Algebraic Thinking Of Grade 8 Students In Solving Word Problems With A Spreadsheet

CERME 7 (2011)
ALGEBRAIC THINKING OF GRADE8 STUDENTSIN SOLVING
WORD PROBLEMS WITH ASPREADSHEET
Sandra Nobre, Escola E.B. 2, 3 Professor Paula Nogueira
Nélia Amado & Susana Carreira, FCT, Universityof Algarve &UIDEF
João Pedro da Ponte, Institute of Education, University of Lisbon
This paper describes and discusses the activity of grade 8studentson two word
problems, using a spreadsheet.We look at particular uses of the spreadsheet, namely
at the students’ representations, as ways of eliciting forms of algebraic thinking
involved in solving the problems ,which entailed dealing with inequalities .We aim to
see how the spreadsheet allows the solution of formally impracticable problems at
students’ level of algebra knowledge, by making them treatable through the
computational logic that is intrinsic to the operating modes of the spreadsheet.The
protocols of the problem solving sessions provided ways to describe and interpret
the relationships that students established between the variables in the problems and
their representations in the spreadsheet.
Keywords: algebraic thinking, inequalities, spreadsheet, representations.
INTRODUCTION
Representations have a dual role in learning and in mathematical communication.
These resources serve the purpose of communicating with others about a problem or
an idea but also constitute tools that help to achieve an understanding of a property, a
concept or a problem (Dufour-Janvier, Bednarz& Belanger, 1987). This is one of the
reasons why we consider students’ use of representations as a lens from which we
can grasp the meaning involved in the mathematical processes of solving a problem.
Spreadsheets have great potential for the construction of algebraic concepts,
including the establishment of functional relationships, the representation of
sequences or the use of recursive procedures in solving mathematical problems. The
use of spreadsheets in problem solving has been deeply investigated by several
authors (e.g., Ainley et al., 2004;Rojano, 2002) and revealed interesting processes in
the development of algebraic thinking, particularly with regard to the transition from
arithmetic to algebra. Within a spreadsheet environment, the symbolic representation
of the relations present in a problem is initiated through the nomination of columns
and writing of formulas. This is considered a stimulating environment that fosters an
understanding of the relations of dependence between variables and encourages
students to submit solutions gradually more algebraic and moving away from
arithmetical methods (Rojano, 2002).These aspects encouraged us to carry out an
analysis of how grade 8 students create their representations, how they conceive and
display the problem conditions on the spreadsheet and how they achieve a solution.

Working Group 3
CERME 7 (2011) 522
PROBLEM SOLVING AND THELEARNING OF ALGEBRA
Contextual problem solving is an important type of task leading to algebraic activity.
According to Kieran (2004) the work in algebra can be divided into three areas:
generational, transformational and global/meta-level activities. Generational
activities correspond to the construction and interpretation of algebraic objects.
Transformational activities include simplifying algebraic expressions, solving
equations and inequalities and manipulating expressions. Finally, global/meta-level
activities involve problem solving and mathematical modelling, including pattern
generalization and analysis of variation.
The nature of algebraic reasoning depends on the age and mathematical experience
of the students. Students at a more advanced level may naturally use symbolic
expressions and equations instead of numbers and operations. But for students who
have not yet learned the algebraic notation, the more general ways of thinking about
numbers, operations and notations, may be effectively considered algebraic (Kieran,
2007). Contexts that involve numbers, functional relationships, regularities, and
other properties, are an essential foundation for the understanding of algebraic
structures. For instance, writing symbolic numerical relations may favour the use of
letters. However, the use of technological tools allows other representations for such
relations, as well as new forms of exploration, which may be seen as analogous to
generational and transformational activities in algebra. Thus, it seems appropriate
that such new representations, and the mathematical thinking associated with them,
are included in the field of algebra (Kieran, 1996).Moreover, Lins& Kaput (2004)
claim that algebra can be treated from the arithmetic field, since there are many
properties, structures and relationships that are common to these two areas.
Therefore, arithmetic and algebra may be developed as an integrated field of
knowledge.In this study we adopt this perspective, considering algebraic thinking as
a broad way of thinking that is not limited to the formal procedures of algebra.This
entails separating algebraic thinking from algebraic symbolism (Zazkis&Liljedhal,
2002).
SPREADSHEETS IN THE DEVELOPMENT OF ALGEBRAIC THINKING
A spreadsheet supports the connection between different registers (numerical,
relational, and graphical). One feature that stands out in this tool is the possibility of
dragging the handle of a cell containing a formula along a column. This action
generates a “variable-column”. Using this tool in problem solving emphasizes the
need to identify the relevant variables and encourages the search for relations of
dependence between variables. The definition of intermediate relations between
variables, that is, the breakdown of complex dependency relations in successive
simpler relations is a process afforded by this tool, with decisive consequences in the
process of problem solving (Carreira, 1992; Haspekian, 2005). As noted by
Haspekian (2005) a spreadsheet also allows an algebraic organization of apparently

