Mathematical Modelling in the Primary School

Lyn D. English
Queensland University of Technology
<[email protected]>
Changes in society and the workplace necessitate a rethinking of the nature of the
mathematical problem-solving experiences we provide our students across the grades. We
need to design experiences that develop a broad range of future-oriented mathematical
abilities and processes. Mathematical modelling, which has traditionally been reserved for
the secondary school, serves as a powerful vehicle for addressing this need. This paper
reports on the second year of a three-year longitudinal study where a class of children and
their teachers participated in mathematical modelling activities from the 5th grade through
to the 7th grade. The paper explores the processes used by small groups of children as they
independently constructed their own mathematical models at the end of their 6
th
grade.
Our ever-changing global market is making increased demands for workers who
possess more flexible, creative, and future-oriented mathematical and technological skills
(Clayton, 1999). Of importance here is the ability to make sense of complex systems (or
models), examples of which appear regularly in the media (e.g., sophisticated buying,
leasing, and loan plans). Being able to interpret and work with such systems involves
important mathematical processes that are under-represented in the mathematics
curriculum, such as constructing, describing, explaining, predicting, and representing,
together with quantifying, coordinating, and organising data. Dealing with systems also
requires the ability to work collaboratively on multi-component projects in which planning,
monitoring, and communicating results are essential to success (Lesh & Doerr, 2003).
Given these societal and workplace requirements, it is imperative that we rethink the
nature of the mathematical problem-solving experiences we provide our students in
terms of content covered, approaches to learning, ways of assessing learning, and ways of
increasing our children's access to quality learning. One approach to addressing these
issues is through mathematical modelling (Lesh, Cramer, Doerr, Post, & Zawojewski,
2003). Mathematical modelling has traditionally been reserved for the secondary school
years (e.g., Galbraith, Blum, Booker, & Huntley, 1998), but recent research (e.g., Doerr &
English, 2003) has indicated that primary school children can participate successfully in
meaningful modelling activities.
This paper draws on data from the second year of a three-year longitudinal study where
a class of children and their teachers participated in mathematical modelling activities from
the 5
th
grade through to the 7
th
grade. The paper explores the processes used by small
groups of children as they independently constructed their own mathematical models at the
end of their 6
th
grade. The activity in question was the final in a series of problem
situations that required students to create usable rating systems in a range of contexts.
Mathematical Modelling for Children
In the past couple of decades, children's problem solving has engaged them in
situations where the givens, the goals, and the legal solution steps have been
specified clearly; that is, the interpretation processes for the child have been minimized or
eliminated. The difficulty for the solver is simply working out how to get from the given
state to the goal state. The solutions to these problems are usually brief answers obtained
from applying a previously taught solution strategy, such as guess and check, or draw a
207
diagram. Furthermore, although these problems may refer to real-life situations, the
mathematics involved in solving them is often not real world and rarely do the problems
provide explicit opportunities for learners to generalize and re-apply their learning (English
& Lesh, 2003). While not denying the importance of these problem experiences, they do
not address adequately the knowledge, processes, and social developments that students
require in dealing with the increasingly sophisticated systems of our society. Mathematical
modelling activities, in the form of meaningful case studies for children, provide one way
in which we can overcome this inadequacy.
As used here, models are systems of elements, operations, relationships, and rules that
can be used to describe, explain, or predict the behaviour of some other experienced
system (Doerr & English, 2003). Model-eliciting activities, which provide the basis for
subsequent model-exploration and model-application activities, engage children in
situations where key mathematical constructs are embedded within the problem context
and are elicited by the children as they work on the problem. Model-eliciting activities
present situations where children (a) are confronted with the need to develop a model, (b)
clearly recognise the need to revise or refine their current ways of thinking about the given
problem situation, (c) are challenged to express their understandings in ways that they can
test themselves and revise as often as necessary, and (d) develop models that can be shared
with others and that can be applied in other problem situations (Lesh & Yoon, 2004).
Engaging students in modelling experiences of this nature is not seen as simply finding a
solution to a given isolated problem. Rather, such engagement involves children in
multiple activities where significant mathematical constructs are developed, explored,
extended, and applied; the end product is a system or model that is reusable in a range of
contexts (Doerr & English, 2003).
