AN ABSTRACT OF THE DISSERTATION OF

Supattra Pativisan for the degree of Doctor of Philosophy in

Mathematics Education presented on May 5, 2006.
Title: Mathematical Problem Solving Processes of Thai Gifted Students.

Abstract approved:

__________________________________________________________________
Margaret L. Niess

The purpose of this study was to examine the problem solving processes of
Thai gifted students when they solved non-routine mathematical problems. The
research questions guiding the study were: (1) What is the nature of the problem
solving processes that Thai gifted students use as they engage in solving non-routine
mathematical problems? (2) What metacognitive behaviors do Thai gifted students
exhibit when engaged in mathematical problem solving?

Five Thai gifted students who were eligible for the Thai Mathematical
Olympiad project and met the selection criteria participated in this study. Each student
practiced the think aloud method before solving three mathematical problems
individually. The problems were non-routine problems that focused on number theory,
combinatorics, and geometry, respectively. The subjects worked on each problem
separately and were interviewed at the end of each problem solving session. Data
sources included videotapes of the think aloud and the interview sessions, students’
written solutions, and researcher’s field notes. These data were analyzed using the
within-case and cross-case techniques. Data gathered were also categorized using a
constant comparative method to conceptualize a model of problem solving process.
 Overall, the participants solved Problem One and Problem Two without
hesitation, although some did not completely solve Problem One. In spite of the fact
that two students had some difficulty searching for a solution for Problem Three, all of
the students eventually succeeded. The results generated a Thai model of problem
solving process that detailed the students’ behaviors in each of four stages:
understanding, planning, executing, and verifying. Their behaviors occurred in this
model cycled back and forth among the four stages. It was apparent in each stage that
the students understood when and how to apply their mathematical knowledge and
strategies in their solutions. They also applied self-evaluation statements to monitor
and evaluate themselves as problem solvers. The findings also provided five
categories of emerging evidence related to the students’ problem solving processes:
advanced mathematical knowledge, willingness to consider multiple alternative
solution methods, recollection and willingness to consider prior knowledge and
experiences, reliance on affect, and parental and teacher support.
©Copyright by Supattra Pativisan

May 5, 2006

All Rights Reserved

Mathematical Problem Solving Processes of Thai Gifted Students

by

Supattra Pativisan

A DISSERTATION

submitted to

Oregon State University

in partial fulfillment of
the requirements for the
degree of

Doctor of Philosophy

Presented May 5, 2006

Commencement June 2006
Doctor of Philosophy dissertation of Supattra Pativisan
presented on May 5, 2006.

APPROVED:

Major Professor, representing Mathematics Education

Chair of the Department of Science and Mathematics Education

Dean of the Graduate School

I understand that my dissertation will become part of the permanent collection of

Oregon State University libraries. My signature below authorizes release of my
dissertation to any reader upon request.

Supattra Pativisan, Author

 ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to my advisor, Dr. Margaret Niess,

for her continuous support, nurturing, and guidance throughout my graduate studies.
Her personal and professional standards will have a lasting influence in my life.
Without her, this dissertation would never have been possible. Her kindness is greatly
appreciated. I would like to express my gratitude to committee members, Dr. Dianne
Erickson, Dr. Barbara Edwards, Dr. Karen Higgens, Dr. Steven Esbensen, and Dr.
Michael Dalton for their invaluable time and much appreciated help in working on this
dissertation. I also want to thank Dr. Larry Enochs, Dr. Larry Flick, Dr. Edith
Gummer, and Dr. Shawn Rowe for their instruction throughout my graduate study.
This study could not have been conducted without the cooperation and support
from many people. Special thanks to the experts, Dr. Vichain Laohakosol, Dr.
Pattanee Udomkavanich, and Ms. Suwimon Hall for their significant advices for this
study. Extreme appreciation is expressed to Dr. Suwattana Utairat and Mr. Nipon
Suppasri for examining the accuracy of translation and their gratefulness support.
Without the excellent cooperation of the students who participated in this study, this
study could not have been carried out; I am grateful for their help. Very special thanks
go to the Gutmann family, the Earhart family, and Carol Odel for not only excellent
critiquing and editing this dissertation, but also their invaluable support all the time.

Most importantly, I am greatly appreciative for the strength support, love, and
encourage which I received from my parents and sisters in my life. They have made
this accomplishment possible and kept me on the right track. A special thanks to all
the instructors of the Thailand Mathematical Olympiad project for their thoughtfulness
and encouragement. I would like to thank all of my friends for their kindness and
support throughout my study. Lastly, I would like to give special thanks to the
Institute for the Promotion of Teaching Science and Technology (IPST) in Bangkok,
Thailand for the financial support during my graduate study. I also thank to the
Laurels supplementary scholarship from Oregon State University for the financial
support on the third year of my study.
 TABLE OF CONTENTS

Page

CHAPTER I: THE PROBLEM…………………………………………..……. 1

Introduction………………………………………………………………... 1
Statement of the Problem………………………………………………….. 4
Research Questions………………………………………………………... 7
Significant of the Study……………………………………………………. 7
Definitions…………………………………………………………………. 8

CHAPTER II: LITERATURE REVIEW…………………………………….... 9

Problem Solving in the Constructivist Paradigm………………………….. 9
Nature of Problem Solving and Its Process………………………………... 12
Metacognition and Problem Solving………………………………………. 15
Methodology Involved in the Problem Solving Process…………………... 18
Mathematically-gifted Students…………………………………………..... 20
Research on Gifted Students’ Problem Solving Processes………………… 23
Conclusions……………………………………………………………..…. 29

CHAPTER III: METHODOLOGY……………………………………………. 32

Design of the Study………………………………………………………... 32
Research Setting………………………………………………………….... 34
Participants……………………………………………………………….... 35
Data Collection Protocols………………………………………………….. 37
Three Problems for Data Collection……………………………………….. 39
Problem One ...…………………………………..…………………….. 39
Problem Two ..……………………………………………………….... 39
Problem Three ………………………………………………………… 40
Data Collection ……………………………………………………………. 42
Data Analysis …………………………………………………………........ 43
 TABLE OF CONTENTS (Continued)

Page

CHAPTER IV: RESULTS ……………………………………………………. 46

Participant Backgrounds ………...………………………………………… 46
Pradya ………………………...……………………………………….. 47
Sira ……………………………...……………………………………... 48
Wude ……………………………...…………………………………… 48
Nipa …………………………………………………………………… 49
Sakda ………………………………………………………………….. 49
Reflections on the Problems………………...……………………………... 50
Analysis of the Think Aloud and Interview Results ……………………… 53
Participants’ Solution Paths to the Problems ………...………………... 53
Description of the Thai Students’ Problem Solving Processes……………. 56
Thai Students’ Problem Solving Model…………………………………… 77
Summary…………………………………………………………………… 80

CHAPTER V: DISCUSSION…………………………………………………. 81
Discussions of Findings…………………………………………………..... 81
Limitations of the Study…………………………………………………… 86
Implications and Recommendations for Future Research……………...….. 88
Concluding Remarks………………………………………………………. 90

BIBLIOGRAPHY …………………………………………………………..… 92

APPENDICES ..……………………………………………………………….. 104

Appendix A. Examples of Participants’ Problem Solving Processes……… 105
Appendix B. Think Aloud Protocol……………..…………………..…...… 111
Appendix C. Mathematical Problem Solving Tasks………………………. 112
Appendix D. Interview Protocol…………….………..………………..….. 113
 LIST OF FIGURES

Figure Page

3.1. Pictorial Representation for Problem Three .….………………..……... 41

4.1. Three Problems of the Study......………………………………………. 51
4.2. Sakda’s Representation of Problem One.……………………………… 68
4.3. Thai Students’ Problem Solving Model .……..……………………….. 78
4.4. Thai Students’ Problem Solving Stage Description..…………….……. 79
 LIST OF TABLES

Table Page

3.1. Background of 7 Gifted Students Meeting the Selection Criteria ….…….. 36

3.2. Weekly Data Collection…………………………………………………... 42
4.1. Participant Backgrounds………….………………………………..……... 47
4.2. Problem Difficulty Identified by the Participants……..………………..… 52
4.3. Examples of Excerpts for Self-Evaluation .……..…………..……………. 76
 MATHEMATICAL PROBLEM SOLVING PROCESSES OF
THAI GIFTED STUDENTS

CHAPTER 1

THE PROBLEM

Introduction

Problem solving has become an important part of learning mathematics

because it helps students understand the mathematical content. It also leads students to
understand how to apply their knowledge in real world problems. Prior to reform
efforts, the National Assessment of Education Progress (NAEP) found that 9, 13, and
17 year old students’ performance on non-routine problems and in problem solving
continues to be unacceptably low. The evidence indicates that students learn
mathematics at a rote level without understanding how to apply what they know while
solving mathematical problems (National Council of Teachers of Mathematics
[NCTM], 1983). These results also show that although many students perform well on
tests of low-level skills, such as arithmetic computation, they perform poorly on tests
of high-level skills such as mathematical problem solving (Mayer & Hegarty, 1996).
The recent reform movement has directed mathematics teachers to provide
students with experiences and opportunities in problem solving throughout the
secondary school mathematics curricula. The Principles and Standards for School
Mathematics emphasizes the need for opportunities in problem solving for all students
from grades K-12 and in all mathematics content (NCTM, 2000). Problem solving is
one of the Process Standards in NCTM’s 2000 recommendations. The Problem
Solving Standard recommends that “by the end of grade 12 students should be able to:
build new mathematical knowledge through problem solving, solve problems that
arise in mathematics and in other contexts, apply and adapt a variety of appropriate
strategies to solve problems, and monitor and reflect on the process of mathematical
problem solving” (NCTM, 2000).
 2

Problem solving requires engaging in a task for which the solution method is
not known in advance (NCTM, 2000). Mayer (1992) defines problem solving as a
cognitive process in which one figures out how to solve a problem that he or she does
not already know how to solve. Most definitions of problem solving emphasize non-
routine problems that require problem solvers to use information and procedures in
unfamiliar ways. Problem solving is an extremely complex human endeavor that
involves much more than the simple recall of facts or the application of well-learned
procedures. Problem solving requires the production of sequential processes that
constitutes strategies besides the execution of the sequence (Hammouri, 2003).
Problem solving also requires coordinating multiple cognitive and metacognitive
processes, selecting and deploying suitable strategies, and adjusting behavior to
changing task demands (Montague, 1991).
Metacognition is the knowledge and awareness of one’s own cognitive
processes along with the ability to monitor, regulate and evaluate one’s thinking
(Flavell, 1976; Brown, 1978). Metacognition in problem solving refers to the
knowledge and processes used to guide thinking directed toward the successful
resolution of a problem (McCormick, 2003). According to Davidson and Sternberg
(1998), metacognitive skills support problem solvers in understanding the problem,
selecting suitable solution strategies, monitoring solution strategies effectively, and
identifying and overcoming obstacles to solving the problems. Metacognition also
plays an important role in regulating and monitoring one’s thinking processes thus
providing a framework for examining the qualitative differences in thinking between
gifted and average students (Dover, 1983). Ultimately, metacognition is an important
component for incorporating appropriate information and strategies during the
problem solving process.
Mathematical problem solving can be conceptualized in two interrelated
stages: problem representation and problem solution (Mayer, 1985; Montague &
Applegate, 1993). Initially, problem solvers build mental representations of the
problem in the problem representation stage and then they devise and carry out the
plan in the problem solution stage. Various models are proposed that describe the
 3

processes that problem solvers use from the beginning until they finish their tasks
(Garofalo & Lester, 1985; Montague & Applegate, 1993; Mayer, 2002; Polya, 1957).
For instance, Polya’s model consists of four stages: understanding the problem,
devising a plan, carrying out the plan, and looking back (Polya, 1957). Polya’s model
is algorithmic in nature and research generated by it focuses purely on heuristics
(Sriraman, 2003). Later, Garofalo and Lester (1985) modified Polya’s model to
include cognitive and metacognitive components. Their model is described in four
stages: orientation, organization, execution, and verification. Montague and Applegate
(1993) presented a model focused on seven cognitive processes (reading,
paraphrasing, visualizing, hypothesizing, estimating, computing, and checking) and
three metacognitive processes (self-instruction, self-questioning, and self-monitoring).
Mayer (2003) proposed another cognitive process model that included translating,
integrating, planning, and executing processes.
Even though these models have been used to investigate problem solving
processes, only two models (Garofalo and Lester’s model; Montague and Applegate’s
model) have been used with gifted students in the literature (Garofalo, 1993;
Montague, 1991, Montague & Applegate, 1993; Sriraman, 2003). Both models
emphasized metacognitive processes and research on these models supports that gifted
students used metacognitive knowledge and a repertoire of problem solving strategies
to process information effectively and efficiently (Sternberg, 1981). Results have
indicated that gifted students continually use metacognitive strategies in their problem
solving. They report substantial use of self-instruction throughout the problem,
frequent self-questioning during and after reading the problem, and effective self-
evaluation and self-monitoring activities (Montague, 1991, Montague & Applegate,
1993). Results have also shown major differences in the processes used by gifted and
non-gifted students in the orientation and organization stages (Sriraman, 2003).
 4

Statement of the Problem

From the constructivist paradigm, students construct knowledge through

interaction with the environment or the task. Complex mental processes can be viewed
and accessed (Seels, 1989). From a cognitive constructivist’s position, Piaget (1971)
purported that individuals construct knowledge continually through the process of
assimilation and accommodation in their mind. When students solve a problem, they
assimilate the problem into their existing knowledge or thought patterns. They adapt
their knowledge to include the given information. If they are satisfied with their mode
of thought, they are in a state of equilibrium. Upon becoming aware of the
shortcomings in their existing thinking, they become dissatisfied with their current
thinking. Finally, they adapt a more sophisticated mode of thought that eliminates the
shortcomings of the old thinking. As a social constructivist, on the other hand,
Vygotsky believed that knowledge is constructed through social interactions with
culture and people (Stone & Reid, 1994). He focused on the connections between
people and the cultural context in which they act and interact in shared experiences
(Crawford, 1996). Therefore, based on Vygotsky’s point of view, students’ cognitive
development is influenced by cultural practices, including the community where they
participate (Chiu & Tron, 2004).
Mathematically gifted students are identified as students who are able to do
mathematics typically accomplished by older students. They are able to employ
qualitatively different thinking processes in solving problems (Sowell, et al., 1990).
They also display curiosity and creativity when assessing a problem situation and
possess a high level of task commitment (House, 1987). In comparison with average
students, gifted students tend to be better mathematical problem solvers (Gallagher,
1975; Renzulli, 1978). At the same time, successful mathematical problem solving
requires students to be able to select and use task-appropriate cognitive strategies for
understanding, representing, and solving problems (Mayer, 1992; Schoenfeld, 1985).
These abilities involve metacognitive knowledge, a knowledge that is necessary for
higher-level learning and problem solving (Brown, 1978). Research has also indicated
 5

that gifted students demonstrate metacognitive attributes with effective problem

solving (Hannah, 1990; Pressley, Borkowski, & Schneider, 1989).
Non-routine problems are the type of problems where the students are not
familiar with problem situations and they are not expected to have previously solved
or have not met regularly in the curriculum. Non-routine problems demand thinking
flexibility and extension of previous knowledge, may involve concepts and techniques
which will be explicitly taught at a later stage, and may involve the discovery of
connections among mathematical ideas (Schoenfeld, et al., 1999). Researchers have
found that more difficult problems have potential to activate metacognitive
functioning to the extent that good problem solvers consciously regulate and control
their cognitive processes (Montague & Applegate, 1993). Additionally, gifted students
possess greater metacognitive understanding than other students and they are also
better able to, through metacognitive thinking, transfer strategies from one situation to
unfamiliar situations (Alexander, Car, & Schwanenflugel, 1995). Gifted students
prefer to solve non-routine problems because of the challenge of working with these
problems (Garofalo, 1993). Thus, non-routine problems are more likely to activate
gifted students to demonstrate their high ability in problem solving.
Many researchers have studied how secondary gifted students solve non-
routine mathematics problems (Garofalo, 1993; Lawson & Chinnappan, 1994;
Montague, 1991; Sriraman, 2003). Results indicate that gifted students spend much
time rereading and translating the problems into their own words (Garofalo, 1993;
Montague, 1991; Sriraman, 2003). This paraphrasing ability supports them in
understanding the problem and indicates one way they differ from other students in
problem solving. They are more verbal than other students and their verbalization
increases when they are confronted with more difficult problems (Sriraman, 2003).
They recall theorems for generating given information, apply prior knowledge in the
problem and use it to access further relevant knowledge (Lawson & Chinnappan,
1994; Sriraman, 2003). Gifted students identify the assumptions in the problem,
frequently set up an equation or algorithm after reading, and generally divide the
problem into sub-problems. They identify a goal before developing their solution
 6

plans. They solve the problems systematically, and use efficient strategies. They redo
the problems by working through the whole problems, rereading them, redoing
computations, and checking steps and processes (Montague, 1991; Sriraman, 2003).
Although these research findings are based on gifted students in the United States, the
results of another study in the Netherlands are similar. These findings indicated that
gifted students spend more time in the orientation stage, have enough prior knowledge
to apply to the problem, and separate the problem into component parts (Span &
Overtoom-Corsmit, 1986).
However, all these participants are from a western culture, a culture that is
distinct from Asian cultures, including the Thai culture. Geary (1996) found that Asian
students receive substantially more exposure to mathematics at school and at home
than U. S. students. Asian cultures emphasize and give priority to mathematical
learning (Hatano, 1990). Asian school differences reflect wider cultural differences in
the relative valuation of achievement in mathematics. Pretz, Naples, and Sternberg
(2003) suggested that social contexts such as culture and language structure play a role
in problem solving ability, especially in recognition, definition, and representation of a
problem. They also suggested that the social context may facilitate understanding the
problem and thinking divergently about the solution. Schoenfeld (1992) argued that
doing mathematics could be considered a social activity, with roots in the cultural and
societal environments. According to Vygotsky, social and cultural sign systems have a
major impact on an individual’s cognitive development (Crain, 1985). Previous
research findings also showed that different cultural practices influence students’
cognitive development (Saxe, 1991). It is logical to assume that cultural differences
may lead students to perform in different ways when engaged in mathematical
problem solving.
In Thailand, the current status of gifted education faces several problems. First,
no national center takes direct responsibility for nurturing the promotion of gifted
abilities. Second, not enough knowledge and resources exist for gifted education, with
especially small numbers of experts supporting this field. These problems result in an
inefficient system and are unsuccessful in nurturing gifted students (Office of the
 7

National Education Commission [ONEC], 2004). The Thai government has

recognized the need for promoting and supporting the high abilities of these students
as mentioned in the strategic action plan of the Ministry of Education in 2004
(Bodharamik, 2004). Considering research on the mathematical problem solving of
Thai gifted students, only two related studies were identified (Klaimongkol, 2002;
Thipatdee, 1996). These studies focused on the development of an enrichment
program, rather than understanding the problem solving process of gifted students.
Consequently, there is a need to examine how secondary Thai gifted students think
and what strategies they employ when solving non-routine problems if Thai educators
are to understand how to advance the educational experiences that support these
students in learning and engaging in mathematical problem solving.

Research Questions

The research questions for this study are:

1. What is the nature of the problem solving processes that Thai gifted
students use as they engage in solving non-routine mathematical problems?
2. What metacognitive behaviors do Thai gifted students exhibit when
engaged in mathematical problem solving?

Significance of the Study

This study provides knowledge about the mathematical problem solving

processes used by Thai gifted secondary students when solving non-routine problems.
Since the participants are in the Thai culture, the study results provide a cultural
component to the development of models of gifted students’ thinking. The study
findings help to extend the current research on student thought processes. The
characterization of gifted students thinking helps teachers plan and adjust problem
solving classroom instruction and expectations to better support gifted students’
development. The results of this study may be applied to the development of teaching
 8

methods and materials to be used in future mathematical problem solving classes to

facilitate the development of students’ abilities and skills for solving non-routine
problems. Further, this research is important for curriculum design in that the results
support the creation of curriculum or special programs that can be more effective and
supportive of gifted students. The conclusions and recommendations of this study
suggest directions for further research that may be undertaken in order to improve
understanding of mathematical problem solving performance.

Definitions

Problem solving refers to a cognitive process in which the student determines

how to solve a problem that he or she does not readily know how to solve (Mayer,
1992).
Problem solving processes refer to actions and strategies that students employ
to solve the problems.
Non-routine problems refer to the type of problems where students are not
familiar with the problem situations, they are not expected to have solved before, or
they have not regularly met them in the curriculum. Non-routine problems demand
thinking flexibility and extension of previous knowledge, may involve concepts and
techniques which will be explicitly taught at a later stage, and may involve discovery
of connections among mathematical ideas (Schoenfeld, et al., 1999).
Gifted students, in this study, refer to students who are able to do mathematics
typically accomplished by older students. They are able to engage in qualitatively
different thinking processes in solving problems (Sowell, et al., 1990).
 9

CHAPTER II

LITERATURE REVIEW

The purpose of this study is to examine the processes that Thai gifted students
use when they solved non-routine mathematical problems. This chapter provides a
review of the literature in several domains. The first section presents the discussion of
problem solving in the constructivist paradigm. The second section discusses the
nature of problem solving and its process. The third section describes the literature
about metacognition and problem solving. The fourth section presents the
methodology involved when investigating problem solving processes and the literature
about the think aloud method. The fifth section illustrates the literature on
mathematically-gifted students. The sixth section summarizes the research on gifted
students’ problem solving processes. The last section presents the conclusions based
on the literature review which guided this study.

Problem Solving in the Constructivist Paradigm

In the constructivist paradigm, knowledge is constructed from experience.

