, J., 87 Wentworth, George A. (“Bull”), 37
Thomas, Gary, 60 Westbury, Ian, 13, 72, 92, 156, 157, 201
Thomas, Isaiah, 27 Whitacre, Ian Michael, 77
Thorndike, Edward L., v White, Richard T., 102, 104, 105, 157, 214
Toh, Mavis, 4 Whitehurst, Grover J., 119
Toulmin, Stephen, 68, 207 Williams, Julian, 207
Townsend, Brian E., 108 Wittman, Erich, 59, 60, 215
Trochim,William M. K., 119 Wolfe, Christian, 23
Turnbull, Herbert W., 26 Woodyard, Ella, v
Twidle, John, 76 Wu, Hung-Hsi, 3, 4, 41, 83, 88, 117, 194
Tzanakis, Constantinos, 46
Y
U Young, Jacob W. A., 35
Ueno, Kenji, 37
UNESCO, 38, 39 Z
University of Melbourne, 13, 32 Zhang, Xiaofen, 104, 154
Usiskin, Zalman, 15, 45 Zurina, Harun, 207
Subject Index
A C
Abbaci and the abbaco tradition, 33 Cai, Jinfa, 80, 95
Algebra, 45. See also “School Algebra” Calculus, 21, 34, 36, 45, 48–49
rhetorical notations, 8 CCSSM common-core curriculum
syncopated notations, 8 (U.S.A.), 31–32, 41, 43, 45, 72,
Apperception, 7, 85, 103, 104, 136, 138, 73, 74, 75–77, 79, 83, 87, 88,
194, 201, 202, 213–214 122, 156, 159, 163, 172, 175,
Applied mathematics, 37 189, 194, 195, 196, 207, 216–217
Arithmetic algebraic structure in, 41, 45, 72, 74,
field properties of, 97 75–77, 80, 87, 88, 94, 97, 101,
generalized, 16, 25–31, 35, 46 102, 156, 159, 163, 175, 189
Asian nations do well in algebra, 4, 31, modeling in, 45, 72, 79–80, 87,
35, 48 101–102, 156, 189
Associative property (Addition), 39, 41, China, 5, 12, 41, 48
42, 74, 75–78, 81–82, 83, 88, Christ’s Hospital School (London),
124, 128, 159, 177–178, 18–21, 23, 34
193–195, 202, 227, 229, 233, Ciphering. See Cyphering
236, 241, 243, 245–246, Clairaut, Alexis-Claude, 3, 23
253–254 Cognitive structure, 67, 106, 138, 157,
Associative property (Multiplication), 203, 212, 214, 219
39, 41, 42, 44, 71, 74, 75–77, Colburn, Warren, 16, 29, 34
81–82, 83, 88, 92–94, 96, Colonialism, 12–13, 27
104–105, 124, 128, 155, 156, Concept image, 67, 104, 138, 159,
159, 178–180, 193–195, 202, 199–200, 214
229, 230, 232, 235, 236, 239, Concepts in Secondary Mathematics and
241, 243, 247–248, 255–256 Science (CSMS), 24, 217
Attitudes to algebra, 6, 104, 105, 157, Concrete learning aids, 12
159–171, 197, 201–203, 206, Conventions in algebra, 82–83, 98–99,
207, 216, 289–293 168, 179–187, 196–197, 198,
Australia, 2, 23, 32, 35 205, 230–231, 234, 238, 261–285
Ausubel, David P., 102, 103, 137, 157, Crossing-the-river task, 207, 264, 268,
214 272, 276, 280, 284
Curriculum
B author-intended, v, 13, 34, 43, 155,
Bézout, Étienne B., 16 156, 157
Books. See Textbooks received, 13, 34, 155, 156, 157–158,
Bourbaki, Nicolas, 16, 38 203
Bourdon, Louis, 16 retained, 203
Brunei Darussalam, 2, 44 teacher-implemented, v, 13, 34, 49,
Bruner, Jerome, 39, 103, 138, 157, 213 155, 156, 157, 187–188

