Developing Simulation Exercises to Enhance Learning

Duncan M. Fraser
Department of Chemical Engineering, University of Cape Town, Rondebosch, South Africa
Contact details: D.M. Fraser, Department of Chemical Engineering, UCT, Rondebosch, 7701 South
Africa. Tel: +27 21 650 2515, Fax: +27 21 650 5501, Email:Duncan.Fraser@uct.ac.za
Abstract
This paper draws on research done into student learning through computer simulations to develop some
guidelines for how to craft simulation exercises to maximize student learning. The first study was on a
computer simulation aimed at extending the students understanding of distillation. In this study we
applied the concepts of learning through variation and the notion of a learning study to identify the
necessary key aspects of the learning situation. Purposeful repetition formed part of the strategy adopted
to extend the learning possibilities for our students. In-depth interviews were used to gain insights into
the learning outcomes of the redesigned simulation experience. The overall finding was that after
redesigning the simulation the students were able to draw on their previous knowledge and expand it in
new directions in ways that made them feel positive about the experience. The second study examined
ideal conditions for simulation learning as identified in the literature. These were also applied to a
distillation simulation. Analysis of students performing the simulation showed us that the conditions
identified were not sufficient to ensure engagement of students with the simulation. Students prior view
of simulations as merely computational tools was found to severely hinder learning, whereas enjoyment
of the simulation promoted learning. The third study developed three Excel-based simulations to
enhance student learning in fluid mechanics. The Fluid Mechanics Concept Inventory (FMCI) was used
to determine areas of difficulty for the students. Three of these were chosen and simulations specifically
developed to address them. After engaging wit the simulations, students re-did the FMCI, as well as
answering a questionnaire on their experience of the simulations. Two of the three simulations led to
significant improvements in the concepts concerned, and also received good comments by the students.
All these studies point to more effective ways of using computer simulations for student learning in
engineering.
Keywords
Student learning, computer simulations, engineering
Introduction
In our chemical engineering programme we often draw on computer simulations to complement our
teaching. A particular example is a computer simulation of the distillation process which was used in our
third year as a laboratory task. Here I had become increasingly concerned about the nature of the
learning experience that this simulation was able to evoke for our students in its existing form the
majority of students appeared to bypass the engagement that we had hoped they would enter into with
the simulation. However, when I looked at the simulation I saw a potentially rich and rewarding learning
experience, and so decided to explore why the students interaction so seldom appeared to move beyond
simply going through the motions in order to complete what was required of them, and how we could
possibly change that experience.
Together with colleagues in the Centre for Research in Engineering Education which is situated in the
Faculty of Engineering and the Built Environment at the University of Cape Town, I started looking into

what kind of developments were emerging in the student learning literature that we could draw on for our
problem. The variation theory of learning seemed to be a productive line to follow (Fraser and Linder,
2006). The ideal conditions for learning through computer simulations seemed to be another. Both of
these lines of enquiry were built around simulation of distillation with chemical engineering students in
their third year of study. These studies pointed to the importance of engaging students in simulations at
an earlier stage of their careers, so the next study moved to an earlier level, looking at learning through
simulations in second-year fluid mechanics. In this paper we will report briefly on the major outcomes of
the three investigations which flowed out of these three lines of enquiry. Full details of all three studies
are available (Fraser, et al, 2006, Streicher, et al, 2005, Fraser, et al, 2007).
Variation Study: Application of Theory
In the first study we used the theory of learning through variation to help us redesign a computer
simulation of a distillation column (Fraser, et al, 2006).
In our third year, distillation is presented to the students in a course on separation processes. A number
of years ago I had set up a distillation simulation for students as an experiment in a parallel laboratory
course. However, tutors reported that students did not really engage with the simulation effectively, but
just seemed to be going through the motions in order to complete what was required of them. Figure 1
shows a typical distillation column with its key input and output parameters.