Working Group 3
CERME 7 (2011) 523
arithmetical solutions and this kind of hybridism, where arithmetic and algebra
naturally cohabit, becomes an educational option that may help students in moving
from arithmetic to algebra(Kieran, 1996).
We want to see how this particular functioning of the spreadsheet is a valid route for
solving problems where the formal algebraic approach is too heavy for the students’
level. More specifically, we aim to understand how far the spreadsheet, while being a
means to promote algebraic thinking, can relieve the burden of formal algebraic
procedures and as such can advance the possibility of solving certain types of
problems. So far, research has shown the value of the spreadsheet in the transition
from arithmetic thinking to algebraic thinking, but less is known about the utility of
the spreadsheet to set up an alternative to formal and symbolic algebra and yet
allowing the development of students’ algebraic thinking in problems that are
formally expressed by inequalities (Carreira, 1992; Haspekian, 2005; Rojano, 2002)
METHODOLOGY
This study follows a qualitative and interpretative methodology. The participants are
three grade 8students (13-14 years). They had some previous opportunities to solve
word problems with a spreadsheet in the classroom, from which they acquired some
basics of the spreadsheet operation.Before the two tasks here presented, students had
worked with the spreadsheet in solving other problems for six lessons. All problems
involved relationships among variables (usually equations) and only one included a
simple linear inequality. The detailed recording of student’s processes was achieved
with the use of Camtasia Studio. This software allows the simultaneous collecting of
the dialogue of the students and the sequence of the computer screens that show all
the actions that were performed on the computer. We were able to analyze the
students’ conversations while we observed their operations on a spreadsheet. This
type of computer protocol is very powerful as it allows the description of the actions
in real time on the computer (Weigand& Weller, 2001).
The two problems
King Edgar of Zirtuania decided to divide their treasure of a thousand gold bars by
his four sons. The royal verdict is:
1- The 1st
son gets twice the bars of the 2nd
son.
2 - The 3rd
son gets more bars than the first two together.
3 - The 4th
son will receive less than the 2nd
son.
What is the highest number of gold bars that the 4th
son of the king may receive?
Figure 1: The treasure of King Edgar
From small equilateral triangles, rhombuses are formed as shown in
the picture. We have 1000 triangles and we wish to make the biggest
possible rhombus.
How many triangles will be used?
Figure 2: Rhombuses with triangles

Working Group 3
CERME 7 (2011) 524
A possible algebraic approach to the problems is presented in table1.Solving these
problems by a formal algebraic approach, namely using in equalities and systems
such as these was beyond the reach of these students. Therefore, it is important to see
which roads are opened by using the spreadsheet.
The treasure of King Edgar Rhombuses with triangles
child
4
1
of
bars
of
number
max
´
2
1000
4
2
4
2
1
3
2
1
4
3
2
1
K
1
nu
4
1
1
´
2
10
i
i
s
s
s
s
s
s
s
s
s
s
s
s
s
i
number
figure
max
1000
2 2
fi
10
n
n
n
Table 1: Algebraic approach to the problems
Problem 1 contains several conditions that relate to each other and the statements“
gets more” and “receives less” involve an element of ambiguity and make the
problem complex, for understanding it, for translating into algebraic language and
for solving it. Problem 2entails a pictorial sequence that can be translated
algebraically into a single condition. However, this condition involves a quadratic
function that does not arise immediately after reading the statement of the problem.
These two problems represent instances of global/meta-level activities considered by
Kieran (2004), insofar as they involve functional reasoning and pattern finding
strategies. They both have in common the search for a maximum value, leading to
some difficulties when a purely algebraic approach is envisioned. However, a
spreadsheet provides alternative approaches to both problems that may make them
clearer to students, facilitating their solution process and efficiently providing a
solution. We examine how students approached these problems in the classroom, the
strategies they used, how they connected the variables involved and expressed that
on a spreadsheet. Excerpts of Excel computer protocols are offered to further clarify
the description of students’ activity.
In solving problem 1, Marcelo assigned and named a column to each of the four sons
and a fifth column for the total of gold bars (table 2). Then, he started writing values
in the cells corresponding to the sons in the following order: 2nd
, 1st
, 4th
and 3rd
, as
follows: choosing a value for the 2nd
son, then mentally doubling it for the 1st
son;
subtracting one unit to the 2nd
son’s number of bars to get the 4th
son’s; add the three
values of the 2nd
, 1st
, and 4th
sons and calculate the difference to 1000tofind the
3rd
son’s number.In another column, the student entered a formula that gives the total
of gold bars and served as control for the total number of bars (1000).

