Thinking Styles

Thinking Styles
Thinking Styles:
Identity, Value, and Malleability
By
Jieqiong Fan
Thinking Styles: Identity, Value, and Malleability
By Jieqiong Fan
This book first published 2020
Cambridge Scholars Publishing
Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK
British Library Cataloguing in Publication Data

A catalogue record for this book is available from the British Library
Copyright © 2020 by Jieqiong Fan
All rights for this book reserved. No part of this book may be reproduced,
stored in a retrieval system, or transmitted, in any form or by any means,
electronic, mechanical, photocopying, recording or otherwise, without
the prior permission of the copyright owner.
ISBN (10): 1-5275-5817-7

ISBN (13): 978-1-5275-5817-5
To Christopher, my beloved son
CONTENTS
List of Tables ............................................................................................. ix
Preface ...................................................................................................... xii
List of Abbreviations ................................................................................ xv
Chapter One ................................................................................................ 1

An Introduction to Thinking Styles
The Field of Styles: An Overview ........................................................ 1
Theory of Thinking Styles: Mental Self-government ........................... 4
Controversial Issues in the Field of Styles ............................................ 6
About This Book ................................................................................. 20
Chapter Two ............................................................................................. 26

Personality and its Relationship with Thinking Styles
The Overlap Between Personality and Thinking Styles...................... 27
The Unique Contribution of Thinking Styles Beyond Personality ..... 32
Comparing The Malleability of Personality and Thinking Styles ....... 37
Chapter Three ........................................................................................... 40

The Socialization of Thinking Styles
Changes in Thinking Styles ................................................................ 41
Family Factors and Thinking Styles ................................................... 57
Learning Environments and Thinking Styles ...................................... 72
The Competitive Influence of Parenting, Learning Environments,
and Personality on the Development of Thinking Styles .............. 88
Chapter Four ........................................................................................... 100

The Role of Thinking Styles in Student Development
Thinking Styles and Academic Development ................................... 101
Thinking Styles and Non-academic Development ............................ 103
viii Contents
Chapter Five ........................................................................................... 139

The Response to Three Controversial Issues
Style Identity ..................................................................................... 139
Style Malleability .............................................................................. 143
Style Value........................................................................................ 149
Chapter Six ............................................................................................. 154

Concluding Remarks
Conclusions....................................................................................... 154
Contributions and Implications ......................................................... 156
Future Directions .............................................................................. 160
Appendices ............................................................................................. 162
References .............................................................................................. 186

LIST OF TABLES
Table 1.1: Descriptions of the 13 Thinking Styles in the Theory

of Mental Self-Government .................................................................. 5
Table 1.2: The Distribution of the Sample in the Main Study .................. 23
Table 2.1: Partial Correlations Between Thinking Styles and Personality

Traits ................................................................................................... 29
Table 2.2: Regressions of Thinking Styles on Personality Traits (EIV) ... 30
Table 2.3: Regressions of CDSE on Personality Traits and Thinking

Styles................................................................................................... 35
Table 2.4: Regressions of SWB on Personality Traits and Thinking

Styles................................................................................................... 36
Table 2.5: Repeated Measures on Thinking Styles and Personality

Traits ................................................................................................... 38
Table 3.1: Mean-Level Change and Rank-Order Consistency

of Thinking Styles ............................................................................... 45
Table 3.2: The Distribution of the Reliable Change Index ....................... 49
Table 3.3: The Distribution of the Reliable Change Index

Among Two Grades ............................................................................ 50
Table 3.4: Zero-Order Correlations Between Socioeconomic Status

(SES) and Thinking Styles .................................................................. 60
Table 3.5: Predicting Thinking Styles from the Three Dimensions

of Parenting Styles (Time 1) ............................................................... 62
Table 3.6: Predicting Thinking Styles from the Three Dimensions

Table 3.7: Style Differences Based on the Types of Parenting

Styles (MANCOVA) (Time 1) ........................................................... 64
x List of Tables
Table 3.8: Style Differences Based on the Types of Parenting Styles

(MANCOVA) (Time 2) ...................................................................... 65
Table 3.9: Multiple Comparisons for Thinking Styles Based on Types

Table 3.10: Multiple Comparisons for Thinking Styles Based on Types

Table 3.11: Predicting Thinking Styles from Learning Environments

(Time 1) .............................................................................................. 82
Table 3.12: Predicting Thinking Styles from Learning Environments

(Time 2) .............................................................................................. 83
Table 3.13: Predicting Thinking Styles from Parenting Styles, Learning

Environments, and Personality Traits (Time 1) .................................. 90
Table 3.14: Predicting Thinking Styles from Parenting Styles, Learning

Environments, and Personality Traits (Time 2) .................................. 92
Table 3.15: Conditional Change Model of Thinking Styles ..................... 98
Table 4.1: Predicting Career Decision Self-Efficacy from Thinking

Styles with Parenting Styles, Learning Environments, Personality
Traits, and Relevant Demographic Factors in Control (Time 1) ....... 108
Table 4.2: Predicting Career Decision Self-Efficacy from Thinking

Styles with Parenting Styles, Learning Environments, Personality
Traits, and Relevant Demographic Factors in Control (Time 2) ....... 110
Table 4.3: Predicting the Change of Career Decision Self-Efficacy from

