ZDM Mathematics Education (2007) 39:355–364

DOI 10.1007/s11858-007-0050-3

ORIGINAL ARTICLE

Curriculum developers and problem solving:

the case of Israeli elementary school projects
Abraham Arcavi Æ Alex Friedlander

Accepted: 8 July 2007 / Published online: 28 July 2007

FIZ Karlsruhe 2007

Abstract Problem solving has been a main focus in Chinese Nine Sections, and others. In a certain sense, these
mathematics education for several decades, yet it seems documents may be considered as ‘‘curricula’’: texts with a
that its definition and classroom implementation are far main goal of teaching, or at least documenting the appli-
from being consensual. We explore the views and cation of a specific solution technique and communicating
approaches of a small community: the project leaders of it to a privileged minority. Thus, one may claim that
five elementary mathematics curriculum development ‘teaching’ problem solving is as old as mathematics itself.
projects in Israel, working within a centralized system, The view of mathematical problem solving as a toolkit of
which dictates the syllabus. We describe and analyze their techniques for different kinds of problems seemed to have
views along six categories: What are problems? What are prevailed during many centuries. ‘‘However, primarily
not problems? Classification of problems, problem solving within the last century, discussions of the teaching of
and individual differences, the ratio of problem solving problem solving have moved from advocating that students
tasks to other tasks in the project, and the role of heuristics simply be presented with problems or with rules for solving
and metacognition in teaching problem solving. We particular problems to developing more general approaches
describe, exemplify, interpret and discuss the (few) points to problem solving.’’ (Stanic and Kilpatrick, 1989, p. 4)
of convergence and the many different approaches. Finally, Many scholars working in different disciplines during
we reflect on the possible role of research in settling those the first half of the twentieth century can be mentioned as
differences. We speculate that our analysis and results go the precursors of the currently 50-year-old (or so) work on
beyond the local and the idiosyncratic. a more general conception of problem solving, its role in
mathematics teaching and its centrality in mathematics
Keywords Problem solving
Curriculum development
education research. Among them, the main pioneering
Elementary mathematics contributions can be attributed to John Dewey, Jean Piaget,
George Polya and William A. Brownell who worked in
diverse fields such as education, philosophy, epistemology,
cognitive psychology, mathematics and mathematics edu-
cation. Some 30 years ago, mathematical problem solving
1 Introduction
has started to play a central role both in teaching and in
research. In teaching, well-known documents have called
Problems were the focus of mathematics since its earliest
to adopt a problem solving approach to the teaching of
times, as recorded in the oldest extant mathematical doc-
mathematics—for example, the National Council of
uments: the Rhind Papyrus, Babylonian clay tablets, the
Teachers of Mathematics (1980, p. 2) and the Cockroft
Report (1982, p. 73). Nowadays international documents
A. Arcavi (&)
A. Friedlander continue to confirm the centrality of problem solving,
Department of Science Teaching, regardless of cultural and policymaking differences across
Weizmann Institute of Science,
Rehovot 76100, Israel
educational systems: ‘‘Mathematical problem solving is a
e-mail: [Link]@[Link] focus of school mathematics internationally’’ (Yeap et al.,