Learning is an active process that develops on the basis of experience. Learning is also
a personal interpretation of the world that should be situated in a realistic setting
(Merrill, 1991). A constructivist paradigm is closely related to the theorems expressed
by Piaget, Vygotsky and other gestalt theorists. A cognitive constructivist, Piaget
(1971) described a cognitive development theory based on the development of
cognitive processes from birth to adulthood through four stages: sensorimotor,
preoperational, concrete operations, and formal operations. In each stage, children
demonstrate new intellectual abilities and an increasingly complex understanding of
the world. The ages at which children progress through the stages are averaged,
varying with the environment and background of individual children.
According to Piaget, secondary students are in the formal operation stage. In
this stage, their cognitive structures are like those of an adult and include conceptual
 10

reasoning. They use symbols related to abstract concepts. They can think about
multiple variables in systematic ways, formulate hypotheses, and think about abstract
relationships and concepts. In addition, Sigel and Cocking (1977) point out that for
Piaget, the process that students exhibit in problem solving is the key to determining
their capacity for learning. At various ages, students handle problems in different ways
as underlying cognitive structures evolve and mature. The mistakes that they display
become a diagnostic tool for understanding the mental process associated with the
behavior observed.
In contrast with Piaget, a social constructivist, Vygotsky suggested that the
development process from birth until death is too complex to be defined by stages
(Driscoll, 1994; Hausfather, 1996). This process is described as dependent on social
interaction and social learning, actually leading to cognitive development. He focused
on the connections between people and the cultural context in which they act and
interact in shared experiences (Crawford, 1996). He described an experimental
development method to create a process of psychological development. This
experiment facilitated a measure called the “Zone of Proximal Development” (ZDP).
He described ZDP as the difference between problem solving that students are capable
of performing independently and their performance on problem solving with guidance
or collaboration. For instance, a student can solve algebra problems individually or
solve some calculus problems with guidance from a teacher. But, that student may not
be able to solve combinatorics problems no matter how much help the teacher
provides.
While Vygotsky emphasized the importance of social interaction to create the
ZDP, Brown and Reeve (1987) suggested that students can create and extend their
own zones of competence without assistance from others if they are given enough time
and are free to work without any external pressure. This idea is similar to results in
Garofalo’s (1993) who suggested that time should not be considered as a significant
factor for identifying the gifted problem solvers’ processes. According to Vygotsky,
students or individuals construct mathematical meaning as they participate in a variety
of communities within which particular mathematical practices, reasoning,
 11

conceptions, beliefs, and interaction patterns are shared (Cobb & Yackel, 1995).
Cultural factors also influence many aspects of cognitive processes that students
deploy in thinking and problem solving, such as knowledge base and structural
organization (Serpell & Boykin, 1994). Hilliday (1973) suggested that at an early age,
students formed mathematical concepts based on a need to systematically organize
ideas around them by using the language. The students have a need to explain their
processes with respect to refining and modifying their own language, and their
understanding of mathematical concepts. When they are encouraged to talk about their
processes to others and to continue practicing their newly-generated notions,
processes, and strategies, their competence for formulating ideas and solutions in
mathematics are developed.
Gestalt theorists believed that there is cognitive processing involved with
learning. Two individuals react differently to the same stimuli because of their past
experiences and perception of stimuli. Gestalt theory suggested that students need to
see a problem as a whole, how components of the problem relate to each other, and
how they can restructure the problem so that they gain knowledge (Kearsley, 1998).
The gestaltists are concerned with the solver rather than the task. Thinking and
problem solving require a dynamic interaction of an active mind with a task. Problem
solving usually involves the formulation of possible solutions by students. Cognitive
restructuring leads to insightful solutions. Gestaltists looked to discover dynamics
more than the product of the encounter (Blosser, 1973).
From a constructivist point of view, mathematical learning is a process of
individual construction. This view can be applied to consider the problem solving
situation. Students solve problems in different ways because of their prior experience,
culture, and community. This study sought to provide a perspective of the problem
solving processes in this paradigm. Participants of this study are native in the Thai
language and have a different culture from those participants in the literature whom
were mostly from western culture and native in different languages. These differences
may lead them to perform in different ways of problem solving.
 12

Nature of Problem Solving and Its Process

Problem solving is a vehicle for students to construct, evaluate and refine their
own theories about mathematics and the theories of others (NCTM, 1989). Engaging
in problem solving is not only to find an answer for a particular problem, but also
encouraging students to develop their own ability to think mathematically (Schifter &
Fosnot, 1993). In addition, Sternberg (1995) and Killen (1996) suggested that problem
solving is a form of inquiry learning where existing knowledge is applied to a new or
unfamiliar situation in order to gain new knowledge. Consequently, problem solving is
considered an integral part of teaching and learning mathematics. The problem solving
process involves knowledge, procedures, strategies, language and reflections (Klein,
2004). Mathematical problems need to be processed systematically, broken down into
the component parts in the process of constructing solutions. These processes require
problem solvers to decide how to work with strategies until getting the solution.
Therefore, it is also necessary to consider the information about problem solving
processes to develop a framework that can explain how students figure out
mathematical problems.
Mathematics educators and psychologists have suggested various problem
solving process models. Polya (1957) suggests a well-known model of problem
solving that consists of four stages:
1. Understanding the problem. Understand the verbal statement of the
problem. Study the data and the condition. Determine the unknown. Draw a figure and
develop a suitable notation for the problem. Separate the parts of the condition in order
to get a better understanding.
2. Devising a plan. Try to find a connection between the data and the
problem. Consider whether the earlier methods can be used now. Develop a plan
considering which calculations, computations, or constructions to perform in order to
obtain the unknown.
3. Carrying out the plan. Examine the details of the plan. Check each step
carefully. Implement the plan step by step.
 13

4. Looking back. Examine the solution and the path to obtain it. Consider how
to apply the result to other problems.
Polya also proposes a variety of strategies that can be used when students solve
problems. His strategies include: using diagrams, looking for patterns, trying special
cases, working backward, guessing and checking, creating an equivalent problem, and
creating a simpler problem.
Garofalo and Lester, (1985) propose a cognitive-metacognitive model. This
model consists of four stages:
1. Orientation refers to strategic behavior to assess and understand the
problem. It includes comprehension strategies, analysis of information, initial and
subsequent representation, and assessment of level of difficulty and chances of
success.
2. Organization refers to identification of goals and planning, global
planning, and local planning.
3. Execution refers to regulation of behavior to conform to plans. It includes
performance of local actions, monitoring progress and consistency of local plans, and
trade-off decisions (speed, accuracy, or degree of elegance).
4. Verification refers to evaluating decisions and outcomes of the executed
plans. It also includes evaluation of actions carried out in the orientation, organization,
and execution.
While Polya’s model describes the process as a linear progression from one
phase to the next phase when solving a problem, Garofalo and Lester (1985) indicate
that the transition from one phase to another occurs when solvers use their
metacognitive decisions. Their model describes the behaviors of problem solvers in
terms of metacognitive actions that occur during problem solving. Whereas in Polya’s
model metacognitive awareness appears only at the looking back stage of this model,
Garofalo and Lester’s model assumes that at each stage students will use
metacognitive knowledge about strategies and they will metacognitively monitor those
processes. For instance, students consider the level of difficulty in the orientation stage,
and they make a plan in the organization stage (Carr & Biddlecomb, 1998).
 14

Schoenfeld (1985) identifies four categories of knowledge/skills that are used

in his model of problem solving:
1. Resources-the mathematical knowledge possessed by a student including
facts and algorithms, which may be used in problem solving.
2. Heuristics-the strategies and techniques for problem solving such as
working backwards, drawing figures, reconstructing the problem, etc. which are used
for working on unfamiliar problems.
3. Control-the metacognitive behaviors, such as planning, monitoring, and
assessing, which are used for implementing resources and strategies.
4. Beliefs-the student’s views about himself, about the topic, and about
mathematics.
Montague and Applegate (1993) proposed a cognitive-metacognitive model of
mathematical problem solving. This model focused on seven cognitive processes and
three metacognitive processes. The cognitive processes included reading, paraphrasing,
visualizing, hypothesizing, estimating, computing, and checking. The metacognitive
processes included self-instruction, self-questioning, and self-monitoring. Self-
instruction consisted of selecting information from a given problem, monitoring
attention, correcting errors, using similar situations to define the problem situation,
and explaining the process or action executed. Self-questioning consisted of evaluating
the process or making a judgment, confirming answers, and self-instruction. Self-
monitoring consisted of evaluating themselves as problem solvers, giving partial
answers in a complex problem, and stating an answer to the problem.
These various models have been used to investigate problem solving processes,
but only two models by Garofalo and Lester’s model as well as Montague and
Applegate’s model have been used with gifted students as a framework to describe
problem solving processes in the literature (Garofalo, 1993; Montague, 1991,
Montague & Applegate, 1993; Sriraman, 2003). Several metacognitive actions during
problem solving were described in each stage of those models. Therefore, it is
necessary to carefully consider this literature related to this construct in order to
understand how metacognition is involved in the problem solving process.
 15

Metacognition and Problem Solving

In problem solving activities, problems differ from each other in terms of skills
required, mental processes used, and domain-specific knowledge required for solving
them. Some problems have only one correct answer. Some have many solutions.
However, the process of solving the problems needs to have two commonalities. First,
it requires the problem solver’s thinking to be directed toward achieving a goal.
Second, it requires metacognition, the awareness and management of one’s mental
process to guide this goal-directed thinking. Every problem consists of three parts:
givens, a goal, and obstacles (Anderson, 1985). The givens are the elements, their
relations, and the conditions that compose the primary from of a problem. The goal is
the outcome or solution. The obstacles are the characteristics of the problem and the
problem solver that make the problem difficult for the solver to consider the givens
with respect to the goal and to recognize when the correct transformation has occurred.
When students solve a problem, metacognition allows problem solvers to identify and
strategically work with these parts of the problem (Davidson & Sternberg, 1998).
In the early 1970s, the term of “metacognition” emerged as an explicit focus of
psychology research. Flavell (1976) defined metacognition as knowledge and
awareness of one’s own cognitive processes and the ability to monitor, regulate and
evaluate one’s thinking. He also argued that metacognition is multifaceted, including
knowledge about strategies, tasks, and persons as well as skills to evaluate strategies.
McCormick (2003) defined metacognition in problem solving as the knowledge and
processes used to guide thinking directed toward the successful resolution of a
problem. Mayer (2002) described metacognitive knowledge as knowledge of when to
use, how to coordinate, and how to monitor various skills on problem solving. For
example, knowing how to summarize is a skill, but knowing that one should take
detailed summary notes from a lecture requires a metacognitive knowledge.
Metacognitive processes are mental operations that direct the cognitive functions of
individuals and support a learning conceptualization (Mevarech & Kapa, 1996;
Nastasi & Clements, 1990).
 16

Research studies found that metacognitive knowledge and processes help

problem solvers become more efficient at handling problems in three aspects: (a)
define what the problem is, and form a mental representation of its elements, (b) select
suitable plans and strategies for achieving a goal, and (c) identify and master obstacles
that facilitate progress (Davidson & Sternberg, 1998). Metacognition in problem
solving involves the processes of planning, monitoring, and evaluating specific
problems - particularly in making the mental representation and the selection of
appropriate strategy (Flavell, 1992; McCormick, 2003). The use of metacognitive
processes supports problem solvers during the solution process and improves their
ability to obtain the goal. The more problem solvers control and monitor the strategies
they use, the better are their abilities to solve a problem (Fortunanto, et.al, 1991; Kapa,
1998; Swanson, 1990).
The relationship between developing conceptual knowledge and metacognitive
knowledge was found in the study of Kuhn, Garcia-Mila, Zohar and Anderson (1995).
The authors examined the development of metastrategic knowledge (knowledge about
where, when, and how to apply strategies and an understanding of structure of the task)
and metacognitive competence (an awareness of the need to reflect on the content of
one’s knowledge). Results indicated that having understood strategies would not
assure the correct use of strategies. Instead, a good strategy use requires knowledge
about when and when not to apply that strategy. Individuals who applied developing
strategies well were more likely to have an understanding of how strategies fit together
and fit to the tasks to which they are applied. Metastrategic knowledge was not found
to develop entirely before or after developing a strategy, but it seemed to occur with
application of the strategy. The findings suggested that metacognition emerges from
knowledge and also guides further learning.
Research conducted on the role of metacognition in mathematical problem
solving considered metacognitive processes not only as driving forces that influenced
all stages of problem solving, but also was connected with a wide range of non-
cognitive factors (Lesh, 1982; Schoenfeld, 1982; Silver, 1982). For the successful
solution of any complex problem solving task, a variety of metacognitive processes is
 17

necessary. Lester, Garofalo, and Kroll (1989) found that students with higher levels of
metacognitive ability performed better in problem solving situations.
Kapa (1998) examined the effect of an instructional intervention designed to
influence students’ metacognition while solving mathematical word problems. The
intervention program supplied metacognitive directive questions during the solution
process. The participants included 441 students from eighth grade who were randomly
assigned to one of four computerized learning environments with different kinds of
metacognitive support during the different phases of the problem solving process:
(1) during the solution process and after the completion of the problem solving process,
(2) during the solution process, (3) at the end of the solution process, and (4) no
metacognitive support. Results indicated that learning environments that provide
metacognitive support during the solution process were significantly more effective
than learning environments that provide metacognitive support only at the end of the
solution process. The directive questions were effective in stimulating students’
previous knowledge and activating relevant thinking functions for the specific solution.
Students with low prior knowledge were more significantly influenced by
metacognitive supports than students with high prior knowledge.
Pugalee (2001) used Garofalo and Lester’s model as a framework to describe
the role of writing in mathematical problem solving processes of 20 ninth-grade
students. The author identified metacognitive behaviors in the four stages of the
problem solving processes: orientation, organization, execution, and verification
stages. Results indicated that several metacognitive behaviors were evident in
students’ written descriptions. Metacognitive actions played an important role for an
initial understanding of the problem and moving the students towards identifying a
solution plan. The results revealed that writing is a vehicle that supports students in
demonstrating their metacognitive behaviors during the process of solving problems.
 18

Methodology Involved in the Problem Solving Process

To investigate the processes that a problem solver employs while solving a

problem, some method of process tracings must be considered. Lester (1980) suggests
several approaches such as thinking aloud, introspection, retrospection, and written
inventories. But the major methodology used is the think aloud method. The think
aloud technique involves the participants’ translation of their thoughts into words and
reciting the translations aloud, and verbalizing aloud all the steps that occurred during
their problem solving. Typically, an audiotape or videotape recording is made, and a
trained observer uses a protocol coding procedure to record certain aspects of the
participants’ behaviors.
The think aloud methodology is recognized as a useful technique that can
provide a window for uncovering psychological mechanisms and knowledge
structures underlying human problem solving activities with regard to specific tasks.
Think aloud techniques have been used as a research technique to access individuals’
thought processes while they perform a task in several content areas, such as reading,
history, and chemistry (Bowen, 1994; Crain-Thoreson, et. al, 1997; Weinburg, 1992).
In mathematics, the think aloud method has been used to examine the processes
involved in the problem solving (Kantowski, 1977; Kilpatrick, 1969), analyze
individual differences (Rowe, 1980) or elucidate self-regulation or control during
problem solving (Artzt & Armour-Thomas, 1992; Schoenfeld, 1985).
When using the think aloud method with problem solving, the participants are
expected to verbalize their thoughts, speaking aloud all that comes to their minds
while performing the task at hand. The participants are not asked to give any
interpretation of their thoughts (Someren, Barnard, & Sandberg, 1994). If they fail to
verbalize for more than five seconds, prompts are used such as “keep talking,” “keep
telling what you are doing,” or “say everything you are thinking and doing” (Lawson
& Chinnappan, 1994; Montague & Applegate, 1993). Most researchers introduce the
think aloud technique to participants with a sample problem. Afflerbach and Johnston
(1984) suggested that training in verbal reports could influence the type of reporting
 19

that students do, and the cueing responses expected by the researcher. Practice
provides participants with more understanding and confidence before using the
technique with the research problems.
A protocol analysis is a procedure to identify psychological processes in
problem solving tasks (Newell & Simon, 1972). The development of the think aloud
protocol was described in the study of Ericsson and Simon (1984). They provide a
rationale and techniques for conducting protocol analysis based on information
processing theory. According to Ericsson and Simon (1993), three assumptions are
made while adopting the methodology of protocol analysis:
1. Only the outputs of the cognitive processes are available to the
consciousness and are verbalized.
2. Performance on any given task can be conceptualized as a set of sequential
processes with intermediate products that are available to working memory. The
intermediate products are conceptualized as being directly associated with observable
indicators.
3. The total time to generate a response is equal to the sum of the individual
cognitive processes.
Hence, concurrent verbal reports do not change cognitive processes although
concurrent verbalization does increase the total time to perform tasks. Think aloud
protocol has an advantage over simple observation as the evaluator may gain valuable
insights into what the participant is thinking on the spot. Less discrepancy in the
verbal response of the participant and what he or she actually thinks exists since the
participant does not need to recall from long-term memory events that have taken
place earlier. Ericsson and Simon (1984) recommend that additional information
should be collected in the form of retrospective reports after the task to avoid any
interruptions of task flow. Think aloud protocol is often used with other
methodologies to gather more in-depth responses from participants, such as individual
interviews.
Individual interviews are used after participants finish solving each problem to
ask them to explain how they solved the problems and what they think about their
 20

solutions. The interview provides a format for participants to express their thoughts as
problem solvers, to afford opportunities for clarification of their think aloud responses,
and as a way for clarifying their strategies used (Berkowitz & Cicchelli, 2004). As
students describe their solutions during the interview, the researcher can closely
follow their processes. A number of research studies have used the think aloud method
with individual interviews with gifted students to access the students’ thought
processes (Gorodetsky & Klavir, 2003; Lawson & Chinnappan, 1994; Montague,
1991; Montague & Applegate, 1993; Montague & Applegate, 2000; Span &
Overtoom-Corsmit, 1986). Students’ verbal protocols are usually captured on
videotape or audiotape for later analysis.
Other techniques for investigating problem solving processes have been used
less frequently: introspection, retrospection, and written inventories. Introspection
differs from thinking aloud in that it requires participants to analyze their thinking as
they attempt to solve a problem. Retrospection requires analysis by the participant
solving the problem and involves a discussion between the participant and the
interviewer. Written inventories attempt to combine the three techniques into a written
form. These techniques eliminate the need for audio taping and protocol coding. With
written instruments, the researcher can collect data from a large number of participants
in a short period of time. However, each technique has inherent weaknesses and
strengths. In order to decide which method to use, researchers need to consider the
constraints and try to eliminate drawbacks that might occur.

Mathematically-gifted Students

In the past, the concept of giftedness was initially associated with high IQ. This
concept is based on the assumption that gifted students are born with high intelligence
and other aspects such as, the ability to achieve high test scores and are able to excel
in all areas of school. Today, with the development of psychology and cognitive
science, and understandings of how learning takes place, giftedness has been redefined
and conceptualized. The earlier assumptions nevertheless continue to be accepted. The
 21

psychometric approach views intelligence as a single cognitive skill, which is mostly

measured by IQ tests. It leaves out much of what intelligence is. Analytical mental
capabilities are measured, but practical and creative aspects are ignored (Gardner,
1983; Sternberg, 1986). On the other hand, the cognitive approach allows for many
types of intelligence of which only some can be measured. This approach focuses on
practical intelligence, which can be measured by the way that people solve everyday
problems. Sternberg (1986) argued that traditional measures of intelligence, such as IQ
tests, capture only a part of what it means to be intelligent, which he defines as the
ability to adapt flexibly and effectively to the environment. He suggests that
intelligences are developing abilities rather than fixed characteristics of an individual.
No universally accepted definition of general giftedness exists (Gagne, 1995;
Morelock, 1996; Sternberg, 1993). This fundamental lack of agreement extends to
mathematics, where differing descriptions of high mathematical performance and
ability are evident in the literature (Sowell, 1993). Krutetskii (1976), a Russian
psychologist, suggested that there are three components of mathematical giftedness:
reception, processing, and retention of mathematical information. Reception refers to
the ability to formalize the perception of material and grasp the formal structure of a
problem. Processing refers to the ability that includes thinking logically, generalizing,
reasoning, and striving for clarity of solution, as well as flexibility of mental
processes. Retention refers to the ability in mathematical memory that involves
generalizing memory for mathematical relationships, schemata for arguments and
proof, and methods of problem solving. He also mentions that computational ability,
memorization of symbols, numbers and formulae, spatial ability, and the ability to
visualize abstract mathematical relationships and dependencies are not absolutely vital
for high mathematical aptitude.
In the United States, the Center for Talented Youth (CTY) at John Hopkins
University assumed that 12-13 year old students with high scores on SAT-M have
enhanced problem transition ability and high ability to represent and manipulate
information in short term memory. The center provides fast-paced math classes during
summer in intensive three-week residential courses for students who scored at or
 22

above a cutoff score on the SAT-M. Participants register and then take an Algebra
Diagnostic Test to place them at the starting point before learning the Precalculus
Mathematics Sequence. The rationale is to support young mathematically gifted
students in swift and improved mastery of concepts within the conventional
mathematics curricula that would not be possible in their regular schools. The CTY
approach provides the gifted with possession of exceptionally large amounts of
mathematical knowledge and outstanding successes in obtaining correct solutions for
conventional problems. Speed of solving problems and memorization of symbols,
numbers, and formulae are not considered to be essential for mathematical giftedness
(Wieczerkowski, Cropley, & Prado, 2000).
Several indicators of mathematical giftedness have been discussed. These
indicators consist of the ability to understand and apply ideas quickly, ability to
transfer mathematical concepts to an unfamiliar situation, and high ability to see
patterns and think abstractly. Mathematically-gifted students also have unusual
curiosity about numbers and mathematical information. In addition, they also use
creative strategies and solutions, use analytical, deductive, and inductive reasoning as
well as persistence in solving difficult and complex problems (Holton & Gaffney,
1994; Miller, 1990).
General cognitive operations such as ability in abstracting concrete problems,
the ability to generalize, reversibility of operations, fluency of thought, or strategic
decision-making have been regarded as major variables for mathematical giftedness
(Wieczerkowski, Cropley, & Prado, 2000). These variables were used to define
mathematical giftedness and the assumption was that general cognitive operations do
not change across specific academic areas. In the case of mathematics, the concept of
general academic giftedness improves by identifying specifically important
characteristics of mathematical talent.
Sousa (2003) stated some attributes of mathematical giftedness that include the
following abilities:
• Learn and understand mathematical ideas very fast
• Demonstrate multiple strategies for solving problems
 23

• Engage other students in their activities

• Keep their concentration and show great persistence in carrying out
solutions
• Switch approaches easily and avoid nonproductive approaches
• Operate with symbols and spatial concepts easily
• Recognize similarities, differences, and patterns quickly
• Consider problems more analytically than holistically
• Work systematically and accurately
• Use charts, tables, and graphs to demonstrate their mathematical
abilities in other subject areas
Additionally, non-cognitive variables were also suggested for considering
mathematical giftedness. These variables included characteristics such as flexibility,
openness, and tolerance for ambiguity. Motivation factors like curiosity, willingness to
take risks, and task commitment were also included in the components of giftedness
(Wieczerkowski, Cropley, & Prado, 2000). These attributes allowed the gifted
students to be better mathematical problem solvers.
With various definitions of giftedness in the literature, there is a need for
research that begins to differentiate the effect of various abilities on mathematical
problem solving. Moreover, the characteristic that separates the gifted from non-gifted
students in mathematics is the quality of their thinking. Gifted students have high
quality of thinking skills, such as induction, deduction, analysis, generalization and
abstraction (Johnson, 1980; Lester & Schroeder, 1983).