Thinking Styles (Time 1) in the Presence of Parenting Styles (Time 2),
Learning Environments (Time 2), and Personality Traits (Time 1) .... 113
Table 4.4: Predicting Subjective Well-being from Thinking Styles

with Parenting Styles, Learning Environments, Personality Traits,
and Relevant Demographic Factors in Control (Time 1) .................. 120
Table 4.5: Predicting Subjective Well-being from Thinking Styles

with Parenting Styles, Learning Environments, Personality Traits,
and Relevant Demographic Factors in Control (Time 2) .................. 122
Thinking Styles: Identity, Value, and Malleability xi
Table 4.6: Predicting the Change of Subjective Well-being from Thinking

Styles (Time 1) in the Presence of Parenting Styles (Time 2),
Learning Environments (Time 2), and Personality Traits (Time 1) .... 124
Table 4.7: Testing Longitudinal Relationships that are Potentially

Mediated by Thinking Styles ............................................................ 132
PREFACE
Thinking styles, one of the latest theories in the field of intellectual styles,
are defined as one’s preferred way of using abilities. Fifteen years ago, I
was so excited when I first learned the concept of thinking styles in a lecture.
It has been claimed that the construct of thinking styles is an additional
factor that can explain individual differences in performance beyond ability
and personality. Furthermore, unlike ability and personality which are at
least half determined by genetic factors, thinking styles were considered to
be more modifiable by environments and experience. Based on these
contentions, it seemed very likely that one can successfully improve
students’ performance by cultivating their thinking styles. It has also been
argued that thinking styles are value free, which means that there are no
good or bad styles and that only “style match” matters. This argument is
also tempting.
A few years later, I began my doctoral study and chose thinking styles as
my research area. I was so surprised when I found out that the above
contentions were mostly conceptual arguments rather than evidence-based
conclusions. Likewise, many arguments with regard to other style models in
the field of intellectual styles were also widely disseminated to educational
practitioners before being fully examined. The hasty dissemination of the
concept of styles combined with the myriad of style models in the field
inevitably attracted plenty of criticism and doubts, which, at one time, made
the field stagnate.
Fortunately, scholars have made tremendous efforts to integrate the field of

styles in recent years. Moreover, many measurement tools for styles have
improved their psychometric properties. Meanwhile, the methodologies
have become more sophisticated. The time is ripe to examine fundamental
assumptions about styles both systematically and rigorously. This book
reviews the relevant literature and introduces empirical evidence addressing
the three major controversial issues in the field: 1) whether thinking styles
are distinct from or a part of personality traits; 2) whether thinking styles
can be changed; and 3) whether thinking styles are value-laden. Based on
solid research findings, this book discusses why thinking styles are
important, which thinking styles should be cultivated, and how to cultivate
them. Furthermore, the hotly-debated issues of style identity, malleability,
Thinking Styles: Identity, Value, and Malleability xiii
and value are revisited and discussed in more detail. In doing so, this book
helps researchers and educators to have a deeper understanding of the nature
of styles and their role in student development; it also provides practical
suggestions for parents, teachers, and other educational practitioners.
Hopefully, this book will contribute to the advancement of the field of styles
and facilitate the application of style theories and research in educational
practice.
This book is aimed at academics and graduate students who wish to research
styles and its relevant fields, such as educational psychology, school
psychology, higher education, student development, curriculum design and
instruction, and career counselling. Educators, teachers, counsellors, and
other practitioners in schools, universities, and other institutions will also
be interested in this book because it discusses the practical implications of
research findings on cultivating adaptive thinking styles and promoting
student development.
It should be noted that this book’s major work is rooted in my PhD research
project and partially draws on some of my recent research articles. Many
people have provided generous help and support during the research process
and the publication of this book. I would like to take this opportunity to
express my appreciation to them. First and foremost, my deepest gratitude
goes to my PhD supervisor, Professor Li-fang Zhang. She possesses all the
attributes required for a fantastic supervisor. Professor Zhang has provided
me with prompt guidance and unconditional help whenever I needed it. Her
feedback and comments on my work are always constructive and insightful.
I am also deeply grateful to Professor Zhang for her constant encouragement
and trust in my abilities. Without her, I would not have made such
tremendous progress. Furthermore, Professor Zhang’s excellent academic
abilities, as well as her conscientious, persistent, and dedicated attitude to
work, have deeply influenced me and will definitely continue to inspire me
throughout my entire academic career.
I would also like to thank my teachers, colleagues, and friends both inside
and outside the Faculty of Education in the University of Hong Kong.
Discussions with them are always illuminating. I am deeply grateful to
Professor Stephen Rayner, Professor Carol Chan, and Dr. W. W. Ki for their
constructive feedback on my PhD dissertation. Special thanks are also given
to Professor Chen Chen and Dr. Yanbi Hong who have provided me with
generous support. Their valuable advice and comments have been a great
help in my research. Their empathy and encouragement have motivated me
on my way forward.
xiv Preface
Moreover, my sincere thanks go to Dr. Guohong Wu, Ms. Shu Gao, and Mr.
Mingbo Liu for their facilitation of my research’s data collection over the
years. I would also like to thank the teachers who arranged time in their
classes for me and the students who participated in my research. In addition,
I am deeply grateful to my Cambridge Scholars Publishing editors, Helen
Edwards and Adam Rummens, as well as my proofreader, Dr. Joanne Ella
Parsons, for their competent assistance when contracting the book,
reviewing the book proposal, and polishing the manuscript into its published
form. I also want to thank my typesetting manager, Amanda Miller, for her
constant help during the publishing process.
Finally, I would like to express heartfelt appreciation to my family,

especially my little son, without whom this book would have been
completed much earlier. It was also my son who have succeeded in changing
my style preference from the hierarchical style to the anarchic style.
LIST OF ABBREVIATIONS
CDSE: Career decision self-efficacy
FDI: Field-dependence/independence
ISPLE: Inventory of students’ perceived learning environment
KAI: Kirton’s adaption-innovation styles
LE: Learning environments
PS: Parenting styles
SCCT: Social cognitive career theory
SES: Socioeconomic status
SWB: Subjective well-being
TS: Thinking styles
TSI: Thinking style inventory