Distillate
D, xd

Feed
F, xf

B, xb
Bottoms
Figure 1. Diagram of a continuous distillation column
To make use of variation theory as a fundamental guide to analyse and redesign the simulation we
needed to identify the necessary key aspects of the learning situation. To do this we drew on the notion
of a learning study as described by Pang and Marton (2003). The aspects of such a learning study to
which we gave attention are now described together with how we addressed them:
1. Choosing the object of learning a capability, appreciation or understanding to be developed.
The original exercise had required students to set up a distillation problem in the simulation
program and use it to explore how a number of different parameters affected the functioning of

the distillation column. The students were asked to change six different parameters affecting the
operation of the simulated distillation column and observe their effects. Our evaluation was firstly
that the dual objectives (setting up and exploring) detracted from the possible learning that could
take place, and secondly that then there were too many parameters being varied in a single
exercise.
2. Gaining insight into students existing understandings, by, for example, an analysis of student learning
research in the area, or some pre-lesson test.
The way our third-year laboratory exercises were scheduled meant that most students would not
have had any formal theory of distillation classwork before the laboratory exercise, and hence
there was no obvious past experience of distillation they could draw on for the simulation learning
experience. A class test before the development of the new simulation showed that few students
had good functional knowledge of distillation.
3. Planning and implementing the learning experience.
Following (1) above we decided to reduce the breadth of possible learning by presenting the
students with a problem already set up and running on the simulation program, and limited the
simulation parameters that needed to be varied from six individual entities to two. Following (2)
above we located this learning experience when the students had all recently covered the theory
of distillation.
4. Evaluating and revising the learning experience by using a post-lesson study, such as interviews with
students, or a test to look at to what extent the students have developed the desired learning
outcome. This evaluation phase should also look at how the object of learning was handled.
The evaluation is discussed in full in the Results section. The approach to this evaluation was
something we gave much consideration to. In the end we decided that for a study such as this a
set of in-depth interviews would offer the greatest potential for teaching insight, both in terms of
learning and in terms of how the object of learning was handled.
Variation Study: Results
The evaluation of our redesigned simulation exercise was done with nine pairs of students who had
recently been taught the theory of distillation. These pairs were selected from a set of volunteers to
represent both the spectrum of ability and the range of backgrounds in the class. Each pair was observed
during the exercise and then interviewed immediately afterwards. Each interview was audio recorded.
The interview recordings of the interviews were transcribed for analysis. The transcribed data together
with the observation data were then analysed to determine the results of the exercise, as described
below.
Here we should point out that the tasks were not prescriptive, and that the participating students were
expected to make their own decisions as to how much to vary the parameters. It should also be noted
that when students interact with a simulation they can constitute what they are learning as they go along.
The following describes how we examined the effectiveness of drawing on variation theory to broaden
what it is possible to learn in a teaching situation such as the one we have described. Because we were
looking for teaching insight, both in terms of learning and in terms of how the object of learning was
handled, we have presented the results in terms of the following themes that emerged from the data:

Experiencing enhanced learning opportunities

What follows is one typical description from what students told us about their learning experiences (all
names are pseudonyms):
Greg: Dead zones, if you increase the number of trays too much you get dead zones, I didnt know
that I didnt know there were dead zones.
From statements like these we would argue that the students identified for themselves that they were
looking at distillation columns with new discernment. The simulation exercise therefore provided an
environment in which enhanced learning could take place.
What is significant here is that the opportunity for learning can be characterised as enhanced by
narrowing the objectives of the exercise.
Drawing on previous knowledge of distillation in a meaningful way
During the simulation exercise students were required to increase the purity of the distillate by varying the
number of trays in the column. Eight out of the nine pairs of students immediately knew that they must
increase the number of trays in order to increase the purity, illustrating how the students drew on their
previous knowledge of distillation columns.
That critical elements of what the students were discovering were being discerned in the light of their
previous knowledge is illustrated in the following quote:
Jason: Our (distillation) knowledge isnt amazing, so it was nice to put things in and then it would
trigger things like oh, if you put the feed stage here it gets a better composition or if you have a high
reflux it will give you a better composition.
It was also interesting for us to note that some students started to appreciate the importance of their prior
knowledge for their learning about distillation. The timing of this exercise relative to their course on
distillation meant that previous knowledge was available to them.
What was focused on?
The concepts mentioned most often in the interviews were feed tray location and number of trays, which
were the parameters that were varied explicitly. These were clearly the concepts that the students
focused on the most.
The recognition of learning
What was surprising was that the students in only two out of the nine groups felt they had learned
something new about distillation columns. Probing of the other seven groups revealed that they clearly
had learned something new. Evidence for this may be seen in the following student comments:
Thandi: See, (the simulation) actually help you understand what you are doing in class much better.
From the discussion above, we would argue that both feed tray location and number of trays (both topics
they were already familiar with) were seen in new ways by the students. This is also true of the
concentration profiles in the column. Dead zones, which is a concept they had not encountered before,
was also something new they had learned.
It appeared at first that these students had a very narrow concept of learning, which did not include
expanding their knowledge about something they already knew. On further reflection, it would seem that