324 Subject Index
Cyphering books Euler, Leonhard, 3, 26

complemented by textbooks, 194 Explicit specifications with algebra, 71,
data from, 29, 30 83, 98–99, 205, 217, 289–291
Cyphering tradition, 29, 49 Expressive mode of communication,
67–68, 77, 78, 79, 107, 124, 135,
D 136, 155, 156, 157, 158, 159,
Davies, Charles, 26, 27, 33 160, 162, 166, 167, 170, 172,
Davis, Robert (“Bob”), 16, 39, 42 188–189, 194, 197, 207, 215, 218
DeGarmo, Charles, 102, 103, 138, 213,
215 F
Del Campo, Gina, 65, 67, 68, 107, 123, France, 35, 38
155, 188, 207, 215, 217, 218 Freudenthal, Hans, 40–41
de L’Hôpital, Guillaume, 49 Functional thinking approach to school
de Saussure, Ferdinand, 93, 98, 102 algebra, 16, 35, 36, 45, 84, 228
Descartes, René, 16, 18 Functions, 25, 36
Design research, 5–6, 60– 62, 88–90, graphs of, 25
109, 115–116, 193, 216, 218 linear, 25
design elements for, 61–62, 89–90 logarithmic, 25, 31
Erich Wittman on, 59–60 polynomial, 25
Kelly, Lesh & Baek on, 60, 62, 115 quadratic, 25
in mathematics education, 60, 109,
216 G
role of theory in, 62–63, 122–124 Gagné, Robert M., 102, 103, 138, 157,
Dienes, Zoltan Paul, 12, 16, 39–40, 48 214
Dieudonné, Jean, 38, 46 Gatekeeper (school algebra as), 16,
Ding, Meixia, 41–42, 83 31–33, 45
Distributive property, 42, 71, 73, 75–77, Generalization, 78, 79, 105–106, 108,
82, 83, 88, 96, 128, 155–156, 193, 205, 295–297
159, 175–177, 193–195, 202, contextual, 100, 173, 174, 295–297
227, 231, 234, 237, 238, 241, factual, 100, 173, 174, 295–297
243, 244, 249–252, 257–259 post-symbolic, 101, 172, 173,
Ditton, Humfrey, 16, 18–20 295–297
Doctoral dissertations in mathematics symbolic, 100–101, 172, 173, 174,
education, 211 295–297
Document analysis, 71–79, 193–194 Geometry, 43
Germany, 32, 33, 35
E Graphs, 16, 36, 46
Effect size, 5, 135–137, 152, 198, 199
Episodes, 104, 105, 157, 159–171, 193, H
197, 206, 216, 278, 289–293 Herbert, Johann Friedrich, 7, 66, 67, 85,
Equations, 4, 8, 25, 34–35, 42–43, 44, 64, 90, 102, 103, 104, 121, 136, 138,
77, 79, 228, 229, 230, 234, 237 157, 158, 193, 194, 201, 202,
general, 14 206, 211, 212, 213–215
linear, 25, 35, 79, 230, 234, 237 apperception, 7, 66, 85, 103, 104,
quadratic, 4, 8, 25, 35, 42, 44 136, 138, 194, 201, 202, 212,
Ethnomathematics, 47 213–214
Subject Index 325
five steps in lesson planning, 66 pilot studies of, 6, 42, 61–62, 71, 78,
Herbatians, 66–67, 102, 136, 138, 82–84, 85, 91, 106, 107,
157, 201, 213 116–119, 124, 125, 155, 194,
Hertel, Joshua, 48 197, 205, 207, 217
History Kant, Immanuel, 213
of school algebra, 3, 5, 11–49, Kart, Kath, 3, 24, 217
211 Kieran, Carolyn, 3, 79, 84, 172
of theory development, 211 Kilpatrick, Jeremy, 1, 14, 34
Hodgson, James, 16, 21–22, 34 Klein, Felix, 14, 16, 35, 37, 43, 44, 46
Høyrup, Jens, 2 Knuth, Eric, 80
I L
Imageries, 104, 157, 159–171, 193, 197, Lacroix, Sylvester, 16, 26–27
206, 216, 289–293 Language factors in algebra learning, 8,
Inequalities, 5, 42–43, 64, 77, 79 13, 23, 28–29, 31, 33–37, 43, 46,
Instrumentation and research design 63–64, 65, 67, 80, 85, 108, 193,
Algebra Test, 124, 196, 197–198, 216, 217
229–239 Laplace, Pierre-Simon, 33
classroom observation schedule, Lean, Glendon A, 65, 217
287–288 Leibniz, Gottlieb, 14, 213
effect size calculations, 135–137, Lim, Ting Hing, 2, 44
152, 198–199 Locke, John, 213
interview procedures, 138
interview protocol, 128–129 M