CHAPTER ONE
AN INTRODUCTION TO THINKING STYLES
Some people have an excellent performance in school but then fail in their
career. A person could be judged as less intelligent in one setting, but may
also be judged as having better abilities in another. A person may not have
excellent performance when he is as an athlete, but he could be a brilliant
coach. In our society, students have to go through fierce competitions in
order to enter the next stage in the educational system: primary school,
junior school, high school, and university. Some students, even though they
have a great performance at the beginning, do increasingly badly in these
successive stages. This is not surprising as the schools in each stage are highly
selective when recruiting students. It is very possible that one person will
become confronted with people in successive stages who have stronger
abilities than those in the last stage. Therefore, it is very difficult for them to
outperform other people every time. However, students can still improve at
every point. Robert Sternberg, who proposed the theory of thinking styles, was
one of them. Like many other scholars, he noticed that the above phenomena
could not be simply explained by abilities or personality. For decades,
researchers, educators, psychologists, and philosophers have been trying to
solve this mystery. During this process, the construct of styles emerged. After
the concept of styles was firstly introduced by Gordon Allport (1937), various
style constructs and theories proliferated (Zhang & Sternberg, 2006). Style
theorists believe that people who have identical abilities may have different
styles, which leads to different performances contingent to their environment.
The Field of Styles: An Overview

In the growing body of literature on styles, there are too many different style
labels. Among them, three high-order terms, which encompass many style
constructs, are frequently used.
One is cognitive styles, which refers to the ways in which people prefer to
process information (Armstrong, Peterson, & Rayner, 2012). Typical
examples of cognitive styles are field-dependent/independent (Witkin,
1954), intuitive-thinking (Myers, 1962), reflective-impulsive (Jerome
2 Chapter One
Kagan, 1965), serialist-holist (Pask & Scott, 1972), and activist-reflector

(Kolb, 1976).
Another term is learning styles, which refers to the ways in which people
prefer to respond to learning tasks (Armstrong et al., 2012). Typical
constructs that are classified into learning styles include instructional
preference (Friedman & Stritter, 1976), learning preference (Rezler &
Rezmovic, 1974), study process (Biggs, 1979), and approach to study
(Entwistle & Ramsden, 1983).
The third term is thinking styles, which was proposed by Sternberg (1988,
1997). Compared with the previous two, thinking styles refer to the ways
that people prefer to think about the information as they are learning it or
after they already know it. In recent studies, thinking styles are considered
to be more general than cognitive or learning styles, because they include
all three traditions (cognition-centered, personality-centered, and activity-
centered) in the field of styles (Zhang & Fan, 2011).
However, the relationships between the various style labels are still puzzling
for many scholars. For example, after a careful review, Zhang and Sternberg
(2006) found cognitive styles and learning styles have much in common,
and some scholars have used them interchangeably (e.g., Campbell, 1991;
Tennant, 1997). In addition, there are still disputes over some style
constructs, such as Kolb’s (1976) learning style (Zhang & Sternberg, 2006).
Aiming to further clarify the relationships between styles, scholars have
made endeavors beyond the aforementioned three terms, which
significantly contribute to the integration of the existing studies in the field.
Among all considerable endeavors to clarify the relationships between

styles, six integrative models of styles particularly stand out. They are
Curry’s (1983) model, Miller’s (1987) model, Riding and Cheema’s (1991)
model, Grigorenko and Sternberg’s (1995) model, Sadler-Smith’s (2009)
model, and Zhang and Sternberg’s (2005) model.
Taking “onion” as a metaphor, Curry’s (1983) model claimed that there are
three layers of styles. The innermost layer involves personality dimensions,
the middle layer involves information processing, and the outermost layer
involves instructional preferences. Miller (1987) proposed a model of
cognitive processes and styles where cognitive styles are seen as individual
differences in three types of cognitive processes: perception, memory, and
thought. He also suggested that all cognitive styles can be analyzed by one
bipolar dimension: analytic-holistic. Riding and Cheema (1991) adopted
An Introduction to Thinking Styles 3
two bipolar dimensions to classify cognitive styles: wholistic-analytic and