123
356 A. Arcavi, A. Friedlander

2006, p. 213). In the last decades, problem solving has also convergent view on problem solving. In particular, this
been at the forefront of research (e.g., Schoenfeld, 1985; may be the case within small sub-communities of mathe-
Silver, 1985; Charles & Silver, 1989; Schoenfeld 1992 and matics education sharing a common goal or working on the
many more1). same task within a small and centralized educational sys-
However, some basic ideas encapsulated under seem- tem at the same level of schooling.
ingly self-explanatory labels such as ‘‘problems’’ and In this paper we describe, exemplify and analyze the
‘‘problem solving’’ can still mean different things to dif- views on problems and problem solving among five teams
ferent people. Therefore, the apparent agreement on the of developers of elementary school mathematics curricula
importance of problem solving does not say much about in Israel, a small country with a centralized educational
what problems and problem solving mean—in fact, it may system. We discuss possible points of convergence, dif-
mask very different views on what constitutes a problem ferences of perspectives and we extrapolate our speculation
and what kinds of problem solving abilities are desirable, about the future of the problem solving in mathematics
teachable and evaluable. These different views are vividly education beyond the local scene.
exemplified in the influential collection ‘‘Teaching and
Learning Mathematical Problem Solving: Multiple
Research Perspectives’’ (emphasis added) edited by Silver 2 Background
(1985). Even when the qualifier ‘‘multiple’’ refers to
research perspectives, the different chapters in that book In Israel, the Ministry of Education (through its Depart-
illustrate a wide variety of views on problem solving. One ment of Curricula, which works together with ad-hoc
may argue that such variety of views is linked to diversity appointed committees) establishes the national syllabus2
of perspectives on research in mathematics education in for all grade levels in the country and the contents of the
general, as suggested by the title of a comprehensive ICMI national exit exam at the end of high school (and of some
study: ‘‘Mathematics Education as a Research Domain: A additional assessments along the way). The national exit
Search for Identity’’ (Sierpinska and Kilpatrick, 1997). exam in mathematics (and in other subjects as well) is a
However, the diversity of views on problems and problem requirement for the high school diploma and a pre-requisite
solving seems to be endemic in the mathematics education for admission to institutions of higher education and to
community at large, far beyond the realm of research. Even many jobs. This exam is uniform all over the country, and
within the same culture or within the same educational in mathematics it can be taken at different levels.
system, curriculum developers, teachers, researchers in Mathematics textbooks, especially in elementary school
mathematics learning/teaching and mathematicians do not (grades 1–6) undergo a process of approval by the Ministry
necessarily share the same views on what constitutes a of Education. However, at the junior-high and high school
problem and what to teach under ‘‘problem solving’’. In a levels the most popular ‘‘textbooks’’ are collections of
thorough literature review which opens his chapter on exercises, which currently do not undergo any process of
mathematical thinking, Schoenfeld (1992, p. 334) reviews approval by the ministry. Until a few years ago, almost all
in detail how the meanings of problem solving range from elementary schools in the country used the same set of
‘‘working rote exercises’’ to ‘‘doing mathematics as a curriculum materials, which included textbooks, teacher
professional’’, including goals for problem solving as guides, manipulative and other teaching aids developed and
diverse as ‘‘to train students to think creatively … to pre- published by a non-profit educational organization.
pare students for problem competitions… to learn standard Recently, six to seven additional elementary school cur-
techniques in particular domains, most frequently in riculum projects were developed and approved (or are in
mathematical modeling … to provide a new approach to the process of being approved) by the Ministry of Educa-
remedial mathematics (basic skills).’’ tion, and schools are free to choose one of them. The
Nowadays, more than two decades after Kilpatrick common characteristics of these curriculum projects are as
(1985), Schoenfeld (1985), Silver (1985) and others, follows:
research results, work on curriculum development and
implementation projects and much collective wisdom have
accumulated. Thus it is reasonable to speculate that the
mathematics education community has a somehow more 2
There are several definitions of the terms ‘‘curriculum’’ and
‘‘syllabus’’ (see, for example, Connelly & Lantz (1991) and Eash
(1991). For us, curriculum is a conglomerate of materials designed to
1
A search in the Educational Resources Information Center (ERIC) teach and learn mathematics. These materials share an implicit or
database for publications bearing the keywords ‘problem-solving’ and explicit rationale on goals, nature of teaching and learning and
‘mathematics’ in their titles yielded more than 500 publications mathematical activity. Syllabus is a list of the topics to be taught and
worldwide between 1993 and 2007. the recommended number of hours for teaching each topic.