possibly the term new might have confused the students and also that the students are not used to
articulating learning in these terms (they normally have their learning assessed in conventional tests that
largely involve problem solving).
Associating playing around with a system with learning
In many of the interviews students mentioned the idea of playing around with the system. This poses the
question of whether the students associate playing around with the system with learning.
Six of the nine pairs said that they would come back and fiddle with the simulation in their own time. Two
more pairs said they would use it for a project, and only one pair said they would not play with it at all.
Most of the pairs also indicated that they would be able to use this simulation for other purposes in the
future.
The comment that students would come back in their own time and play around with the system was
unexpected. From their comments we infer that playing around with the simulation means that they are
wanting to explore the effect of other parameters on the system performance.
Variation Study: Implications
In this part of the paper we described how we drew on the variation theory of learning to redesign a
distillation simulation exercise done by third year chemical engineering students in order to open up
discernment as a way to enhance the possibility of learning of distillation concepts. Results in terms of
teaching insight, both in terms of learning and in terms of how the object of learning was handled,
produced a number of insightful themes that pointed towards achieving better learning outcomes. For us,
the most interesting additional insights were:
(1) that we needed to narrow the object of learning in order to increase the possibility for learning; and,
(2) that the students did not easily recognise that exploration using a computer simulation was significant
learning, even though they had clearly learned much through it.
At the same time we recognise the method we used for our evaluation could be improved in the following
ways:

By taking videos of the students handling the object of learning.

By undertaking a much more in-depth study of the key conceptual difficulties in the object of learning
from the students perspective. This could also be accompanied by the development of a set of
questions which would probe students understanding of key concepts associated with the object of
learning, to be used to test improvements in understanding.

I would like to highlight here that we found variation theory to be a potentially powerful tool for helping us
to improve student learning. In this regard, the particular aspects of variation theory we found most
fruitful were:

The nature of the variation used is it just varying numbers in problems, or is it varying the approach
to solving particular types of problems?

The object of learning has it been clearly defined, and is it too broad or too narrow?

The sort of repetition being used is it merely rote or is it purposeful (bearing in mind that a certain
amount of repetition may be needed to develop a particular skill)?

Are we allowing students to access previous knowledge and bring it to bear in this situation?

Are we making use of the power of bringing different aspects of a phenomenon into students focal
awareness at the same time?

Have we considered what students may be overlooking, as well as what they may be taking for
granted, so that we can most effectively bring what we would like into their focal awareness?

The use of variation theory in higher education appears to have an untapped potential that we can thus
recommend to our engineering educator colleagues.
Conditions Study: Factors encouraging learning though computer simulations
In this study we built on what we had learned in the first study, and shifted the emphasis to approaching
the ideal conditions for learning through computer simulations as identified in the literature (Streicher, et
al, 2005). Table 1 lists the factors identified from the literature as encouraging student learning through
computer simulations, as well as how closely we were able to meet these specifications.
Feature

Incorporated in
Design

Simulations seen as a
learning tool

Yes

Complexity approaching
reality

No

Peer interaction learning

environment

Yes

Thorough accompanying
instructions

Yes

Using a familiar package

Yes

Sufficient time for

engagement

No

Learning history

Physical not
automatic

Open-endedness

No

Table 1. How closely did we achieve the features facilitating simulation learning?

Our search of the literature also uncovered relevant work on approaches to learning and what students
focus on in computer simulations, which we linked together. These links are shown in Table 2.
The approach taken in this study was to observe students in pairs undertaking the simulation, while video
recording their interaction with the simulation, as well as capturing what was happening on the computer
screen.
All this data was later analysed and certain parts of it were used in later interviews of the students about
their experience of the simulation, in what is known as stimulated recall. Seven pairs of students did the
simulations.