verbal-imagery. According to Grigorenko and Sternberg’s (1995) model, all
existing style constructs fall into three traditions: cognition-centered,
personality-centered, and activity-centered. The styles that can be seen as
representatives in these three traditions are Witkin’s (1962) theory of field
dependence/independence (cognition-centered), Myers and McCaulley’s
(1988) theory of personality types (personality-centered), and Biggs’s
(1978) theory of learning approaches (activity-centered). Grigorenko and
Sternberg (1995; Sternberg, 1997) also pointed out that the studies on the
styles in each of these three traditions all have their own limitations. With
the purpose of overcoming the limitations in previous studies, Sternberg
(1988, 1997) proposed a theory of thinking styles––the theory of mental
self-government––which covers all of these three traditions. Sadler-Smith
(2009) proposed a duplex model of cognitive style where he classified two
basic modes of processing information: intuitive and analytic. He argued
that some people have relatively stable preference for one of the two modes
while some people are more flexible and are able to use either mode.
All of the five models mentioned above have made significant contributions
to the integration of the field’s massive style constructs. However, there are
still some limitations in these five models, such as the lack of empirical
support (apart from Riding and Cheema’s model), and the fact they place
too much emphasis on cognitive processes (e.g., Miller’s model, Riding and
Cheema’s model, and Sadler-Smith’s model). Furthermore, it has to be
pointed out that none of them explicitly provided reliable answers to the
major controversial issues in the field, such as the ones concerning style
malleability and the value of styles.
The sixth model that integrated works on styles is Zhang and Sternberg’s
(2005) threefold model of intellectual styles. It overcomes all of the
limitations mentioned in the previous five models. Therefore, Zhang and
Sternberg’s (2005) threefold model was selected to be the integrative style
model on which the present research is based.
The threefold model of intellectual styles (Zhang & Sternberg, 2005) was
established based on ten influential theories of styles: Biggs’s (1978)
learning approaches, Holland’s (1973) career personality types, Torrance’s
(1988) modes of thinking, Myers and McCaulley’s (1988) personality
types based on Jung’s (1923) work, Gregorc’s (1979) mind styles,
Kirton’s (1961, 1976) adaption-innovation decision making and problem
solving styles, Kagan and his colleagues’ (1964) reflective-impulsive
styles, Guilford’s (1950) divergent-convergent thinking, Witkin’s (1962)
4 Chapter One
field-dependence/independence, and Sternberg’s (1997) thinking styles. In

this model, intellectual styles, which are defined as people’s preference for
processing information and undertaking tasks, are used as an encompassing
term that includes various style constructs, such as cognitive styles, learning
styles, and thinking styles. Based on a meticulous review of various style
theories and their relevant empirical evidence, Zhang and Sternberg (2005)
found that style constructs included in the model of intellectual styles can
be generally reconceptualized into three types. Type I intellectual styles are
normally characterized by cognitive complexity, nonconformity, autonomy,
and low degrees of structure. Examples include the deep learning approach,
field independence, divergent thinking, reflectivity, innovation, holistic
thinking, and the legislative thinking style. In contrast, Type II intellectual
styles normally feature cognitive simplicity, conformity, authority, and high
degrees of structure. Examples include the surface learning approach, field
dependence, convergent thinking, impulsivity, adaptation, analytic
thinking, and the executive thinking style. Type III intellectual styles
“manifest the characteristics of both Type I and II styles, depending on the
stylistic demands of a specific task and on an individual’s level of interest
in the task” (Zhang & Sternberg, 2005, p. 36). Examples include the
achieving learning approach, integrative thinking, and the anarchic thinking
style. More details can be found in Appendix 1.
The Theory of Thinking Styles: Mental Self-Government

Among various models of styles, the model of thinking styles (Sternberg,
1988, 1997) (also known as the theory of mental self-government) is seen
as the most general one that adopts all of the three traditions (cognition-
centered, personality-centered, and activity-centered) (Zhang & Fan, 2011).
Furthermore, it is also the starting point for the establishment of the
threefold model of intellectual styles (Zhang & Sternberg, 2005). Therefore,
it is believed that great progress in clarifying the major controversial issues
in the field of intellectual styles can be made by studying thinking styles.
For this reason, this book selects the theory of mental self-government as its
theoretical foundation.
Taking “government management” as a metaphor, Sternberg (1988, 1997)
pointed out that, just as the government adopts various ways to manage the
society, people also have various ways to govern or manage their everyday
activities. These ways can be seen as their thinking styles. Sternberg (1988,
1997) proposed 13 thinking styles that fall along five dimensions: functions,
forms, levels, scopes, and leanings. The description of each thinking style
is presented in Table 1.1.
Table 1.1: Descriptions of the 13 Thinking Styles in the Theory of Mental

Self-Government
Thinking
Dimension Key Characteristics
Styles
One prefers to work on tasks that require creative
Function Legislative (I) strategies; one prefers to choose one's own
activities.
One prefers to work on tasks with clear
Executive (II) instructions and structures; one prefers to
implement tasks with established guidelines.
One prefers to work on tasks that allow for one's
Judicial (I) evaluation; one prefers to evaluate and judge the
performance of other people.
Hierarchical One prefers to distribute attention to several tasks
Form
(I) prioritized according to one's valuing of the tasks.
Monarchic One prefers to work on tasks that allow complete
(II) focus on one thing at a time.
One prefers to work on multiple tasks in the
Oligarchic
service of multiple objectives, without setting
(III)
priorities.
One prefers to work on tasks that allow flexibility
Anarchic (III)
as to what, where, when, and how one works.
One prefers to pay more attention to the overall
Level Global (I)
picture of an issue and to abstract ideas.
One prefers to work on tasks that require working
Local (II)
with concrete details.
One prefers to work on tasks that allow one to
Scope Internal (III)
work as an independent unit.
One refers to work on tasks that allow for
External (III)
collaborative ventures with other people.
One prefers to work on tasks that involve novelty
Leaning Liberal (I)
and ambiguity.
One prefers to work on tasks that allow one to
Conservative
adhere to the existing rules and procedures in
(II)
performing tasks.
Note. Extracted from Zhang, 2003, p. 630; I = Type I thinking style; II = Type II
thinking style; and III = Type III thinking style
Based on solid empirical evidence, Zhang (2002b) reconceptualized these