123
Curriculum developers and problem solving 357

– The curriculum materials include a collection of 2. Provide an example of a task from your project that
textbooks, exercise books, teacher guides, manipulative you consider as not being a problem-solving task.
aids (especially designed or easily available objects), 3. Specify the approximate ratio between problem solv-
and student assessment questionnaires. ing and other tasks in your project.
– The projects are conducted by teams of mathematics 4. To what extent are students of different mathematical
educators, curriculum developers, and experienced abilities exposed to problem solving in your project?
teachers and, in some cases, professional mathemati-
This questionnaire was sent to the curriculum leaders a few
cians. All teams have project leaders or consultants
days before conducting individual interviews with each of
from academic institutions, or are related to an
them. The interviews were 60 to 90 min long and were
academic institution.
audiotaped. The purpose of these questions was to learn
– All projects provide ongoing in-service teacher pro-
about the developers’ explicit and implicit views on
grams and on-site teacher support.
problem solving through their choice and analysis of
examples and non-examples (items 1 and 2), the promi-
nence attached to problem solving in their projects (item 3)
3 The study and their views of problem solving being (or not being) an
activity for all (item 4).
The main question explored in the present study is: What Since our main purpose for this study was neither to
are the explicit and implicit views on problems and prob- characterize each project nor to compare among them, we
lem solving held by the team leaders of elementary decided to pool all the collected data without identifying
mathematics curriculum development projects in Israel? the originating project. This pooling of data does not allow
Our decision to focus on curriculum developers of ele- to learn about the degree of cohesiveness and richness of
mentary school mathematics was based on the following each project, yet it allows to explore the issues proposed. In
reasons: the following, we report our findings by presenting selected
quotes and examples from our interviews. The analysis of
– In the Israeli mathematics education scene, the most
these themes and their interpretations are based on the
organized, regulated and intensive activity is being
quotes, the examples, the context of the quotes and
currently conducted in the development and imple-
knowledge from other sources. Finally, we discuss our
mentation of elementary school curricula. During the
findings and their local and possibly more general
2005–2006 school year a new syllabus was adopted and
implications.
published by the Ministry of Education, and the
development teams started to adjust their existing
materials accordingly.
– It is agreed that the mathematics encountered by 4 Findings
students in elementary school leaves a strong imprint
on many of their mathematical thinking habits, on their All the interviewed curriculum project leaders were very
most basic beliefs about mathematics and mathematics responsive to our requests. During our interviews, all of
learning, and on their self-image with respect to them were open, willing to share their views with us and
mathematics. generous in terms of time devoted both to preparations and
– The authors had the opportunity to analyze in depth interviews. They all referred to the questions submitted,
most first and second grade mathematics curriculum and (as can be seen below) went spontaneously far beyond
projects, and to meet and interview their developers in them. Frequently, they embedded their answers into larger
an evaluation study commissioned to us by the Chief issues related to problem solving, curriculum development,
Scientist of the Israeli Ministry of Education, which views of mathematics and didactics, limitations posed by
finalized when this study was in its planning stages working within a somewhat strict and prescriptive educa-
(Arcavi et al. 2006). Our acquaintance with these tional system, and to their dissatisfaction with the state of
projects and their leaders served us both as back- mathematics education at the level of school implementa-
ground and as a way of triangulating data and tion. All project leaders attached an important role to
interpretations. problems and problem solving in the teaching and learning
of elementary school mathematics.
For the purpose of this study, we prepared the following
In the following, we organize our findings into six cat-
four-item questionnaire:
egories. In each category, we include selected quotes
1. Provide four examples from your project of problem (translated from Hebrew), a sample of problems presented
solving tasks. to us as examples, and our interpretations and comments.

123
358 A. Arcavi, A. Friedlander

4.1 What are problems? What is problem solving? We conclude this section by stating that, in general, the
project leaders agreed on the essence and meaning of
During the interviews, some project leaders defined infor- problems and problem solving. However, it is interesting to
mally what problems and problem-solving mean for them, note that such an agreement does not seem to hold as widely,
as reflected in the following quotes. when the project leaders characterize what is not a problem.
1. ‘‘A problem consists of data, a goal and a way to
achieve this goal. While you solve a problem, you do 4.2 What kind of tasks are not problems?
not know in advance the way to achieve the goal.’’ What kind of activity is not problem solving?
2. ‘‘A problem is a situation which the student did not
encounter previously. It is possible to approach it in 1. ‘‘What is not a problem is an exercise, and an exercise
several ways and to ‘wet your hands’ while coping is the application of a rule.’’
with it. At this stage, children don’t have an algorith- 2. ‘‘Exercises and straightforward word problems.’’
mic way to deal with it.’’ 3. ‘‘It all depends on who is the solver.’’
3. ‘‘A situation in which students have some tools to cope 4. ‘‘There are plenty of activities which are not problem
with and other tools are constructed during the solving: defining and enhancing the meaning of
problem solving process.’’ concepts (for example, identifying and classifying
4. ‘‘The characteristics of a problem are: there is no different meanings of subtraction), formulating and
‘paved way’ to a solution, it makes sense to the solver explaining ideas…’’
who has at least partial tools to cope with it, the solver
There seems to be a general agreement that exercises are
can understand the constraints, the solver is aware
not problems. Exercises are characterized by a straightfor-
that she has some tools to solve it, and she can know
ward application of a rule or a technique, which is
whether she solved it correctly or not.’’
‘algorithmic’ in nature. However, for some there seems
5. ‘‘Any non-algorithmic mathematical activity is prob-
to be a dichotomous classification of mathematical tasks
lem solving.’’
into exercises and problems (including those who claimed
For all our interviewees, problems and problem solving that the same task can be a problem for some students and
entail a situation for which students have some entrance an exercise for others), whereas for other project leaders
knowledge or tools (straightforward techniques or algo- the classification is less clearcut. Consider the last quote
rithms) to approach and to solve it, and lack others. Implicit above, stating that some non-algorithmic or non-straight-
in these views is that the situation presented in the problem forward applications of rules are not regarded as problems.
is understood and makes sense to students, and that what According to this view, reflective thinking, deepening the
constitutes a problem at a certain point in time may not understanding of a concept, working toward the correct
necessarily be a problem later on. This point was made formulation of an idea, and classifying arithmetical oper-
clear by one of the project leaders who presented a fifth ations by their meaning within a certain context are
grade classroom episode she witnessed a few days before important activities, but should be regarded neither as
the interview. The class was presented with the following exercises nor as problem solving activities.
task: ‘‘A teacher plans to devote 1/4 of her lesson to revise
previous materials and 1/3 for diverse activities. Is there 4.3 Classification of problems
any time left to work on problems from the book?’’ The
project leader considered this situation to be a simple During the interviews, all the developers engaged, either
application exercise that requires students to estimate that spontaneously or after being prompted by us, in classifying
the sum of 1/4 and 1/3 is less than 1, and was surprised problems. The classifications reflected further aspects of
when the problem could not be solved. She then realized their views on problem solving, and especially exposed
the source of difficulty for these students: they insisted on differences among them. We analyzed the leaders’ classi-
knowing the unknown length of the lesson. In her opinion, fications of problems according to the following three
the students could not initially identify that the lesson itself categories.
was the unit (regardless of its length), and that the solution
does not require any additional quantitative information.
Once that was understood, the situation was no longer a 4.3.1 Problems, ‘word problems’ and other
problem. Thus she concluded that, sometimes, the inter- representations
pretation of the data can be ‘problematic’ for students but
once this is cleared, the task is no longer a ‘problem’, but a The first category we found refers to classifying problems
simple exercise. according to the way they are presented, or in other words,

