World Applied Sciences Journal 28 (4): 449-456, 2013

ISSN 1818-4952
© IDOSI Publications, 2013
DOI: 10.5829/idosi.wasj.2013.28.04.841

Science Teaching and Learning with System Dynamics Approach:

The Effects of Attitude and Success of Causality Relationships

Hasret Nuho lu

Faculty of Education, Maltepe University, stanbul, Turkey

Abstract: The aim of this study is to 1) apply the “system dynamics approach” in a “Science and Technology”
course that is being instructed to 7th grade middle school students in Istanbul, Turkey; 2) determine problems
of field applications; 3) improve negative attitudes of students towards specific skills (like understanding causal
relationships 4) provide more effective tools of learning and teaching, respectively for students and teachers.
In this research, the experimental design pre-post test with a control group is employed. The study was
conducted with 81 students in middle schools in Istanbul, Turkey. Assessment tool is used: “Cause-Effect
Relationship Scale”. Improvements that stem from system dynamics are observed in the tests that measure
capability of understanding causal relationships. However, in the boundaries of the research, no significant
level of improvements was observed in the perceived understanding of causal relationships.

Key words: System dynamics approach Science and technology course Causality relationships

INTRODUCTION courses even after school. At some instances, students

became so enthusiastic with the subjects that they made
System dynamics has been applied in many diverse their parents to take part in the projects too.
fields. The most widely known system dynamics study is There was a remarkable increase in the interest and
the Limits to Growth book which was commissioned by understanding levels of students with respect to the
the Club of Rome in the 1970’s [1, 2]. This study claimed courses they take. This increase led to anticipation by the
that the natural balance in the earth would be unsettled practitioners that this approach will enter the general
until the year 2000 if no precautions are taken. The study educational system in USA. Nevertheless, the researchers
triggered a lot of controversy, but the discovery of the observed that the extent to which system dynamics
Antarctic ozone hole in 1985 created a big effect on the applications have diffused into the educational system
public opinion of the world citizens, as well as political have not reached the intended level in the aftermath of
leaders. This discovery urged them to take precautions these studies [3]. Various explanations have been
for the problems mentioned in the study Limits to Growth. suggested on the causes of this gap. The significant of
As a result of these actions, many countries around the these explanations are applications of the system
world started to take decisions in order to prevent dynamics approach in the K-12 education has not focused
emission of gases that cause ozone depletion. on the development of lesson plans and applications that
System dynamics studies led to important changes are based on the pedagogical methods that enhance
in the fields of management and economy as well. learning. The practicing teachers focused on the rules of
Having inspired by successful policy changes in a lot of system dynamics and they neglected practical principles
fields, system dynamics researchers aimed to apply the for the successful applications [3, 4]. Therefore a few
system dynamics approach also in educational fields. curriculum projects have been developed that are based
Early educational applications showed that important on system dynamics (Stacin, Cc-Stadus, Cc-Sustain,
improvements could be obtained in this field [3]. In the Science Ware). By using these projects, a lot of new ideas
schools where system dynamics approach is employed, and useful models are provided in order for practicing
students ran voluntary projects in relation to their school teachers to apply system dynamics in the classroom.

Corresponding Author: Hasret Nuho lu, Faculty of Education, Maltepe University, stanbul, Turkey.