null and research hypotheses, MacGregor, Mollie, 2, 65, 217
129–135, 137, 193–197 “y= 8z” task, 2, 65, 212
questionnaire for students, 239 Ma, Liping, 5, 41, 121
workshop notes, 243–244 Mathematical Association of America
workshops, 77, 138 (MAA), 1
Intellectual skills, 104, 105, 157, Mathematicians, 88, 212
159–171, 193, 197, 206, 216, Mathematics educators, 5, 101
289–293 theory development, 215
Interviews, 5, 42, 71, 116, 160–170 U.S. mathematics education
data from, 160–170 dissertations, 211–213
protocol for, 128, 218, 227–268 Melbourne (Australia), 2, 11, 23, 32
IRCEE genre, 29 University of, 13, 23, 32
Izsák, A., 1, 14, 34 Milgram, R. James, 31
Modeling, 16, 33, 62, 78–80, 83, 98–99,
J 195, 207
Japan, 36, 37, 48 with algebra, 16, 33–34, 78–80, 83,
195
K Model lessons, 124, 125
Kanbir, Sinan Modes of Communication, 85, 107
doctoral dissertation of, 11, 61, 72, Del Campo & Clements, 7, 85, 107
84, 88–90, 116, 119–120, 212, Moore, Eliakim Hastings, 14, 16, 35,
213, 216 46
326 Subject Index
N Q
National Council of Teachers of Qualitative analysis of data, 139,
Mathematics, 1, 12, 41, 48, 156. 155–190, 194, 196, 206, 208
See NCTM random allocation to groups, 61, 207
Fifty-Sixth Yearbook, 12
First Yearbook, 1 R
Seventieth Yearbook, 1, 12 Radford, Luis, 91, 95, 96, 98, 99–101,
Standards, 5, 156 108, 172–174, 206, 295–296
National Governors Association Center Received curriculum, 155
for Best Practices, & Council of Receptive mode of communication, 7,
Chief State School Officers. See 67–68, 77, 79, 107, 124, 135,
CCSSM 136, 188–190, 194, 197, 212,
National Mathematics Advisory Panel 215, 218
(U.S.A.), 24–25, 31 Recursive specification, 78, 83, 101, 155,
NCTM, 1, 5, 12, 41 156, 167, 169–170, 187, 193,
Newman, M. Anne, 217 205, 217, 289–293
New Math(s), 37, 38–40, 42, 44, 46 Research design, 124–125, 204–205
Newton, Isaac, 3, 16, 18, 21, 25, 33, Research questions, 65, 108–109,
46 193–206
New Zealand, 96
S
O School algebra, 8, 44–47
Owens, Kay D., 48 enrollment patterns for, 1, 3–4,
34–35, 44–45
P essential for higher studies, 16, 64
Papua New Guinea, 39, 48 as functional thinking, 16, 35, 36, 45,
Pedagogical content knowledge, 46, 64, 71, 78–80, 82–83, 84,
218–219 95, 98–99, 101, 166–174, 203,
Peirce, Charles Sanders, 66, 68, 85, 90, 207
98, 102, 156, 193, 211, 212 as a gatekeeper, 16, 31–33
theory of, 7, 85, 91–95, 102, as generalized arithmetic, 16, 25–31,
156–157, 206, 212 35, 46, 64, 72
PEMDAS (“Please Excuse My Dear history of, 3, 5, 8, 11–49, 66, 212
Aunt Sally”), 42, 85, 93–94, 102, and mathematicians, 47–48, 212
104–105, 156, 195, 206 problems with, 1–5, 8, 59
Perry, John, 14, 16, 35, 37, 46 purposes of, 11, 13, 14–16, 63
and Perryism, 14, 35 structure, 37–39, 41, 64, 71–72, 95,
Pestalozzi, Johann Heinrich, 34, 157 96, 121, 158–166, 207
Piaget, Jean, 157, 213 students’ difficulties with, 1–3, 8, 23,
Pike, Nicolas, 16, 27, 29 31, 98–99, 115
1793 Abridgment, for schools, 27 as the study of variables, 8, 42–44, 64
1788 Arithmetic, 27–28 teachers’ difficulties with, 4–5
Pilot studies, 6, 42, 71, 82–84, 85, 91, unidimensional trait?, 44–47, 87
106, 107, 116–119, 155, 194 word problems for, 2, 33–34, 78,
Purposes of school algebra, 11, 13, 126–128, 182, 184, 228, 229,
14–49, 63 231–234, 236, 238
Subject Index 327
School Mathematics Study Group Thomas, Isaiah, 28