123
Curriculum developers and problem solving 359

the representation in which the problem is given (words, into a mathematical model. Thus, in one of the examples,
diagrams, tables, etc.). It turned out that this classification presented to us as a problem-solving activity by the leader
entailed a deeper characterization of problems than just the of this project, the heights of eight different trees are given
representation in which the formulation is provided. and students are requested to formulate different questions
in which they have to use expressions such as: ‘‘taller than
1. ‘‘Traditionally, people refer to word problems, but we
… by …’’, ‘‘… is … times taller than… ’’.4 This problem
also need to include other problems.’’
concentrates on ‘inverse modeling’: students are provided
2. ‘‘We should not identify problems with word problems
with sets of numerical data, from which they have to
… In geometry, problems can be different. There are
establish relationships and attach to them, the correspond-
problems related to ‘hands on’.’’
ing verbal expressions. The project leader claimed that
3. ‘‘Our curriculum project devotes an ample place to
problems of this type enhance, in a non-algorithmic way,
word problems. These problems appear both in special
students’ proficiency in modeling (quote 3 above). Another
word problem booklets and in booklets related to other
project leader also attached great importance to ‘word
topics as well. Word problems are characterized by
problems’, especially in the lower grades, and included
their form: they are given in words in a certain
many of them in his curriculum. Also in this project, great
context, they require a passage from natural language
importance is attached to inverse modeling, which in this
to mathematics and there is no algorithm—there is an
case consists of requesting students to create ‘‘arithmetical
elaboration of the text into a mathematical model. We
stories’’.
strongly oppose an ‘algorithmic translation’ (…times
The prominence of ‘word problems’ led other project
larger = multiply, larger than by … = addition)
leaders to declare explicitly that they do not want to fully
because this leads to many mistakes.’’
identify problems with word problems (quotes 1 and 2).
4. ‘‘In our program, we take great care to diversify the
The following three tasks were given as examples of other
situations and the representations (verbal formulation,
types of problems.
tables, graphs).’’
5. ‘‘Problems can be classified according to their content • ‘‘Which colored part in the two equally sized squares is
(integer numbers, fractions, operations), their type larger?’’ (See Fig. 1)
(verbally or numerically given), the type of formula- • ‘‘Use a piece of paper to build a pyramid.’’5
tions, the type of stories they involve…’’ • ‘‘Draw a quadrilateral with exactly two equal sides.’’6
Many (if not most) problems have a main verbal compo- These problems have predominant visual components.
nent in their formulation. However, for many developers, However, this is not the only difference between them and
the expression ‘word problems’ has a distinctive meaning, ‘word problems’. In these three problems, students have to
which is not necessarily related to the verbal representation ‘‘operate’’ visually: to observe, to think about and to
in which these problems are given. The expression ‘word manipulate forms and figures rather than translating from
problems’ seems to be reserved for the problems which are ‘real life’ situations presented verbally into mathematical
usually presented in a ‘real life’ context and their solution models, operate with them, find results and then go back to
requires (a) to identify the relevant data, (b) to ‘strip off’ the situation.
contextual details, (c) to translate the relevant verbal
information into a ‘mathematical model’, (d) to operate
within the model and to obtain a mathematical result, and 4.3.2 Inquiry, discovery and (other) problem solving goals
(e) to interpret the result in terms of the given situation. All
projects include in their materials other problems in The goals and nature of student activity during problem
addition to these ‘word problems’. However, the emphasis solving were important characteristics that guided the
on word problems that require ‘modeling’ varies among the classification of problems by our interviewees.
projects. One project (see quote 3) includes ‘word prob-
lems’ not only as an integral part of all its learning units in
the curriculum but also in special booklets devoted entirely
to these problems.3 In this case, teaching problem solving 4
In the textbook, this problem appeared under the heading ‘‘Com-
also includes teaching of the related skills, the main one parison Problems’’.
5
being the translation from a situation presented verbally This problem is given to students before learning about the
unfolding of a pyramid.
6
It is interesting to note that, the project leader who presented this
3
Devoting special booklets to ‘word problems’ may send the problem reported that most students draw quadrilaterals with two
unintended message that problem solving is a topic in itself rather equal opposite sides, and only a few construct quadrilaterals with two
than an approach to the learning and teaching of mathematics. equal adjacent sides.