Research on Gifted Students’ Problem Solving Processes

Since gifted students generally have a greater potential in problem solving than
other students, processes that occur when they solve problems have been considered in
order to understand this different ability. In the literature, most students’ problem
solving processes were investigated by comparing gifted students with average
students. Not many studies have worked in this particular area. This review focused on
 24

research studies that discussed the processes of secondary gifted students when they
attempted to solve non-routine mathematical problems.
Garofalo (1993) compared the differences between two problem preferences,
the meaning-oriented and number-oriented students in their problem solving.
Meaning-oriented students referred to students who approached problems by trying to
form meaningful interpretations of conditions and quantities in the problems. They
constructed meaningful relationships between quantities and then used these
relationships to develop an action plan. In contrast, number-oriented students referred
to students who approached problems by focusing on the magnitudes of the numbers
in the problem to determine what to do. They used many calculations more often to
understand the problem.
In order to identify students in both groups, the judgments of four mathematics
educators were used from a focus group in the larger studies concerned with strategic
and metacognitive aspects of students’ mathematical problem solving (Lester,
Garofalo, & Kroll, 1989; Garofalo, 1992). Participants in this study included 11
students in grade seven from a middle school. Three of them were identified by their
school districts as gifted in mathematics. These students had a mathematics percentile
rank of 99 on the Iowa Test of Basic Skills (ITBS), and had at least one distinguishing
characteristic relevant to mathematics achievement. They were attending a summer
math program for the gifted at the University of Virginia. Another three students were
in an advanced math course and had ITBS mathematics percentile ranks between 87
and 99. These six students were strictly meaning-oriented in their approaches. Five of
the students were in a regular seventh grade math course and had ITBS mathematics
percentile ranks between 36 and 82. They tended to be number-oriented in their
approaches to solving the problems. The author developed a set of eight problems that
included routine, multi-step, and non-routine math problems. Each participant was
asked to solve 4-8 problems that were taken from this set for 45 minutes with an
interviewer. After finishing their task, they were asked to describe their problem
solving strategies and affects. These interviews were videotaped and analyzed using
the frameworks described in Garofalo and Lester’s model.
 25

Results indicated that both groups preferred the easier routine problems in
graded situations. However, in non-graded situations, the responses were different.
The meaning-oriented students preferred a non-routine problem or a multi-step
problem because that challenged them and they had not done the problem previously.
Yet, the number-oriented students still preferred an easier routine problem or a one-
step problem because it was easy, short, and simple. They did not like multi-step and
non-routine problems because they had more difficulty understanding those problems.
They preferred to work on the problems that did not require much analysis, planning,
and strategy. The meaning-oriented students viewed problem solving as an
opportunity for understanding, figuring, and accomplishment. They reread the
problem many times, spent time thinking about the solution before writing it down,
and checked again when finished. They tried to find a general solution to the
problems.
On the other hand, the number-oriented students viewed problem solving as
something to get over in almost any way possible. They did not reread the problem,
spent a lot of time doing calculations, and did not produce meaningful solutions. These
findings showed that the successful problem solvers spent more time in reading,
planning, and evaluating than the less successful problem solvers. Thus, time should
not be considered as a significant factor for identifying the gifted problem solvers.
Montague (1991) examined strategies used for solving problems by eighth
grade students, three gifted and three learning-disabled gifted. Results indicated that
gifted students read the entire problem, rereading it part-by-part while solving. They
paraphrased the problem while reading and progressing through the problem. They
accessed a variety of problem representation strategies to facilitate linguistic and
numerical information processing. They were aware of the strategies they used. They
also showed the ability to translate information to mathematical equations and
algorithms. They formulated goal statements before developing solution plans, worked
systematically, quickly, with little hesitation. They engaged in self-instruction, self-
questioning, and self-monitoring. They redid the problem by working through the
whole problem, rereading the problem, redoing computations, and checking steps and
 26

processes. Learning-disabled gifted students decided on a solution path after reading

the problem once, then wrote the algorithms, and computed the answer. They also
used mental imagery and verbalization while reading to remember information in the
problem, but there was no evidence that they actually paraphrased or translated the
problem into their own words. They reported using estimation to predict answers,
whereas only one gifted student estimated his answers. All six students used
visualization such as pictures, diagrams, or tables representing the problem to help
when they were having difficulty solving the problem. This study showed that gifted
students use paraphrasing as a process of translation that supports them in successful
problem solving.

In the field of geometry, Lawson and Chinnappan (1994) compared solutions

of geometry problems between high-achieving and low-achieving students.
Participants were eleventh grade students, five high-achieving and five low-achieving.
They were selected at random from a pool of volunteers at the top and bottom levels
of their mathematics examination at the end of grade 10. Each participant was
videotaped while solving four problems involving circles and triangles using the think
aloud method. These problems were in the appropriate difficulty for students at this
level of mathematics based on pilot tests with other students, teachers, and
mathematicians. After students completed four problems, they were asked to recall
theorems and knowledge used that were relevant to the problems that had just been
solved, as well as any other geometry theorems known. Students’ responses were
transcribed and coded in five categories: identification of given information, control or
management, generation of new information, metacognitive use, and error.
Results indicated that the high-achieving students recalled significantly more
theorems than did the other group. They not only accessed a greater knowledge of
geometry, but they also used that knowledge more effectively. For difficult problems,
they generated more information and used it to access further relevant knowledge.
They showed more frequent management of their processing behavior. Although both
groups had problem difficulty, the low-achieving students accessed a smaller body of
available knowledge and were less effective in activating that knowledge to be useful
 27

in the path to a solution. The coding protocols indicated that the high-achievers
process more events with a higher level of activity. This greater activity occurred in
identification of given information and management of processing activity.
Sriraman (2003) examined how ninth-grade students approached problem
solving, abstracting, and generalizing mathematical concepts. Abstraction is a process
that occurs when students focus on specific properties of a given object and then
consider these properties in isolation from the original. Generalization is the process
where students derive or induce from particular cases. It includes identifying
commonalities in the structure of problems and their solutions. It is inseparably linked
to the operation of abstracting. Nine students participated in this study. Four of them
were identified as gifted in their primary schools based on many factors, including the
IQ scores (over 124), the Stanford Achievement Test (95 percentile), teacher
recommendations, and counselor recommendations. Participants were in an
accelerated Algebra class taught by the author. They were routinely assigned a
problem every week, to be solved in their journals so they had experience in journal
writing. For this study, they were asked to solve five non-routine combinatorial
problems in their journals. These problems were assigned over the course of three
months at increasing levels of complexity. They were chosen because they would
facilitate representation, reasoning, abstraction, and formulation of generalization. The
generality that characterized five problem solutions used the pigeonhole principle. The
researcher gave three cues to students for their writing:
(1) What is the problem asking?
(2) How would you begin solving the problem?
(3) Solve the problem and write a summary of what worked and what did not
work.
These cues were to initiate the four stages of Garofalo and Lester’s model.
Students had 7-10 days to solve each problem. After collecting and reading students’
solutions, the researcher recorded in his journal possible questions for the interview.
Students were asked to verbalize their processes while solving a given problem. The
journal writing and transcribed interview data were coded and analyzed.
 28

Gifted students started solving the problems by restating them in their own
words. They invested time in understanding the problem situation and identifying the
assumptions clearly. They began working with simpler cases that modeled the given
problem situation. They made a list, used a particular case and looked for patterns.
They made a plan, executed it, and performed generalizations. Finally, they checked
for accuracy and consistency. They also verified the solution and identified the
similarity in the structure of all problems. These processes showed major differences
between gifted and non-gifted students in the orientation and organization stages of
Garofalo and Lester’s model. Moreover, all gifted students succeeded in discovering
and verbalizing the generality of the five problems, whereas five of the non-gifted
students were unable to discover the hidden generality.
Span and Overtoom-Corsmit (1986) compared the differences of information
processing in mathematical problem solving between gifted and average students. In
order to select the subjects, three standardized tests (a test for general information
from a Dutch research project, the Raven Progressive Matrices, and a test for
creativity) were administered to 399 students. Finally, 14 students of each group in
grade eight participated in this study. They were asked to solve seven math problems
individually using the think aloud method with the experimenter. These problems
covered the areas of combinatory, transformation, permutation, and chance
calculation. After students finished solving each problem, they were interviewed to
describe their solutions. All students’ responses were videotaped and transcribed for
analysis.
This study focused on three information processes: orientation, execution, and
evaluation. Results indicated that overall the gifted solved more problems and were
faster in getting a correct solution. They processed information in a different way and
at higher levels than average students. In the orientation stage, the gifted took more
time to orient themselves to carry out the problem and did the process more
thoroughly. They reflected beforehand on what to do and planned their approaches.
The average students immediately started to try a solution before they tried to
understand what they were required to do. In the execution stage, average students
 29

tried to solve problems by trial and error for nearly all problems, while the gifted tried
to solve problems with other strategies that were useful and efficient in the execution.
In the evaluation stage, gifted students were apt to control themselves automatically
and constructed more mathematical problems. They were also more capable of
discussing their strategies, while the average students failed to consider the possibility
of evaluation. These findings were similar to the study of Garofalo (1993) in the
context that the gifted spent more time in the orientation stage than average students.
This result might lead them to understand the problem clearly and succeed in solving
the problem.

Conclusions

This chapter discussed the literature related to problem solving, metacognition,

and gifted students. The previous research is critical in understanding how gifted
students solved mathematical problems and processes that they demonstrated. The
studies in this review commonly assumed that it is important to investigate the
problem solving process of these students in order to promote students’ existing
processes and improve their mathematical problem solving ability.
To access students’ processes, the research literature employed a qualitative
design with the think aloud method and individual interviews. Training the students in
the think aloud method was found to be necessary before collecting data. An audiotape
or videotape was also used to record the students’ responses. This design was
appropriate to gather descriptive information about what actually occurs during the
problem solving process session rather than using some merit scale typically used in a
quantitative design. The think aloud method and individual interviews were used
because these approaches facilitated obtaining qualitative data. An improvement of
these methodologies should be considered in future research. With the complexity of
the problem solving process and individual differences, researchers should be closely
involved with the students’ processes as observer participants using a clinical
investigation, a case study, or multiple cases study.
 30

The clinical approach is particularly necessary in considering the process of

problem solving. This approach allows researchers to observe how a subject operates
and to determine the level of sophistication instead of only the number of correct
solutions (Kantoski, 1977; Lester, 1980). Moreover, case studies of subjects selected
because of giftedness in mathematics also provide useful information for
understanding the processes of problem solvers. Collecting data to describe students’
processes with the same set of problems particularly provides insight to specific
information of individual differences that are specific to how individuals solve
problems naturally without instruction.
Results in the literature indicated that metacognitive components might
enhance the students’ ability to be a better problem solver. Several metacognitive
actions were evident during problem solving processes as addressed in many models.
Gifted students also used metacognitive processes effectively in their problem solving.
Therefore, it would be reasonable to explore metacognitive behaviors that students
demonstrate while solving the problems in order to improve them gets more efficient
work. The findings also indicated that the successful problem solvers spent more time
reading, planning, and evaluating than the less successful problem solvers.
Consequently, time allowed should not limit or be a constraint when students work on
problems.
Many criteria were used to identify the subjects as gifted students for
investigating their processes in mathematical problem solving. Most research used IQ
scores on the intelligence tests (Montague, 1991), the scores on standardized
achievement tests (Span & Overtoom-Corsmit, 1986; Sriraman, 2003), and the
mathematics achievement scores on the school district test (Garofalo, 1993). The
evidence of characteristics of giftedness according to a standard scale and
documentation of the need for a special program were used (Montague, 1991).
Additionally, participating in a specific class for the gifted, teacher recommendations
and counselor recommendations were also mentioned as criteria (Sriraman, 2003).
Therefore, the variety of identification of giftedness emphasizes the need for research
to clarify which criteria have direct effects and promote problem solving ability. The
 31

identification should depend on specific characteristics of this particular group and

meet the purpose of the study.
Although mathematical problems in the literature were non-routine problems,
they were related to different mathematical content areas. The variety and the
complexity of the problems may have affected the students’ processes and their
solutions. Students might not succeed in solving problems because they may lack
some appropriate abilities such as reading ability, translating ability, and representing
ability to understand the problems. They may also not have enough knowledge and
experience to solve the problems. Consequently, researchers should validate that the
problems are suitable for participants and meet the purpose of the study. Gathering the
problem solving processes depends on not only the problem difficulty and complexity,
but also students’ learning styles and their strategy preferences; both need to be
considered when conducting further investigations about problem solving processes.
 32

CHAPTER III

METHODOLOGY

This chapter presents the design and methods selected to answer two research
questions:
1. What is the nature of the problem solving processes that Thai gifted
students use as they engage in solving non-routine mathematical problems?
2. What metacognitive behaviors do Thai gifted students exhibit when
engaged in mathematical problem solving?

Design of the Study

A phenomenological qualitative design was selected for the study. The

strengths of this design are in the ability to explore how and why phenomena occur in
the study of dynamic processes. This design is also useful for describing complex
phenomena, such as problem solving thought processes (Johnson & Onwuegbuize,
2004; Patton, 2002). In fact, Kilpatrick (1978) recommended qualitative studies over
experimental studies in research on problem solving. He mentioned that experimental
studies where all variables are under control are not likely to be of much value in the
present stage of understanding how students solve complex mathematical problems.
Since this study was focused on students’ thought processes, the study can be
considered as one that makes obvious a complicated and dynamic thought-based
phenomenon. Problem solving processes are viewed as being personal in nature and
may vary from student-to-student. A qualitative design is appropriate for this study in
order to be in a position to gather descriptive information about what actually occurs
during the problem solving process, rather than using a merit scale typically used in a
quantitative design. Although a small sample size suggests a question of
generalizability of the study findings in reflecting the nature of the population, this
qualitative research is not focused on generating generalizeable findings; instead, this
 33

study is about identifying processes of problem solving that occur with each of these
gifted students. These samples are not generalized to gifted students as a whole;
however, the results and implications of the study are important for understanding the
problem solving processes of gifted students in Thailand.
Within a qualitative design, a multiple-case study that utilized the think aloud
method, individual interviews, and researcher’s field notes from observations were
used for gathering data. Merriam (1998) describes a case study as an examination of a
specific phenomenon or process like problem solving. A qualitative case study also
seeks to describe that specific phenomenon in depth and detail, holistically, and in
context (Patton, 2001). A case study approach also encourages analysis of an
individual case in personal ways. Kroll (1988) argues that for individuals, a case study
with interviews is the most appropriate research design. Since this study investigated
how each of the gifted students solved problems, rather than examining outputs and
outcomes, the nature of the case study approach allowed for insights into their
personal thinking. One of the study’s purposes was to propose a model of the problem
solving process used by Thai gifted students. This purpose was in line with the main
purposes of case study research that works toward a comprehensive understanding of
the cases, and developing general theoretical statements about regularities in social
structure and processes (Becker, 1968).
This research study used multiple methods including a think aloud procedure
as the students were engaged in solving problems, individual interviews following the
problem solving processes, researcher’s observation of the students when engaged in
solving the problems, and an analysis of students’ solution papers to investigate
students’ thought processes during problem solving. Prior to the data collection, the
participants practiced the think aloud technique with a sample problem. The procedure
provided participants with important practice for understanding and developing
confidence prior to utilizing the technique with the research problems.

Over the period of data collection, the participants were given three
mathematical problems to solve individually using this think aloud method. They
continuously spoke aloud while working on the problems describing their thinking.
 34

They also had unlimited time for solving each problem. The individual interviews took
place after the participants finished solving each problem. In the interviews, they were
asked to explain their thinking as they solved the problems and what they thought
about their solutions. As they described their solutions, the researcher closely followed
their processes. From the researcher’s experience, gifted students usually do not write
many explanations in their solutions. Some parts of the participants’ solutions were
not clear because they skipped some details of the strategies that they were thinking.
They also figured that their thinking was clear in their written explanations. The
interviews helped the researcher clarify the solutions and understand their reasoning as
they solved the problems.

Since misinterpretations of the data might have resulted with a single

researcher as the data collector, the researcher maintained a record of field notes
explaining her reflections about the activities. The field notes included the descriptions
of questions, reactions, and behaviors that occurred during data collection.

Research Setting

In Thailand, secondary mathematically-gifted students are usually in the same

classroom with average students in regular schools. They generally work on the same
curriculum during the class period; however, schools with enough resources to
promote the high abilities of these students, provide extra-curricular activities for the
gifted students in the evening or on the weekend. Most training emphasizes teaching
advanced content in mathematics by university professors. One of Thailand’s ways of
promoting mathematically-gifted students’ abilities is through participation in the
International Mathematical Olympiad (IMO). The IMO is the annual world
championship mathematics competition for secondary students. This event is
recognized as a strategy of fostering mathematical talent (Wieczerkowski, Cropley, &
Prado, 2000).
The Thai government has given the Institute for the Promotion of Teaching in
Science and Technology (IPST) the responsibility of selecting the Thai representatives
 35

for this contest since 1989. For the Thai Mathematical Olympiad (TMO), secondary
gifted students from around the country complete in two rounds of examinations for
entrance into the project each year. Both examinations focus on non-routine
mathematical problems with the mathematical content limited to grade 11 according to
the national curriculum. In June 2005, a total of 7,982 gifted students applied for the
TMO project. Among these students, 6,861 students took the first round examination
that examined their mathematical problem solving abilities through multiple-choice
questions and short answer questions. Only 42 students were selected from the first
round exam to take the second round examination, using all open-ended mathematics
questions where students are required to show their written solutions. Finally, 25
students were selected from the second examination to participate in the TMO project.
These 25 students enrolled in the TMO training camp at IPST for three rounds from
October 2005 to June 2006. The rigorous schedule of the training is meant to enrich
students’ problem solving abilities in advanced mathematical content. During these
trainings, students are selected through examinations in each round until six
mathematically-gifted students are selected as the Thai representatives for the IMO.

Participants

According to Sankar-DeLeeuw (2004), when studying a subpopulation of

gifted students, it is difficult to select samples of gifted students from a large
population. A random selection method was impossible. Some studies of this
population had been conducted while the students were actually participating in
various events, such as after-school programs, home schooling, and other influences
that contribute to their overall development. Like other research, this study was
conducted while 24 gifted students, who were eligible for the TMO training camp,
were participating in the TMO training camp at the IPST from October 13 to
November 4, 2005. The researcher used purposeful sampling to select the participants
out of the pool of 24 gifted students within the camp.
 36

Although these students were selected by the entrance examinations where the
mathematical content was limited to grade 11, their grade levels varied from grades 8
to 12. Thus, the students had different levels of mathematical background. However,
they have had prior experiences in solving non-routine mathematical problems from
the entrance examinations that also required them to provide written solutions. With
the definition of gifted students used in this study, student participation was limited to
students in grades 8-10 even though these students were able to do mathematics
typically accomplished by older students (Sowell, et al., 1990). Based on these factors,
the researcher considered the following criteria from the IPST database in to the
process of identifying students with similar mathematics background; students were
considered if they
• had similar scores on the second round of the entrance examination to the
TMO project
• did not participate in the training camp in the previous year
• were in grade 10 or below.
Seven gifted students meeting these criteria were assigned pseudonyms for describing
their backgrounds shown in Table 3.1.

Table 3.1. Background of 7 Gifted Students Meeting the Selection Criteria

Number Pseudonyms Age Grade School

1 Pradya 13 years 3 months 8 A
2 Tana 13 years 5 months 8 A
3 Sira 14 years 9 B
4 Wude 14 years 5 months 9 C
5 Nuttapong 14 years 10 months 9 A
6 Nipa 15 years 1 month 9 A
7 Sakda 16 years 6 months 10 D
 37

Five of the seven gifted students who matched the selection criteria were
selected to assure a diversity of school, grade, gender, and age. The researcher met
these students and described an overview of the study on the first day of training. The
students received the informed consent letter. After they read the letter, the researcher
explained the details in the Thai language. The students asked some questions before
agreeing to participate. Voluntary participation was emphasized with no pressure for
the students to be involved.

Data Collection Protocols

Three protocols were used for gathering data for this study: (1) the think aloud
protocol, (2) the mathematical problem solving tasks, and (3) the interview protocol.
The think aloud protocol (Appendix B) included the instructions for practicing the
think aloud method with a sample problem.
The mathematical problem solving tasks (Appendix C) included three non-
routine problems selected and modified from a variety of sources, including
mathematical journals, textbooks and examination contests (ApSimon, 1991;
Covington, 2005; Gardiner, 1987; Krantz, 1996; Posamenteir & Schulz, 1996;
Posamenteir & Salkind, 1996; Schoenfeld, 1985). The researcher also considered the
training curriculum before developing a pool of problems that meet the following
criteria:
• The problems are non-routine; that is, the students are not familiar with
the problem situations, they are not expected to have solved the problems
previously, nor have they worked on them in the mathematics
curriculum. Non-routine problems demand thinking flexibility and
extension of previous knowledge, may involve concepts and techniques
that will be explicitly taught at a later stage, and may involve discovery
of connections among mathematical ideas (Schoenfeld, et al., 1999).
 38

• The problems cover mathematical content areas in number theory,

combinatorics, and geometry, which are the major subjects in training
students for the IMO.
• The problems should challenge students’ thought processes; that is, the
problems must examine the upper levels of the cognitive domain
according to Bloom’s Taxonomy: analysis, synthesis, and evaluation
(Bloom, et al., 1956).
• The solutions should not require mathematical concepts and skills that
students have not learned according to the national curriculum.
A pool of problems consisted of 13 problems: 6 number theory problems, 3
combinatorics problems, and 4 geometry problems. The pool was examined for
content validity by seven experts with expertise in mathematics and mathematics
education. Three of the experts were the instructors and former team leaders of the
TMO project. The experts selected one problem in each area and suggested important
feedbacks for the study. After modifying problems according to the experts’
suggestions, one problem was selected for the study from each mathematical area
(number theory, combinatorics, and geometry).

The interview protocol (Appendix D) consisted of two parts. The first part was
intended to elicit the participant’s background, such as age, grade, and how they felt
about problem solving. The second part proposed to elicit each participant’s thought
processes in solving a given problem. The interview protocol was also examined by
experts similar to those examining the mathematical problem solving tasks. The
general protocol questions were used as a guideline of the interview. Additional
questions were added to further meet the varying participants’ thought processes.

Before using all three protocols with participants, the researcher translated
them into the Thai language. Two experts (Thai-English speakers and specialists in
mathematics education and education) examined the accuracy of the translation. They
also translated protocols from Thai to English and English to Thai to check for
consistency. After modifying the protocols, the think aloud protocol was typed on one
 39

page for the participants to read. The instructions included the attachment of blank
papers for scratching and writing the solution of the sample problem. For the
mathematical problem solving task, three problems were typed on single sheet of
paper, one problem on each page.

Three Problems for Data Collection

In this study, the participants individually solved three problems using the
think aloud method. The researcher asked the participants to explain their solutions
and the processes they used as they worked on the problem. The following section
describes each problems used in the study.

Problem One

Problem One is: Does a Friday the 13th occur every year? Explain your
reasons.
This problem is related to the mathematical concept in number theory. The
problem requires the student to apply the concept of division by 7 and the remainders.
The problem can be solved by considering two cases: when there are 28 days in
February and there are 29 days in February. One method for approaching this problem
is to assume January 1 is day 1, January 2 is day 2, and so on. Continuing with this
method, the days of the year that represent a 13th day of a month are 13, 44, 72, 103,
133, 164, 194, 225, 256, 286, 317, and 347. For a leap year, the numbers become 13,
44, 73, 104, 134, 165, 195, 226, 257, 287, 318, and 348. When dividing these numbers
by 7, seven remainders from 0 to 6 are possible. Therefore, the conclusion is that a
Friday the 13th does occur every year.