449
 World Appl. Sci. J., 28 (4): 449-456, 2013

System dynamics provides means of communication qualifications are more valuable for their societies because
between mental models and simulations. Mental models they can discover hidden problems in life and they can
of human beings are powerful in terms of knowledge, but provide effective suggestions to solve such problems.
they are weak in terms of calculating the behavioral The aim of this study is to 1) apply the “system
results of the models. Simulation models that are based on dynamics approach” in a “Science and Technology”
mental models complement the inadequacy of human course that is being instructed to 7th grade middle school
minds to calculate complex and dynamic behaviors [3]. students in Istanbul, Turkey; 2) determine problems of
Another benefit of simulation models is that it helps field applications; 3) improve attitudes of students
experimentations, which improves learning process [5]. towards understanding causal relationships skills; 4)
Using system dynamics approach, modelers produce provide more effective tools of learning and teaching,
simulation tools that are called “micro worlds”. Students respectively for students and teachers.
use these tools to make certain experiments. These tools
are actually replicas of the real world. That is why they are MATERIALS AND METHODS
called micro worlds. The experiments in these micro
worlds can be repeated easily, using varying parameters Research Models: In this research, the experimental
and alternative scenarios. This allows the student to see design with pre-post test with a control group is
how the dynamics of the system work, by experiencing it employed. Independent variable is the “system dynamics
in the virtual world. Usually, there is no other way of approach”. Experimental group was instructed according
observing the results of the experiments outside of the to the system dynamics approach. Control group was
micro worlds. These experiments are conducted with the instructed according to the standard syllabus. Students
help of easily managable simulation software. Dynamo, in the experimental and control groups were selected
Powersim, Vensim, Stella, ithink, Extend and Anylogic are randomly.
some of the system dynamics software [5, 6]. Stella is the It was critically analyzed whether there were
most widespread tool for K-8 students [3, 7]. statistical differences among scientific successes and an
System dynamics approach makes it easy for attitude of students in understanding cause-effect
students to focus on the causes of events. Moreover, relationships.
students understand that there are usually more than one
cause-effect relationships in complex systems. In addition, Research Sample: The study was conducted with a total
students realize that the result of combined interactions of of 81 students (40 in experimental group, 41 in control
all cause-effect relationships cannot be analyzed by group) who are in 7th grade in two different middle
superficial studies. schools in Istanbul, Turkey. Research sample was
Undoubtedly, the goal of education is beyond selected randomly.
teaching students certain courses. It is not sufficient to
reach the goals of the education system when the Assessment Tools: In order to assess the sub problems of
students can correctly answer the questions they are the study, assessment tool is used: “Causal Relationship
asked. Education system also aims that students should Scale” developed by researchers [8], in order to determine
be able to construct problems by themselves. how students evaluate the relationship between cause
Constructing problems requires a more sophisticated way and effect on sample events and to learn their attitude
of comprehension, than merely answering pre-determined towards cause and effect relationships.
questions. To construct problems, it is necessary to
observe the environment. Moreover it is necessary to Causal Relationship Scale: Causal relationship scale
have a critical view on the issues at hand and to see the measures students’ perceptions of their abilities to
world from unusual perspectives. The individuals understand the causality relationships between
that gained this perspective are more flexible, events. The scale has been developed by the researcher
tolerant, productive and valuable for the society they live [8]. It consists of two parts. First part of the scale
in. The reason for these qualifications is that students are measures attitudes towards causal relationships.
aware that there is not a unique truth. They are aware of Its Cronbach’s Alpha value is 0.88. Therefore it is
the fact that truth can change with respect to the assumed to be reliable in statistical sense. Its validity is
conditions and time. The individuals that have these verified by expert opinions and statistical factor analysis.

450
 World Appl. Sci. J., 28 (4): 449-456, 2013

Tablo 1:Contents of the Causality 2 Scale

Questions Contents
1 Detect the cause of graph that plotted as a result of an experiment
2 Thinking the cause of movements of a frog in boiling water
3 Find cause-effect relationships in an event clearly
4 Determine cause-effect relationships between more than one event, respectively
5 Determine cause-effect relationships among the spring mass system types, force and the amount of elongation.
6 Realize an event consist of cause-effect relationship and occurred in their environment.