Working Group 3
CERME 7 (2011) 525
xls file xls file (command “show formulas”)
Table 2: Print screen of Marcelo’s representation
Although Marcelo did not display the relations between the number of bars of the
four brothers -using formulas or otherwise-he kept them always present in his
thinking. The task required a greater effort for the student, since in each attempt he
had to recall the relations, while carrying out the calculations mentally.
Marcelo: Teacher, I found the best! [The value 139 was obtained in cell G6]. If I
choose 150 [for the 2nd
son]it won’t do. I’ve tried it.
Teacher: But this is not the maximum number of bars for the 4th
son, is it?
Marcelo: I went from 100 to 150, and it turns out that 150 gets worse because the
other gets over 450 and the last one falls to 99.
The teacher asked Marcelo to do more experiments to which he replied that he had
already made some, for example 160 and 170. So she made another suggestion:
Teacher: Here you already got an excellent value and it increased significantly from
130 to 140 [referring to column E]. So, try around these values.
The student continued to experiment, always doing the calculations mentally. He
found 141, confirming that it was the best. As an answer the student wrote:“I solved
this problem taking into account the conditions of the problem, making four
columns, one for each child, and trying to find a higher number”.
In our view, Marcelo has developed algebraic thinking by focusing on dependence
relationships between different variables to finding the optimal solution. As he
stated, he took into account the five conditions of the problem and expressed them in
the spreadsheet columns. From the standpoint of an algebraic approach, the student
began by choosing an independent variable (the 2nd
son’s number of bars) and
established relationships to express the number of bars for each remaining son.
2
s
2
1 2s
s 1
2
4 1
s
s
4
2
1
3 1000 s
s
s
s s
s
s
10 1000
4
3
2
1 10
s
s
s
s
1
s1 2
2

Working Group 3
CERME 7 (2011) 526
The diagram above summarizes the translation of the student’s algebraic thinking in
solving the problem and shows how Excel allowed dealing with simultaneous
manipulation of several conditions, by means of numbers, rather then with letters and
symbolic algebra. It is important to note that the condition set for the 4th
son
demonstrates an understanding of looking for the highest possible value, given that
the difference down to the 2nd
wasonly one bar.
Maria and Jessica (Table 3) started to solve the problem like Marcelo, with the
allocation of columns to the number of gold bars for each son and another column
for the total of bars. Then, they created a column of integers for the number of bars
of the 2nd
son; the number of bars of the 1st
son was obtained by doubling the 2nd
son’s;
the number of bars of the 3rd
son was found by adding a unit to the sum of the 1st
and
2nd
sons’ bars; the number of bars of the 4th
son was obtained by subtracting one unit
to 2nd
son’s; finally, the last column computed the sum of bars of the four sons.
Table 3: Print screen of Maria e Jessica’s representation
At one point the students got a value higher than1000 in the last column and
concluded that it was necessary to remove a bar from one of the sons. Yet, it was
necessary to realize that one bar could only be taken from the 4th
in accordance with
the terms of the problem.
Maria: It shows 1001, it is wrong!
Teacher: And now?
Maria: Take one out! Take one out from the 4th
!... The largest number of bars that
the 4th
can receive is 141.
The diagram below shows the relations as they would be expressed symbolically in
algebraic language:
These students formulated two conditions intended to obtain the optimal solution:
first, the difference between the number of bars of the 4th
and the 2nd
son must be one
2
s
2
1 2s
s 1
2
4 1
s
s
1
1
2
3 1
s
s
s 1000
4
3
2
1 10
s
s
s
s
s1
s 2