A conceptual test was also developed to gauge students understanding of distillation, to be used both
before and after the simulation, to gauge improvements in understanding and how the students changed
in their approach to learning. These tests were administered at the start of each of the two sessions with
the students (the first being when the simulation was done, and the second the interviews).
Passing the Test / Interaction
APPROACH
Surface approach
Procedural surface
approach

STRATEGY

SIMULATION
FOCUS
Following the
task

Memorising

Simulation as
Problem Solving
representations

Understanding / Engagement
APPROACH

SIMULATION
FOCUS

Procedural deep
approach

Manipulation
of
simulations

Deep approach

Exploration of
Concepts
phenomenon

Problem
Solving

Table 2. Approaches to Learning and Simulation Foci

Conditions Study: Results

Here we will discuss the major findings of this study.
The first finding was how students approached the simulation exercise. This is shown in Table 3, from
which it will be seen that most student pairs used one form of surface approach, with none using the full
deep approach.
Focus

Description

Occurrence

Following the task

Simulations as a
representation

Manipulation of the
simulation

Exploration of the
phenomenon

Table 3. Breakdown of Student Simulation Learning Foci

A strong correlation was found between enjoyment and engagement with the simulation. This is shown in
Table 4, where engagement is observed through progression in understanding as measured by the
conceptual tests.

Names

Improvement

Enjoyment

-ve

+v
e

Net
t

Yes

Partia
l

Bongani

Gershwin

Nandi

Mpho

Sizwe

Thandiwe

Arkash

Delia

-1

Vani

-1

Carey

-1

Jarrod

-2

Olivia

-2

Richard

-3

Derek

-3

No

x
x

x
x
x

Table 4. Overall Improvement and Students Expressed Enjoyment

Conditions Study: Implications
In this section we will draw out the implications of our findings concerning the ideal conditions for learning
through computer simulations, in terms of those features which facilitated learning and those which
seemed to be problematic as far as learning was concerned.
Factors which facilitated learning

Familiarity with package. Using a familiar package improved the learning potential for
working with the simulation, since students were able to focus on the conceptual issues being explored
as opposed to worrying about navigational aspects. The decision to use a different simulation package
was vindicated as none of the students had any navigational difficulties. This also reduced the time
required for the exercise.

Peer interaction learning environment. The majority of the students thought that performing
the simulations in pairs was helpful because it allowed them to share ideas and build on each others
knowledge.

Enjoyment. This factor noted above has nort previously been identified as a factor
necessary for promoting learning.

Factors which did not necessarily help learning

Simulations seen as learning. It is evident from this study that it takes more than the effort
of briefing students to change their view of simulations as engineering tools. It seems that a long-term
mindset change, developed throughout the undergraduate programme, is required to enable them to
start to see simulations as an opportunity for gaining understanding.

Complexity. The particular simulation package used in this study did compromise the
approach of the simulation to reality. Approaching reality would have required the use of a package
such as AspenPlusTM, but this was not familiar to the students. In this case we had to choose familiarity
over complexity.

Thorough instructions. Many of the students felt that the exercise may be improved with
fewer, more focused questions, consisting of baby steps that would guide them to the final answer (a
similar structure to their tutorials). Clearly the students did not like the open-endedness of the
questions.

Open-ended. It appears that there is a fine line between allowing for an open-ended
structure and helping students to explore all the variables necessary for developing a conceptual
understanding.

Time for engagement. In this simulation exercise time was limited, but it seems that the
view students held about simulations (as engineering tools rather than learning tools) was the major
factor hindering engagement, rather than time pressure.

Learning history. The physical learning history was not beneficial. The majority of students
abandoned it relatively quickly in favour of memory. A physical learning history proved not to be a
substitute for the stop/rewind/restart functionality.

Consolidation. The students expressed a need for consolidation of the knowledge acquired
during the simulation, in order to cement the understanding they had acquired.

Conceptual tests. Students also indicated that it would be helpful to have access to their
notes during the conceptual tests, in order to emphasize that it was understanding and not recall that
was being tested.

Visualisation Study: Fluid Mechanics Concepts

This study was motivated by the need identified in the previous studies of introducing computer
simulations to students at an earlier stage than their third year of study. As with the first study, we
approached this as a learning study with the following elements:

choosing an object of learning;

ascertaining students existing understandings;

planning and implementing classroom activities; and

evaluating and revising the activities.