13 styles into three types, which later became the foundation for the
threefold model of intellectual styles. Similarly, Type I thinking styles are
6 Chapter One
characterized by creativity-generating and high levels of cognitive

complexity, including the legislative, judicial, hierarchical, global, and
liberal styles. Type II thinking styles denote a norm-favoring tendency and
low levels of cognitive complexity, including the executive, local, monarchic,
and conservative styles. Type III thinking styles are circumstantially
characterized by the features of either Type I styles or Type II styles,
including the anarchic, oligarchic, internal, and external styles.
The theory of thinking styles has been selected to be the theoretical

foundation of the present research because of the following five strengths.
First and most importantly, the styles in this theory cover all three traditions
(cognition-centered, personality-centered, and activity-centered) (Grigorenko
& Sternberg, 1995) in the study of styles (Sternberg, 1997; Zhang &
Sternberg, 2005, 2006). Second, thinking styles can be applied to both
academic and non-academic settings (Sternberg, 1988, 1997). Third, the
theory classifies styles based on five dimensions rather than merely the one
bipolar dimension references in most of the other style theories (Zhang &
Sternberg, 2005). Fourth, a profile of styles can be described for each
individual according to the theory of thinking styles, rather than merely a
delineation of a single style (Zhang & Sternberg, 2005). Fifth, as the
foundation for the threefold model of intellectual styles, this theory of
thinking styles provides a desirable window through which a deep
understanding of the nature of intellectual styles can be achieved.
Controversial Issues in the Field of Styles

Despite of the extensive theories of styles, the idea of styles per se seems to
intuitively appeal teachers, educators, and other practitioners. However, it
is surprising that many assumptions have not been fully examined by
empirical research. When teachers and other educational practitioners want
to enhance students’ performance by improving students’ styles, they are
confronted with three questions: 1) “Is it necessary to pay attention to
styles?”; 2) “Can styles be cultivated?”; and 3) “Is it more worthwhile to
cultivate some styles than others?”. These three questions pertain to the
three major controversial issues in the field of styles. The first one is about
the uniqueness of styles compared with other salient constructs in individual
psychology, such as personality. The second is whether styles can be
socialized/changed. The third one is whether styles are value-laden.
Is it necessary to pay attention to styles?

The field of styles has been criticized for the lack of reliable and valid
empirical evidence (Coffield, Moseley, Hall, & Ecclestone, 2004). Despite
the fact that the empirical evidence has been accumulating in recent
decades, there are still some obstacles when demonstrating the identity of
styles in the broader context of education and psychology. One of them is
distinguishing styles from other existing constructs of individual differences,
such as personality (Zhang, 2013). The debate over the relationship between
personality and styles began with the rise of style theories and is still
ongoing. Grigorenko and Sternberg (1995) contended that style is the
interface between intelligence and personality. Messick (1994) claimed that
styles are manifestations of personality. Some scholars also argue that styles
are subordinate constructs of personality (e.g., Furnham, Jackson, & Miller,
1999; Jackson & Lawty-Jones, 1996), while others insist that personality
and styles are distinct constructs and that each one makes unique
contributions to the understanding of individuality (e.g., Busato, Prins,
Elshout, & Hamaker, 1999; Larson, Rottinghaus, & Borgen, 2002; Riding
& Wigley, 1997; Zhang, 2003c, 2006b). However, some scholars are still
skeptical about the uniqueness of styles beyond personality (e.g., Daniel von
Wittich, 2011). The controversy of the relationship between personality and
styles is partially due to the large number of style constructs without a
consensus for their definition, and also due to the lack of rigorous studies
that examine this issue. Empirically, the relationship between personality
and styles can be systematically examined from three perspectives.
First, how do styles and personality overlap? With regard to this aspect,
many studies have been undertaken to examine the correlations between
personality and styles or the percentages of the variance in styles that can
be explained by personality (e.g., Chamorro-Premuzic & Furnham, 2009;
Kwang & Rodrigues, 2002; Larson et al., 2002; Zhang, 2006b). The extent
of the overlap between personality and styles found in these studies vary
depending on the different personality constructs and style models adopted.
However, apart from several dimensions, such as the extraversion-
introversion dimension in Myers and McCaulley’s (1988) personality types
(one model of intellectual styles), the correlations found between
personality and styles are moderate at most (see Chapter 2 for more detail),
which means that the overlap of personality and styles is limited.
Second, do styles have a unique explanatory power for individual outcomes

beyond personality? In this research area, there are also a few studies that
predicted a third variable for both personality and styles simultaneously
8 Chapter One
(e.g., Chamorro-Premuzic & Furnham, 2008; Chamorro-Premuzic, Furnham,