123
360 A. Arcavi, A. Friedlander

for quite an extended period of time in a country which has

no lack of issues competing for headlines, and involved
mathematicians, mathematics educators, educational
administrators, parents and the public at large. A main
issue of this debate was the central role that acquiring basic
skills should play in the teaching of elementary mathe-
matics. Although problem-solving has not been a direct
Fig. 1 Which shaded part is larger? target of that debate, open-ended inquiry was attacked as a
time-consuming and unproductive activity, conducted at
the expense of acquiring and mastering basic skills, and not
1. ‘‘There is a difference between inquiry tasks and
always related to the curriculum core.
problem solving tasks. In problem solving, someone
Our data do not allow for a detailed discussion on dif-
states the problem and the students know that they
ferences among our interviewees’ conceptions of inquiry
must connect it to mathematics.’’
and discovery and their relations to problem solving.
2. ‘‘We distinguish between discovery activities and
However, we regard the discussed differences as another
inquiry activities as two different types of problem
illustration of how seemingly self-explanatory words can
solving. Discovery activities focus on what we want to
bear allied, but very different meanings, and subsequently
teach and place the child in a situation in which there
different attitudes toward them. Noticeably, quote 4 objects
is a need to learn something new, the child looks for
most explicitly to the openness of some inquiry problems.
something, proposes ideas and tries to implement
Open-endedness is viewed by this project leader as
them. The inquiry activities set a goal; the data is
potentially damaging to the primordial goal of promoting
clear, the problem is longer and is given after learning
systematic thinking and order in the learning of mathe-
(a topic). These are different types of problem solving.
matics. It was explained to us that such opposition does not
There is yet another type of problems in our curric-
exclude ‘‘complex problems’’, which should be an integral
ulum, they are relatively short, aimed at enrichment
part of the curriculum. The following complex problem
and extension, it could be a riddle or a puzzle. Thus
was presented to us as an example:
there are problems that are solved before learning a
mathematical topic (discovery activities) and there are ‘‘The difference of two numbers is 2,184. One of
problems that are solved after learning a topic (inquiry them is three times larger than the other. Find the sum
activities and riddles). During a learning process, of these numbers.’’
students solve simple exercises or straightforward
At the elementary level, students obviously lack algebraic
word problems for the purposes of immediate appli-
symbolism in order to solve this problem. The curriculum
cation of what was learned.’’
materials in this project lead students to use line segments
3. ‘‘We distinguish between inquiry problems and non-
as a graphical representation, which supports the visuali-
traditional problems. Inquiry problems need to include
zation of the relationships (in an orderly and systematic
a component of generalization; non-traditional prob-
way), and guides students toward the choice and use of the
lems present a non-traditional context, which does not
operations needed to understand and solve the problem.
relate directly to a curriculum topic.’’
For one project leader, the inquiry/discovery nature of
4. ‘‘In our program we do not include many ‘open-ended’
certain problems was tightly related to the different roles
problems. Open problems do not relate to the learned
these problems should play in the learning process: some
topics and are prone to cause damage, since they
problems (called ‘‘discovery’’ by this interviewee) are used
stimulate nonsystematic thinking, and the most impor-
for the introduction of a topic, others (called ‘‘inquiry’’) are
tant thing is to promote systematic thinking and order.
used after the learning a topic, whereas simple exercises
But, we do include complex problems.’’
are used during the learning process to consolidate content
The distinctions made by some of our interviewees (see quote 2 above). The following two examples relate to
between inquiry problems and other types of problems the above mentioned characterization.
were unexpected to us. With hindsight, we attribute the
• ‘‘Take a grid paper and cut out a rectangle (along the
need for making such distinctions (and, to some extent, the
grid lines). Ask your neighbor to find its area.’’
caution with which inquiry problems were mentioned), at
• ‘‘The calculator key ‘5’ is broken. Find two ways to
least in part, to the very intensive public debate which has
obtain 55 on the display of this calculator.’’
taken place recently in Israel on the causes of the ‘‘national
failures’’ (i.e., relatively low achievements) in international The first item was characterized by this project leader as a
examinations. This debate reached the media, occupied it ‘‘discovery’’ problem presented to students toward their