Unlike traditional non-routine problems, modelling activities are inherently social
experiences, where students work in small teams to develop a product that is explicitly
sharable. Numerous questions, issues, conflicts, revisions, and resolutions arise as students
develop, assess, and prepare to communicate their products. Because the products are to be
shared with and used by others, they must hold up under the scrutiny of the team members
(Zawojewski, Lesh, & English, 2003).
Design and Methodology
The present study took the form of a three-year, longitudinal teaching experiment
involving multilevel collaboration (English, 2003; Lesh & Kelly, 2003). At the first level
of collaboration children work on activities involving constructing, refining, and applying
mathematical models. At the second and third levels, participating preservice teachers
(university undergraduates) and the classroom teachers work collaboratively with the
researchers in designing and implementing the childrens activities. These activities also
serve as challenging and thought-provoking experiences for the teachers as they explore
the nature of the mathematical ideas being developed, consider appropriate implementation
strategies, and promote learning communities within their classrooms. The focus of the
present paper, however, is level 1, that is, the childrens model construction.
During the second year of the study (2002; from which data have been drawn for this
paper), a series of modelling activities was implemented in the 6
th
grade classroom from
May through to the end of October. One session per week was conducted, with the sessions
ranging from 50 minutes to an hour. Meetings with the class teacher were held prior to the
introduction of each new modelling activity. The class teacher introduced each modelling
activity (described next) in a whole class format, which was followed by small group work.
208
The teachers and researchers observed the children as they worked the activities. Where
appropriate, the observers might ask the children to explain or justify a response but
mostly, the observers remained in the background. No explicit teaching was given to the
groups. At the end of the activity, each group of children shared with the class their
approaches to working the activities, explained and justified the model they had developed,
and then invited feedback from their peers. This group reporting was followed by a whole
class discussion that compared the features of the mathematical models produced by the
various groups.
Modelling Activities
The series of model development activities that was implemented across the year
comprised an initial model-eliciting activity (Sneakers Problem, described in Doerr &
English, 2003), a model exploration activity (Weather Problem, also described in Doerr &
English, 2003), and two model-application/model-adaptation activities (Consumer Guide,
and Car Problem, the latter of which appears as the appendix). The last two activities
provide opportunities for children to transfer and refine the mathematical constructs they
have developed in the previous tasks, as well as to refine any representational systems.
The activities involve the core mathematical ideas of ranking, weighting ranks, and
selecting and aggregating ranked quantities. These ranking processes entail analysing and
transforming entire data sets or meaningful portions thereof, rather than single data points.
The sequence is designed such that children can engage in meaningful ways with the
problem situation and can create, use and modify quantities (e.g., ranks) in ways that are
meaningful to them and in ways that can be shared, generalized, and re-used in new
situations. The children in the present study had no specific formal exposure to or
instruction on these core mathematical ideas prior to commencing the activities.
Data Collection and Analysis
Data sources included videotapes of whole class sessions and audio- and videotapes of
small group responses to each of the modelling activities. Field notes, children's work
sheets, and final reports detailing their models and how they developed them were also
important data sources. The tapes were transcribed and, for the present paper, analysed for
evidence of the processes that the children used in constructing their mathematical models.
Results
Consideration is given first to the modelling processes displayed by one representative
group of children (referred to as Jasmines group) as they worked the Car Problem. Next,
a brief description of the models that were produced by other groups is presented.
Model Development by One Group
An analysis of the interactions of Jasmines group as they worked the Car Problem
indicated a number of diverse processes through which the children cycled as they
progressed towards their final model. More specifically, the children displayed the
following processes: interpreting and re-interpreting the problem statement, clarifying and
revisiting the goal, making decisions subjectively versus objectively, applying and refining
mathematical procedures, posing hypotheses and problems, making assumptions (false or
otherwise), posing arguments and counterarguments, considering all options (versus
209
considering limited options), asking for justification, thinking metacognitively, and
applying previous models.
In the following excerpt, Jasmines group is commencing the Car Problem with
Jasmine restating the goal. In doing so, she makes an incorrect assumption regarding Carls
mothers monetary input; this was not detected by the other group members. Charlotte and
Jasmine then suggest beginning with a process of elimination, which instigates a comment
from Rachel regarding the need to consider all factors:
Jasmine: Our job is to find a car for Carl and his mothers paying half of it and she doesnt want it
to cost much and we have to make a list and decide which one is best to buy.