& Lewis, 2007; Furnham, Crump, Batey, & Chamorro-Premuzic, 2009).
Most of these studies indicated that intellectual styles had made unique
contributions to individual outcomes beyond personality to different
degrees (see Chapter 2). However, the style constructs and outcome
variables covered in this part of research are limited; therefore, more
research is needed to consolidate the findings on the unique contributions
of styles beyond personality.
Third, another difference between styles and personality that scholars argue
about is the question of malleability. Some scholars believe that styles are
more changeable than personality (e.g., Chamorro-Premuzic & Furnham,
2009; Zhang & Sternberg, 2005). However, there is a lack of empirical
studies that directly examine this argument.
In summary, as Armstrong and his colleagues (2012) pointed out, the

clarification of the relationships between styles and other individual
constructs, such as personality, would be beneficial to the advancement of
the field of styles. However, the consensus about whether styles are distinct
from personality has not been reached. Based on the above limitations, an
examination of the relationship between styles and personality from the
three perspectives (i.e., overlap, their unique contributions to other
variables, and a comparison of the malleability) simultaneously is needed to
help to draw a more compelling conclusion. Therefore, in the research
introduced in this book, all of the three perspectives have been adopted so
that the relationship between personality traits and thinking styles can be
examined in a more comprehensive manner.
Can styles be cultivated?

Nature versus nurture is a classic controversy in the field of developmental
psychology. It also seems to be an unavoidable issue when examining the
development of styles, which has caused many heated debates. Can styles
be socialized or modified? If so, how? If the answers to these questions can
be provided with convincing evidence, it will undoubtedly greatly
contribute to the optimization of people’s style development, and thus
enhancing their performances. In contrast, if styles are not malleable, efforts
to study them would be pointless for educational practice (Zhang, 2013).
However, the issue of style socialization is still one of the most controversial
in the field (Zhang & Sternberg, 2006). Some scholars consider styles to be
a product of nature. For example, Riding and Rayner (1998) argue that
styles are inborn characteristics and relatively static. However, more

scholars believe that styles are a joint function of nature and nurture, and
that they can be socialized. For example, Sternberg (1997) pointed out that
styles are malleable to some extent. Mandelman and Grigorenko (2012) also
inferred from the effect of genes on intelligence and personality that its
effect on styles is likely to be much less than 50%, which implies that the
effect of socialization on styles might be even larger than that of genetic
factors. The results from empirical studies have also been inconsistent.
Nevertheless, after a careful review of previous empirical evidence, it was
found that the view that styles can be socialized seems to be much stronger.
There are three main lines of research that have contributed to the
socialization issue of styles. The first line examines style differences based
on certain personal factors (e.g., age, gender, socioeconomic status, and
personal experience, etc.). The second line explores the relationships
between certain environmental factors (e.g., disciplines, culture, and work
environment, etc.) and styles. The third line consists of longitudinal studies
that mostly examine the effects of training (e.g., program, course, and
therapy, etc.) on styles. The details of these studies are described below.
The first line: Style differences based on personal factors
The most widely examined socialization factors in the field of styles are age
and gender. In Zhang and Sternberg’s (2006) and Fer’s (2012) reviews,
although there are some inconsistent results from empirical studies, they
found that most of them had concluded that age and gender play an important
role in the development of styles. For example, Jonassen and Grabowski
(1993) found the developing curve of field dependence/independence. Style
differences based on other theoretical frameworks were also found to be a
function of age, such as Biggs’s (1978) learning approaches (e.g., Hilliard,
1995; Richardson, 1995; E. Sadler-Smith, 1996), Kirton’s (1961,1976)
adaption-innovation decision making and problem solving styles (e.g.,
Hayward & Everett, 1983; C. M. Jacobson, 1993), and Sternberg’s (1988,
1997) thinking styles (e.g., Zhang, 1999, 2004d; Zhang & He, 2003). By the
same token, gender differences in styles are also found in many kinds of
styles (see the review in Zhang and Sternberg, 2006; Fer, 2012). Though
there were different findings on the specific relationships between gender
and particular styles, it was consistently proved that gender plays an
important role in the development of styles. Zhang and Sternberg (2006)
even argued that the inconsistent results on the gender differences can also
be seen as evidence of socialization, because it reflects the influence of
different cultures on gender stereotypes and socialization.
10 Chapter One
Another important socialization factor is socioeconomic status. For

example, Cakan (2003) found that students achieved higher scores on field
independence if their fathers have a higher education level. Studies on
thinking styles also found that a higher socioeconomic status (SES) was
positively related to higher cognitive complexity and more creative styles
(Ho, 1998; Sternberg & Grigorenko, 1995; Tse, 2003; Yang & Lin, 2004;
Zhang & Postiglione, 2001). Zhang and Sternberg (2006) explained that
students from higher SES families have more opportunities to experience
different issues and situations, and therefore their previous cognitive
structures have more opportunities to be chanllenged, which could enable
students to develop more complex and creative styles. This argument is
supported by empirical studies, which directly examine the relationship
between styles and personal experience.
The style studies on personal experience include three major aspects. First,
style differences related to education or training have already been studied.
For example, In Fer’s (2012) review of the relationships between
educational level and intellectual styles, she found that people with higher
levels of education tended to have more creativity-generating styles (Type
I) than their counterparts. Lee (2002) conducted a study among teachers and
found differences in their thinking styles by comparing teachers with and
without professional training. Second, in addition to learning experience,
studies on non-academic settings and styles show that work experience
plays an important role in the socialization of individuals’ styles. Examples
include the length of work experience (e.g., P. A. Holland, Bowskill, &
Bailey, 1991; Sternberg & Grigorenko, 1995; Zhang & Sachs, 1997), work
positions (e.g., R. T. Keller & Holland, 1978a; Zhang & Higgins, 2008),
and job functions (e.g., Foxall, Payne, & Walters, 1992; Kirton & Pender,
1982).
All of these studies on the relationship between styles and personal