123
Curriculum developers and problem solving 361

learning of the sides’ multiplication formula for the area of problem solving is for everyone.’’ The following quotes
a rectangle, and is intended to promote the ‘‘need to learn illustrate some opinions on this issue.
something new’’ through exploration, proposing ideas,
1. ‘‘These activities require patience; not all students are
implementing and discussing them with others. The second
bright, but these activities enable systematic work and
item was characterized as an example of an inquiry
if working slowly many can succeed.’’
problem assigned to students after learning a topic.
2. ‘‘Problem solving does not have to be a difficult
This characterization of problems according to the roles
activity. And not all the activities, which are difficult
that different problems play at different stages of the
are problem solving. In other words, difficulty =
learning process, brings to mind the categorization
problem solving. It follows that problem solving
described by Stanic and Kilpatrick (1989) as ‘‘problem
activities can and should be also for students with
solving as motivation’’, ‘‘problem solving as vehicle’’,
difficulties.’’ [The inequality sign was indicated by
‘‘problem solving as recreation’’ and ‘‘problem solving as
hand movements]
skill’’.
3. ‘‘In our curriculum, we distinguish between problems
for all, problems for the advanced students and
problems for students with difficulties. The degrees of
4.3.3 Scope
difficulty lay in the size of the numbers involved (the
number size determines the working strategy), in
Problems were also classified by some project leaders
‘reversibility’ (more than half the students in second
according to their scope—as expressed in the following
grade cannot understand that addition and subtraction
quotes.
are inverse operations) and level of logical
1. ‘‘There is a distinction between wide problems and inferences.’’
local (specific) problems.’’
Some curriculum projects assume that all students, espe-
2. ‘‘Problems should include a demand for logical
cially in earlier grades, should master the basics together
thinking; they could include several or just a small
and their materials do not differentiate tasks according to
number of constraints.’’
ability levels. Other projects offer different problems for
In other words, another classification of problems refers to students of different ability levels even in the earlier grade
the number of constraints it includes and to the amount of levels (quote 3).
time and effort needed to solve the problems. Some are As indirectly implied, it seems that some of the issues
large (or mid-size) projects; others can be ‘‘small’’ riddles can be addressed by task designers (number size, com-
or puzzles. We were given the following examples of a plexity, complexity of the logic required and so on), but
wide-scope task (first item below) and of a small-sized others (patience, managing of difficulties) seem to go
riddle (second item): beyond what designers and materials can do. The latter
should be addressed by teaching practices and classroom
• ‘‘Out of 721 airport maintenance workers, 257 go to
management. Thus, some of the projects prescribe specific
work by car, 210 by bike, and 176 by either car or bike.
and detailed methods of teaching and classroom manage-
30 senior workers should have reserved parking places.
ment as an integral part of the implementation of their
The parking lot has 360 parking places. One parking
curriculum.
place can accommodate a car or 8 bikes. Plan the
workers’ parking lot: Decide how many places should
be allotted to cars, and how many to bikes. Describe
your thinking.’’ 4.5 The ratio of problem solving tasks to other tasks
• ‘‘5 children plant 5 flowers in 5 min. How long will it in the projects
take 100 children to plant 100 flowers?’’
According to the project leaders, the occurrence of problem
For at least two of the team leaders, both types of problems
solving tasks in their curriculum projects, ranges from 40 to
are desirable and should be included in the curriculum.
80%.
1. ‘‘The time and the timing [for the teaching of certain
4.4 Problem solving and individual differences topics] are dictated by the official syllabus which
forces us to compromise. Ideally, I would devote 70%
All project leaders believed that all students are capable of of the time to problem solving, because knowledge
solving problems. In Stanic and Kilpatrick’s words (1989, construction has to be done by problem solving. In
p. 20) our interviewees took ‘‘seriously the notion that reality, I estimate our project devotes only 40%.’’