Charlotte: I think we should do a process of elimination. (A brief discussion ensued regarding the
mother wanting the car to be reliable.)
Jasmine: Maybe first, maybe we should do a process of elimination so work our way down the
list or work our way up.
Rachel: We have to consider all the factors though.
Jasmine: Yeah, but were going to.
Rachel: Before you start the elimination process you should number the things first then do it. Then
do the eliminations.
The group subsequently moved off task for a short period, discussing what they
considered to be the best car listed in the table and which car the mother would like.
Charlotte reminded the group that the car is just for Carl, with Jasmine responding,
Yeah, but he gets to choose. Shes not going to make him. Douglas brought the group
back on task, indicating that they needed to be objective just as they needed to be in the
last modelling activity (The Consumer Guide problem involving the snack chips):
She (the mother) is helping him. How about we just judge off what the thing says, not by what we
think. It goes with the chips too you know. Lets all read through this again and underline all the
details that help us.
The group spent a number of minutes revisiting the goal and re-interpreting the
problem information. The issue of what Carl needs versus what he wants occupied their
discussion (You cant just think about what he wants, you have to think about what he
needs as well.). This discussion led the group to reconsider their elimination process and
what they should eliminate first. Considerable argumentation took place as to whether or
not the most expensive car (the Honda Legend) should be eliminated. While Charlotte and
Jasmine claimed They are not going to buy the most expensive car so we could cross out
the most expensive car but cross out no more than one, Rachel felt they shouldnt delete it
because, The Honda Legend might just have one bad thing and the rest is good. Douglas
and Charlotte tried to resolve this dilemma by revisiting the problem information and again
emphasising the need to be objective rather than subjective: Yeah, but we are judging off
what this thing (problem description) says. They dont want it to cost that much and its
like the most expensive car. The argumentation continued however, with Rachel claiming
that It (the Honda Legend) could have the best things in it. Rather than entertain Rachels
comment, Douglas repeated his statement regarding objectivity: We dont know that.
Were just judging on facts.
Rachel persisted with her claim that there are other factors to consider before
eliminating the Honda Legend and persuaded the group to consider its fuel consumption
and mileage, followed by safety factors. Argumentation again followed on which were the
important safety factors for the Honda Legend. Douglas, however, was still concerned
about the problem criteria: They also said they dont want something expensive and thats
210
the most expensive car. weigh the choices up. The remaining group members in the
meantime were hypothesising that Carls mother might think the car is worth the money
because its got all those safety things and its going to help. During the argumentation,
Jasmine had been examining the table of data herself and declared that the Ford Turbo and
Honda Legend were possibilities. Rachel demanded that Jasmine justify her claims:
Jasmine, tell us why you think these things. We need to know why you think them.
While accepting Jasmines explanation, Rachel felt the group was not making
sufficient progress. Hence, another group member, Lisa, suggested that the members vote
to resolve their disagreements. The others, however, did not agree: We cant just vote.
Thats not logical. Jasmine then continued to explain the ranking system that she had
developed: The year.the later the year, probably the newer the model so its more in
and so I just did a ranking system of 1,2,3,4,5,6,7, 89and yes, the best one of that was
the Hyundai Excel GX X3. Rachel, however, reminded Jasmine that We have to write a
list. It says your job is to create a list for Carl and for his mother. The group then reverted
to argumentation over cost factors versus leisure features of the cars, with Lisa suggesting
that they consider their own perspective to help resolve their disagreement:
Maybe we should decide if we were buying a car what would we like.I know were not buying
the car but sometimes that actually helps to think about what we would buy.
Once again, Douglas emphasised:
Were not deciding on what we like, were deciding on the facts.we have to look at these factors.
In the next session, however, Jasmines group decided to resolve this dilemma by
voting on what they considered to be the most important features to address. They then
constructed Tables 1 and 2 (reproduced from the groups computer file). The groups
explanation of their model development appears below Table 1. Notice how the group
referred back to the system they used in the previous modelling activity (the chips, i.e.,
the Consumer Guide problem).
Our group used a rating system like the chips. We voted the most important features and then we
put them into a table. We put them into categories as safety features, leisurely features, and extras.