Problem Two

Problem Two is: In a tournament, there are 15 teams. Each team plays with
every other team exactly once. A team gets 3 points for a win, 2 points for a draw, and
 40

1 point for a loss. When the tournament finishes, every team receives a different total
score. The team with the lowest total score is 21 points. Explain why the highest total
score team has at least one draw.
The mathematical concept underlying this problem is a basic in counting. The
problem indicates that each team plays with every other team exactly once. So, the
total number of games played is 14+13+12+…+3+2+1 = 105 times. Each game
generates a total of 4 points, which are shared between the two teams in that game.
Thus, the total number of scores is 4×105 = 420 points. If every team receives a
different total score and the team with the lowest total score gets 21 points, the team
with second lowest total score gets at least 22 points. The team with third lowest total
score gets at least 23 points, and so on. So, the highest total score team gets at least 35
points because 21+22+23+...+33+34+35 = 420 points.
Since this result is exactly the total number of points, it must be that the team
with the second lowest total score gets exactly 22 points. The team with the third
lowest total score gets exactly 23 points, and so on. Therefore, the highest total score
team gets exactly 35 points. If any team scores more points, the sum of the total score
of all teams would be more than 420 points, which is impossible. The maximum
numbers of points that a team could have received is 3 points × 14 games = 42 points.
Thus, the highest total score team lost only 42 – 35 = 7 points. For each loss, 2 points
are subtracted (a win is 3, but a loss is 1). If the highest total score team does not have
any draws, then the total number of points lost is an even number, not 7. Therefore,
the highest total score team has at least one draw.

Problem Three

Problem Three is: Let ABC be an isosceles triangle with AB = BC. Angle ABC
equals to 20 degrees. Point D is on AB such that angle ACD equals to 60 degrees.
Point E is on BC such that angle EAC equal to 50 degrees. Find the value of angle
CDE.
Problem Three is a geometry problem that can be solved in several ways. The
pictorial representation in Figure 3.1 indicates one possible solution by extending a
 41

line from point A to point F. To identify this line, considering point C as the center of
a circle with CD as its radius; the circle intersects the line generated from AC at F and
BC at G. For the isosceles triangle ABC, angle ABC = 20, angle EAC = 50, and angle
ACD = 60. Thus, angle EAD = 30 and angle DCE = 20. Draw FA, FD, and DG. Thus,
angle FAD = 100, angle ADC = 40, and angle AEC = 50. Since CF = CD and angle
DCF = 60, then DCF is an equilateral triangle. In ∆ FDA, angle AFD = 60 and angle
FDA = 20. In ∆ CDB, angle BCD = angle DBC = 20. Thus, CD = BD (equal sides of
isosceles triangle). And CD = DF (sides of equilateral triangle), so BD = DF. In ∆
AEC, CA = CE (equal sides of isosceles triangle). CF = CG (radii), so FA = GE
(subtraction). ∆ DCG is isosceles and angle GDC = DGC = 80. Therefore, angle BGD
= 100 and ∆ DBG ≅ ∆ FDA (side-angle-side). Then, FA = DG because they are
corresponding sides. Therefore, DG = GE and angle EDG = angle GED = 50. So,
angle CDE = 80 – 50 = 30.

Figure 3.1. Pictorial Representation for Problem Three

20

G
D

50 60
F A C
 42

Data Collection

Data sources for this study consisted of the following: (a) videotapes of the
think aloud sessions, (b) videotapes of the interview sessions, (c) students’ written
solutions to the problems, and (d) the researcher’s field notes of observations. Data
were collected when participants were in the TMO training camp at IPST from
October 13 to November 4, 2005. Since participants were native Thai speakers, all
processes were proceeded in the Thai language. Data collection was in a one-to-one
setting between the participant and the researcher. The researcher videotaped all the
processes to record the participants’ behaviors, how they responded to the problems
and what mathematical language they used. Each participant made three appointments
for solving the problems with the think aloud method followed by individual
interviews. Since participants were in the training session from 9.00 A.M. to 16.00
P.M., the appointments were conducted before or after the training time. The
appointments were weekly during the training; the first appointment was on the first
week of training, and so on. Scheduling was arranged so that the participants were not
under time constraints. Table 3.2 summarizes the detail of data collection for this
study.

Table 3.2. Weekly Data Collection

Appointment Data collection for each participant Duration

1: During week 1 of training 1. Practice the think aloud method 15 minutes
2. Do Problem One with think aloud 20 minutes
3. Interview 15 minutes
2: During week 2 of training 1. Do Problem Two with think aloud 20 minutes
2. Interview 15 minutes

3: During week 3 of training 1. Do Problem Three with think aloud 20 minutes

2. Interview 15 minutes
 43

For the first appointment, the researcher described an overview of procedures

for the participant to follow including the think aloud protocol. The student read
through the instructions and asked any question before started to work on the sample
problem. When the student was ready, the researcher started videotaping. During the
practice period, the researcher also noted the positive and negative student reactions
and helped them to improve their work with the method. Practice with the think aloud
method took around 15 minutes. The researcher replayed the videotape and discussed
ways to improve the participant’s skill with this technique. Next, the participants were
asked to solve Problem One using the think aloud method. After receiving Problem
One, the participants began by reading the problem aloud. They asked any questions
they had to make sure they understood the wording in the problems before beginning
to work on the problem. However, not all students asked questions at this point.
Participants spoke aloud describing their thinking while also writing their solutions on
the paper. They used as much time as they needed in solving each problem. This
unlimited time decreased pressure and motivated students to make a more complete
solution. On average, the participants took around 20 minutes per problem, followed
by a 15-minute interview. Before beginning the interview, the participant and the
researcher had copies of the participant’s solution paper. This same procedure was
used for Problem Two and for Problem Three. At the end of each appointment, the
researcher confirmed that the participant was not to disclose the nature of the problems
with the other participants.

Data Analysis

In multiple-case study research, Merriam (1998) suggested two stages of

analysis: the within-case analysis and the cross-case analysis. For the within-case
analysis, each case was treated as a comprehensive case. The researcher gathered as
much data as possible that had a bearing on the case. For this study, this within-case
technique was used to describe the processes that each participant presented while
solving each problem. When each case analysis was completed, cross-case analysis
 44

began. Since in the qualitative multi-case study the researcher seeks to build
abstraction across cases, the researcher built a general explanation for each case, even
though the cases varied in detail (Yin, 1994). The level of cross-case analysis resulted
in a more unified description across the cases; this process led to categories, themes,
or typologies that were conceptualized data from all the cases to build a substantive
theory that integrated multiple cases (Merriam, 1998). This cross-case technique was
used to create general explanations of problem solving processes that fit each problem
for all participants.
All data from the think aloud session, the interview session, participants’
solution papers, and researchers’ field notes were transcribed for analysis by the
researcher. Then they were categorized using a constant comparative method that
consists of
• comparing incidents and generating categories,
• integrating categories,
• delimiting the theory, and
• writing the theory (Glasser & Strauss, 1977).

To generate the categories, the researcher read through all transcribed data
sentence-by-sentence and identified words or phrases that described the participants’
responses. Then, the codes were applied based on a review of the data and the
concepts emerging from the data. The responses of one student were compared with
those of other students in the same problem, as well as the same student across other
problems. Multiple data sources were used to triangulate and confirm patterns that
were emerging. Each response was compared with other responses with the same idea,
regarding the source of the responses. For example, a segment from a think aloud
session of one student was compared with a segment of an interview of another
student if the two responses involved the same concept. The codes were grouped into
categories. At this point, preliminary categories were developed. Responses were
compared across categories in terms of similarities and differences. Next, the
researcher revised categories with transcribed data again and again until the final
categories were confirmed. The final categories were also reviewed against the
 45

transcribed data for the last time. Through this form of analysis, the researcher finally
developed a model that described the paths of problem solving processes and
responses that occurred in each category.
 46

CHAPTER IV

RESULTS

This chapter reports the results of the study that was conducted to answer two
research questions:
1. What is the nature of the problem solving processes that Thai gifted
students use as they engage in solving non-routine mathematical problems?
2. What metacognitive behaviors do Thai gifted students exhibit when
engaged in mathematical problem solving?
The chapter is organized into four sections. The first section illustrates the
backgrounds of the participants in mathematical problem solving. The second section
summarizes the participants’ reflections on the different problems. Since the study
proposes to investigate the nature of gifted students’ problem solving processes and
their metacognitive behaviors, an analysis of the results in the third section describes
the participants’ solution paths to the problems that specifically describe their
behaviors in solving the mathematical problems. This section ends with a Thai
students’ problem solving model developed from the research on their problem
solving. Finally, a brief summary of results is also presented.

Participant Backgrounds

Five gifted students participated in this study, 4 males and 1 female. Their ages
at the time of participation ranged from 13 years 3 months to 16 years 6 months. They
were studying in Grades 8, 9, and 10 of four different schools. Three schools (Schools
A, B, and, C) were in Bangkok, the capital of Thailand, and another school (School D)
was outside Bangkok. All participants were assigned pseudonyms as Pradya, Sira,
Wude, Nipa, and Sakda for reporting results. Their backgrounds are shown in Table
4.1.
 47

Table 4.1. Participant Backgrounds

Number Pseudonyms Age Grade School

1 Pradya 13 years 3 months 8 A
2 Sira 14 years 9 B
3 Wude 14 years 5 months 9 C
4 Nipa 15 years 1 month 9 A
5 Sakda 16 years 6 months 10 D

The following description provides a more detailed look at their mathematical

problem solving backgrounds.

Pradya

Pradya expressed why he liked to solve mathematical problems as he

mentioned “I have fun and feel challenged when I solve mathematical problems.” He
also liked to solve difficult problems. When he reached a difficult problem, which he
had never seen before, he tried to find different approaches to solve it. He stated that,
“When I cannot solve it, I leave it until I have thought about it for many hours. I will
solve it again when I have time or after I have seen similar problems that may be
easier. Sometime my friends teach me. Eventually, I can solve it.” In general, Pradya
typically left many unsolved mathematical problems for his spare time. He liked
mathematics very much because there are many mathematical problems in everyday
life, which are not always the same, and he believed people could always find some
mathematical problems to think about every day.
In the area of mathematics, Pradya preferred geometry because in this area he
was able to draw pictures. He talked about geometry content saying “no imagination
used because they already have pictures. But, I may not solve as many geometry
problems as other areas.” He did not like mathematical content like patterns or
notations, such as inequalities in algebra. In Pradya’s view, he believed he had both
good and bad characteristics that contributed to his problem solving skill. He believed
 48

that a good mathematician should be patient and able to come up with multiple
approaches. He said, “When I solve a problem, I have a lot of patience. I try to solve it
until I get the answer. I try to think in many ways. This helps me think of the problems
easier. Every time I solve a problem, I try to think of various ways to get many
solutions.” Characteristically, he was easily frustrated by some problems; as he said,
“When I cannot solve a problem, I would feel quite frustrated. Then I would say
something discouraging out loud which ends up interfering my friends’ progress.”
Pradya also believed that “To be a good problem solver, one should practice many
problems and come up with many ways to solve them.”

Sira

Sira liked to solve mathematical problems because it challenged him and

demonstrated his mathematical ability. The problems that he liked to solve were in
geometry, in particular, the problems that required him to exhibit mathematics proofs.
But, he did not like two-column proofs. He preferred to write proofs in paragraph
form. In general, he was not sure if he was a good problem solver or not “Sometimes I
can solve problems and sometimes I cannot.” When he was unable to solve difficult
problems, he continued thinking about the problem for sometime. If he still was
unable to solve the problem, he left it and continued doing other things. When he had
time, he returned to work on the problems. If he still was unable to solve them, he
asked his friends or teachers. Sira demonstrated much confidence about problem
solving. “I feel intelligent when I can solve difficult problems.” He thought having
enough knowledge, frequently solving difficult problems and reading a lot of
textbooks helped someone become a good problem solver. From his experience, he
indicated, “I believe that good attitude in problem solving is the important criteria that
motivates me to solve mathematical problems.”

Wude

Wude liked to solve mathematical problems. He said, “It challenges me as well

as stimulates my brain.” He also said that sometimes he was unable to think of a
 49

reason he loved solving problems. He loved to solve mathematical problems in all

subjects, even though he was not good in all. He thought he was able to solve
inequality problems well because he had more skills in this area. Overall, as a problem
solver, he thought he was a high to mid-level problem solver. “I need a lot more
practice than I have been having.” What he wanted to acquire in the future also
included learning new theories and techniques that he had not seen previously. He
stated that he improved his skill and knowledge by reading textbooks or taking
courses. Wude talked about his experiences and support from his parents and teachers.
He said “My mother finds mathematical problems for me to practice. My teachers also
help me a lot.” Moreover, he liked to read mathematics textbooks on his own. All of
these techniques helped him in becoming a good problem solver.

Nipa

Nipa liked to solve mathematical problems because she enjoyed the process
and she wanted to know the answers. She did not prefer to solve all kinds of
mathematical problems. As she said “I like to solve geometry problems because there
are pictures.” She expressed that when she was unable to solve difficult problems, she
left them and continued with other things. Then, she returned to work on them. She
believed she was a mid-level problem solver. “Because there are some problems I
cannot solve after reading the questions.” Nipa suggested that practicing problem
solving every day, as she did, would help someone in becoming a good problem
solver. “I try to solve and think about a problem everyday.” She also added that many
problems she gathered were from many textbooks.

Sakda

Sakda liked to solve mathematical problems because he enjoyed working on

them and felt the need to know the answers when he solved them. He liked to solve
combinatorics problems because “This area relates to basic of counting and it focuses
on integers.” He did not like to solve geometry problems. He stated that he believed he
did not have an “imagination” for doing them. At first, he had difficulty explaining
 50

what he meant by “imagination.” In order to explain his definition of “imagination”,

he tried providing an example. “It is hard to explain,” he said. “For example, in this
problem (referring to Problem One in the problem solving think aloud), I understood
after reading the question that I must use number theory and find the remainders.”
This problem was different from geometry problem. Eventually, he clarified that his
definition of “imagination” was seeing a way to solve the problem. It was difficult for
him to transfer geometric descriptions, visualize, and transform them into images in
his head.
When Sakda was unable to solve a problem, he, like the others, left it and
returned to it at a later time. Rating himself, Sakda believed he may be a good problem
solver. He was convinced that he had good techniques. “When I solve a problem, I am
more systematic than average students. But, when I compare myself with other
intelligent and knowledgeable students, I am really not good enough.” Nevertheless,
he felt his drawback was a lack of “imagination” which he may have been referring to
mentally transferring geometric descriptions, visualizing diagrams, and transforming
images. “Some problems require ‘imagination’, such as geometry problems. For me, I
do not have this particular ‘imagination’ skill.” He thought of how someone could
become a good problem solver. “When I was young, I thought I needed to know many
theorems and content.” His opinion changed with age and education. “Now, I believe I
need to practice and solve more problems. Knowing only content will not help solving
problems.” He thought having “imagination” was more important in problem solving
than knowing theorems ahead of time.

Reflections on the Problems

The three problems the students worked with in the think aloud sessions are
displayed in Figure 4.1. In reflecting on the three problems, the participants
considered which ones were the most difficult for them, and which were their favorite
ones.
 51

Figure 4.1. Three Problems of the Study

• Problem One: Does a Friday the 13th occur every year? Explain your reasons.
• Problem Two: In a tournament, there are 15 teams. Each team plays with every
other team exactly once. A team gets 3 points for a win, 2 points for a draw, and
1 point for a loss. When the tournament finishes, every team received a different
total score. The team with the lowest total score sis 21 points. Explain why the
highest total score team has at least one draw.
• Problem Three: Let ABC be an isosceles triangle with AB = BC. Angle ABC
equals 20 degrees. Point D is on AB such that angle ACD equals 60 degrees.
Point E is on BC such that angle EAC is equal to 50 degrees. Find the value of
angle CDE.

The participants had never seen the three problems in Figure 4.1 prior the
study. Sira, Wude, and Nipa first believed that they might have seen a problem similar
to Problem One, but later they realized the ones they had previously seen were not
exactly same. The problems they had previously seen had no effect on them while they
worked on Problem One, because they had not actually worked the specific Problem
One previously. They had only seen similar to it. In observing the participants work
with the three problems, Sakda liked Problem One the most. Pradya and Nipa, who
attended the same school, liked Problem Two. Sira liked Problem Three. And, Wude
did not prefer any problem: “I do not like one more than the others.”
For the level of difficulty, the participants rated the three problems from the
most difficult to the easiest as summarized in Table 4.2. Sira thought Problem One
was the most difficult, Problem Three was the next, and Problem Two was the least
difficult. Pradya ranked Problem Two as the most difficult one for him because he
could not come up with a way to solve the problem after reading the problem. He also
said “I was confused by the problem where it asked for the highest total score team
who had at least one draw.” Pradya thought the level of difficulty of Problem Three
 52

was in between Problem One and Two. He commented that Problem One was easy
because he recognized a way to solve it after reading the problem.
Wude and Nipa rated Problem Three as the most difficult, the next was
Problem Two, and Problem One was easy. Even though Wude thought he was not
good in combinatorics, Problem Two was not as difficult for him as he said, “When I
solved it, I got lost. But only a little lost. It seemed to contradict with what I first had
in mind. However, I later went back to read the problem again. After reading the
problem again, I found the correct way to solve it which helped me to solve the
problem faster.” Sakda also ranked Problem Three as the most difficult for him. He
said, “I could not figure it out. I had tried many ways.” The second most difficult
problem was Problem One, and Problem Two was the least difficult.

Table 4.2. Problem Difficulty Identified by the Participants

Participants Most difficult Middle Easiest

Pradya Problem Two Problem Three Problem One
Sira Problem One Problem Three Problem Two
Wude Problem Three Problem Two Problem One
Nipa Problem Three Problem Two Problem One
Sakda Problem Three Problem One Problem Two

Pradya commented that Problem One was a very short problem that challenged
him to solve it; but, he took a long time to solve it. “I think most short problems are
more difficult than long problems. Short problems are easy to read and understand, but
it is difficult to work on.” Sira reflected that Problem One and Problem Two are
tricky. “If we can understand the trick, we can solve it.” He thought Problem One was
a very short problem but it could take too much time to think about. Nipa thought that
Problem One was a good problem and she stated her reasons. “I did not like a problem
like this before. However, I believed I could solve it this time. It was not as difficult as
I thought before.”
 53

Nipa described her thoughts on Problem Three. “This problem could be solved
in several ways. It depends on which way we choose.” Sakda made a comment on
Problem Three that “It was a problem which we needed to find a relationship between
the two angles - 50 and 60 degrees. But, I could not see this the first time. So, I spent
too much time solving it rather than finding the relationship and understand the
problem itself.” He did not know how to solve Problem Three after reading the
problem the first time. He said, “I could not see anything at that time. I had to draw a
picture first.” He also said “I thought it was difficult for me.” Sakda also said Problem
Three seemed easy to him at first because it related to isosceles triangles. The figure
was not too complex. But, he changed his mind later, “It was too difficult after I
solved it.” After solving Problem Three, Sira and Nipa commented that they found a
way to solve geometry problems by extending lines and that this strategy may be
useful for solving future problems.

Analysis of the Think Aloud and Interview Results

Participants’ Solution Paths to the Problems

The first research question was concerned with identifying the nature of the
problem solving processes that Thai gifted students use as they engage in solving non-
routine mathematical problems. This section presents the evidence in support of this
question. The results are discussed in terms of how participants approached each
problem, and what in their thinking led to their final solutions. The following details
focus on only key ideas at the moment the participants determined a solution path to
each problem. Examples of selected participants’ problem solving processes to each
problem are described in the Appendix A.

Problem One. All five students worked on this problem with the concept of
remainders, but they used different approaches. Their thinking began with the
recognition that there were 365 days in a year and that when dividing 365 by 7, the
remainder was 1. Then they found the remainders after dividing the number of days in
 54

each month by 7. These remainders were the key idea for determining which day of
each month would be the 13th day. Wude and Sakda approached this problem by
considering that January 1st, 2nd, 3rd, 4th, 5th, 6th, and 7th were Fridays for the first seven
years. They then showed all 14 cases (7 for the years that had 28 days in February, and
other 7 cases for 29 days). While Wude illustrated by counting week-by-week for each
month, Sakda showed that there were 7 remainders from 0 to 6 in all cases. Pradya
used a variable x to represent the day of the 13th January from Sunday, Monday,
Tuesday, …, to Saturday. He then considered the day of the 13th for the other months
in two cases (28 days and 29 days in February). He found the distribution of all the
13ths in these two cases were the same, that is, x, x+1, x+2, x+3, x+4, x+5, and x+6.
Nipa supposed the 13th of January was Friday and found that the 13th of the other
months fell on the other six days. Neither Sira nor Nipa considered the possibility of a
leap year and thus their solutions were actually incomplete.

Problem Two. All five participants solved this problem correctly. Their
solutions were similar, except for the last part in proving why the highest total score
team had at least one draw. At first, they interpreted the given information of this
problem (Each team plays with every other team exactly once.) to find the number of
games that each team played; that was equal to 14 games. They found the total games
played of this tournament by using the binomial coefficient formula
C (n, r) = n! / [r! (n – r)!]. Thus, C (15, 2) = 105 games, and each game played produced
4 scores. They then calculated the total scores 4×105 = 420 points. From this point,
they returned to what the problem said “Every team receives a different total score.
The team with the lowest total score is 21 points.” This given information made them
start counting the numbers one-by-one from 21 until they obtained 15 teams, so the
last number was 35. Therefore, the highest total score team made at least 35 points.
They also calculated the sum of a sequence, 21, 22, 23, …, 35 to be 420. This sum was
equal to the total score that was previously calculated. It confirmed that the highest
total score team made exactly 35 points.
 55

The participants proved that this team had at least one draw in different ways.
Three participants (Sira, Wude, and Nipa) used variables to represent the number of
times that this team won, lost and made a draw. They devised equations, solved them,
and found the contradiction. In contrast, the other two participants (Pradya and Sakda)
did not use algebra as their reasoning in their proofs. Pradya said, “Each time when the
highest total score team plays, its score would be an odd number (Pradya meant when
this team got a win or a loss). The sum of odd numbers was an even number. But, 35
is an odd number. At that time, I knew that the highest total score team has at least one
draw.” Sakda considered that the score jumped from 42 to 35 and if the highest total
score team did not have any draw (i.e. they gained a loss), its score would decrease 2
by 2 from 42. Thus, it was not possible that its score would be 35 points.