The scale consists of 10 5-point Lickert-type questions After the pre-tests, system dynamics approach was
where 5 of the questions are positive and the other 5 introduced to experimental groups in 4 study hours.
negative. The scale has a KMO value of 0.86 and a During the introductory course, students learned the
Bartlett’s value of 543.5. Therefore the scale is appropriate basics of system dynamics approach and they built
for factor analysis. All the questions are clustered models presented as exemplary scenarios (example:
in the first factor that has an Eigen-value greater than 1. bathtub dynamics).
The explained variance is 50%. The topics “spring-mass systems”, “work and
Second part of the scale consists of open ended energy”, “energy types” and “preservation of
questions on the opinions of the students about energy” were studied in both schools and groups.
exemplary events. Second part of the scale (Causality 2) Instruction of the curriculum material was same for
consists of open ended questions on the opinions of the both control and experimental groups. There were 4
students about exemplary events. study hours available in each week. 2 of them were
used for the instruction of curriculum material in both
Application Steps of the Research: control and experimental groups. The remaining 2
hours were used for exercises in control group,
Pre-application of the experimental study was whereas these hours were used for system dynamics
conducted with a group of 20 students. The aim of approach in experimental group. Design of experiment
pre-application was to discover probable is shown at Table 2.
problems that would be encountered during the Post-tests were applied and qualitative interviews
study. Real application was designed by taking the were done
findings obtained during pre-application into Data were analyzed statistically.
account. Total duration of the experimental study was 8
Application schools (two different middle school in continuous weeks.
stanbul) and the experimental and control groups
within these schools were determined by random Learning Topics with System Dynamics Approach:
assignment. Experimental group students attended introductory
Both schools have experimental and control groups. system dynamics lectures. During these 4 study hour
In one of the schools, control group was instructed lectures, they learned the relationship between system
by the actual teacher. In the other school control and dynamics, basic elements of a system (stock and flow,
group was instructed by the researcher. Experimental feedback loops, causality relationships). Also they
groups were instructed by the researcher in both learned using the Stella program to build and test models
schools. Instruction was conducted in the Fall and interpret the dynamic behavior of a model. Students
semester in the form of 4 weekly study hours in a learned system dynamics tools with 4 different scenarios
“science and technology” course. in the introductory lesson.

Table 2: Design of Experiment

1. School 2. School
----------------------------------------------------------------- -----------------------------------------------------------------------
Experimental group Control group Experimental group Control group
Curriculum material 2 hours/week 2 hours/week 2 hours/week 2 hours/week
Teached by Researcher school teacher school teacher school teacher
Supplementary material 2 hours/week 2 hours/week 2 hours/week 2 hours/week
Teaching method system dynamics modeling exercises system dynamics modeling exercises
Teached by Researcher school teacher researcher school teacher

451
 World Appl. Sci. J., 28 (4): 449-456, 2013

Table 3: The activities and topics of the lectures in two groups

Courses Topics Activities
1. introduction to spring mass systems - Playing with springs
- Making a spring
- Designing a dynamometer
2. work and energy - Do we work in which situation
- Is there any work?
3. types of energy - Velocity, mass and kinetic energy relationships
- What is the potential energy depending on?
- What is the spring flexibility potential energy depending on?
4. conservation of energy - How does the energy transfer?

The students built models of spring-mass systems Independent t-tests are applied for comparing the two
step by step using concepts they have recently learned, groups on scale. The aim is to show whether the
as well as causal relationships between these concepts. system dynamics approach indicates an improvement
They found out what kind of changes happened to in comparison to the traditional supplementary
springs after applying force. They decided how the materials.
mathematical equations can be applied on the model.
They achieved to produce the graphical output of Findings
dynamic behavior of the system where more than one Statistical Values of Mean and Standard Deviation:
variable is shown together in the graphics. At each The results of pre- and post-test score made on success
lecture, students made additions to the model. They of figuring out the causal relationships skills and attitude
discussed the model with their friends. of students in experimental and control groups are shown
The curriculum material of the Turkish Ministry of on Table 5.
National Education [9] for the year 2006 was instructed in When Table 5 is checked, it can be seen that pre-test
both control and experimental groups according to scores of experimental groups and control groups are not
constructivist learning methods. Students learned the
equal, but are quite close. The understanding scale of
topics through activities and class discussions. The only
causal relationship has two sections. In the first section
difference between the control and experimental groups
(causality 1) the interests and attitudes of students
were supplementary materials. Supplementary materials of
towards causal relationships are evaluated, while in the
the control group were exercise problems given in the
second section (causality 2) their skills of commenting on
workbook. This material was replaced with system
and discerning causal relationships in some sample cases
dynamics modeling activities in the experimental group.
are evaluated. When analyzing the first section by taking
The activities and content of the lectures are shown in
into consideration the mean values, which were taken
Table 3. The final model developed by the experimental
group is shown in Table 4. from the students’ data, it is seen that the initial scores of
experimental and control groups are different in the
Statistical Techniques Used in the Research: Collected beginning, whereas their mean scores are equal in the end
data are analyzed with 3 statistical methods: of the empirical study. When analyzing the second
section, in the end of the empirical study it is understood
Descriptive statistics of the means and standard that causal relationship skills of the students have
deviations are found for scale and group. increased.
Paired t-tests are applied for comparison of pre- and
post-tests for scale and group. Pre-test of the group The Effects of the System Dynamics Approach
is compared with the post-test of the same group to on Students’ Skills of Understanding Causal
see whether there is any change in a group between Relationships: The results of pre and post test
pre- and post study. The aim is to show whether scores of the tests that measure students’ skills
teaching activities indicates an improvement in the of understanding causal relationships in
skills and attitudes of the students in comparison to experimental and control groups are shown on
doing nothing. Table 6.