Working Group 3
CERME 7 (2011) 527
unit and, second, the difference between the number of bars of the 3rd
and the sum of
the 1st
and 2nd
must also be one unit.
In both solutions, the data show that students use relationships between variables but
they do it with numbers through the use of the spreadsheet. The fact that they are
working with numbers does not deviate them from the mathematical structure of the
problem. On the contrary, it helps them to better understand the problem and to deal
with a set of simultaneous conditions of different nature: equations, inequalities and
a free variable. We believe that the thinking involved in either approach is consistent
with the perspective of Kieran (2007) and Lins& Kaput (2004) on genuine algebraic
thinking development.
For the second problem, Marcelo (Table 4) started to introduce the inputs 2, 8, 18.
Then, he selected these three cells as a cluster and tried to drag them (Figure 3),
noticing that the numbers generated were not all integers.
Table 4: Print screen of Marcelo’s representation
He eventually abandoned the dragging and called the teacher:
Marcelo: I don’t know if this works... How do I do this? Is there an easier way?
Teacher: To move from the 1st
to the 2nd
how much did you add?
The student writes in cell E4 the number 6.
Teacher: And from the 2nd
to the 3rd
how much do you add?
The student wrote in cell E5 the number 10, followed by 14 and 18.
Teacher: What are you going to do now?
Marcelo: If I pull it down [referring to column E] and then by adding this column plus
this one [referring to column C and column E]...
The student inserted the formula “=C4+E4” (below the first term of the sequence)and
generated a variable-column:
Marcelo: 968! It’s what we will use from 1000. We have 1000, so it can’t be more
than 1000 and 1058 already exceeds.

Working Group 3
CERME 7 (2011) 528
The student tried to find a pattern in the number of triangles. The construction of
additional figures did not help the student to find a pattern based on the figure. One
useful approach was to look at the differences between the consecutive terms.
From an algebraic point of view, this student is using a recursive method to generate
the sequence of triangles with the help of the arithmetic progression which gives the
difference between consecutive terms. Excel easily allows handling a recursive
approach. Somehow it was no longer necessary to find nth
element to solve the
inequality, although the mathematical structure of the problem remained visible.
Maria and Jessica (table 5) addressed the problem with a similar strategy, noticing
that dragging the values 2, 8 and 18 did not produce the sequence of rhombi
presented in the problem. At one point they called the teacher:
Table 5: Print screen of Maria and Jessica’s representation
Jessica: From this one to this one it goes 6 and from this one to this one it goes 10.
Maria: From 2 to 8 it goes 6…From 8 to 18it goes 10.
Jessica: Oh teacher, we don’t know how to continue.
Teacher: Have you already drawn the next figure to see if there is any relation?
They drew it on paper, but only half of the picture.
Maria: 14, 15 and 16. Then 16 plus16 is…32
Teacher: And now?... How many will the next one have?
Maria: 50.
The students were still looking for a relation between the numbers.
Maria: I know what that is... Look... The link is…
n
n
n
n
y
x
x
n
y
x
1
1
2
4
2

Working Group 3
CERME 7 (2011) 529
Jessica: The number plus 4.
This was the decisive moment to build the column with the differences between
consecutive rhombuses, and then a formula for the number of triangles.
CONCLUDING REMARKS
Our main aim was to understand the role of the spreadsheet in solving two word
problems, which are expressed by inequalities, and examine how the solutions reflect
students' algebraic thinking, regardless of the use of algebraic symbolism. It was not
our intention to consider what students have done without the use of technology,
since any of the problems demanded an algebraic knowledge that was beyond the
level of students. In any case pencil and paper solutions could certainly come up with
methods based on trial and error. We interpreted the students’ processes based on the
spreadsheet in light of what would be a possible use of symbolic algebra. Thus we
intended to make clear students’ algebraic thinking in establishing the relationships
involved in the problems. In the first problem, four columns corresponded to the four
sons and the column for the 2nd
son was reserved for the introduction of initial values
(the input), serving as a column for the independent variable. The remaining columns
were constructed through relations of dependence. For the second problem, the
students were not able to express the general term of the sequence, but by counting
the number of triangles in the sequence of rhombuses they used the differences
between consecutive terms to generate the former sequence recursively. As reported
in some studies students when confronted with more demanding sequences tend to
use the difference method (Orton & Orton, 1999).We found that the spreadsheet
helped the students to establish relations between variables, expressed through
numerical sequences generated by the computer, and also with the use of formulas to
produce variable-columns. We claim that algebraic thinking was fostered by the
affordances of the Excel in generating the rules of the problems. This result resonates
with other investigations such as Ainley et al. (2004) but it also highlights the
structure of students’ algebraic thinking expressed in a particular representation
system. It provided a clear indicator of how students interpreted the problems in light
of their mathematical knowledge and their knowledge of the tool. The analysis
allows us to make inferences about what is gained in using Excel to solve algebraic
problems, and helps to understand the relationship between the symbolic language of
Excel and the algebraic language. The use of Excel can be seen as means to fill the
gap between the algebraic thinking and the ability to use algebraic notation to
express such thinking. The lack of algebraic notation and formal algebra methods
does not prove the absence of algebraic thinking. The kind of algebraic thinking that
emerges from the use of the spreadsheet is the kind that belongs to global algebraic
activities (Kieran, 2004).We highlight the following features of the spreadsheet in
algebraic problem solving: (i) It was a way to anticipate complex algebraic
problems; our study shows how the spreadsheet was a tool that allowed 8th
grade
students to solve two problems that were impracticable from the point view of formal

