Mixed Method Research Analysis:

Studying Virtual Manipulatives Paired with Explicit Instruction to Teach

Algebraic Equations to Students with Learning Disabilities

ETEC 500
Yi (Winnifer) Kong
December 9, 2021
Introduction:

The purpose of this paper is to analyze the study “Studying virtual manipulatives paired
with explicit instruction to teach algebraic equations to students with learning disabilities”
(Satsangi, et al., 2018). The analysis will summarize and critique the mixed methodology used in
the study, and the study design, the study purpose, the study participants, and setting. It will also
describe and critique methods of data collection, the research findings, the strengths, and
weaknesses as they relate to the study purpose, as well as the implications for educational
technology.

The selection of this research article was because of the interest in exploring how
integration of technology in mathematics education will support students, especially ELLs and
students with learning disability (LD) in math. Analysis of this research will help secondary math
educators decide how to meaningfully integrate technology in their classroom. This mixed
method study is mostly quantitative with some qualitative content.

Research purpose:

The study did not mention its research hypothese or alternative ones. However, the
research purpose was laid out clearly by the researchers and it was easy to identify. The goal of
the study was to “advance this line of research by studying the benefits of a virtual manipulative
balance paired with explicit instruction to teach high school students with learning disabilities in
mathmatics how to solve multistep linear equations” (Satsangi, et al., 2018). They aimed to find
out the relation between the use of a virtual manipulatives paired with explicit instuction on
accuarcy, independence and duration to complete multistep linear equations. The researchers also
tried to find out the views of the participants and their teachers about using virtual manipulatives
in class.

As this study is trying to evaluate effectiveness of an intervention (using virtual

manipulatives among students with LD in math), it is fair to say that the study purpose is
consistent with the findings. The research questions were explicitly stated right following the
research purpose. They were consistent with the research purpose.
Literature review:

The literature review section was made clear that there were several gaps in knowledge
that need to be explored and reasons for current study. There’s a need to address the current LD
students’ needs since majority of them struggle with math.

The authors also saw the significance for algebraic instruction, especially in higher level
math and science. The study explored relevant and up-to-date research regarding the promising
benefits of explicit math instruction and graduated teaching (CRA), paired with the use of
concrete math manipulatives. It pointed out the limitations for concrete manipulatives, which
resulted in researchers’ shifted interests in virtual manipulatives considering its potential.
However, there’s a limited amount of research based on students with LD in math to demonstrate
the effectives of a virtual manipulative balance paired with explicit instruction among high
school students.

Study design:

This problem-based action research used the study design of single subject multiple probe
as a type of quantitative research. It is a popular design in the fields of special education and
counselling. It is said to be useful when the researcher is attempting to change the behaviour of
an individual (in this case, students with LDs in math) and to document the changes to see how
effective is the virtual manipulatives program.

Multi probe design consists staggered introduction of participants to the intervention so

each participant serves as control for themselves and each other. The rationale of selecting this
particular study design was not explained in depth in the study.
The independent variable in this study was the use of virtual math manipulatives. Their
performance in solving multistep linear questions were measured as the following three
constructs: the accuracy, independence, and duration to complete multistep linear questions.
They were the three dependent variables in this study measured repeatedly in each of the three
phases: baseline, intervention, and maintenance, to compare changes before and after
intervention.
Sample:

In this single subject study, the 3 participants were selected based on the criterion-based
purposeful sampling from one public high school in a suburban area outside of a major Mid-
Atlantic city. The criteria were listed clearly by the researchers. It is a non-random sampling
technique that utilizes a specific criteria to select a particular sample. It is an effective method in
this single subject research as it helps the researchers to collect in-depth information
(effectiveness of virtual manipulatives in algebra) from the right respondents (students with LD
and struggling in math).

There are some limitations of criterion-based purposeful sampling. The non-random

nature of the sampling can cause bias and makes it less reliable and harder to be generalized to a
bigger population. In addition, how researchers developed those speicifc criteria and any
potential implications were not explained in the study.

A detail description for each participant was mentioned in the study. All of them fulfilled
the requirements of purposeful criterion-based sampling process with LD in math. It was
applicable to the research purpose and research questions to study the benefits of using virtual
manipulatives on students with LD in math. All three students provided parental consent and
assent to participate in this study.

Data collection:

In single subject studies, the same students served both as control and experimental group
to measure changes before and after the treatment. Setting of the study was clearly explained for
understanding of the big picture, including demographics of the area where the participants were
chosen, the room where this study took place, the setting of the instructional sessions, the length
of the sessions as well as a description of the virtual manipulatives program used.

This study collected both quantitative and qualitative data. Quantitative data included
independent variables: use of virtual manipulative balance paried with explicit instruction. The
virtual manipulatives used and explicit instruction were clearly defined. The dependent variables
were percent accuracy, percent independence, and duration to complete linear equaition
questions. All were clearly defined.

A recording system was used to measure the dependent variables, and the system was
explained in the study (Satsangi et al., 2018). There were three phases (baseline, intervention,
and maintenance) and each session in each phase was measured and data recorded. Results were
compared among the three phases. In this case, the participants served as their own control group
using the asessment results during the baseline phase. The treatment group were also the same
participants, but after they received interventions.

Semi-structured interviews with open-ended questions were conducted before and after
the treatment with the three participants and their math and ELL teachers to investigate the social
validity of the intervention. Triangulation occurred here to determine if the quantitative survey
results were in line with the qualitative interview finding. Interobserver agreement and treatment
integrity data were collected to show no inter-rater bias in the aseesments in each of the three
phases.