452
 World Appl. Sci. J., 28 (4): 449-456, 2013

Table 4: The final model developed by the experimental group

Models of Systems

Introduction to Spring Mass Systems Graphics

Work and Energy

Energy types
(kinetic, potential, spring potential energy) Models of Systems

Conversation of the energy Graphic

453
 World Appl. Sci. J., 28 (4): 449-456, 2013

Table 5: Statistical Values of Mean and Standard Deviation

Mean Standard deviation
--------------------------------------------------------------- ------------------------------------------------------------
E1 C1 E2 C2 E1 C1 E2 C2
Causality Causality 1 38.6 35.2 40.8 40.8 7.7 14.2 7.5 7.4
Causality 2 -0.1 -0.6 2.2 0.4 0.9 0.7 1.8 1.1
E1: experiment pre-test C1: control pre-test
E2: experiment post-test C2: control post-test

Table 6: Statistical Values of Causality 1 and Causality 2

Tests Groups h p (%) t Df sd(1) sd(2) is difference significant? SD useful or not? Experiment biased
CAUSALITY 1 E1 E2 1 2.26 -2.07 38.00 5.88 Yes
C1 C2 1 1.95 -2.14 37.00 11.22 Yes
C1 E1 0 8.83 -1.37 61.70 14.16 7.69 No No
C2 E2 0 49.86 0.00 74.96 7.42 7.45 No No
CAUSALITY 2 E1 E2 1 0.00 -8.85 38.00 1.69 Yes
C1 C2 1 0.00 -4.64 37.00 1.30 Yes
C1 E1 1 0.92 -2.41 77.91 0.72 0.85 Yes Yes
C2 E2 1 0.00 -5.47 64.72 1.12 1.76 Yes Yes

The students’ answers have been evaluated as primary and middle schools. Students can learn the basics
causality 1 and causality 2 (Table 6). In the causality 1 of system dynamics very efficiently [4]. In this research it
scale, the ideas and attitudes of the students about causal was determined that students developed some skills like
relationships are evaluated; in the causality 2 scale, skills realizing and interpreting causal relationships,
of commenting on and discerning causal relationships in competence in problem solving, competence in drawing
some sample cases are evaluated. Statistically there is not and analyzing graphics by using system dynamics
any significant difference in the interests and attitudes of approach. These results are similar to general findings in
the students towards causal relationships (causality 1) in the literature.
the end of empirical study (t = 0.00; p>.05). In the classes Everyone who teaches System Dynamics modeling
where the system dynamics approach is performed, it is has reported how difficult it is, even though the benefits
obvious that ideas of the students about causal are great [3, 4, 23, 24]. There are some common difficulties
relationships did not change. The results gained from when applying system dynamics in school. Students have
causality 2 scale shows that there is an increase in pre- difficulty ib distinguishing stocks from flows. Instead of
and post-empirical study in both groups. Causal understanding the system under study, they try to fit the
relationships are understood better by the students in the mathematical formulas they have in mind without thinking
experimental group according to mean scores. It can be thoroughly. When simulation doesn’t produce expected
said that system dynamics provides the benefit of behavior, they include “fudge factors”. Fudge factors are
understanding and commenting on causal relationships. variables specifically designed to fix the problem
artificially. They are not representations of any reality.
DISCUSSION Students don’t test their models well. Therefore the
models don’t work under different conditions. They build
Many researchers and teachers [6, 10-22, 26], applied unnecessarily complex models. They try to build
system dynamics approach in their classes or experimental imitations of textbook models. The teachers don’t realize
studies and attained positive results. that building a realistic model requires really long time.
A common finding of these teachers is that by using Patience is a pre-condition for students and teachers
system dynamics tools, learning becomes more learner- when building system dynamics models [3, 24].
centered and cooperative. System dynamics lead students In this research, middle school students learned
to discover problems by putting puzzle pieces together, various topics of science and technology course with
looking for similar patterns and working with their supplementary activities of system dynamics approach.
friends to actively generate questions in various The topics taught are spring-mass systems, work and
disciplines. All these activities are coherent and make a energy, energy conversion. The students learned
great benefit for children. The work is interdisciplinary in basic concepts (stock-flow, causality relationships and