Problem Three. Although all five participants were able to find the correct
solution for this problem, some of them had difficulty as they tried several methods
before getting to their final solutions. Four students solved the problem using
Euclidean geometry, but Wude solved it with trigonometry. Pradya, Sira, and Nipa
drew a line from point C to AB that divided 60 degrees angle ACD into 20 and 40
degrees, respectively. From this point, they continued finding many isosceles triangles
and finally showed the result. Sakda used CD as a radius for drawing a circle cutting
BC at the point G and cutting an extended line of AC at the point F. He then found
many isosceles triangles that led to the answer. On the other hand, Wude used the Law
of Sines to identify trigonometric equations. He did not solve these equations for his
final answer, but used trial and error to substitute possible answers in the equations.
Wude gave reasons as to why he used trigonometry to solve this problem. “I thought I
could get the answer by extending a line as well. But, using trigonometry gave me
more confidence than drawing the line.” He also feared that he could not find the
answer using the Euclidean method. So, his belief in solving this type of problems led
him to use trigonometry. However, he said his next strategy would be to try to draw
any line if he was not able to find the solution for the equations.
 56

Description of the Thai Students’ Problem Solving Processes

The second research question sought the metacognitive behaviors that students
exhibited during their problem solving actions. To answer this question, the researcher
considered two models (Garofalo and Lester’s model (1985) and Montague and
Applegate’s model (1993)) that described in Chapter 2 as frameworks when coding
the students’ responses. While Garofalo and Lester’s model included the terms
orientation, organization, execution, and verification, Montague and Applegate’s
Model consisted of seven cognitive processes (reading, paraphrasing, visualizing,
hypothesizing, estimating, computing, and checking along with three metacognitive
processes: self-instruction, self-questioning, and self-monitoring). These two models
were used as references for analyzing the metacognitive aspects since both reference
models focused on metacognitive behaviors and have been used to describe gifted
students’ problem solving processes in the literature. For example, Stillman and
Galbraith (1998) used Garofalo and Lester’s model as a framework in their study
because this model set out to specify key points where metacognitive decisions were
likely to influence cognitive actions and this model also included affective aspects.
The distinction between cognitive and metacognitive processes has been discussed as
being problematic because cognition is inherent in any metacognitive activity and
metacognition may be present in many cognitive activities. It is difficult to classify
behaviors as exactly cognitive or metacognitive processes. Therefore, some
researchers have used the term cognitive-metacognitive because of this problem
(Garofalo & Lester, 1985; Artzt & Armur-Thomas, 1992; Pugalee, 2001).
For this study, the behaviors correlated to the problem solving stages were
assigned as metacognitive processes. Besides considering these reference models, the
researcher also used the nature of participants’ responses when identifying the
students’ problem solving processes. Finally, after careful comparisons of the student
behaviors with all the problems, similarities were identified and categorized in four
stages: understanding, planning, executing, and verifying. The following discussions
present the metacognitive behavior evidence for each of the stages followed by
 57

examples of the self-evaluative statements in this Thai model of the problem solving
processes of gifted students.

Stage: Understanding. During this stage, the Thai students identified the
specific problem. They started by reading the problem aloud. They looked for the
question or stated some words in the questions to make sure of what was asked or they
restated the questions in their own words. Typically, they first stated the given
information and then restated the information in their own words. The students
analyzed the problem by representing the given information with pictures or tables.
Then, they reread the problem to ensure they made the correct representations. They
used their prior knowledge in mathematics for interpreting the given information and
referred to relevant mathematical concepts that might be used in the problem. They
also connected the problem with their prior experience or the current situation. The
students organized the given information into a systematic format before attempting to
develop the solution plan. They reflected on the problem in terms of familiarity and
difficulty of the problem. The following descriptions highlight these Thai students’
behaviors in solving the specific problems in this stage.
The students demonstrated specific behaviors in trying to understand the
problem after reading it aloud. Their actions included reading the problem, stating
and/or restating the given information, stating the question using some words in the
question, and restating the question. Since Problem One was in the form of a question
and was very short, they did not state the given information. Some of them stated
some words in the question.
Nipa: Friday the 13th.

Wude: Friday the 13th in each year.

Sira: Hmm, Friday the 13th.

They restated the question in their own words or stated the question as in the
following statements:
Pradya: Hmm… a Friday the 13th is in the calendar every year or not.
 58

Sakda: Is there a Friday the 13th?

Nipa: The question asked, “Does a Friday the 13th occur every year?”

When the participants were faced with more given information in the longer
problems, they tried to state the given information from their understanding. In
Problem Two, students stated the given information part-by-part at different times.
Their actions showed that the students did not consider all of the given information as
a whole, at a single time.
Pradya: The problem said every team receives a different total score. The
lowest total score team gets 21 points.

Sira: The team with the lowest total score gets 21 points.

Students also thought about what the problem was asking and why it was asked.
Wude: What does the problem ask? It asks about the highest total score.

Wude: The problem needs to know the reason why there is at least one
draw.

Sakda: Why does the lowest total score team gets 21 points? Why? And,
why the highest score team has at least one draw?

For a geometry problem like Problem Three, the students stated all the given
information while drawing a picture in order to make sure that they had everything
that the problem provided before continuing with their solution. At the same time, the
figures students drew indicated that they used their representations in their process. A
picture representation was important not only for understanding a geometry problem
but also helped them to devise a plan for the solution. Some students drew more than
one picture; they drew a new picture again and again when they could not figure out a
way to solve the problem. They also suggested they needed a big picture for better
visualization.
Wude: Okay, I’m going to draw a picture.

Sira: Let’s draw a new picture. It may be better.

 59

Nipa: Let’s draw a new picture again. This is the fourth picture. Make it
very big.

Pradya: Draw a very big picture first. I do not know I can solve it or not. I
try and guess first. I draw a picture as possible as I can.

Sakda: Let’s draw a new picture again. This is the fourth picture.

The evidence showed other representations that students used in their

processes to help them understand the problem, such as making a table or writing a
calendar.
Nipa: I am thinking about the calendar. There are 7 days in a calendar. I
am writing a calendar now.

Pradya: Make a year table, first.

After students stated the given information, they tried to clarify the given
information as much as they could; in fact their explanations were based on their prior
knowledge in mathematics. As in Problem Two, all of them carefully considered the
sentence “Each team plays with every other team exactly once.” Their interpretations
indicated that they had a concept in basic of counting.
Pradya: Okay, each team plays 14 times.

Sira: Amm… 15 teams play with every other team one time. It means
the first team meets the other teams 14 times.

Nipa: Each team plays 14 times.

Wude: Let’s see team 1. Team1 must meet the other teams 14 times.

Sakda: That is, a team meets other teams 14 times. A team plays 14
times.

In Problem Three, the students stated all the given information part-by-part
while drawing a picture. At the same time, they integrated their prior knowledge about
an isosceles triangle with the given information and calculations to find other angles
and length of the sides while drawing.
 60

Sira: The base angles are 80 and 80 degrees. So, the angle BCD is 20
degrees. The angle DAE is 30 degrees. The angle DBC is 20
degrees, and the angle DCB is 20 degrees too. So, CDB is an
isosceles triangle that has DB = DC.

Nipa: The angle DAE is 30 degrees. The angle DCE is 20 degrees. The
angle AEC is 50 degrees.

Pradya: The problem said angle B is 20 degrees. Thus, two base angles are
80 degrees. Then, angle DCB is 20 degrees. … Since angle AEC
is 50 degrees, angle DAE is 30 degrees.

Sakda: So, angle B is 20 degrees and base angles are equal to 80

degrees…. Ohh…I got something. Angle DCB is 20 degrees.
Since angle B and angle DCB are 20 degrees, DCB is an isosceles
triangle. …So, angle BAE is 30 degrees.

Wude: So, base angles are equal to 80 degrees…. Angle DCB is 20

degrees. … Angle BAE is 30 degrees.

The students provided evidence that they linked their prior experience in
problem solving to areas of mathematics that other problems they had seen were
related.
Pradya: It seems to be a number theory problem.

Sakda: This is a math problem. It should relate to number theory.

Nipa: Is this problem a combinatorics problem or a number theory

problem?

Additionally, Sakda also talked about what he had done before about a Friday
the 13th and the problem that he had previously seen.
Sakda: Actually, I always look for a Friday the 13th in new calendars
every year. There exists Friday the 13th in every calendar year.
For example, it will be in January next year, Friday January
the13th. I have found it to occur every year. This is a math
problem. Friday the 13th should relate to number theory. I think
there is another problem I have read. That problem asked about
the New Year day.
 61

When Pradya was interviewed, he responded that after he finished reading

Problem One, he thought about the problem part-by-part. “I thought that there is a 13th
and there is a Friday, sure.” This statement demonstrated that he was confident that
those two things can occur together. He then tried to think about and identify how a
13th and a Friday could happen in a particular month as he mentioned, “How do these
two things come together?”
During the understanding stage, the students reflected on the problems in terms
of their familiarity with the problem and their difficulty with it as they worked to
understand in Problem One and Problem Three.
Sakda: For this problem, I had never seen before. Friday the 13th.

Sira: How do we solve this? I’m confused. Friday the 13th.

Nipa: Friday the 13th. I do not know how to think.

Nipa: I cannot solve it. How many degrees for the angle CDE?

Sakda: I think this problem is not too difficult.

Although the students did not indicate the same evidence of difficulty with
Problem Two in the think aloud session, Pradya responded in the interview about
problem difficulty and what he thought after he finished reading the problem.
Pradya: For this problem, I did not know what to do next after reading the
question. I am confused why the problem asked about at least one
draw.

Pradya: I thought we did not have enough scores. I thought the highest
total score team cannot win all the games. Next, I thought it is
possible to have some losses. At that time, I did not know how to
think about it.

Stage: Planning. During this stage, the Thai students developed their plans by
selecting given information and generating new information. They also represented the
information with pictures, symbols, or tables before entering them in a solution plan.
When assessing a plan, the students applied relevant prior knowledge from number
 62

theory, basic of counting and geometry in solving the problems. They set the
conditions and stated a formula relying on their prior knowledge. The students solved
the problems using efficient strategies, such as drawing pictures, making tables, or
looking for patterns. They demonstrated their understanding in mathematical
knowledge and strategies that had helped them in solving problems. Students
rechecked their ideas to determine the plan in terms of making sense. They changed
their plans or looked for other plans as their original plans were unsuccessful.
In this stage, students searched for solution plans by selecting given
information and generating new information from the selected ones. When they
devised their plans for Problem Two, they focused on a particular sentence: “A team
gets 3 points for a win, 2 points for a draw, and 1 point for a loss.” This statement
helped them generate the same information; each game produced 4 points.
Sira: In a game, the score of both teams is exactly 4 points because a
team gets 3 points for a win and another team gets 1 point for a
loss, or each team gets 2 points for a draw.

Nipa: In any game the total score is exactly 4 points.

Pradya: If we select any two teams play together. The total score of both
teams is 3+1 or 1+3 or 2+2. So, in any game the total score is
exactly 4 points.

Sakda: When a game finishes, the score of two teams always increase 4
points. Because when team A plays with team B and they get a
draw, team A gets 2 points and team B also gets 2 points. If team
A plays with team B, one team wins and another team loss, their
scores will be 3 and 1 points. However, the total score of two
teams is always 4 points.

Wude: When two teams play together, they gets 3+1 or 2+2 which equals
to 4 points.

The next given information (When the tournament finishes, every team
receives a different total score. The team with the lowest total score is 21 points.) was
considered and provided the constraints for their plans. Manipulating between the
given and new information led students to identify the following plans to find the total
 63

games played and how many scores the highest team received before proving what
was asked in the question (i.e., why the maximum score team must have at least one
draw).
Students described their reasons behind the plans of Problem One as follows.
Wude’s plan started by setting January 1st as Friday and then he looked for a Friday
the 13th in the same year. He continued his search by moving January 1st to Saturday,
Sunday, and so on until he identified 7 cases for the year (years with 28 days in
February), and then similarly for the other 7 cases for years with 29 days in February.
Researcher: Why did you know that you should move the 1st of January from
Friday to Thursday?

Wude: Because there are 7 days in a week. After the day passes for 7
days, it will be on the same day as before.

At first, Nipa planned to solve this problem using a contradiction as she said,
“How do I solve it? I think a Friday the 13th occurs every year. Or I assume that a
Friday the 13th does not occur every year. Then, I find a contradiction.” While she did
not end of using this method, she considered various types of mathematics problems to
identify a way to solve this problem.
Nipa: Is this problem a combinatorics problem or a number theory
problem? Now, I am finding the way to solve it.

Nipa also talked about her thinking in the interview. After reading the problem,
she indicated that she did not know the way to solve. She thought about the calendar
for some time, trying to think about anything else that might be related to this
problem.
Researcher: When did you find the way to solve?

Nipa: First, I started setting the 13th of January as Friday. I got it when
I looked for the day of [for] February 13th. At that time, I used
my imagination to think that the 13th should be on every day
from Monday to Sunday.

Researcher: What caused you to think that the 13th fell on every day?
 64

Nipa: I read the problem again and it asked that a Friday the 13th occurs
every year or not. The word “every year” helps me to know that.

Pradya developed a plan for Problem One by thinking about the 13th and how
many 13th there were in a year. At this point, his realized a way to solve the problem,
as he described in the interview.
Pradya: It is more difficult if we consider what date for Friday because
there are many dates in a year. I mean from 1 to 30 or 31. It is
easier if we consider only the 13th, which is only 12 days in a
year. Then I thought [that] a Friday the 13th will fall on what date.

Researcher: What date or what day?

Pradya: Ohh… I said wrong. I want to say what date. I mean it will fall on
Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, or
Saturday.

Some students had difficulty developing plans for Problem Three. At the same
time, the evidence indicated that they searched for a variety ways to solve this
problem. Sakda tried to solve it with both Euclidean geometry and trigonometry. He
searched for several ways to solve the problem, such as extending a line, drawing a
circle, and using the Law of Sines. He also recalled what he had learned the previous
day when finding a solution plan that used a compass to draw a circle that passed
through points G and D and met BC at the point F.
Sakda: Do we need to draw lines from point A and point D meet
together? Okay, draw an extended line from point D to point G
and let its length equals to BD. What will happen if we have a
circumscribed circle and an equilateral triangle? What will
happen if we think about the equilateral triangle? Start to draw an
equilateral triangle GDC. Try to use trigonometry. Does it work?
Try to use the Law of Sines….Try to set equations. Let’s draw a
perpendicular line from point D to BC. Try to think about what I
just learned in geometry class. What did I learn yesterday? The
Law of Sines and the areas of triangles. This is a geometry
problem. Ohh… what will happen if I use a compass to draw a
circle passing through points G and D and meet BC at the point
F?
 65

Nipa also had problem difficulty as she considered many plans for Problem
Three. In her plans, she drew both a perpendicular line and a parallel line, but those
lines did not work. Finally, she found a way to solve the problem as she drew a line
CF and found an answer.
Nipa: Try to draw a perpendicular line from point D to BC. Let this line
meet BC at the point X.

Nipa: Try to draw another line. Draw a line from point C to Y which
parallel to DE. In the former picture, I drew a parallel line, but it
did not make any help. Anyway, try other ways again. Try to
draw a line from C to meet AB at point F such that angle ACF is
20 degrees.

Nipa: Draw CF that makes angle ACF is 20 degrees.

Students predicted the answer when they set up the plan for Problem One. For
instance, Nipa, Sira and Pradya guessed that Friday the 13th occurred each year and
Sakda knew from his prior experience that it occurred every year.
Nipa: I seem to me that there is a Friday the 13th every year.

Sira: It seems to me that it occurs every year. Why?

Pradya: At that time, I guess a Friday the 13th occur every year.

Sakda: I had ever found it occurs every year.

Although Wude did not predict the answer in the think aloud session as the
others did, he mentioned it in the interview. He believed that a first guess helps
someone in problem solving.
Wude: In general, what we expect or guess at the first time before
solving the problem helps us a lot. But if we make a wrong guess,
it will show by itself. Finally, we will know we make a wrong
guess or we cannot solve that problem.

When the students developed the plans, they referred to mathematical concepts
that they considered might be used in the plan. For Problem Two, Sakda thought about
the Pigeonhole Principle first, but he did not use it in the solution plan. Wude
mentioned modulo 7 for Problem One, and trigonometry for Problem Three. He then
 66

used both concepts in his solution plans. These techniques indicated that the students
used their knowledge of mathematics in their solutions.
Sakda: Or we need to use Pigeonhole Principle.

Wude: I consider the date with modulo 7.

Wude: I’m thinking about how to use trigonometry to solve this.

When assessing a plan, the students applied their relevant prior knowledge in
number theory, basic of counting, and geometry. For Problem One, they used the
concept of remainders to show that a Friday the 13th occurred every year. Wude
initially set January 1st as Friday and then looked for other Friday the 13ths in the same
year. He continued this process by moving January 1st to Saturday, Sunday, and so on
until he got 7 cases for the year that has 28 days in February, and the other 7 cases for
29 days.
Researcher: Why did you know that you should move the 1st of January
from Friday to Thursday?

Wude: Because there are 7 days in a week. After the day passes for 7
days, it will be on the same day before.”

Two students referred to “a counterexample” as a way to prove Problem One,

but finally they did not use this process in their solutions.
Sakda: We need to give a counterexample if a Friday the 13th does not
occur every year.

Wude: Do we need to find a counterexample to show that a Friday the

13th does not occur every year?

In Problem Two, the students had a goal to prove why the highest total score
team had at least one draw. Although they proved this result with a method of proof by
contradiction, their reasons varied. Pradya considered that the highest total score team
played 14 times and got 35 points. He reasoned that if this team did not get any draw,
the team could only get wins and losses. Thus, the score needed to be an odd number
because there were 3 points for win or 1 point for loss. When an even number was
multiplied by an odd number, he recognized the result was an even number. Since, 14
 67

is an even number, the total score needed to be an even number too. But, 35 was an
odd number.
Sira thought that if this team did not get any draw, the team must win 11 times
and lose 3 times resulting in the scores 33 + 3 = 36 points. This score meant the
highest score team with no draws must get the score of at least 36 points, but this team
only received 35 points. Nipa supposed that the team did not have any draws. For her,
the equation that represented the scores when this team played 14 times and won x
times would be 3x + (14 – x) = 14 + 2x = 2(7 + x) points. She noticed that 2(7 + x)
was an even number, but 35 was an odd number.
Wude considered the value of x and y to represent the number of wins and
draws. After solving equations, he got y = 21 – 2x. He also knew that 21 – 2x was an
odd number and x was more than zero. So, he concluded that y ≥ 1. Sakda considered
that if this team had only wins, its score would be 42 points. Each time, when the team
got a loss, its score decreased 2 points from 42 points. But, its score was 35 points.
This result was impossible if the team was to get 35 points as it decreased 2 by 2 from
42.
In Problem Three, Wude applied the Law of Sines in his plan. With this
application, he found the trigonometry equations and tried to find the value of angle
CDE. However, he did not solve these equations. Rather, he used trial and error as he
substituted the value for the angle in the equation until he found the answer.
During the planning stage, the students used efficient strategies, such as
making tables, using symbols, and looking for patterns to represent the information. In
Problem One, Pradya assumed the day for January 13th and used a variable x to
represent it. The variable x represented Sunday, Monday, Tuesday, Wednesday,
Thursday, Friday, or Saturday. He also put the remainders that he found in a table.
Sakda made a table that consisted of months, number of days in that month, and
remainders (see Figure 4.2).
 68

Figure 4.2. Sakda’s Representation of Problem One

Month 1 2 3 4 5 6 7 8 9 10 11 12
Days 31 28 31 30 31 30 31 31 30 31 30 31
Remainders 3 0 3 2 3 2 3 3 2 3 2 3

Three students (Sira, Wude, and Nipa) used symbols when they set equations
for Problem Two.
Sira: From this condition, it makes the highest score team get at least
35 points. So, we get the first equation 3a + 2b + c ≥ 35.
But a + b + c = 14 where a, b, c ≥ 0 and they are integers.

Wude: Set the equations 3x + 2y + z = 35 and x + y + z = 14.

Nipa: If team A did not have any draw, team A wins x times gets 3x
points, team A losses 14 – x times gets 14 – x points. So, team
A gets 3x + (14 – x).

The students also used variables to represent the value of the angle they were
trying to find. Then they started comparing this angle with the others.
Wude: Let angle CDE be θ degrees.

Nipa: Let angle YCD be x degrees.

Pradya: In order to find angle CDE, let assume it is an x degrees angle.

Sakda: The problem asks for angle CDE. Let it be x degrees.

Students looked for patterns when they solved Problem One. For example, Sira
tried to make the conclusion of a Friday the 13th from the sum of remainders he got.
Sira: Let the 13th of February this year be Friday. Next, find the sum
of remainders until we get 6. Let’s see the remainders of
February, March, April, May, June, and July. The sum of these
remainders is 0+3+2+3+2+3 = 13. Divide 13 by 7, we get the
remainder of 6. So, if the 13th of February this year is Friday,
then the 13th of August next year will be a Friday. Thus, a
 69

Friday the 13th occur again on the next year. Is it necessary to

get the remainder of 6 for this? Try again in the same way with
other months. Find the sum of remainders from:

March to July, we get 3+2+3+2+3 = 13.

April to August, we get 2+3+2+3+3 = 13.
May to September, we get 3+2+3+3+2 = 13.
June to October, we get 2+3+3+2+3 = 13.
July to November, we get 3+3+2+3+2 = 13.
August to December, we get 3+2+3+2+3 = 13.

(Sira meant that if the 13th of April this year is Friday, then the 13th of September next
year will be a Friday. If the 13th of May this year is Friday, then the 13th of October
next year will be a Friday and so on.)
Sira: How about September? To get the remainder of 6, we need to
subtract 1 from the sum of remainders because we count
forward one year. So, the sum of remainders from:

September to May is 2+3+2+3+3+0+3+2+3 = 21.

October to March, we get 3+2+3+3+0+3 = 14.
November to July, we get 2+3+3+3+0+3+2+3+2+3 = 21.
December to May, we get 3+3+0+3+2+3 = 14.

Subtract 1 from 14 and 21, then we get 13 and 20. Both 13 and
20 have remainders of 6 when they are divided by 7.

Wude looked for a pattern of a Friday the 13th after he set January 1st to a
Friday and found the other Fridays in January by counting day by day. He showed all
14 cases that January 1st fell on Friday, Saturday, … , Thursday for years with
Februarys that have 28 days and those with 29 days.
Wude: So, use this method for the other days. A Friday the 13th will
occur every year if I show all cases that have a Friday the 13th.

Pradya found a pattern of a Friday the 13th when he considered the remainders.
He found that January13th, January 20th, and January 27th are on the same day. He
represented the days from January 28th to February 2nd with x+1, x+2, …, x+6.
February 3rd was x+7 which was the same day as x. So, the day of February 10th was x
and February 13th was x+3. At this point he concluded that when the day of
January13th was x, the day of February 13th was x+3. He found to cases when he
 70

considered from February to March. The day of March 13th depended on the number
of days in February, 28 or 29 days. He found the days of the 13th in other months of
both cases by considering the remainders when dividing the number of days in that
month by 7. His example was that if there were 28 days from February 13th to March
13th the remainder was 3. So, the day of March 13th was x+3, that was the same day as
February 13th. For another case, the day of March 13th was x+4. He used the same
method until he got all 12 months. He found that in both cases the distributions were
the same, in other words that there were x, x+1, x+2, x+3, x+4, x+5, x+6.
There was evidence that the students verified that their plans made sense, that
they looked for efficient plans, and that they changed their plans during this stage.
Sakda: Only 14 cases. Is it good to write a calendar? Not good.

Sakda: Let’s try this way. It may easier. Try to think about the dates of
Friday in the first month.

Wude: Are there any other ways that easier than this?”

Wude: First, try to guess the value of θ. It may easier to solve. Amm…
not guess the value of θ. Try to solve the equation……

Wude: Amm…try to put the value of θ in this equation. It may be

better.