Working Group 3
CERME 7 (2011) 530
algebra. On the other hand it anticipated forms of algebraic reasoning involved in the
problems that were elicited by the representation systems embedded in the
spreadsheet; (ii)It helpedto understand the conditions in the problems; students
clearly understood the relations between the several variables involved and were able
to express such conditions and restrictions appropriately. These were not expressedin
algebraic notation but instead with the language of Excel (iii)It led to a numerical
approach of an algebraic problem; students found ways to represent the problem
through numerical variable-columns without loosing the structure of the
problems.Our perspective of algebraic thinking stresses the distinction between
algebraic notation and algebraic structures, separated by a gap that is often
underestimated. We suggest that this gap can be gainfully filled with suitable
spreadsheet activities.
Rather than insisting on any particular symbolic notation, this gap should be accepted and
used as a venue for students to practice their algebraic thinking. They should have the
opportunity to engage in situations that promote such thinking without the constraints of
formal symbolism (Zazkis&Liljedhal,2002, p. 400).
REFERENCES
Ainley, J., Bills, L., & Wilson, K. (2004). Construting meanings and utilities within
algebraic tasks. In M. J. Høines& A. B. Fuglestad (Eds), Proceedings of the
28th
PME Conference (Vol. 2, pp. 1-8). Bergen, Norway.
Carreira, S. (1992). A aprendizagem da Trigonometria num contexto de aplicações e
modelação com recurso à folha de cálculo (Tese de Mestrado). Lisboa: APM.
Dufour-Janvier, B., Bednarz, N., & Belanger, M. (1987). Pedagogical considerations
concerning the problem of representation. In C. Janvier (Ed.), Problems of
representation in the teaching and learning of mathematics (pp. 109-122).
Hillsdale, NJ: Lawrence Erlbaum.
Haspekian, M. (2005). An ‘instrumental approach’ to study the integration of a
computer tool into mathematics teaching: The case of spreadsheets. International
Journal of Computers for Mathematical Learning,10(2), 109-141.
Kieran, C. (1996). The changing face of school algebra. In C. Alsina, J. M. Alvares,
B. Hodgson, C. Laborde& A. Pérez (Eds.), International Congress on
Mathematical Education 8: Selected lectures (pp. 271-290). Seville: SAEM
Thales.
Kieran, C. (2004). The core of algebra: Reflections on its main activities. In K.
Stacey et al. (Eds). The future of teaching and learning of algebra: The 12th
ICMI
Study (pp. 21-34). Dordrecht: Kluwer.
Kieran, C. (2007). Developing algebraic reasoning: the role of sequenced tasks and
teacher questions from the primary to the early secondary school levels.
Quadrante, 16(1), 5-26.

Working Group 3
CERME 7 (2011) 531
Lins, R.,&. Kaput, J. (2004). The early development of algebraic reasoning: the
current state of the field. In K. Stacey et al. (Eds). The future of teaching and
learning of algebra: The 12th ICMI Study. (pp. 47-70). Dordrecht: Kluwer.
Orton, A.,& Orton, J.(1999). ‘Pattern and the approach to algebra’. In A. Orton (ed.),
Pattern in the Teaching and Learning of Mathematics. (pp. 104-120) London:
Continuum.
Rojano, T. (2002). Mathematics learning in the junior secondary school: Students’
access to significant mathematical ideas. In L. English, M. B. Bussi, G. A. Jones,
R. A. Lesh& D. Tirosh (Eds.), Handbook of international research in mathematics
education (pp. 143-161). Mahwah, NJ: Lawrence Erlbaum.
Weigand, H.,& Weller, H. (2001). Changes in working styles in a computer algebra
environment: The case of functions. International Journal of Computers in
Mathematical Learning, 6(1), 87-111.
Zazkis, R., &Liljedhal, P. (2002). Generalization of patterns: The tension between
algebraic thinking and algebraic notation. Educational Studies in Mathematics, 49,
379-402.

Algebraic Thinking Of Grade 8 Students In Solving Word Problems With A Spreadsheet

More Related Content

Similar to Algebraic Thinking Of Grade 8 Students In Solving Word Problems With A Spreadsheet (19)

More from Holly Fisher (20)

Recently uploaded (20)

Algebraic Thinking Of Grade 8 Students In Solving Word Problems With A Spreadsheet