454
 World Appl. Sci. J., 28 (4): 449-456, 2013

feedback loops) of system dynamics in applying them in feedback loop, causality relationships and graphics
those science topics. Students had difficulties at first on drawing and reading abilities need to be introduced to the
what the stock and flow variables were, when they were students as well, with exemplary scenarios.
modeling spring-mass systems. Reminding them the An ideal learning environment would include
stock-flow relationship in the bath-tube example helped discussion of a topic, student-directed research,
them to build the model of spring-mass system. Adding laboratory experimentation, model building and
causal feedback loops to the model was another difficulty exploration and computer simulation to verify the link
for students. Teachers have to be patient in order for the between model behavior and experimental observations.
students to build better models with time. The overall goal is to teach students critical thinking skills
and a methodology for dealing with complex problems
RESULTS AND SUGGESTIONS that they can use later in life as managers, company
presidents, journalists, generals, pilots and engineers [5].
Results of analysis show that the system The use of computers in the classroom in contrast to
dynamics approach has some benefits in learning science computer lab has important results. Students learn and
topics. build simulation models in parallel. They work in groups.
In the boundaries of this research, system dynamics This helps communication between students [25].
approach had no effect in the perceived understanding
causality relationships (causality 1 scale). Since the study REFERENCES
was conducted in a setting of 4 weeks, these results did
not deviate from estimates. According to the researcher’s 1. Meadows, D.H., D.L. Meadows, J. Randers and
observations, the interests and attitudes of students W.W. Behrens III., 1972. The Limits to Growth. New
towards the system dynamics approach increased steadily York: University Books. ISBN 0-87663-165-0
during the study. However, according to the pre- and 2. Forrester, J.W., 1973. World Dynamics (2 ed.).
post-tests of the attitude scale, any change in the Waltham, MA: Pegasus Communications, pp: 144.
perceived attitude towards causality relationships were 3. Forrester, J.W., 1996. Road Map 1: System Dynamics
not detected. This is not surprising since the attitude and K-12 Teachers. MIT System Dynamics in
towards a subject depends on more than one factor. Education Project. http:// web.archive.org/ web/
Therefore in order to increase the positive attitude, these 20060430061546/http:// sysdyn.clexchange.org/ sdep/
factors should be studied in detail. papers/D-4665-4.pdf
Second part of causality relationship scale consists 4. Lyneis, D.A., 2000. Bringing System Dynamics To A
of open-ended questions. These questions help us to School Near You Suggestions For Introducing And
assess whether the students have the ability to determine Sustaining System Dynamics In K-12 Education.
the relationships between causes and effects correctly. International System Dynamics Society Conference
The answers of the questions were evaluated as true or Bergen, Norway.
false. After the study, both control and experimental 5. Martin, L.A., 1997. Road Map 2: The First Step. MIT
groups had an increase in comparison to their pre-study System Dynamics in Education Project.
levels. The increase in the mean of the experimental group http://sysdyn.clexchange.org/sdep/Roadmaps/RM
was higher than the control group. 2/D-4694.pdf
Therefore it is reasonable to conclude that the system 6. Alessi, S.M., 2000. Designing Educational Support In
dynamics approach has a positive impact on the ability of System-Dynamics-Based Interactive Learning
understanding causal relationships for students although Environments. Simulation&Gaming, 31(2): 178-196.
the perceived attitude towards causality did not increase 7. Brown, G.S., 1992. Improving Education in Public
in the experimental group. The steadiness of attitude may Schools: Innovative Teachers to the Rescue. System
be due to external factors that are not handled in the Dynamics Review, 8(1): 83-89.
research. 8. Nuho lu, H., 2008. The Effect Of The System
A spring-mass system is an ideal topic to apply Dynamics Approach On Students’ Attitude,
system dynamics approach in a science course. Before Sc entific Success And Different Skills In Middle
starting with the modeling of the spring-mass system, the School Science And Technology Course.
students need to be introduced to the Stella software and Unpublished PhD Dissertation, Gazi University,
basics of model building. Basic concepts like stock-flow, Turkey.