In this study we chose the object of learning to be students understanding of basic phenomena in fluid
mechanics. We then determined the particular difficulties students had in understanding fluid mechanics
phenomena.
Next, we decided to use simulations as the classroom activities for improving
understanding of these phenomena. The absence of suitable existing simulations led us to develop our
own simulations, programmed to run in Excel. Implementation involved the majority of students in the
second year class engaging in these simulations. The improvement in their understanding was gauged
by repeating the test used to determine their difficulties. The improvement (or lack of it), together with
student feedback, was used to evaluate the simulations and indicate changes needed to them.

The theoretical grounding for this study combined a constructivist perspective on learning with the
phenomenographic theory of learning through variation. The constructivist viewpoint has led to the
development of concept inventories, and the Fluid Mechanics Concept Inventory (FMCI) was thought to
be the ideal tool for us to use in determining students understanding of our chosen object of learning.
This inventory was reduced appropriately to fit time constraints and the material covered in our fluid
mechanics course.
It was decided that the simulations should focus on the three areas identified as being the weakest by the
FMCI, namely pressure measurement, pipe flow with changing diameter, and velocity profiles between
flat plates. We found that there were no simulations available that covered these three areas, and which
were readily usable by students, so we had to develop our own simulations. These were written in Visual
Basic in Excel in that it would be easy for students to engage with them because of their familiarity with
Excel.
In each of the simulations, certain parameters were invariant while others were varied, and the students
could change the varying parameters to see what effect they had. The simulations were accompanied by
a worksheet of instructions and questions regarding what the students were seeing in the simulations and
the relationships between the variables in the system concerned. These questions were all qualitative,
based on our finding that students are generally good at calculations but weak on the conceptual side.
The simulations were all highly visual in nature. The first simulation (Simulation A, Figure 2) covered
pressure measurement using a Pitot tube. The second simulation (Simulation B, Figure 3) showed the
velocity and energy changes through a pipe as the diameter is decreased. The third simulation
(Simulation C, Figure 4) showed the velocity profile for a fluid between two flat plates.

Figure 2. Screenshot of Simulation A

Figure 3. Screenshot of Simulation B

Figure 4. Screenshot of Simulation C

Visualisation Study: Results
Here we will discuss the major findings of this study. Table 5 shows the change in proportions of students
getting the correct answers in different categories of the reduced FMCI. It will be noted that there was
virtually no change for the pressure measurement area covered by Simulation A, as well as for the areas
not covered in the simulations (Force balance and General fluid mechanics theory), whereas there was
what seemed to be a significant improvement in the areas covered by Simulations B and C (Flow through
a pipe and Velocity profiles).
% of students passed
Concept Category
Pre-test

Post-test

Shift

Pressure-measuring instruments

25.0

25.4

0.4

Flow through a pipe (changing diameter)

23.9

42.9

18.9

Velocity profiles (boundary conditions)

22.1

47.1

25.0

Force balance

35.4

37.7

2.3

General fluid mechanics theory

69.1

68.3

-0.9

Table 5. Comparison of results per category

In order to see if these changes were significant we performed a question-by-question statistical analysis
for differences in proportion. The z values indicated significantly different understanding of the concepts
dealt with in Simulations B and C (see Fraser, et al, 2007 for further details).
After the post-test students also completed a questionnaire about their experience of the simulations.
This shed some light onto why the impact of Simulation A was negligible in comparison to Simulations B
and C. It seems from the students comments that the way the water flow was depicted might have been
confusing, and also the way the velocity was made to increase from zero till it settled down at the final
chosen value, with the height and pressure differences also changing accordingly.
Other feedback from the questionnaire students filled in indicated a strong desire for such simulations to
be an integral part of the teaching and learning in their fluid mechanics course, at the appropriate time in
the course. They also indicated ways in which each of the simulations could be improved, and
particularly wanted to be able to play more with them and be able to change more variables (although this
would possibly be counter-productive in terms of bringing desired aspects of the phenomena being
simulated into their focal awareness). The impact of the simulations was judged by subsequent responses
to the same FMCI test used beforehand. These indicated that two of the simulations had indeed met their
objectives, but that one of the simulations had been less than successful Improvements to both the
simulations and the accompanying worksheets have been identified that will enhance their effectiveness
when they are incorporated into the teaching of the fluid mechanics course taken by the sophomore
students.

Visualisation Study: Implications

The major implications of this study are as follows.

Students did not see simulations in Excel as computational tools, in contrast with computer
simulation packages that are primarily engineering tools.

Activities that aid student visualisation of abstract concepts will foster student understanding;
these could be concrete activities or computer simulations, or both.