Pradya: Which lines do I need to extend? The extended line, which is

the most useful to get the answer? Line. Line. Which is the best
line? Okay, this line may be good.

Interview responses confirmed that all the students followed the plans that they
created. However, sometimes, they did not know a way to solve the problem after the
initial reading of the problem.
Researcher: Did your solution follow what you thought in your plan?

Wude: Yes, they are the same. What we plan depending on what kinds
of problems we get.

Sira: They are the same. But I’m not quite sure when I plan to solve it
at the first time.
 71

Nipa: Yes, I did even though I do not know what to do at the first
time.

Pradya: Yes, they are the same.

Sakda: Yes, they are the same.

Stage: Executing. During the stage labeled executing, the Thai students were
directed by their goal to find the final answer. They applied mathematical formulas
and carried out computations as called for in their solution plans. They made logical
mathematical statements to support their plans. They also stated the conclusions or the
final answer.
Students carried out computations with the application of mathematical
formulas to achieve their goals. As in Problem Two, three students stated “the
combination of 15 and 2” when they calculated how many games were played in this
tournament. This statement indicated that they knew the binomial coefficient formula
C (15, 2). Nipa did not mention to these words, but she calculated the result in the same
way. Sira calculated this result by finding the sum of a sequence.
Wude: The combination of 15 and 2 is … 105 times.

Pradya: The combination of 15 and 2 is equal to (15×14)/2 = 105 times.

Sakda: The combination of 15 and 2 is (15×14)/2 = 105 games

Nipa: So, the total number of games play is = (15×14)/2 = 105 times.

Sira: So, the total number of games play is = 14+13+12+…+3+2+1 =

(14×15)/2 = 105 times.

The students used the sum of arithmetic sequence formula for calculating the
sum of a sequence 21, 22, 23, …, 35.
Sira: Find the sum of 21+22+23+…+35. How many? 28 multiply by
15. I get 420.

Wude: Find the sum of 21+22+23+…+35. How many? Ohh… 420,

exactly.
 72

Nipa: Thus, the sum of each total scores for the team is
21+22+23+24+25+26+27+28+29+30+31+32+33+34+35 =
[(21+35)×15]/2 = 28 ×15 = 420

Pradya: From 21 to 35, add them together. The sum is 15×28 = 420.

Sakda: Add them together. How many scores? Ohh…I am lazy to think
about the sum. I do not like sequences. We get
20×15(1+2+3+…+15). How many? Then we get 300
(1+2+3+…+15). How many? I am so confused. We get
(15×16)/2 = 120. So, the sum is 420.

During the execution stage, some students solved equations to find an

unknown variable as Wude and Sira did in Problem Two.
Wude: From these two equations, we get 2x + y = 21.

Sira: Subtract the second equation from the first equation; we get
2a + b ≥ 21. Oh…wait a minute if we suppose the highest
score team did not have any draw, that is, b = 0.
Then we get 3a + c ≥ 35 and a + c = 14.
Since, 2a ≥ 21 then a ≥ 10.5.

Although Wude tried to solve trigonometric equations in Problem Three, he

could not find the value of θ by solving the equations.
Wude: Try to solve the equation. Okay, from the equation
sin 60 sin 50 sin(40 + θ) = sin2 80 sin (110 – θ). We get
(sin 60) 2 sin 50 sin(40 + θ) = sin 80 sin 80 sin(110 – θ).
Sin 60[cos (10 – θ) – cos (90 + θ)] = 2 sin2 80sin (110 – θ).
Sin 60[cos (10 – θ) – cos (90 + θ)] = sin 80[2 sin 80 sin (110 – θ)].
Sin 60 [cos (10 – θ) – cos (90 + θ)] = sin 80[cos (30 – θ) – cos (90 – θ)].
Sin 60[cos (10 – θ) – sin θ] = sin 80[cos (30 – θ) + cos (10 – θ)].

Students usually made logical mathematical statements that supported their

plans before stating the conclusions or the final answer.
Nipa: It occurs every year because when I suppose January 13th is
Friday, I found that the 13th of the other months fall on the rest
six days. Therefore, there is a Friday the 13th in every year.

Pradya: The distribution of two cases are the same; that is, there are x,
x+1, x+2, x+3, x+4, x+5, x+6; where x represents Sunday,
Monday, Tuesday, Wednesday, Thursday, Friday, or Saturday.
So, the Friday13th will fall on Sunday, Monday, Tuesday,
 73

Wednesday, Thursday, Friday, and Saturday in any year.

Therefore, a Friday the 13th occurs every year.

Sira: If this team did not have any draw, it must win 11 times and
loss 3 times and gets the scores 33 + 3 = 36 points. This means
the highest score team with no draw must get the score at least
36 points, but it get only 35 points, contradiction. Therefore, the
highest score team has at least one draw.

Particularly in Problem Three, students often made logically mathematical

statements when they found the value of any angle or the length of side.
Sira: In the triangle AXC, angle C is 20 degrees, and angle A is 80
degrees. So, angle X is also 80 degrees.

Sakda: Let’s see the triangle EYC. Since angle ECY is 20 degrees and
angle EYC is 110 degrees, angle YEC is 50 degrees.

Pradya: From what we did before, we know that EFC is an equilateral

triangle. So, CE = EF = FC. We also get CE = EF = FC = AC
too.

Stage: Verifying. During the final stage, verifying, the Thai students
sometimes checked what was done to make certain that the solution made sense. They
usually revised the solution plans when the plans did not work. The students
rechecked what was done and were able to explain reasons for their solutions. When
they verified the solutions, they reread the problem and examined all their written
responses in cyclical processes as they attempted to verify local plan as to its
usefulness for solving the problems.
Only one student, Nipa showed that she paid attention to verifying the solution
for Problem Two. She stopped speaking during this stage for 3-4 minutes. So, the
researcher asked her to speak aloud what she was thinking.
Researcher: What are you thinking?

Nipa: I am checking my solution. I examine what I wrote. It is correct

or not. I return to read the question and read my solution again.
I am done.
 74

Although the other students did not directly demonstrate checking their
solutions, obviously Nipa had. The students did examine what was done in the plan
before confirming their answers.
Wude: I have already checked all cases, 14 cases. A Friday the 13th
occurs in all cases. Therefore, it occurs every year.

Pradya: We got them all. We have all 12 months in a year.

Wude also checked his computations again after he got the result that used the
binomial coefficient formula to find the total numbers of games played. Then he
restated the result to confirm it was correct.
Wude: Is it correct? Correct. It is the total number of games, 105 times.

After Sakda carried out his calculations to find how many games were played,
he noticed that the result was too much. At this point, he checked to make certain that
the answer made sense in his mind. However, he did not specifically show a way to
verify the solution.
Sakda: It is so much. Did I do something wrong? There are 105 games
of 15 teams.

Self-Evaluation. In addition to each of the above stages, students frequently

exhibited self-evaluative statements that helped them continue working on the
problem until they had finished the tasks. The term “self-evaluation” was used as a
category for coding in the study of Montague and Applegate (1993) when the students
evaluated themselves as problem solvers or used “I” statements about their
performance. For this study, self-evaluative statements were divided into two types.
First, students demonstrated self-monitoring as they monitored their work on the
problems until they got complete solutions. For instance, the participants usually asked
themselves with the sentences such as: What’s next? or What am I going to do?.
Second, the students demonstrated affect statements as they evaluated themselves as
problem solvers in terms of how much confidence they had, their difficulties and
frustrations, and their efforts while solving the problems. The statements that students
 75

made were identified for these self-evaluation categories. Examples of self-evaluation

statements in each of these categories are provided in Table 4.3.
The following interview data confirmed that students had confidence in their
solutions and supported their responses. However, some students showed some
hesitation about their calculations or wording they used in explaining their solutions.
Researcher: How can you be sure that your solution is correct?

Sira: I’m sure because I showed that each sum of remainders can be
divisible by 7.

Nipa: I think it’s correct. I found that the 13th fall every day. I also
already checked all of each step and each sub step.

Wude: I’m sure because I showed all of 14 cases. If there is any wrong,
it may cause by wrong calculations.

Pradya: I am sure it is correct because the reasons are correct. If there is

anything wrong, it might be because of wrong calculations.

Sakda: I am sure it is correct because I have much confidence in the

reasons that I used. But, I am not sure about my wording.
 76

Table 4.3. Examples of Excerpts for Self-Evaluation

Self-Evaluation Examples of Excerpts

Self-monitoring What’s next?

What am I doing now?
Sira: How to show that?
Nipa: How do I solve it?
Sira: What am I going to do next?
Sira: Does this works for what?
Sakda: Where is the way to solve? How to find the answer?
Sakda: Keep trying. Keep going.
Affect Nipa: Ohh…I see. I got it.
Pradya: Therefore, a Friday the 13th occur every year, sure.
• Confidence Wude: Is it correct?
Wude: Ohh… I see.
Pradya: I do not know I can solve it or not.
Pradya: I don’ know to get this for what.
Pradya: From what I did, I cannot find anything wrong.
Sakda: I’m happy. But, I cannot see anything more. I think I
can figure it out.
Sakda: I’m exciting that we get an equilateral triangle.
Sakda: But, I can do it.
Affect Nipa: I don’t know what I’m going to do.
Nipa: I cannot solve it.
• Difficulty/
Sira: How do we solve this? I’m confused.
Frustration Sakda: Ohh…I cannot think about wording to explain.
Pradya: I do not know what to do?
Sakda: Did I do something wrong?
Sakda: I am so confused.
Nipa: Draw a new picture. This picture is too small. It makes
me so confuse. I cannot think when I see it.
Nipa: I almost die.
Sakda: I cannot figure it out. I’m so confused.
Sakda: I’m very stress.
Sakda: Why I cannot think about them? I cannot find
anything.
Sakda: I cannot think about it.
Affect Wude: We must think slowly. Think slowly.
Wude: Amm… it needs to think a lot.
• Effort Sakda: Ohh…I am lazy to think about the sum.
Pradya: I am very lazy now.
 77

Thai Students’ Problem Solving Model

The Thai students’ behaviors discussed through the four stages were
conceptualized into a four-stage model. This model presented the framework, as
illustrated in Figure 4.3, of the Thai students’ processes while solving non-routine
mathematical problems. The four stages were labeled as understanding, planning,
executing, and verifying. These stage names were modified from the study of Garofalo
and Lester (1985). The first stage, understanding, required the problem solvers to
identify and analyze the problem. They were looking for the given information and
what was being asked. The second stage, planning, engaged the problem solvers in the
search for a solution plan relying on their prior knowledge. They manipulated the
given information and generated information using a variety of strategies to assess the
plan. They also revised the plan to make sense until they arrived upon the final plan.
The third stage, executing, involved the problem solvers in finding a solution through
computation or proving the conjectures. The last stage, verifying, required the problem
solvers to check the results for reasonable. In each stage, the term self-evaluation from
Montague and Applegate (1993) was used to label when the problem solvers
monitored their thinking and efforts and demonstrated affective behaviors as they
worked in solving the problem. In other words, self-evaluation affected the
participants’ actions in each of the stages of the model. See the stage description of the
model in Figure 4.4.
 78

Figure 4.3. Thai Students’ Problem Solving Model

Thai Students’ Problem Solving Model

Reading a Problem

Understanding

Stage

Verifying Planning

Stage Stage

Executing

Stage

Linear flow
Self-evaluation flow
 79

Figure 4.4. Thai Students’ Problem Solving Stage Description

Thai Students’ Problem Solving Stage Description

Stage: Understanding
(1) Identify the problem
• Read/reread/restate the problem, the given information, and the
question
(2) Analyze the problem
• Represent the problem with pictures or tables
• Clarify/interpret/organize the given information
• Connect with prior experience
• Reflect on the problem
(3) Self-evaluation

Stage: Planning
(1) Devise a plan
• Manipulate the given and generating information
(2) Assess a plan
• Apply prior knowledge/ mathematical concepts/ theorems
• Use strategies (look for pattern/make a table)
• Predict possible answers/use estimation
(3) Revise a plan
• Determine the plan makes sense
• Change the plan if it is not working
(4) Self-evaluation

Stage: Executing
• Carry out computations
• Make logical mathematical statements
• State the conclusion/the answer
• Self-evaluation

Stage: Verifying
• Check results for reasonableness
• Reread the problem and solutions for checking
• Move to a new plan based on verifying results
• Self-evaluation
 80

Summary

The first aim of this study was to investigate the nature of the problem solving
processes that Thai gifted students used as they engaged in solving non-routine
mathematical problems. The supporting evidence demonstrated characteristics and
backgrounds of problem solving through five students by monitoring their reflections
to the problems as well as how they determined their approaches to each problem.
The second aim sought the metacognitive behaviors that students
exhibited while solving the problems. Therefore, students’ responses were presented
that illustrated behavioral evidence that actually occurred as students spoke and
thought about the problems in the think aloud sessions and in the interview sessions.
Examples excerpts were presented to explain the students’ processes and their
behaviors, resulting in the final model for the Thai students that consisted of four
stages: understanding, planning, executing, and verifying. The behaviors within each
stage were identified and described.
As students were engaged in solving the problems, their thinking processes did
not proceed in a strictly linear order from the understanding stage to the verifying
stage. Their efforts showed that while they did display each of the stages, they did not
proceed linearly through the stages from the first stage to the last one. Thus, an
important result for this study was that this model is not linear. Furthermore, the self-
evaluation aspect identified in each of the stages may have been the activity that
initiated them in vacillating among the stages as needed to work on the problem (as
represented in Figure 4.3). With the level of complexity of the problem solving
process, no exact model can describe the actual processes for all students.
Nevertheless, this four-stage model captured the actions, behaviors, and thinking
exhibited by these Thai mathematically-gifted students to the extent possible through
this research method.
 81

CHAPTER V
DISCUSSION

The purpose of this study was to examine the problem solving processes of
Thai gifted students when they solved non-routine mathematical problems. The
research questions for this study were:
1. What is the nature of the problem solving processes that Thai gifted
students use as they engage in solving non-routine mathematical problems?
2. What metacognitive behaviors do Thai gifted students exhibit when
engaged in mathematical problem solving?
This investigation was considered a multiple-case study. Five Thai gifted
students who were eligible for the Thai Mathematical Olympiad project and met the
selection criteria participated in this study. Each student practiced the think aloud
method before solving three mathematical problems individually. The problems were
non-routine problems that focused on number theory, combinatorics, and geometry,
respectively. The subjects worked on each problem separately and were interviewed at
the end of each problem solving session. Data sources included videotapes of the think
aloud and the interview sessions, students’ written solutions, and researcher’s field
notes. These data were analyzed using the within-case and cross-case techniques. Data
gathered were also categorized using a constant comparative method to conceptualize
a model of problem solving process. The following sections present the discussions of
findings and the limitations of the study. Finally, the implications and
recommendations for future research and concluding remarks are addressed.

Discussions of Findings

Overall, the gifted students in this study demonstrated good success in solving
the non-routine mathematical problems. Basically, the students solved Problem One
and Problem Two without hesitation although some did not completely solve Problem
One (forgetting about the changes resulting from leap years). Although, at first, two
 82

students had some difficulty searching for a solution for Problem Three, all the
students eventually succeeded. Their solution processes were based on logical analysis
and systematic strategies. They showed high ability in verbalizing and explaining their
thoughts and reasoning for their solutions. This ability demonstrated their
understanding of mathematical structure and strategies, similar to those described in
Heinze’s study (2003). The study findings were consistent with other studies in the
literature, where participants were from Western cultures. In particular, the results
were consistent within the context that gifted students accessed problem representation
strategies, such as drawing pictures, making tables, or looking for patterns in order to
facilitate their understanding (Gorodetsky & Klavir, 2003; Montague, 1991; Montague
& Applegate, 2000; Sriraman, 2003). The Thai gifted students and the students in the
Western culture research applied their prior knowledge to a problem or to an
unfamiliar situation; they made use of a variety of mathematical knowledge, recalling
and relying on theorems to generate additional relevant information (Gorodetsky &
Klavir, 2003; Lawson & Chinnappan, 1994; Overtoom-Corsmit, Dekker & Span,
1990). The Thai gifted students were observed to increase their conversations as they
were confronted with more difficult problems; this result was similar to the actions of
the American students in Sriraman’s study (2003).
Based on an analysis of the results, major emerging evidence related to the
students’ problem solving processes consists of five categories: advanced
mathematical knowledge, willingness to consider multiple alternative solution
methods, recollection and willingness to consider prior knowledge and experiences,
reliance on affect, and parental and teacher support. This evidence was considered as
factors that influenced the demonstration of the students’ problem solving processes
and their behaviors in terms of their successful actions in problem solving, and why
they exhibited these processes.
First, advanced mathematical concepts were integrated in students’ solutions
such as the Law of Sines, the binomial coefficient formula, and the sum of arithmetic
sequence formula. All these concepts were usually taught in grades 11-12 of the Thai
national curriculum. Yet, these students who were not yet in those grades had in some
 83

way gained this prior knowledge. This result indicated that these participants had
conceptual understandings in mathematical concepts in higher levels, a characteristic
of gifted students. It also suggested a high ability in problem solving for applying
these advanced concepts toward the completion of a correct solution.
Second, the students searched for alternative solution methods as they worked
on the problems. These students tried to understand and worked with the problem
using a variety of approaches when they had difficulties. As in Problem Three, Sakda
and Nipa demonstrated their willingness and abilities to consider different paths in
their solution rather than insisting on the unproductive paths. However, the ways they
searched for alternative paths depended upon their prior experiences and beliefs about
working with mathematical problems.
Third, the influence of the students’ prior knowledge was evident. As Klein
(2004) suggested, problem solving activity is affected by what takes place in prior
problem solving lessons, by the environment, and by what happened outside the
classroom. Students applied techniques and strategies that they had used previously.
Wude reported that he usually used a trigonometry approach in a geometry problem
because, from his experience in solving this type of problems, more than 60% of the
time this method led him to the solution. He also feared that he could not solve the
problem if he used Euclidean geometry. However, he planned to try drawing some
lines if he could not find the answer by using trigonometry.
Sakda tried to connect Problem Three with the topics he had learned. “Try to
think about what I just learned in geometry class. What did I learn yesterday?” He also
thought of a tool for finding solutions, “This is a geometry problem. Ohh… What will
happen if I use a compass to draw a circle pass through points G and D and meet BC at
the point F?”
For Problem One, Sakda had experienced finding a Friday the 13th in the
calendar in every year as he stated below.
Actually, I always look for a Friday the 13th in new calendars
every year. There exists Friday the 13th in every calendar year.
For example, it will be in January next year, Friday January
the13th. I have found it to occur every year. This is a math
 84

problem. Friday the 13th should relate to number theory. I think

there is another problem I have read. That problem asked about
the New Year day.

Fourth, affective behaviors (behaviors relying on emotion more than cognition,

thought and action) played an important role in the problem solving process of all five
gifted students. This result was consistent with the findings of Carlson and Bloom
(2005) and DeBellis (1998). As Goldin (2000) described, the ways that affect is or can
be utilized by mathematical problem solvers to guide their steps, and to their influence
cognition in a constructive manner increased their problem solving power. In this
study, affect was evident in terms of self-confidence, frustration, and effort. The
students relied on their confidence to monitor their frustration and anxiety, turning
these feelings into motivation that ultimately led them to success in finding a solution.
Their motivation maintained their interest, encouraging them to continuing working
efficiently on the problems. Overall, the students expressed positive emotions while
they attempted to solve the problems. One student indicated that he was stressed due
to the level of difficulty in one of the problems, even though he was not under a time
constraint.
Researcher: In your scrap paper, you tried many ways in order to find the
answer. How did you feel at that moment?

Sakda: I felt stress. However, this is not an examination, and I can use
as much time as I need. So, I just kept going.

Fifth, parental and teacher support were clearly notable as Wude responded to
the interview question. He believed that their strong support helped him in becoming a
good problem solver.
Researcher: What makes someone a good problem solver in your opinion?

Wude: Everything, I think. For me, I have excellent support from my

parents. My teachers also helped me a lot. My mother always
looked out for any good math problems out there for me to
solve. I also really like to read math textbooks.

This student’s response provided not only an example of teacher and parental
involvements, but also illustrated Thai cultural influences. This evidence is consistent
 85

with the literature; in Asian cultures, parents consider their children's education their
highest priority in their upbringing. Their children’s education is emphasized and
priority is given to mathematical learning (Hatano, 1990; Geary, 1996). This cultural
value in mathematics also reflected differences in the investment of children, parents,
and teachers in learning mathematics (Geary, 1994; Stevenson & Stigler, 1992).
Furthermore Tocci and Engelhard (1991) investigated the relationships of attitudes
toward mathematics with parental support, mathematics achievement, and gender. The
subjects were eighth grade student in Thailand and United States who participated in
the Second International Mathematics Study (SIMS). The authors used students’
perceptions of parental behaviors toward mathematics as parental support. Results
indicated that in both countries parental support, achievement, and gender were
significant predictors of attitudes toward mathematics. Hence, the evidence from
literature confirmed that parental involvement influenced students in learning
mathematics. Perhaps the focus of the Thai culture on importance of children’s
education is an important factor in these students success in mathematical problem
solving.
Additionally, this study described a model that captured the solution processes
of the Thai gifted students in four stages: understanding, planning, executing, and
verifying. Understanding was the first important stage in guiding the gifted students
toward success in solving the problem set. After the students read the problems aloud,
they identified the questions. The given information was stated, interpreted, and
represented with pictures or tables as well as organized into a systematic format.
Rereading the problem was used for checking the correct representations. Students’
prior knowledge was necessary when they interpreted the given information and
referred to any relevant concepts before developing a solution plan. Reflections on
problem difficulty and familiarity were also established in this stage.
The second stage was planning. Students generated new information and
represented the problems with pictures, symbols, or tables as well as organized them
into a plan. Efficient strategies, such as drawing pictures, making tables, or looking for
patterns were employed with the application of relevant mathematical concepts in
 86

number theory, basic of counting and geometry in solving the problems. The plans
were reevaluated and determined whether they were valid. New plans were derived if
the current plans were determined to be invalid.
The third stage was executing. The students proposed a final answer by
carrying out their computations in this stage. The students made logical mathematical
statements to support their plans and finally stated their conclusions. The last stage,
verifying, involved checking all written solutions and they may have also reread the
problem again in this stage.
The students’ behaviors occurred in this model cycled back and forth among
the four stages. Perhaps the metacognitive behaviors evident throughout the solution
processes were the reason for this activity. It was apparent in each stage that the
students understood when and how to apply mathematical knowledge and strategies
into their solutions. They also applied self-evaluation statements to monitor and
evaluate themselves as problem solvers. These findings are in agreement with other
expert’s view on how metacognitive behaviors are driving forces that influence and
connect cognitive and non-cognitive behaviors at all stages of problem solving,
especially their beliefs and attitudes (Lester, Garofalo, & Kroll, 1989, Schoenfeld,
1985).