Then we ticked the cars that had those features as shown. (Hold up picture of table.) Next we rated
the mileage. The best was the highest and the worst was the lowest rating. After we came up with
the results we did a draft list for Carl and his Mother and also a separate one. We did this by adding
the features and the mileage together and the highest was the best. After we got the scores from the
table we numbered them 1st, 2nd and so on then we put them in a list for carl and his mum. After
doing that we wrote the total so that there was a combined list for carl and his mum. The best for
both of them and each of them was the Daewoo Lanos (see Table 2).
211
Table 1
Car Features Table
CAR Antilock
Brakes
Air bags Air
conditioner
Alloy
wheels
Power
Windows
Electric
Sunroof
Nissan Silva N/A N/A N/A YES YES N/A
Ford Capri
Turbo
N/A N/A N/A YES YES N/A
Audi 90 Sport N/A N/A YES YES N/A YES
Ford falcon
EAS
N/A N/A YES N/A YES N/A
Nissan Pulsar
LX
N/A N/A YES N/A N/A N/A
Hyundai
Excel GX
N/A N/A YES N/A N/A N/A
Daewoo
Lanos SE
N/A N/A YES N/A N/A N/A
Honda
Legend
YES YES N/A N/A YES YES
BMW 318i
E36
YES N/A YES YES N/A YES
Table 2
Car Preferences for Carl and His Mum
Name of Car Carl Mother Total
Nissan silver 4
th
4
th
4
th
Ford Capri turbo 4
th
4
th
4
th
Audi 90 8
th
4
th
6
th
Ford falcon 3
rd
3
rd
3
rd
Nissan pulsar 5
th
5
th
5
th
Hyundai excel 2
nd
2
nd
2
nd
Daewoo lanos 1
st
1
st
1
st
Honda legend 7
th
6
th
7
th
BMW 6
th
7
th
7
th
Model Development by Other Groups
The remaining four groups varied in their interpretation of the problem information, in
particular, in their interpretation of the given criteria (safe, reliable, fun etc) and in
identifying factors to define those criteria. While all groups considered the criterion of
safe to be significant, only two groups also specifically addressed reliable. One group
defined reliable by the year of make and mileage, and listed the cars in descending order
with respect to each of these factors. This same group assigned points to the ranked cars as
follows: 1
st
= 5 points, 2
nd
= 4 points, 3
rd
= 3 points, 4
th
= 2 points, 5
th
= 1 point. To
determine the car suitable for Carl, the group totalled the points assigned to the cars for
each of the factors, fun, cheap, and good fuel mileage. They did the same for Carls
mother (i.e., totalling the points for safe and for reliable).
The other group began with two categories, namely, Carl (Fun) and Mother (Safe)
and, as they explained, rated the cars according to how many features they had. They
212
then created a third category, Cars for Both, which comprised four cars. The group
subsequently eliminated three of the four cars by mainly looking at the mileage and age.
Concluding Points
Modelling problems of the present type provide rich opportunities for children to
engage in a range of mathematical processes, along with the development of important
mathematical constructs that are embedded within the problem. Because modelling
activities are designed for small group work, they are ideal vehicles for developing
collaborative problem-solving skills. This paper has provided examples of the
mathematical and social processes that children display when working meaningful
modelling problems. These processes include interpreting and re-interpreting the problem
information (including the goal statement), making appropriate decisions, justifying ones
reasoning, posing hypotheses and problems, presenting arguments and counterarguments,
applying previous learning, and acting metacognitively. The Car Problem was the final
activity in a series of problems that focused on the mathematical ideas of ranking,
weighting ranks, and selecting and aggregating ranked quantities. The children applied
these ideas in different ways to generate models independently of instruction.
References
Clayton, M. (1999). What skills does mathematics education need to provide? In C. Hoyles, C. Morgan,
& G. Woodhouse (Eds.), Rethinking the mathematics curriculum (pp. 22-28). London: Falmer Press.
Doerr, H., & English, L. D. (2003). A modeling perspective on students' mathematical reasoning about
data. Journal for Research in Mathematics Education,34(2), 110-136.
English, L. D. (2003). Reconciling theory, research, and practice: A models and modelling perspective.
Educational Studies in Mathematics, 54, 2 & 3, 225-248.