(SMSG), 16, 42, 43 Thorndike, Edward Lee, v
Scotland, 32 Trends in Mathematics and Science
Selleck, Richard, 157 Study (TIMSS), 3, 31
Semiotics, 7–8, 41, 63, 65, 71, 80, 83, 90, Trigonometry, 22, 25
97, 98–99, 102–103, 108, 137–138,
156, 195, 204, 212, 213, 216 U
Service mathematics, 47 Understanding algebra, 49, 65, 194
Simpson, Thomas, 22 United Kingdom, 37
Singapore, 4, 31, 48 United States of America
Smith, David Eugene, 1, 14, 33, 35–36, Kanbir’s study within, 59–208, 212
37, 46 and school algebra, 5–6, 14, 16, 27,
Springer’s series on “History of 29, 33, 34, 35–36, 37, 41–42,
Mathematics Education”, 6, 11 43, 44–45, 48, 205, 212, 213,
Stacey, Kaye, 217 218
Sterry, Consider and John, 29 Usiskin, Zalman, 15, 17, 45
Structure, 16, 37
of real numbers, 16, 37 V
Students-and-professors problem, 65 Vaiyavutjamai, Pongchawee, 2, 44, 49,
Subscript notation (for sequences), 78, 154
79, 99, 128, 168, 170, 196, 197, Variables, 16, 44, 45, 193, 195
204, 205, 206, 227–229, 233, Verbal knowledge, 104, 157, 158–171,
235, 236, 239, 261–285 193, 197, 206, 216, 289–293
Vinner, Shlomo, 102–104, 118, 138,
T 157, 214
Teachers, 4, 41, 49, 207–208, 212 Vygotsky, Lev, 157, 201, 213
Technology, 207
Textbooks, 3, 12, 18, 21, 23, 29–31, 34, W
36, 42–43, 46, 49, 80–83, 212 Westbury, Ian, 13, 156, 157
with American authors, 26–29, 34, 36, Wittman, Erich, 59–60, 215
37, 42–43, 80–83, 156, 194 Wolfe, Christian, 22
with British authors, 18, 21, 23, 27, Workshops, 77, 138, 193, 203, 205, 218,
34 243–244, 245–259, 261–285
with Continental authors, 12, 18, 21, Wu, Hung-Hsi, 3, 4, 41, 83, 88, 194
23, 26–27, 29
and historical research, 31, 42–43, 46, Z
212 Zhang, Xiaofen, 154