experience indirectly reflect the role of environment in the socialization of
styles. In addition, studies that infer the influence of environmental factors
more directly have also been conducted.
The second line: Style differences based on environmental factors
The most widely studied factor that has been used to infer environmental
effects on styles is people’s disciplines. Student styles vary based on
different majors (e.g., Bin, 2009; Fer, 2007; Jones, Reichard, & Mokhtari,
2003; Skogsberg & Clump, 2003; S. N. Smith & Miller, 2005; Tucker,
1999) or different specialized colleges (e.g., Barnhart, 2003; Mitchell &
Cahill, 2005); teachers’ styles vary based on different academic disciplines

or the subjects that they teach (e.g., Lam, 2000; Sternberg & Grigorenko,
1995; Zhang & Sachs, 1997); and people’s styles who work in non-
academic settings vary based on the type of jobs (e.g., Foxall et al., 1992;
Hommerding, 2003; Kaufman, 2001; Kirton & Pender, 1982). For example,
students majoring in psychology were found to use the deep learning
approach more often than those majoring in biology (Skogsberg & Clump,
2003). Zhang and Sachs (1997) found that teachers who taught science and
technology tended to use the global and legislative styles more often than
those who taught social sciences and humanities. In addition, Kaufman
(2001) also found significant style differences between writers and
journalists. Zhang and Sternberg (2005, 2006) believe that these findings
support the fact that different environments shape and facilitate the
development of different styles. However, besides the environmental effect,
it has to be admitted that there is another alternative explaination for these
style differences based on disciplines, which is that people tend to choose
the disciplines that are in line with their style disposition. Torbit’s (1981)
study provided further evidence for the former argument. He found that,
although students who wanted to be counselors had more divergent learning
styles, this tendency became stronger after they entered the counseling
discipline. This suggests that the role of environment in shaping people’s
styles cannot be overlooked.
Furthermore, the style studies on some of the environments that people

choose less voluntarily than disciplines (e.g., school bands and residential
locations) provided more convincing envidence on the socialization effect
of environment on styles. For example, the style differences found between
students from different bands of secondary schools (Cheung, 2002) and
students from different types of universities (Fer, 2007) implied the role of
school culture in student development of styles. On a macro level, the styles
of people from different residential locations were also compared. For
instance, style differences have been found between rural and urban regions
(e.g., Verma, 2001; Wu & Zhang, 1999), between different cities (Zhang &
Postiglione, 2005), and different countries or nationalities (e.g., Bagley &
Mallick, 1998; Foxall, 1990; N. F. Skinner, Hutchinson, Lukenda, Drake,
& Boucher, 2003; Tang, 2004; Volet, Renshaw, & Tietzel, 1994; You & Jia,
2008). For example, Skinner and his colleagues (2003) showed that
Canadian undergraduates seem to be more adaptive than their American and
British peers when Kirton’s (1961, 1976) adaption-innovation styles were
examined. Similarily, significant differences in learning approaches were
also found between Chinese and American pre-service teachers (You & Jia,
2008). Tang’s (2004) study also revealed that Chinese Canadians had higher
12 Chapter One
scores in the legislative, executive, local, and liberal thinking styles than
people in Hong Kong. All of these studies supported the important role of
culture in the socialization of people’s intellectual styles.
However, the studies examining environmental effect based on the above

factors (e.g., disciplines, schools, and countries) are still providing relatively
indirect envidence because they do not reveal what specific features or
dimensions of environments were related to the development of styles.
Recently, more studies seem to take a much closer look at the role of
environmental factors in the socialization of styles, as they directly examine
specific features in environments and their roles in the development of
specific intellectual styles.
In academic settings, most studies explore the relationship between specific

dimensions of learning environments and learning approaches. The type
of assessment is one of the most examined dimensions in learning
environment (e.g., Gijbels & Dochy, 2006; Gordon & Debus, 2002;
Gulikers, Kester, Kirschner, & Bastiaens, 2008; Seddon, 2008; Segers,
Gijbels, & Thurlings, 2008; S. N. Smith & Miller, 2005; Tian, 2007).
However, the results are not consistent, although researchers all predicted
that the deep learning approach would be improved by the adopted
assessements (see Chapter 3 for more details). Some scholars argue that it
is the way that assessment is implemented during students’ learning process
rather than how it is designed that influences the development of different
learning approaches (Segers et al., 2008). Besides the type of assessment,
the varying nature of learning environments, such as constructivist,
cooperative, problem-based, and action learning based, has also been
examined (e.g., Gijbels, Segers, & Struyf, 2008; Groves, 2005; Klinger,
2006; McParland, Noble, & Livingston, 2004; K. Wilson & Fowler, 2005).
The results from these studies are as complicated as the ones found in the
studies of the relationships between learning approaches and assessments.
Some of these studies found that students’ learning approaches became
deeper, which met the researchers’ expectations (e.g., K. Wilson & Fowler,
2005), while some of them found that a more constructivist or problem-
based environment led to a more surface learning approach, which was
contrary to their predictions (e.g., Gijbels et al., 2008; Groves, 2005) (see
Chapter 3 for more details). However, regardless of the different directions
of style change in these studies, it has to be admitted that learning
approaches can be modified by different learning environments. More
details of the empirical research on the relationships between learning
environments and intellectual styles can be found in Chapter 3.
Besides learning environments, a few features in work environments have