Appropriate use of computer simulations in the early years of study can aid student
unbdestanding of difficult engineering concepts.

Computer simulations need to be carefully designed to meet a particular learning objective.

How well students engage with a particular computer visualisation of a phenomenon is

unpredictable and needs to be carfully evaluated.

Links with theory need to be explictly built into student engagement with computer
simulations, otherwise students may not relate what tey are expwereincing to the theory they have
learned.

Conclusions
In this paper we have presented three studies which point to ways to improve student learning through
computer simulations.
In the first study the value of using the theory of learning through variation to design a simulation exercise
was demonstrated, especially in order to bring things into students focal awareness.
In the second study we showed that enjoyment and consolidation of understanding are additional factors
in facilitating learning through computer simulations.
In the third study we demonstrated the power of using visualisation to help students grasp basic concepts.
From all these studies we have inferred the following general guidelines for enhancing student learning
through computer simulations:

Special attention needs to be paid to helping students appreciate the learning potential of
computer simulations.

The earlier such simulations are introduced to students the less likely they are to view the
simulation as only a technical tool.

Important factors in choosing a package for the computer simulation are as follows:
o

Familiarity with the package (so that navigation and manipulation of the package is not a
hindrance to their learning);

Ease of use of the package (which does not necessarily equate to familiarity);

Complexity approaching reality (although this may have to be sacrificed for ease of use);

Ease of visual depiction of the phenomenon;

Ease of comparison of results; and

A learning history (in the form of an undo feature), so students can easily repeat
actions.

Before designing a computer simulation learning task, it is critical to first identify the object of
learning. This can be done by considering what aspect(s) of the phenomenon being considered
we want to bring into the focal awareness of the students. In doing this it is important not to try to
achieve too many objectives in a single exercise.

The following factors are strongly conducive to learning, and need to be built into way the
learning task is conceived:

A peer interaction environment;

Enjoyment (which promotes engagement in the exercise);

Time to play around with the simulation (both during and after the exercise); and

An ability to explore beyond the questions being asked.

In helping students to engage with the object of learning by using a computer simulation, the
following questions need to be considered:
o

Where repetition is being used is it purposeful?

How can we best encourage students to access previous knowledge and bring it to bear
on the situation being examined?

What students may be overlooking, and what they may be taking for granted, so that we
can most effectively bring what we would like into their focal awareness?

How can visualisation be used to help students grasp difficult concepts?

How can thorough instructions best be balanced with open-endedness?

How can we help students consolidate their understanding as they progress thorough the
simulation exercise (possibly through the use of a learning history as well as reflective
questions built into the simulation)?

In developing a computer simulation learning task, you also need to consider how you will
determine the effectiveness of the task (you cannot presume that it will work). This is most easily
done if there is an appropriate conceptual test available that can measure improvement.

We strongly recommend the use of these guidelines in designing computer simulation learning exercises.
Acknowledgement
I hereby acknowledge the input of my colleagues Cedric Linder and Jenni Case, as well as the work done
by the students who did these three studies as final year research projects: Heather Coombes and
Sharrol Allison (Variation), Samantha Streicher and Kate West (Conditions), and Reneta Pillay and
Lawrence Tjatindi (Visualisation). I also wish to acknowledge financial support for this work from the
South African National Research Foundation (NRF) and the Swedish International Development Agency
(SIDA).
References
Fraser, D.M. and Linder, C. (2006). Learning Through Variation: A New Way to View Learning, SEFI2006
Conference, Uppsala, 28 June 1 July.
Fraser, D.M., Linder, C.J., Allison, S., Coombes, H., and Case, J.M. (2006). Using Variation to Enhance

Learning in Engineering, International Journal of Engineering Education, 22(1), 102-108.

Fraser, D.M., Pillay, R., Tjatindi, L., and Case, J.M. (2007). Enhancing the Learning of Fluid Mechanics
Using Computer Simulations, Journal of Engineering Education, 96(4), 381-388.
Pang, M.F. & Marton, F. (2003). Beyond lesson study Comparing two ways of facilitating the grasp of
economic concepts, Instructional Science, 31 (3), 175-194.
Streicher, S.J., West, K., Fraser, D.M., Case, J.M., and Linder, C.J. (2005). Learning through Simulation:
Student Engagement, Chemical Engineering Education, 39(4), 288-295+301.