Data analysis and quality

Data was analyzed using visual anlysis and two effect-size indices to show level and
trends (Satsangi et al., 2018). Researchers were able to demonstrate a functional relation between
three students' use of virtual manipulatives and their percent accuracy scores solving equations.
Researchers conducted two effect size measures: percentage of non-overlapping data (PND) and
Tau-U were calculated to determine the strength of the intervention (Satsangi et al., 2018).

PND statistical analysis and its indications were explained in the study. PND is calculated
by comparing the data points between baseline and intervention assessments and divided by
number of total intervention data points and multiplied by 100. The researchers explained that
this is the effect size of the treatment as a percentage. 90%-100% shows a very effective
treatment, 70%-89% indicates an effective treatment (Scruggs et al., 1987).
 Secondary effect size measure (Tau-U: a nonparametric statistical evaluation of effect
size) to measure trend between phases (baseline, intervention, maintenance). Tau-U provides a
score between 0-1 (0.93-1: a large effect; 0.66-0.92: a medium effect; 0-0.65: a small effect)
(Parker et al., 2011). The statistical analysis method of Tau-U and its indications were clearly
explained in the study.

Findings:

Quantitative data weighted more heavily in this study. Graphs and tables were used for
visual presentation of information to help the reader understand the changes, thus the effect of
intervention. Data showed that all students performance significantly improved in solving
multistep linear equations using virtual manipulatives when compared to their baseline scores.
The findings were consistent and reflective of data. The study successfully showed an in-depth
analysis of the findings.

The findings highlighted that student performance across the three dependent variables of
percent accuracy, percent independence, and duration were in favour of the intervention
treatment (use of virtual manipulations with one-on-one explicit instruction under the design
principles of the CRA-I sequence).

A small section was used to discuss the qualitative data to show social validity of using
virtual manipulatives in math education. Student and teacher responses from the interviews also
showed a consistent response to the treatment. They all provided positive feedback. The virtual
manipulatives helped the participants understand and comprehend a math concept. It boosted
students’ levels of confidence in math. Both teachers and students would consider using the
virtual manipulatives in class in the future.

Limitations:

The researchers laid out limitations in the study. They discussed the constraints of the
virtual manipulative program itself, such as the exclusion of some values to use in solving
questions. They also mentioned that the intervention only taught problems with positive
coefficients in algebra. As a result, the program had a limited potential generalization of the
findings. There was a lack of formal generalization phase to assess students on questions with
paper and pencil. Consequently, reserachers couldn’t decide if the three students developed the
procedural and conecptual skills needed to solve problems without the aid of the program
(Satsangi et al., 2018). Additional research was recommended.

Implications:

Based on the study, integration of tech in math with explicit education can strengthen
students' understanding of math concepts (Satsangi et al., 2018). Students and teachers also
support these findings. Teachers should integrate tech supporting visual representations of math
concepts during explicit instruction through repeated modelling, practice, and feedback to
students. Althought teachers can look at integrating the tech in a range of settings; for students
with LD in math, they suggested indiviaulized practices might be more beneficial, such as
offering students a device to interact with the program independently in class or at home
(Satsangi et al., 2018). Teachers were also encouraged to look into advanced programs offering
explicit instruction and lessons throught the problem-solving process as a new approach to math
instruction in secondary schools.

Summary:

After critiquing the study “Studying virtual manipulatives paired with explicit instruction
to teach algebraic equations to students with learning disabilities”, one can say that it provided
excellent insight into the benefits of a virtual manipulative balance paired with explicit
instruction to teach three high school students with learning disabilities in mathmatics how to
solve multistep linear equations. The researchers could have provided more explanation on the
sampling criteria, validity and reliability of the assessments used to show students’ math abilities
in solving multistep linear equations. The generalization of the results to a bigger population is
limited. Overall, with a combination of qualitative and quantitative findings, the study offered
good research evidence for integrating technologies in teaching math among students with
learning disability. It also laid out good recommendations for implications in integrating tech in
secondary math education. The study may be used by math educators to help support students
with needs.
 References

Parker, R. I., Vannest, K. J., Davis, J. L., & Sauber, S. B. (2011). Combining Nonoverlap and
Trend for Single-Case Research: Tau-U.Behavior Therapy, 42(2), 284-299.
10.1016/j.beth.2010.08.006

Satsangi, R., Hammer, R., & Hogan, C. D. (2018). Studying Virtual Manipulatives Paired With
Explicit Instruction to Teach Algebraic Equations to Students With Learning
Disabilities. Learning Disability Quarterly, 41(4), 227-242. 10.1177/0731948718769248

Scruggs, T. E., Mastropieri, M. A., & Casto, G. (1987). The Quantitative Synthesis of Single-
Subject Research. Remedial and Special Education, 8(2), 24-33.
10.1177/074193258700800206
Reflection:
The feedback I got was detail oriented. They identified some grammar mistakes in my
writing, such as spelling mistakes. They also pointed out some awkward wordings in my draft
for me to rewrite to avoid repetition and confusion. In addition, they gave some suggestions to
remind me to be more concise in terms of summarizing the study. I tried to add more details for
analyzing and critiquing instead of just describing the study. They also noticed some missing
components required for this research analysis. I was able to use their suggestion and the rubric
to add those parts to my final draft. Besides, they pointed out some mistakes in my in-text
citation, as well as the reference list. I was able to use their feedback to update that. I found the
feedback very beneficial, and it was helpful to have another set of eyes to double check my
writing.