123
362 A. Arcavi, A. Friedlander

2. ‘‘Problems occupy about 80% of our curriculum.’’ three given alternative solution approaches, requiring
explicit formulations.
The lowest percentage (quote 1) was mentioned by one
We note that the project leaders did not expand further
project leader who regarded the new syllabus as both
on strategies and reflection. These components of problem
packed with topics and tight in the time it allots to each
solving do not seem to be as important as mathematical
topic. The highest percentage (80) was quoted by one of
content per se. But there could be other possible reasons.
the project leaders who attributed a high prominence to
On the one hand, the public debate already mentioned
‘word problems’, incorporated other problems as well and
leaned toward the importance of acquiring basic skills,
viewed exercises as the only non-problems. Thus, the
relegating the role of strategies and reflection. On the other
differences of perspectives on the density of problem-
hand, the density and demands of the new syllabus as
solving activities in the curriculum materials may not
viewed by some project leaders (see previous section) does
necessarily stem from the relative importance attached to
not allow for addressing more intensively both strategies
problems and problem solving, but rather from what is
and reflection.
perceived as constraining external conditions or from
views of what is or is not a problem.
5 Discussion
4.6 The role of heuristics and metacognition
Kilpatrick (1985, p. 13) concludes his retrospective account
of 25 years of research on teaching mathematical problem
The terms heuristics and metacognition, associated to the
solving by pointing to promising developments: the
theoretical and empirical literature on mathematical prob-
‘‘changing approaches to research’’, ‘‘the converging
lem solving (e.g. Polya, 1945; Schoenfeld, 1985, 1992),
interests of cognitive scientists and mathematics educa-
were not explicitly mentioned in any of the interviews.
tors’’, and ‘‘the emerging view of the mathematics teacher
However, all project leaders alluded to the ideas behind
as a full partner … rather than an imperfect system for
these terms and referred to strategies and reflection (during
delivering instruction’’. The closing sentence in that chapter
or after problem solving activities). For example:
is: ‘‘Such developments suggest that by 2008 the harvest
1. ‘‘There are many types of problems, some can require from research will be rich indeed.’’ On the verge of 2008,
‘give all the possibilities for…’ or ask ‘why, in your we can say that research in mathematics education has
opinion,…’.’’ indeed flourished, incorporating theoretical approaches,
2. ‘‘One type of problems consists of presenting the child which go beyond classical cognitive science, and offering
with three alternative solutions, and ask to explain many valuable results.
them, demanding correct formulations.’’ In this modest study, we found some (local) traces of the
3. ‘‘A problem is supposed to guide towards a concept or realization of Kilpatrick’s prediction and its application to
a procedure, but it stands on its own right, namely it curriculum development. Our data led us to the following
does not have to be followed immediately by the conclusions about some points of convergence of our in-
teaching of a concept, it can seed ideas for the future, terviewees’ views and approaches to problem solving:
and in any case it can be a goal in itself, just to engage
– As opposed to the wide range of interpretations of
with it and with the strategies it affords.’’
problem solving in the past, and in spite of some
4. ‘‘The solution process includes experimentation and
persisting differences of approaches in the present,
the use of the ‘methodology of the subject matter’.’’
‘‘working rote exercises’’ and the mere drill of solution
5. ‘‘Problems can be different according to the strategies
techniques are no longer considered as problem
they require…’’
solving.
We propose that implicit in these quotes is the assumption – Problem solving is viewed as important even at the
that both strategies and reflection can be taught. Strategies very early stages of elementary school and should be
can be elicited by confronting students with different types included as a rather central component of instruction.
of problems (quote 5), by including some problems that are – Problems can be of very different sorts, and can be
not necessarily presented for the purpose of teaching a classified in different ways – for example, by repre-
specific topic (quote 3 and the first part of quote 1) and by sentation, by nature, by goals, by scope and by place in
supporting experimentation and the use of the ‘‘methodol- the curriculum sequence.
ogy of the subject matter’’. According to these project – Problem solving is considered an activity for all and, as
leaders, reflection can take different forms and goals, for a result, it is a desirable practice to pose demands to
example, providing an opinion, inspecting and explaining students and to expose them to complex problems.