Limitations of the Study

A number of limitations affected the findings of this study. The first limitation
was with the mathematical problems used in this study. Although these non-routine
problems met the criteria of this study and students had never seen them previously,
there was a possibility that the results may be biased by the students’ prior knowledge.
Their mathematical background, knowledge, and problem solving experiences prior to
their participation in the Olympiad training camp might have affected how they
approached the problems and their solutions. Furthermore, the number of problems
presented to the students was small and limited to some content areas in mathematics,
such as number theory, combinatorics, and geometry. The specific types of problem
 87

may have influenced their performances if they were not comfortable or did not
specialize in these areas of content. Some problems may have activated a higher
ability in problem solving of this particular group than other problems.
However, the researcher had carefully considered these limitations when
selecting problems for this study. For instance, a pool of problems in each area was
considered by American and Thai experts. These experts were selected because they
were specialists in mathematics and mathematics education. In particular, the Thai
experts had many years of experience in training Thai gifted students in number
theory, combinatorics, and geometry. Experts selected the problems and advised
useful suggestions. Therefore, the problem selection procedures were an important
concern in this study. The researcher tried to limit those constraints that might have
occurred and affected the results as best as possible before collecting data.
A second limitation was the method used in data collection. This study was
based on the assumption that students’ thought processes can be verbalized into words
through the think aloud method. Without time constraint pressure, the participating
students did not have any difficulty verbalizing their thought process, even though it
was their first experience exercising the think aloud method. It is possible that the
think aloud method may have improved their thinking processes and better assisted
their approaches compared to the conventional thinking method. The participating
students may not have reported all of their thoughts because they had to spend extra
effort applying the think aloud technique. In addition, an assumption was made that
the students had answered all interview questions honestly without bias or concern for
self-esteem.
A third limitation was with the small number of participants and the fact that
they were selected from only students in the training camp at the Institute for the
Promotion of Teaching in Science and Technology (IPST), Thailand. Because of the
uniqueness in design and methodology, purposeful sampling was used to select
specific rich cases that meet the selection criteria and can provide a more description
of problem solving processes. However, generalization of study findings to other
settings was not the purpose of this study. The study results and implications are
 88

important for understanding the problem solving processes of Thai gifted students and
provide information for other researchers to conduct future investigations in this
particular area.

Implications and Recommendations for Future Research

The findings of this study raise many issues for future research to improve the
understanding of the mathematical problem solving processes of Thai gifted secondary
students while engaged in solving non-routine mathematical problems. Since little
research has been conducted with Thai gifted students, this study serves as an initial
stage for other studies in this area. The present study has generated a model that was
derived from five gifted students using the think aloud method and individual
interviews. The study was conducted during the four weeks of a training camp, and
students schedules were before and after the training times. Results may be different
for a larger number of participants extending over a longer period of time. Future
study is needed to observe and compare the processes and behaviors when students are
treated in their school environment, rather than in a training camp experience.
The findings are a result of no time limit while solving problems, as a
necessary condition for examining the problem solving process. Having no time
constraint helped decrease students’ pressure and supported them in activating their
problem solving ability. As Sakda responded how he felt when searching the solution
for Problem Three, “I felt stress. However, this is not an examination, and I can use as
much time as I need. So, I just kept going.” Finally, he found the answer. Thus, future
research is needed to consider time as a factor when students work on non-routine
mathematical problem in order to promote students searching for their own solution
path.
The results indicated that the Thai gifted students applied prior knowledge
effectively, used a variety of knowledge, and had a high ability in verbalizing and
explaining their reasons for their solutions. Thus, future research must recognize these
factors and examine how they affect the process and ability in problem solving of
 89

gifted students. The findings indicated that gifted students are capable of integrating
advanced mathematical concepts in their problem solving. This result makes a crucial
recommendation for teachers and educators to create enrichment programs, learning
environments and learning opportunities that provide opportunities for these students
to meet their high potential abilities as well as encourage them to achieve their goals.
As the findings indicated, students’ prior knowledge in problem solving led
them to approach the problems in different ways. Thus, this factor needs consideration
when selecting a problem for training and nurturing this particular group of students in
order to improve their high potential in problem solving through adjustments in
classroom instruction. In addition, the think aloud method together with the interview
proved to be an effective tool that helped the researcher understands the students’
thinking processes and their reasoning as they solved the problem the ways they did.
This strategy might be useful for classroom teachers as they instruct their students in
problem solving.
The results indicated that gifted students effectively used pictorial
representations in solving a geometry problem. These participants also used other
representations to help them succeed in solving non-routine mathematical problems.
Additionally, representation is one of the Standards in NCTM’s 2000
recommendations, and it was mentioned as an essential element that should be treated
in supporting students’ understanding of mathematical concepts and relationships
(NCTM, 2000). The Representation Standard suggested that students should be able
to solve problems by selecting, applying, and translating among mathematical
representations. Thus, future research is needed to investigate a number of questions
about teaching and learning to use multiple representations of gifted students:
• How do teachers help students use multiple representations in solving
problems?
• What factors encourage students to demonstrate the use of various
representations in solving mathematical problems?
• How do these representations impact students’ processes in solving
problems?
 90

• Does the use of representations in problem solving rely on different

mathematical content?
• How does the use of pictorial representation support students in succeeding
in problem solving?

The findings indicated that gifted students employed self-evaluation statements

during the think aloud session. These comments helped the gifted to be successful
problem solvers. It would be reasonable to explore other variables related to these
processes and ways to improve them for more efficient work during problem solving.
Furthermore, research should expand to examine non-cognitive variables such as
motivation, affect, and self-efficacy that might explain some of the variation between
individual students in their problem solving processes.

Concluding Remarks

Problem solving is a complex mental process, which requires the solvers to

engage in organizing, analyzing and reflecting activities (Kirkwood, 2000). Many
researchers have attempted to describe problem solving processes with several models
as mentioned in the literature. This study intended to provide insight to the processes
used by Thai gifted students when they were engaged in solving non-routine
mathematical problems. The study results generated a Thai model of problem solving
process that detailed the students’ behaviors in each of four stages: understanding,
planning, executing, and verifying. Their behaviors described according to this model
not only demonstrated the complex processes but also helped in understanding student
actions during the thinking processes.
The study results may be useful for the Thai Mathematical Olympiad project in
developing future training programs and teaching methods in order to improve the
high abilities of Thai gifted students. Due to the fact that Thai gifted students are
usually in the same classroom with average students, the study findings may help
teachers plan problem solving classroom instruction to facilitate students’ problem
 91

solving abilities. Since the study was conducted with Thai gifted students, the results
also provided examples of the importance of parental support in influencing students
in learning mathematics. Because not much research has examined the problem
solving processes of Thai gifted student, this research serves as initial evidence to
motivate Thai educators in creating enrichment programs or learning materials for
gifted students. The findings are also important for guiding directions for researchers
to extend future research on student thought processes.
 92

BIBLIOGRAPHY

Afflerbach, P., & Johnston, P. (1984). Research methodology: On the use of verbal
reports in reading research. Journal of Reading Behavior, 16, 307- 322.

Alexander, J. M., Carr, M., & Schwanenflugel, P. J. (1995). Development of

metacognition in gifted children: Direction for future research. Developmental
Review, 15, 1-37.

Ampra, K., & Thaithae, C. (2000). Curriculum planning, development and reform in
Thailand. In Central Board of Secondary Education (Ed.), Final report of the
Sub-Regional Course, organized by the International Bureau of Education and
the Indian Ministry of Human Resources Development, New Delhi, (pp. 126-
130). Paris, France: UNESCO.

Anderson, J. R. (1985). Cognitive psychology and its application. New York: W. H.

Freeman.

Anderson, L.W., & Krathwohl, D. (2001). A taxonomy for learning, teaching,

and assessing: A revision of Bloom's taxonomy of educational objectives. New
York: Longman.

ApSimon, H. (1991). Mathematical byways in ayling, beeling, and ceiling. Oxford,

NY: Oxford University Press.

Artzt, A. F., & Armour-Thomas, E. (1992). Development of a cognitive-metacognitive

framework for protocol analysis of mathematical problem solving in small
group. Cognition and Instruction, 9, 137-175.

Becker, G. (1968). Crime and punishment: An economic approach. Journal of

Political Economy, 76, 169-217.

Berkowitz, E., & Cicchelli, T. (2004). Metacognitive Strategy use in reading of

gifted high achieving and gifted underachieving middle school students in New
York City. Education and Urban Society, 37, 37-57.

Bloom, B., Englehart, M. Furst, E., Hill, W., & Krathwohl, D. (1956). Taxonomy of
educational objectives: The classification of educational goals. Handbook I:
Cognitive domain. New York: Longman.

Blosser, P. (1973). Principles of Gestalt psychology and their application to teaching

junior high school science. Science Education, 57, 43-53.
 93

Bodharamik, A. (March 2004). Strategic action plan of the Ministry of Education,

Thailand. Paper presented to Officials of the Ministry of Education, Thailand
at the Bangkok International Trade and Exhibition Center (BITEC), Thailand.

Boekaerts, M., et.al. (1995). Solving math problems: Where and why does the solution
process go astray. Educational Studies in Mathematics, 28, 241-262.

Bogdan, R. C., & Biklen, S. K. (1998). Qualitative research for education: An

introduction to theory and methods. Boston, MA: Allyn and Bacon.

Bowen, C. W. (1994). Think-aloud methods in chemistry education: understanding

student thinking. Journal of Chemical Education, 71, 184-190.

Brown, A. (1978). Knowing when, where and how to remember: A problem of

metacognition. In R. Glaser (Ed.), Advances in instructional psychology, (pp.
77-165). Hillsdale, NJ: Lawrence Erlbaum Associates.

Brown, A.L., & Reeve, R. A. (1987). Bandwidths of competence: The role of

supportive contexts in learning and development. In L. S. Liben (Ed.),
Development and learning: Conflict or congruence?, (pp. 173-223). Hillsdale,
NJ: Lawrence Erlbaum Associates.

Car, M., & Biddlecomb, B. (1998). Metacognition in mathematics from a

constructivist perstructive. In D. J. Hacker, J. Dunlosky, & A. C. Graesser
(Eds.), Metacognition in educational theory and practice, (pp. 69-92).
Mahwah, NJ: Lawrence Erlbaum.

Carlson, M. P., & Bloom, I. (2005). The cyclic nature of problem solving: An
emergent multidimensional problem solving framework. Educational Studies
in Mathematics, 58, 45-75.

Ceci, S. J., & Ruiz, A. I. (1993). Insert context into our thinking about thinking:
Implications for a theory of everyday intelligent behavior. In M. Rabinowitz
(Ed.), Cognitive science: Foundation of instruction, (pp. 173-188). Hillsdale,
NJ: Lawrence Erlbaum.

Chiu, S., & Tron, M. O. (2004). Classroom discourse and the development of
mathematical and analogical reasoning. In L. D. English (Ed.), Mathematical
and analogical reasoning of young learners, (pp. 75-96). Mahwah, NJ:
Lawrence Erlbaum Associates.

Clement, K., & Ellerton, N. F. (1996). Mathematics education research: Past, present
and future. Bangkok, Thailand: UNESCO Principal Regional office for Asia
and the Pacific.
 94

Cobb, P., & Yackel, E. (1995). Constructivist, emergent, and sociocultural

perspectives in the context of developmental research. In D.T. Owens, M.K.
Reed, & G. M. Millsaps (Eds.), Proceedings of the 17th annual meeting of the
North American chapter of the International Group for the Psychology of
Mathematics Education, ( pp. 3-29). Columbus, OH: ERIC Clearinghouse for
Science, Mathematics, and Environmental Education.

Covington, J. (2005). Solutions to January calendar. Mathematics Teacher, 98, 334-

336.

Crain, W. C. (1985). Theories of development concepts and applications. Englewood

Cliffs, NJ: Prentice Hall Inc.

Crain-Thoreson, C., Lippman, M. Z., & McClendon-Magnuson, D. (1997),

Windows on comprehension: Reading comprehension processes as revealed by
two think-aloud procedures. Journal of Educational Psychology, 89, 579-591.

Crawford, K. (1996). Vygotskian approaches to human development in the

information era. Educational Studies in Mathematics, 31, 43-62.

DeBellis, V. A. (1998). Mathematical intimacy: Local affect in powerful problem

solvers. Proceedings of the 20th annual meeting of the North American Group
for the Psychology of Mathematics Education, ( pp. 435-440). Columbus, OH:
ERIC Clearinghouse for Science, Mathematics, and Environmental Education.

Davidson, J. E., & Sternberg, R. J. (1998). Smart problem solving: How

metacognition helps. In D. J. Hacker, J. Dunlosky, & A. C. Graesser (Eds.),
Metacognition in educational theory and practice, (pp. 47-68). Mahweh, NJ:
Lawrence Erlbaum Associates.

Dover, A. (1983). Computers and the gifted: Past, present, and future. Gifted Child
Quarterly, 27, 81-85.

Driscoll, M. (1994). Psychology of learning for instruction. Needham Heights, MA:

Allyn & Bacon.

Ellerton, N. F. (1986). Children’s made-up mathematics problems-a new perspective

on talented mathematicians. Educational Studies in Mathematics, 17, 261-271.

Ericsson, K. A., & Simon, H. A. (1984). Protocol analysis: Verbal report as data.
Cambridge, MA: The MIT Press.

Ericsson, K. A., & Simon, H. A. (1993). Protocol analysis: Verbal reports as data
(Revised edition). Cambridge, MA: The MIT Press.
 95

Fennema, E., & Sherman, J. (1977). Sex-related differences in mathematics

achievement, spatial visualization, and affective factors. American Educational
Research Journal, 14, 51-71.

Flavell, J. h. (1976). Metacognition aspects of problem solving. In L. Rensick (Ed.),

The nature of intelligence, (pp. 231-236). Hillsdale, NJ: Lawrence Erlbaum
Associates.

Flavell, J. h. (1992). Metacognitive and cognitive monitoring: A new area of cognitive

development inquiry. In T. O. Nelson (Ed.), Metacognition-core readings, (pp.
3-8). Library of Congress.

Fortunanto, I., Hecht, D., Tittle, C. K., & Alvarez, L. (1991). Metacognition and
problem solving. Arithmetic Teacher, 39, 38-40.

Gagne, E. (1995). From giftedness to talent: A developmental model and its impact on
the language of the field. Roeper Review, 18, 103-111.

Gallagher, J. J. (1975). Teaching the gifted child. Boston, MA: Allyn and Bacon, Inc.

Garderen, D. V., & Montague, M. (2003). Visual-spatial representation, mathematical

problem solving, and students of varying abilities. Learning Disabilities
Research & Practices, 18, 246-254.

Gardiner, A. (1987). Mathematical puzzling. Oxford, England: Oxford University

Press.

Gardner, H. (1983). Frames of mind. New York: Basic Book Inc.

Gardner, H. (1999). Intelligence reframed: Multiple intelligences for the 21st century.
New York: Basic Book Inc.

Garofalo, J. (1992). Number-consideration strategies students use to solve word

problems. Focus on Learning Problem in Mathematics, 14, 37-50.

Garofalo, J. (1993). Mathematical problem preferences of meaning-oriented and

number-oriented problem solvers. Journal for the Education of the Gifted, 17
26-40.

Garofalo, J., & Lester, F. K. (1985). Metacognition, cognitive monitoring, and

mathematical performance. Journal for Research in Mathematics Education,
16, 163-176.
 96

Geary, D. C. (1996). Sexual selection and sex differences in mathematical abilities.

Behavioral and Brain Sciences, 19, 30.

Geary, D. C. (1994). Children’s mathematical development: Research and practical

applications. Washington, DC: American Psychological Association.

Geary, D. C. (1996). Biology, culture, and cross-national differences in mathematical

ability. In R. J. Sternberg, & T. Ben-Zeev (Eds.), The nature of mathematical
thinking, (pp. 145-171). Mahwah, NJ: Lawrence Erlbaum Associates.

Glasser, B., & Strauss, A. (1977). The discovery of grounded theory: Strategies for
qualitative research. San Francisco, CA: University of California Press.

Goldin, G. A. (2000). Affective pathways and representation in mathematical problem

Solving. Mathematical Thinking and Learning, 23, 209-219.

Gorodetsky, M., & Klavir, R. (2003). What can we learn from how gifted/ average
pupils describe their processes of problem solving?. Learning and Instruction,
13, 305- 325.

Hammouri, H. (2003). An investigation of undergraduates’ transformational problem

solving strategies: Cognitive/metacognitive processes as predictors of holistic/
analytic strategies. Assessment & Evaluation in Higher Education, 28, 571-
586.

Hannah, C. (1990, April). Meacognitive strategies used by learning-disabled gifted

students. Paper presented at the annual meeting of the American Association
for Educational research Association, Boston, MA.

Hatano, G. (1990). Toward the cultural psychology of mathematical cognition.

Monograph of the Society for Research in Child Development, 55, 108-115.

Hausfather, S. J. (1996). Vygotsky and schooling: Creating a social contest for

learning. Action in Teacher Education, 18, 1-10.

Heinze, A. (2003, July). Mathematically gifted elementary students’ problem solving

strategies: Significant differences to “non-gifted” students. Paper presented at
the biennial conference of the World Council for Gifted and Talented Children.
Adelaide, Australia.

Holton, D., & Gaffney, M. (1994). Teaching talented students. In J. Neyland (Ed.),
Mathematics education: A handbook for teacher, (pp. 397-409). Wellington,
New Zealand: Wellington College of Education.
 97

House, P. A. (1987). Providing opportunities for the mathematically gifted. Reston,

VA: National Council of Teachers of Mathematics.

Humenberger, H., & Reichel, H. (1996). Problem solving as a continuous principle of

teaching. In A. S. Posamentier, & W. Schulz (Eds.), The art of problem
solving, (pp. 229-240). Thousand Oaks, CA: Crown Press.

Johnson, M. L. (1983). Identifying and teaching mathematically gifted elementary

school children. Arithmetic Teacher, 30, 25-26; 55-56.

Johnson, R. B., & Onwuegbuzie, A. J. (2004). Mixed methods research: A research

paradigm whose time has come. Educational Researcher, 33, 14-26.

Kantowski, M. G. (1977). Processes involved in mathematical problem solving.

Journal for Research in Mathematics Education, 8, 163-180.

Kapa, E. (1998). A metacognitive support during the process of problem solving in a

computerized environment. Educational Studies in Mathematics, 29, 317-336.

Kearsley, G. (1998). Explorations in learning & instruction: The theory into practice
database: Gestalt theory. George Washington University [On-line]. Available:
http://www.gwu.edu/~tip/wertheim.html

Killen, R. (1996). Effective teaching strategies: Lesson from research and practice,
Sydney, Australia: Social Science Press.

Kilpatrick, J. (1969). Problem solving in mathematics. Review of Educational

Research, 39, 523-534.

Kilpatrick, J. (1978). Variables and methodologies in research on problem solving. In

L. L. Hatfield, & D. A. Bradbard (Eds.), Mathematical problem solving, (pp. 7-
20). Columbus, OH: ERIC Clearinghouse for Science, Mathematics, and
Environmental Education.

Klaimongkol, Y. (2002). The development of an instructional process by applying a

problem-based learning approach to enhance mathematical competencies of
prathom suksa five gifted students in mathematics. Unpublished Doctoral
Dissertation, Chulalongkorn University, Bangkok, Thailand.

Kirkwood, M. (2000). Infusing higher-order thinking and learning to learn into content
instruction: A case study of secondary computing studies in Scotland. Journal
of Curriculum Studies, 32, 509-535.

Krantz, S. G. (1996). Techniques of problem solving. Providence, RI: American

Mathematical Society.
 98

Kroll, D. L. (1988). Cooperative mathematical problem solving and metacognition: A

case study of three pairs of women. Unpublished Doctoral Dissertation,
Indiana University.

Krutetskii, V. A. (1976). The psychology of mathematical abilities in school children.

Chicago, IL: University of Chicago Press.

Kuhn, D., Garcia-Mila, M., Zohar A., & Anderson, C. (1995). Strategies of knowledge
acquisition. Monographs of the Society for Research in Child Development, 60,
42-45.

Lawson, M. J., & Chinnappan, M. (1994). Generative activity during geometry

problem solving: Comparison of the performance of high-achieving and low-
achieving high school student. Cognition and Instruction, 12, 61-93.

Lesh, R. (1982). Metacognition in mathematical problem solving. Unpublished

manuscript.

Lester, F. K. (1980). Research on mathematical problem solving. In R. J. Shumway

(Ed.), Research in mathematics education, (pp. 286-323). Reston, VA:
National Council of Teachers of Mathematics.

Lester, F. K. (1994). Musing about mathematical problem-solving research: 1970-

1994. Journal for Research in Mathematics Education, 25, 660-675.

Lester, F. K., Garofalo, J., & Kroll, D. (1989). Self-confidence, interest, beliefs, and
metacognition: Key influences on problem solving behavior. In D. McLeod, &
V. Adams (Eds.), Affect and mathematical problem solving: A new
perspective, (pp. 324-355). New York: Springer-Verlag.

Lester, F. K, & Schroeder, T. L. (1983). Cognitive characteristics of mathematically

gifted children. Roeper Review, 6, 26-28.

Mayer, R. E. (1985). Mathematical ability. In R. J. Sternberg (Ed.), Human abilities:

information processing approach, (pp. 127-150). San Francisco, CA: Freeman.

Mayer, R. E. (1992). Thinking, problem solving, cognition. New York: Freeman.

Mayer, R. E. (2002). Cognitive, metacognitive, and motivational aspects of problem

solving. In H. J. Hartman (Ed.), Metacognition in learning and instruction,
(pp. 87-102). Dordrecht, The Natherlands: Kluwer Academic Publishers.

Mayer, R. E. (2003). Mathematical problem solving. In J. M. Royer (Ed.),

Mathematical cognition, (pp. 69-92). Greenwich, CT: Information Age
Publishing Inc.
 99

Mayer, R. E., & Hegarty, M. (1996). The process of understanding mathematical

problems. In R. J. Sternberg, & T. Ben-Zeev (Eds.), The nature of
mathematical thinking, (pp. 29-53). Mahwah, NJ: Lawrence Erlbaum
Associates.

Mevarech, Z. R., & Kapa, E. (1996). The effect of a problem-solving based Logo
environment on children’s information processing components. The British
Journal of Educational Psychology, 66, 181-195.

McCormick, B. C. (2003). Metacognition and learning. In W. M. Reynolds, & G. E.

Miller (Eds.), Handbook of Psychology, (pp. 79-102). New York: John Wiley
& Sons, Inc.

Merriam, S. B. (1998). Qualitative research and case study applications in education.

San Francisco, CA: Jossey-Bass Publishers.

Merrill, M. D. (1991). Constructivism and instructional design. Educational

Technology, 31, 18- 22.

Miller, R. C. (1990). Discovering mathematical talent. Reston, VA: Council for

Exceptional Children, ERIC Clearinghouse on Disabilities and Gifted
Education.

Montague, M. (1991). Gifted and learning disabled gifted students’ knowledge and
use of mathematical problem-solving strategies. Journal for the Education of
the Gifted, 14, 393-411.

Montague, M., & Applegate, B. (1993). Middle school students’ mathematical

problem solving: An analysis of think-aloud protocols. Learning Disabilities
Quarterly, 16, 19-32.