455
 World Appl. Sci. J., 28 (4): 449-456, 2013

9. MEB, 2006. New Curriculum Program in 19. Coffin, S., 1999. Getting Started with Systems
Turkey.http://ttkb.meb.gov.tr/ogretmen/modules.ph Thinking in the Primary Grades.
p?name=Downloads&d_op=viewdownload&cid=48 www.clexchange.org/ftp/documents/
10. Clauset, K.H., 1982. Effective schooling: a system Implementation/IM1999-03GettingStartedST.pdf
dynamics policy study. Unpublished Ed D. Boston 20. Ticotsky, A., R. Quaden and D. Ve Lyneis, 1999. The
university school of education. In And Out Game: A Preliminary System Dynamics
11. Hassell, D.J., 1987. The Role Of Modelling Activities Modeling Lesson. http://sysdyn.mit.edu/cle/
in The Humanities Curriculum, With Special 21. Lyneis, D.A. and D. Fox-Melanson, 2001. The
Reference To Geography: An nvestigate Study. Challenges of Infusing System Dynamics into a K-8
London: King‘s College. Curriculum. International System Dynamics Society
12. Evans, J.K., 1988. Application of system dynamics as Conference, Atlanta, Georgia.
a strategy for teaching management concepts. 22. Shaffer, C.S., 2006. Toward a System Dynamics
Unpublished Ed D. Boston University. Model of Teaching Computer Programming Via
13. Webb, M.E., 1988. An investigation of the Distance Education. Unpublished PhD dissertation.
opportunities for computer based modelling and the The Pennsylvania State University.
possible contributions to children‘s learning, in 23. Forrester, J.W., 1992. Road Map 1: System Dynamics
secondary school science. London: King‘s College. and Learner-Centered-Learning in Kindergarten
14. Draper, F. and M. Ve Swanson, 1990. Learner-directed through 12th Grade Education. MIT System
systems education: a successful example System Dynamics in Education Project.
Dynamics Review, 6(2): 209-213. 24. Alessi, S., 2005. The Application of System Dynamics
15. Hopkins, P.L., 1992. Simulating Hamlet in the Modeling in Elementary and Secondary School
Classroom. System Dynamics Review, 8(1): 91-98. Curricula. http:// web.archive.org/ web/
16. Davidsen, Bjurklo, Wikström 1993. Introducing 20060304015136/ http:// www.c5.cl/ ieinvestiga/ actas/
System Dynamics in Schools: The Nordic Experience. ribie2000/ charlas/alessi.htm
17. Ossimitz, G., 1996. Projekt "Entwicklung vernetzten 25. Brown, G.S., 1990. The Genisis of the System
Denkens" (Endbericht an die For- Thinking Program at the Orange Grove Middle
schungskommission). Translated from Germany; School, Tucson, Arizona. Personal report.
Ossimitz, G. (2000). Teaching System Dynamics and 26. Cruz, M., M.T. González, M.P. Restrepo and M.L.
Systems Thinking in Austria and Germany. System Zuluaga, 2007. Colombian Classroom Experiments: A
Dynamics Conference in Bergen, Norway. Preliminary Report. CLE Newsletters, 16(1).
18. Zaraza, R. and D. Ve Fisher, 1997. Introducing System
Dynamics into the Traditional Secondary Curriculum:
The CC-Stadus Project's Search For Leverage Points.
www.clexchange.org/ftp/documents/Implementatio
n/IM1997-07SDLeveragePoints.pdf

456