Using Design Research and History To Tackle A Fundamental Problem With School Algebra

Uploaded by

Copyright:

Available Formats

Using Design Research and History To Tackle A Fundamental Problem With School Algebra

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Using Design Research and History To Tackle A Fundamental Problem With School Algebra

Uploaded by

Copyright:

Available Formats

History of Mathematics Education

Using Design Research

More information about this series at http://www.springer.com/series/13545

“Algebra is powerful—but it can also be frightening. It demands a shift of attention

Using Design Research

Foreword by A. Eamonn Kelly

ISSN 2509-9736 ISSN 2509-9744 (electronic)

Library of Congress Control Number: 2017945773

© Springer International Publishing AG 2018

Printed on acid-free paper

This Springer imprint is published by Springer Nature

Foreword (by A. Eamonn Kelly) .......................................................................................... v

6. Research Design and Methodology .............................................................................. 115

7. Quantitative Analyses of Data ...................................................................................... 141

8. Qualitative Analyses of Data ......................................................................................... 155

9. Answers to Research Questions, and Discussion ........................................................ 193

Figure 2.1 Page 10 from Ditton’s (1709) appendix. .............................................................. 19

Table 1.1 Percentages Correct, 328 Mathematics Teacher-Education Students

Individual Chapter Abstracts

Chapter 2: Historical Reflections on How Algebra Became a Vital Component of

Chapter 3: Framing a Classroom Intervention Study in a Middle-School Algebra

Chapter 4: Document Analysis: The Intended CCSSM Elementary- and Middle-School

Chapter 5: Review of Pertinent Literature

Chapter 6: Research Design and Methodology

Chapter 7: Quantitative Analyses of Data

Chapter 8: Qualitative Analyses of Data

Chapter 9: Answers to Research Questions, and Discussion

Chapter 10: Postscript: Framing Research Aimed at Improving School Algebra

Recently, David M. Bressoud, a former President of the Mathematical Association of

Sinan Kanbir M. A. (Ken) Clements

Keywords: Mathematics education research, History of school algebra, Student difficulties

A Fundamental Problem with School Algebra

© Springer International Publishing AG 2018 1

Some Performance Data, and Associated Critiques of Practices in School Algebra

In the twentieth and twenty-first centuries, results from international comparative

How Well Do Middle-School Teachers Understand Algebra?

x2 + 6 = 0 No real 69 (21%) x2 + 2 > 0 All real 53(16%)

The Background to How and Why this Book was Written

Overview of this Book

attempt to respond realistically to the fundamental problem. The issue addressed is

Ellerton, N. F., & Clements, M. A. (2011). Prospective middle-school mathematics teachers’

Keywords: Algebra as a gatekeeper, Algebra as generalized arithmetic, History of

Providing Historical Frameworks for Mathematics Education Research

There is a tradition in mathematics education of requiring graduate students who are

© Springer International Publishing AG 2018 11

Algebra in Secondary School Mathematics: The Debate Over Purpose

Figure 2.1. Page 10 from Ditton’s (1709) appendix.

Algebra as Generalized Arithmetic

School Algebra as a Gatekeeper for Entry to Higher Studies

In the course of criticizing the common-core mathematics sequence (National

School Algebra as a Language for Modeling Real-Life Problems

Algebra as an Aid for Describing Basic Structural Properties of Real Numbers

Figure 2.5. Comments on a theory of relations (Vredenduin, 1962, p. 105).

School Algebra as the Study of Variables

mappings-of-the-plane, approach which emphasized a variables perspective (Banchoff,

Is School Algebra a Unidimensional Trait?

The History of Mathematics and the History of School Mathematics

Ruthven, K. (2008). Mathematical technologies as a vehicle for intuition and experiment: A

Applying Principles of Design Research in Mathematics Education