English, L. D., & Lesh. R. A. (2003). Ends-in-view problems. In R. A. Lesh & H. Doerr (Eds.), Beyond
constructivism: A models and modelling perspective on mathematics problem solving, learning, and teaching
(pp.297-316). Mahwah, NJ: Lawrence Erlbaum.
Galbraith, P. L., Blum., W., Booker, G., & Huntley, I. D. (1998). Mathematical modelling: Teaching
and assessment in a technology-rich world. West Sussex: Horwood Publishing Ltd.
Lesh, R. A., & Doerr, H. (2003). Foundations of a Models & Modeling Perspective on Mathematics
Teaching and Learning. In R. A. Lesh & H. Doerr (Eds.), Beyond constructivism: A models and modelling
perspective on mathematics teaching, learning, and problem solving (pp. 3-34).. Mahwah, NJ: Erlbaum.
Lesh, R. A., & Kelly, A. E. (2000). Multi-tiered teaching experiments. In R. A. Lesh & A. Kelly (Eds.),
Handbook of research design in mathematics and science education (pp. 197-230. Mahwah, NJ: Erlbaum
Lesh, R., & Yoon, C. (2004). What's Distinctive in (Our Views about) Models & Modeling Perspectives
on Mathematics Problem Solving, Learning and Teaching? In H. Henn & W. Blum (Eds.), ICMI Study 14:
Applications and Modeling in Mathematics Education. Pre-Conference Volume.
Lesh, R., Cramer, K., Doerr, H., Post, T., & Zawojewski, J. (2003). Model development sequences. In R.
A. Lesh & H. Doerr (Eds.), Beyond constructivism: A models and modelling perspective on mathematics
teaching, learning, and problem solving (pp. 35-58). Mahwah, NJ: Lawrence Erlbaum.
Zawojewski, J. S, Lesh, R., & English, L. D. (2003). A models and modelling perspective on the role of
small group learning. In R. A. Lesh & H. Doerr (Eds.), Beyond constructivism: A models and modelling
perspective on mathematics problem solving, learning, and teaching (pp. 337-358). Mahwah, NJ: Erlbaum.
213
Appendix
What Car to Buy?
Carl and his mother have been out shopping for cars. Carl wants a car that will be fun to
drive around in, gets good gas mileage, but doesn't cost too much. But Carl's mother, who
is going to help pay for the car, wants him to have a car that is reliable and safe. Your job
is to create a list for Carl and a list for his mother showing which cars are the best. Then
they will have to decide which one to buy!
Car Information
Car Year Cost Color Mileage
litres/
100 km
City
Features Body Style
Nissan
Silva
1992 10,000
Navy
Blue
96,000 10
Rear Spoiler, Power
Windows, Power Steering,
CD Player, Alloy Wheels,
Alarm
Coup
Ford Capri
Turbo
1989 8,200 Red 105,000 9
Rear Spoiler, Power
Windows, Power Steering,
CD Player, Alloy Wheels,
Alarm
Convertible
Audi 90
Sport
1991 9,500 Silver 97,500 10.5
Rear Spoiler, Power
Windows, Power Steering,
CD Player, Alloy Wheels,
Electric Sunroof, AC
Sedan
Ford
Falcon EA
S
1988 5,200
Pale
Blue
113,500 11.5
Power Steering, Radio
Cassette, Tow Bar, AC
Sedan
Nissan
Pulsar LX
1993 7,950 Gold 125,000 7.5
Power Steering, Radio
Cassette, Tinted Windows,
Bull Bar, AC
Sedan
Hyundai
Excel GX
X3
1999 9,500
Dark
Blue
49,000 7.6
Power Steering, CD Player,
Rear Spoiler, AC, Tinted
Windows
Hatchback
Daewoo
Lanos SE
1997 7,250
Azurite
Blue
74,118 8.8
6 Stacker CD player, 6
speakers, Amplifier, Tinted
Windows, AC
Hatchback
Honda
Legend
1993 17,200
Dark
Green
154,000 12.5
Dual Airbags, Antilock
Brakes, Alarm, Cruise
Control, Electric Sunroof,
Power Windows, Power
Steering,
Sedan
BMW
318i E36
1991 15,000 Blue 164,000 9.5
Radio Cassette, Power
Steering, Electric Sunroof,
AC, Alloy Wheels,
Antilock Brakes,
Sedan
214