Montague, M., & Applegate, B. (2000). Middle school students’ perceptions,

persistence, and performance in mathematical problem solving. Learning
Disabilities Quarterly, 23, 215-227.

Morelock, M. J. (1996). On the nature of giftedness and talent: Imposing order on

chaos. Roeper Review, 19, 4-12.

Nastasi, B. K., & Clements, D. H. (1990). Metacomponential functioning in young

Children. Intelligence, 14, 109-125.

National Council of Teachers of Mathematics. (1983). Report of National Assessment

of Education Progress (NAEP). Reston, VA: Author.
 100

National Council of Teachers of Mathematics. (2000). Principles and standards for

school mathematics. Reston, VA: Author.

Newell, A., & Simon, H. A. (1972). Human problem solving. Englewood Cliffs, NJ:
Prentice-Hall.

Office of the National Education Commission, Ministry of Education, Thailand.

(2004). Education in Thailand. Bangkok, Thailand: Author.

Overtoom-Corsmit, R., Dekker, R. & Span, P. (1990). Information processing in

intellectually highly gifted children by solving mathematical tasks. Gifted
Education International, 6, 143-148.

Patton, M. Q. (2002). Qualitative research & evaluation methods. Thousand

Oaks, CA: Sage Publication, Inc.

Piaget, J. (1971). Biology and knowledge: An essay of the relation between organic
regulations and cognitive processes. Chicago, IL: University of Chicago Press.

Piriyanuwat, S., & Anantrasirichai, A. (November 2002). Curriculum and learning

reform in Thailand. Paper presented at Invitational Curriculum Policy Seminar:
School Based Curriculum Renewal for the Knowledge Society Developing
Capacity for New Times, Hong Kong.

Polya, G. (1957). How to solve it: A new aspect of mathematical method. Garden City,
NY: Doubleday & Company, Inc.

Posamenteir, A. S., & Schulz, W. (1996). The art of problem solving: A resource for
mathematics teacher. Thousand Oaks, CA: Corwin Press, Inc.

Posamenteir, A. S., & Salkind, C. T. (1996). Challenging problems in geometry. New

York: Dover Publications.

Pressley, M., Borkowski, J., & Schneider, W. (1989). Good information processing:
What it is and how education can promote it. International Journal of
Educational Research, 13, 857-867.

Pretz, J. E., Naples, A. J., & Sternberg, R. J. (2003). Recognizing, defining, and
representing problems. J. E. Davidson, & R. J. Sternberg (Eds.), The
psychology of problem solving, (pp. 3-30). Cambridge, MA: Cambridge
University Press.

Pugalee, D. K. (2001). Writing, mathematics, and metacognition: looking for

connections through students’ work in mathematical problem solving. School
Science and Mathematics, 101, 236-245.
 101

Reed, C. F. (2004). Mathematically gifted in the heterogeneously grouped

mathematics classroom: What is a teacher to do?. The Journal of Secondary
Gifted Education, 15, 89-95.

Renzulli, J. S. (1978). What makes giftedness?. Phi Delta Kappan, 60, 180-184.

Rowe, H. (1980). Problem solving and intelligence. Hillsdale, NJ: Lawrence Erlbaum
Associates.

Sankar-DeLeeuw, N. (2004). Case studies of gifted kindergarten children. Roeper

Review, 26, 192-207.

Saxe, G. B. (1991). Culture and cognitive development: Studies in mathematical

understanding. Hillsdale, NJ: Lawrence Erlbaum Associates.

Schifter, D., & Fosnot, C. T. (1993). Reconstructing mathematics education: Stories of

teacher meeting the challenge of reform. New York: Teachers College Press.

Schoenfeld, A. H. (1982). Some thoughts on problem solving research and

mathematics education. In F. K. Lester, & J. Garofalo (Eds), Mathematical
problem solving: Issues in research, (pp. 27-37). Philadelphia, PA: Franklin
Institute Press.

Schoenfeld, A. H. (1985). Mathematical problem solving. Orlando, FL: Academic

Press.

Schoenfeld, A. H. (1992). Learning to think mathematically: Problem solving,

metacognition, and sense making in mathematics. In D. Grouws. (Ed.),
Handbook of Research on Mathematics Teaching and Learning, (pp. 369-408).
New York: MacMillan.

Schoenfeld, A. H., Burkhardt, H., Daro, P., Ridgway, J.,Schwartz, J., & Wilcox, S.
(1999). High school assessment. White Plains, NY: Dale Seymour
Publications.

Schroeder, T. L., & Lester, F. K. (1989). Developing understanding in mathematics

via problem solving. In P. R. Trafton (Ed.), New directions for elementary
school mathematics (pp. 31-42). Reston, VA: National Council of Teachers of
Mathematics.

Seels, B. (1989). The instructional design movement in educational technology.

Educational Technology, 29, 11-14.
 102

Serpell, R. & Boykin, A. W. (1994). Cultural dimensions of cognition: A multiplex,

dynamic system of constraints and possibilities. In R. J. Sternberg (Ed.),
Thinking and problem solving, (pp. 369-408). New York: Academic Press.

Sigel, I. E., & Cocking, R. R. (1977). Cognitive development from childhood to

adolescence: A constructivist view point. Holt, NY: Rinehart & Winston.

Silver, E. A. (1982). Knowledge organization and mathematics problem solving. In

F. K. Lester, & J. Garofalo (Eds.), Mathematical problem solving: Issues in
research, (pp. 15-25). Philadelphia: Franklin Institute Press.

Simon, D., & Simon, H. (1978). Individual differences in solving physics problems. In
R. Siegler (Ed.), Children’s thinking: What develop? , (pp. 325-348). Hillsdale,
NJ: Lawrence Erlbaum.

Someren, M. W., Barnard, Y. F., & Sandberg, J. A. (1994). The think aloud method.
London, Great Britain: Academic Press.

Sousa, D. A. (2003). How the gifted brain learns, Thousand Oaks, CA: Corwin Press
Inc.

Sowell, E. J. (1993). Programs for mathematically gifted students: A review of

empirical research. Gifted Child Quarterly, 37, 124-132.

Sowell, E. J., Zeigler, A. J., Bergwell, L., & Cartwright, R. M. (1990). Identification
and description of mathematically gifted students: A review of empirical
research. Gifted Child Quarterly, 34, 147-154.

Span, P., & Overtoom-Corsmit, R. (1986). Information processing by intellectual

gifted pupils solving mathematical problems. Educational Studies in
Mathematics, 17, 273-295.

Sriraman, B. (2003). Mathematical giftedness, problem solving, and the ability to

formulate generalizations. The Journal of Secondary Gifted Education, 14,
151-165.

Sternberg, R. J. (1981). A componential theory of intellectual giftedness. Gifted Child

Quarterly, 25, 86-93.

Sternberg, R. J. (1986). A triarchic theory of intellectual giftedness. In R. J. Sternberg,

& J. E. Davidson (Eds.), Conceptions of Giftedness, (pp. 223-243). Cambridge,
MA: Cambridge University Press.
 103

Sternberg, R. J. (1993). The concept of giftedness: A pentagonal implicit theory. In G.

R. Block, & K. Ackrill (Eds.), The origins and development of high ability,
(pp. 5-16). Chi Chester, England: Wiley.

Sternberg, R. J. (1995). In search of human mind, Orlando, FL: Harcourt Brace

College Publishers.

Stevenson, H. W., & Stigler, J. W. (1992). The learning gap: Why our schools are
failing and what can we learn from Japanese and Chinese education. New
York: Summit.

Stillman, G. A., & Galbraith, P. L. (1998). Applying mathematics with real world
connections: Metacognitive characteristics of secondary students. Educational
Studies in Mathematics, 36, 157-195.

Stone, C. A., & Reid, D. K. (1994). Social and individual forces in learning:
Implications for instruction of children with learning difficulties. Learning
Disability Quarterly, 17, 72-86.

Swanson, L. H. (1990). Influence of metacognitive knowledge and aptitude on

problem solving. Journal of Educational Psychology, 82, 306-314.

Thipatdee, G. (1996). The construction of an enrichment curriculum developing

complex thinking ability of the upper secondary school students with high
achievement. Unpublished Doctoral Dissertation, Chulalongkorn University,
Bangkok, Thailand.

Tocci, C. M., & Engelhard, G. (1991). Achievement, parental support, and gender
differences in attitudes toward mathematics. Journal of Educational Research,
84, 280-286.

Weinburg, S. S. (1992). Probing the depths of students' historical knowledge.

American Historical Association Perspectives, 30, 1-24.

Wieczerkowski, W., Cropley, A. J., & Prado, T. M. (2000). Nurturing talents/gifts in

mathematics. In K. A. Heller, F. J. Monks, R. J. Sternberg, & R. F. Subotnik
(Eds.), International handbook of giftedness and talent education, (pp. 413-
425). Oxford, United Kingdom: Pergamon.

Yimer, A. (2004). Metacognitive and cognitive functioning of college students during

mathematical problem solving. Unpublished Doctoral Dissertation, Illinois
State University.

Yin, R. K. (1994). Case study research design and methods, Newbury Park, CA: Sage
Publications Inc.
 104

APPENDICES
 105

Appendix A

Examples of Participants’ Problem Solving Processes

The following discussion presents what happened during the think aloud
session when the participants solved each problems. In order to describe participant’s
problem solving process, the researcher selected three of the participants to present
his/her process for either Problem One, Problem Two, and Problem Three.

Pradya’s Problem Solving Processes for Problem One

Pradya began his solution by reading Problem One aloud. He linked his prior
experience in problem solving to this problem as he said “It seems to be a number
theory problem. Then he restated what the problem was asking into his own words
said “Amm… a Friday the 13th is in the calendar every year or not.” Pradya planned to
solve Problem One using the concept of remainders. He searched for the day of the
13th in each month by beginning with the day of January 13th successively to each day
of the week (Sunday, Monday, Tuesday, Wednesday, Thursday, Friday, or Saturday).
He made a table to represent days from Sunday to Saturday and set Sunday as the day
one. So, Friday was the sixth day. He assumed the day for January 13th and used
variable x to represent it. The variable x can be represented Sunday, Monday,
Tuesday, Wednesday, Thursday, Friday, or Saturday. He found that January13th,
January 20th, and January 27th were on the same day. He represented the days from
January 28th to February 2nd with x+1, x+2, …, x+6. February 3rd was x+7, the same
day as x. So, the day of February 10th was x and February 13th was x+3. At this point
he concluded that when January13th was x, the day of February 13th was x+3. As he
considered the months of February and March, he recognized 2 cases. The day of
March 13th depended on the number of days in February, 28 or 29 days. He found the
days for the 13th in the other months for both cases by considering the remainders
when dividing the number of days in that month by 7. For example, if there were 28
days from February 13th to March 13th and the remainder was 3. So, the day of March
 106

13th was x+3, or the same day as February 13th. For another case, the day of March
13th was x+4. He continued this method until he completed all 12 months. He found
that in both cases the distribution were the same, that is, x, x+1, x+2, x+3, x+4, x+5,
x+6. He concluded that Friday 13th fell on Sunday, Monday, Tuesday, Wednesday,
Thursday, Friday, and Saturday in any year.

Wude’s Problem Solving Processes for Problem Two

After reading Problem Two aloud, Wude tried to find the reasons that
supported what the problem asked as he said “This problem must be analyzed first.
The problem needs to know the reason why there is at least one draw. He then thought
about what he was going to assume for prove this, “Okay, if we assume that this team
has not any draw, it has only win and loss.” At that time, he realized that this problem
needed more thought. Although he began his plan to find that the highest total score
team had no draws, he eventually determined that it was not possible. He monitored
himself to continue thinking and stated some given information by saying:
Wude: Must to think. The problem said the team with the lowest total
score gets 21 points. We must think slowly. Think slowly.

Next, Wude restated the given information that there were 15 teams. He
quickly calculated in his mind the total games played with the combination of 15 and 2.
That, he found, was equal to 105 times. He also asked himself to check the number
that he got. He considered what happened for each game that resulted in a win, loss, or
draw. He assumed that a team drew x times, won and lost y times. He decided to rank
the scores of 15 teams from the maximum score team to the minimum team score. He
thought that team 15 had the maximum score team and linked what he supposed with
the given information as team 1 had 21 points.
Wude returned to read the other given information; “The problem said a team
got 3 points for a win, 2 points for a draw, and 1 point for a loss.” He found that team
1 met the other teams 14 times before planning to find the scores that each team got
and the minimum time this team won. At this point, he considered:
 107

Wude: Suppose that team 1 has not any win. So, team 1 has 10 draws and
1 loss. But, it plays 14 times. It is impossible. How about team 1
has 3 wins 1 draw and 10 losses? We can distribute the results of
team 1 in three cases. It gets at least 8 times in order to have 21
points.

Next, Wude looked for what the problem asked and stated some given
information; the problem asked about the highest total score team. It also said every
team had a different total score. The lowest total score team got 21 points. He found
that the next team got 22 points and thought about the score of the highest total score
team. Finally, he found that the highest total score team needed at least 35 points. He
also considered that when two teams played together, they got 3+1 or 2+2 which
equaled 4 points. Since he had already known the total number of games played was
105 times, the total score was 420. He calculated the sum of a sequence 21, 22, 23, …,
35 in his mind which equaled 420. Thus, he concluded that the highest score a team
might get was 35 points. He mentioned that “When we know this, it is not hard to do
next.”
He used the variables x, y, and z to represent the number of times each team
won, lost, and drew. He set two equations 3x + 2y + z = 35 and x + y + z = 14 and
solved them to get 2x + y = 21. While he thought about the possible answers for this
equation, he rearranged the equation into y = 21 – 2x. He noticed that from this
equation, y was the number of times which this team earned a draw and the highest
total score team wins as more than zero (x > 0). So, he found y > 0 too. To prove the
highest total score team got one draw, he considered the equation y = 21 – 2x and
provided the following reasons.
Wude: Let’s see 21 – 2x is an odd number. In this term, x is an positive
integer, 21 is an odd number, 2x is an even number.
Therefore, y ≥ 1.
Finally, he also stated his conclusion that answered this problem; the highest total
score team had to have at least one draw.
 108

Nipa’s Problem Solving Processes for Problem Three

After Nipa read Problem Three aloud, she drew a picture. She stated the given
parts while drawing the picture. This action indicated that she used the picture
representation for this problem. She then thought about how to find the answer for the
angle CDE. From the given information, she found the value of the other three angles.
Nipa: The angle DAE is 30 degrees. The angle DCE is 20 degrees. The angle
AEC is 50 degrees.

At this point, Nipa decided to draw a new picture. She thought the picture that
she drew was too small. This picture confused her and she was unable to determine a
solution path when she saw the picture. With a new picture, she asked herself what she
was going to do next. She used her prior knowledge about isosceles triangles and the
given information to identify other angles in the picture. For example, she found that
angle ADC was 40 degrees because angle DAC and angle ACB were 80 degrees. She
also got CDB as an isosceles triangle because the base angles DCB and DBC were 20
degrees. To find the angle CDE, Nipa drew a perpendicular line from point D to BC
and met BC at the point X. Then, she got angle CDX as 70 degrees because angle DCE
was 20 degrees and angle CXD was 90 degrees. In this way, she was unable to figure
how to solve the problem and asked herself; “How do we find the angle CDE that the
problem asks?”
She decided to draw another line from point C to Y which was parallel to DE.
She used a variable x to represent the value of angle YCD and got angle DYC as
(140 – x) degrees and angle AYC as (40 + x) degrees. She still did not find a solution
plan and said, “I don’t know what I’m going to do.” So, she decided to draw a new
picture and filled the values of the angles that she knew from the previous picture.
Nipa: In the former picture, I drew a parallel line, but it did not make any
help. Anyway, try other ways again.

She searched for a new plan by drawing a line from point C to meet AB at
point F such that angle ACF was 20 degrees. Then, she got angle AFC as 80 degrees.
Therefore, ACF was an isosceles triangle where CA = CF. Since angle AEC was 50
 109

degrees, ACE was an isosceles triangle with AC = EC. She drew another line EF. At
that time, she still was unable to figure out this problem and drew a new picture. She
commented that this fourth picture needed to be as big as possible so she could work
on it. For a new plan, she drew a line CF that made angle ACF as 20 degrees. Then,
she got angle AFC as 80 degrees and ACF as an isosceles triangle where CA = CF.
Next, she considered the triangle ACE and found that angle AEC was 50 degrees
because the base angles were 50 and 80 degrees. So, ACE was another isosceles
triangle with AC = CE. The angle ADC was 40 degrees because angle DAC was 80
degrees and angle ACD was 60 degrees. In a triangle DCF, angle FDC and angle DCF
were equal to 40 degrees. So, DCF was an isosceles triangle with FC = FD.” She
drew FE Line, and since AC = CE, ∆CEF was an isosceles triangle with the top angle
60 degrees, the angle FEC was 60 degrees. Therefore, CEF was an equilateral triangle
and CE = EF = CF. Then, angle DFE was 40 degrees because 180 – 80 – 60 = 40.
She drew a new line DE. The Line FE then met DC at the point X where the angle
DXF was 100 degree. The angle DXE was 80 degrees because of an opposite angle.
The angle BDC was 140 degrees because 180 – 40 = 140. By this time, she realized
she could not solve the problem with the current information. She asked herself again,
“How many degrees for the angle CDE?” Suddenly, it appeared to her that FD = FE.
Thus, FDE was an isosceles triangle. She uttered to herself, “Ohh… I got it.” And she
started to write the solution down on another piece of paper. She was relieved and
said, “I almost died.”
By this time, Nipa started to draw a new and even bigger picture from the
problem statement description. She then set point F on line AB, drew CF which
resulted in the angle ACF becoming 20 degrees. She set the angle DCF 40 degrees and
the angle AFC was 80 degrees. So, ∆ACF was an isosceles triangle with sides CA =
AF. At this time, looking at the triangle ACE, she set the angle AEC 50 degrees
because the base angles were 50 and 80 degrees. Hence, AC = CE and ACE was an
isosceles triangle. After that, she drew line EF. Because AC = CE, ∆CEF was an
isosceles triangle with the top angle 60 degrees. Another word, the angle FEC was 60
degrees. Conclusively, CEF was an equilateral triangle because all three angles were
 110

60 degrees. She also set CE = EF = CF. Moving on to the triangle ADC, she set the
angle ADC 40 degrees because the angle DAC was 80 degrees and the angle ACD was
60 degrees. In a triangle DCF, angle CDF and angle DCF were equal to 40 degrees.
So, ∆CDF was an isosceles triangle resulting in FC = DF. She continued by drawing
line DE, convinced that line FC equaled line EF and line FC was equal to line DF.
She then concluded that EF = DF.
By then, it was apparent to her that ∆DFE was an isosceles triangle which
confirmed the fact that DF = EF. She continued that because CEF was an equilateral
triangle, angle EFC must be 60 degrees. Because the fact that sum of three angles
must be 180 degrees, the sum of angles DFE, EFC, and CFA had to be 180 degrees.
Therefore, angle DFE was 40 degrees. Furthermore, since ∆DFE was an isosceles
triangle with angle EFC equal to 40 degrees, angle FDE was 70 degrees as a result.
Lastly, she concluded that the sum of angles FDC and CDE had to be 70 degrees and
angle FDC was 40 degree. Therefore, angle CDE was 30 degrees. She assured herself
that she was completely done with this problem and thought aloud with the statement,
“I’m done.”
 111

Appendix B
Think Aloud Protocol

Instructions

The title of this study is “Problem Solving Processes of Thai Gifted Students.”
I am interested in the processes you use when solving problems. I cannot read your
mind; however, the think aloud method will help me understand your ideas while
solving problems. In total, I will give you three mathematical problems to solve.
Please always speak out loud while you are working on these problems and describe
how you are solving them. Your participation will be videotaped. The videotape will
be erased when the study is completed.

Now, we will practice this method with a sample problem. Here are paper and
pencils for you to use. Remember to speak out loud as you work on the problem. To
solve this problem, you can use as much time as you need. Do you have any questions
before we start to practice?
(The researcher will give a sample problem to the participant.) Please read this
mathematical problem aloud.

Sample problem:

It takes 1140 digits to number the pages of a book. How many pages are there in
the book?

(The participant reads the problem. The researcher will encourage the participant
when he/she pauses for an extended period of silence; the researcher will use the
following prompts to encourage the participants to speak his or her thoughts: “keep
explaining aloud what you are thinking,” or “say everything in your mind” or “speak
out loud what you are doing”)
Now, you know how to use the think aloud method.
Do you have any questions about this?
 112

Appendix C
Mathematical Problem Solving Tasks

Show all your work to get the solutions.

1. Does a Friday the 13th occur every year? Explain your reasons.

2. In a tournament, there are 15 teams. Each team plays with every other team
exactly once. A team gets 3 points for a win, 2 points for a draw, and 1 point
for a loss. When the tournament finishes, every team receives a different total
score. The team with the lowest total score is 21 points. Explain why the
highest total score team has at least one draw.

3. Let ABC be an isosceles triangle with AB = BC. Angle ABC equals to 20

degrees. Point D is on AB such that angle ACD equals to 60 degrees. Point E is
on BC such that angle EAC equal to 50 degrees. Find the value of angle CDE.
 113

Appendix D
Interview Protocol

Instructions: Before starting the interview, the participant will be given copies of the
problem and the participant’s solution paper from the think aloud session to be
reviewed with the interviewer. The following questions will be used as a guideline of
the interview. Other questions may be added to further prompt participants’ thought
processes when they answer in varying ways. If the participant does not understand
what they are being to asked, the interviewer may clarify the question.

Part 1: Participant’s background

1. What is your name?
2. How old are you?
3. What is your birth date?
4. What is your school name?
5. What grade are you in?
6. Do you like to solve mathematical problems? If so, why?
7. What kind of mathematical problems do you like to solve?
8. Do you think you are a good problem solver? Why? Why not?
9. What do you think makes someone a good problem solver?

Part 2: Participant’s problem solving processes

Understanding
1. Can you tell me about this problem?
2. Have you ever seen a similar problem like this before?
If yes, did it affect how you solved this problem?
3. Did you have difficulty understanding the given information in the problem?
If yes, explain what parts confused you?
4. What did you do after reading the problem?
5. Before you started working, did you think the problem was difficult to solve?
Planning
6. How did you plan to solve this problem at the beginning? Explain.
7. What did you do to overcome any difficulties?
8. Did you plan to use any known mathematical content to solve the problem?
9. Did you plan to use any known mathematical procedure to solve the problem?
10. Please explain your solution plan to me?

Executing
11. How did you decide to carry out your solution plan?
12. What mathematical content did you use? Explain why and how.
 114

13. What mathematical strategies and procedures did you consider as potentially
useful for solving the problem?
14. Did you follow your solution plan? If no, explain why not.
15. How did you know you solved the problem correctly?
16. What did you do when you got stuck on the problem?

Verifying
17. How can you be sure that your solution is correct?
18. Did you check that your solution with your plan and the given conditions of
the problem?
19. If you and a friend got different answers to the same problem, what would you
do?
20. Is it possible to get the correct answer and still not understand the problem?
Explain.

Others
21. What did you learn from solving this problem?
22. Do you have any other comments about your work and thoughts while working
on this problem?