123
Curriculum developers and problem solving 363

Thus, we are tempted to suggest that possibly theory and • Further explore the uncharted territory of problem
research had, after all, certain subtle yet discernible role in solving still populated by beliefs not fully substantiated
the shaping of some common characteristics of the by empirical studies.
curriculum materials developed by the project leaders we • Help educators in making better informed choices,
interviewed. between either complementary or opposite views.
However, beyond the general commonalities described • Observe, document and discuss at a very fine grain
above, we found that there are many noticeable differ- level of detail how teaching problem solving in
ences among the various theoretical and practical classrooms may look (e.g., Arcavi et al., 1998) or
approaches explicit and implicit in our data. More how an intended curriculum is implemented (e.g.,
extensive dialogues and a detailed analysis of curriculum Arcavi et al., 2006).
materials would have revealed even deeper differences • Help educators to become more aware of and account-
than those described in this article. Thus, we claim that able for the options they choose and above all, make
considerable differences in the approach to problem them less sanguine about deep rooted convictions.
solving are pervasive even within the same community (of
Perhaps in 2058, when our community will come back to
curriculum developers) focusing on the same target pop-
reassess the state of theory, research, development and
ulation (elementary schools) within a very centralized
practices of problem solving, we will be able to point to
educational system (in Israel) with a uniform syllabus.
further progress.
The different approaches manifest themselves in those
aspects which are not regulated by the syllabus—for
example: how to introduce a topic, what kinds of tasks are References
more important than others in order to support the
learning of a topic, what is the role of inquiry, and which Arcavi, A., Kessel, C., Meira, L., & Smith, J. (1998). Teaching
lesson format is preferable. mathematical problem solving: An analysis of an emergent
Some of the detected differences may be considered classroom community. Research in Collegiate Mathematics
Education, III, 7, 1–70.
compatible and even complementary, whereas others seem Arcavi, A., Friedlander, A., Fresko, B., & Carmeli, M. (2006).
to be contradictory. Complementary aspects would pose Evaluation of six mathematics curriculum projects for first and
less of a conundrum than contradictory approaches. Many second grade classes. Report (in Hebrew) submitted to the Chief
Israeli mathematics educators, educational administrators, Scientist of the Ministry of Education. Rehovot: The Weizmann
Institute of Science.
policy makers and laymen welcome a variety of approa- Charles R. I., & Silver E. A. (1989). The teaching and assessing of
ches. Moreover, such variety would be considered as the Mathematical problem solving. Reston: The National Council
realization of a much needed pluralism, that provides of Teachers of Mathematics and Lawrence Erlbaum
educators with the freedom to choose curriculum materials Associates.
Conelly, F. M., & Lantz O. C. (1991). Definitions of curriculum: An
that suit their preferences and at the same time allows them introduction. In A. Lewy (Ed.), The International Encyclopedia
to abide by the requirements of a uniform syllabus. Two of Curriculum (pp. 15–30). New York: Pergamon Press.
assumptions underlie the support of pluralism: (1) there is National Council of Teachers of Mathematics (1980). An agenda for
no single approach in mathematics education that has been action. Reston: NCTM.
Cockroft W. H. (1982). Mathematics Counts. London: HMSO.
proved to be better than all the others, and (2) diversity Eash Maurice J. (1991). Syllabus. In A. Lewy (Ed.), The International
within a system creates opportunities to learn from each Encyclopedia of Curriculum (pp. 71–74). New York: Pergamon
other and ultimately will enrich the system, in the spirit of Press.
the Talmudic saying that ‘‘literary rivalry can contribute to Kilpatrick, J. (1985). A retrospective account of the past 25 years of
research on teaching mathematical problem solving. In E. Silver
wisdom.’’ (Ed.), Teaching and learning mathematical problem solving:
And what about contradictory approaches? In much of Multiple research perspectives (pp. 1–15). Hillsdale: Lawrence
the Israeli public discussions (and also in many academic Erlbaum Associates.
discussions as well), research is not always invoked, and Polya, G. (1945). How to solve it. Princeton: Princeton University
Press.
strong-held beliefs about learning and about mathematics Sierpinska A., & Kilpatrick J. (Eds.) (1997). Mathematics education
seem to have a more prominent role when coming to settle as a research domain: A search for identity. Dordrecht: Kluwer.
contradictions between approaches or when deciding which Schoenfeld, A. H. (1985). Mathematical problem solving. New York:
curriculum materials to choose. Academic.
Schoenfeld, A. H. (1992). Learning to think mathematically: Problem
What would then be the role left to research on teaching solving, metacognition and sense making in mathematics. In D.
mathematical problem solving? We conclude with a few A. Grouws (Ed.) Handbook of research on mathematics teaching
suggestions for some general goals for (the much needed and learning (pp. 334–370). New York: MacMillan Publishing
and ongoing) research on problem solving: Company.

123
364 A. Arcavi, A. Friedlander

Silver E. (Ed.) (1985). Teaching and learning mathematical problem Yeap, B., Ferrucci, B. J., & Carter, J. A. (2006). Comparative study of
solving: Multiple research perspectives. Hillsdale: Lawrence arithmetic problems in Singaporean and American mathematics
Erlbaum Associates. textbook. In F. K. S. Leung, Graf, K. D., & Lopez-Real F. J.
Stanic G. M. A., & Kilpatrick, J. (1989). Historical perspectives on (Eds.) Mathematics education in different cultural traditions—A
problem solving in the mathematics curriculum. In R. I. Charles, comparative study of east Asia and the West. The 13th ICMI
& E. A. Silver (Eds.), The teaching and assessing of mathemat- Study (pp. 213–226). New York: Springer.
ical problem solving (pp. 1–22). Reston: The National Council of
Teachers of Mathematics and Lawrence Erlbaum Associates.

123