also been found to be related to one’s development of thinking styles. For
example, the degrees of freedom teachers perceived in their work
environments have been found to be associated with their thinking styles
(Zhang & Sternberg, 2002). Lee’s (2002) study on kindergarten teachers
also confirmed the role of perceptions of work environment in teachers’
development of styles. There is less style research on family settings than
work enviroments. For example, in terms of thinking styles, there is only
one study relevant to the family environment. Zhang’s (2003a) study found
that parents and students’ thinking styles are significantly related. However,
to what extent the socialization effect or genetic effect contribute to this
relationship needs to be further explored. Therefore, more style studies in a
family environment are needed to deepen the understanding of styles’
socialization process.
The third line: Longitudinal studies on style malleability
In fact, the most direct and most convincing pathway to examine the issue
of style malleability is via a longitudinal study. However, there has been
limited attention paid to this type of study.
Among the existing longitudinal studies on styles, two types can be

identified. The first type is to examine style differences based entirely on
the passage of time, which means that the researchers examine the natural
changes in styles over a certain period of time. For example, Wilding and
Andrews (2006) followed undergraduate students for one year and found
that they adopted more surface learning approaches after this time. A ten-
year qualitative study on a family (Helwig & Myrin, 1997) also showed that
Holland’s personality types can be changed as a function of certain familial
movement and development. Volet and his colleagues (1994) found that
South-east Asian students’ learning approaches became similar to
Australian students after one semester’s study in Australia.
The second type of longitudinal studies examines style changes after

certain interventions (e.g., programs, courses, training, and therapies). In
terms of learning approaches, most longitudinal studies are related to the
programs or courses implemented in schools and universities, which were
intended to promote the deep learning approach (e.g., Baeten, Dochy, &
Struyven, 2008; Balasooriya, Hughes, & Toohey, 2009; Gijbels et al., 2008;
Groves, 2005; McParland et al., 2004). Interestingly, many of these studies
found that students did not become more deep-oriented as expected (e.g.,
McParland et al., 2004), and some studies even showed they became more
14 Chapter One
surface-oriented (e.g., Baeten et al., 2008; Gijbels et al., 2008; Groves,

2005). These unexpected results might be attributable to the way in which
these programs and courses were actualized in practice. Despite the
inconsistent results, these studies supported the idea that learning
approaches can be modified. The longitudinal studies that use other style
inventories are more various in terms of their types of interventions. Taking
research on field dependence/independence, for example, studies showed
that both yoga and meditation could enhance individuals’ field
independence (Dillbeck, Assimakis, Raimondi, & Orme-Johnson, 1986;
Rani & Rao, 2000; Sridevi & Rao, 2003; Sridevi, Sitamma, & Rao, 1995).
Moreover, training relevant to spatial tasks, formal operations, coping with
stress, and even chess instruction were also proved to lead to more
independent styles (Collings, 1985; Lapidus, Shin, & Hutton, 2001; J. P.
Smith & Cage, 2000; Stericker & LeVesconte, 1982). O’Leary, Donovan,
and Kasner’s (1975) study even found that treatment for alcoholics
improved the participants’ scores on field independence. Besides, as the
Matching Familiar Figures Test for reflective-impulsive styles was
frequently used as a criterion of attention-deficit hyperactivity disorder and
some other psychiatric and cognitive diseases during 1970s and 1980s,
abundant studies showed that some therapies or treatments can help patients
to be less impulsive (e.g., Brown, 1980, 1986; N. J. Cohen, 1981; Feindler,
Ecton, Kingsley, & Dubey, 1986; Holmes, 1981; Kendall & Finch, 1978).
In summary, the research findings identified in the literature so far have

provided more support for the argument that styles can be socialized, but
they are still not strong enough to provide specific suggestions in practice
because many research gaps remain. First, there are not many studies that
directly explore what specific features or dimensions of environments are
related to the development of styles. Most of the studies simply compared
the styles of students from different majors, schools, hometowns, and
cultures. Second, even among the limited studies that directly explore
specific features or dimensions of environments, the results are still not very
consistent, which indicates a need to examine them more closely and
carefully in the future. Third, another important socialization setting, the
family environment, seems to have been ignored compared with the school
and work environments. Fourth, there are also a limited number of studies
that that examine different environmental and personal factors
simultaneously to compare the extent of these socialization factors’
influence. Fifth, though longitudinal studies can be seen as a very effective
way to examine the issue of style malleability, only some style constructs
have been examined using this method. In other words, longitudinal studies
are still sparse with regard to some other style constructs, such as thinking
styles. Based on all of the above considerations, this book’s research

examines the effect of specific environmental factors on styles with more
elaborate concerns; it also pays attention to the family environment.
Meanwhile, these environmental factors are examined simultaneously with
personal factors to achieve a comprehensive understanding of the
socialization process of styles. In addition, longitudinal design has been
adopted.
Is it more worthwhile to cultivate some styles than others?

Like the issue of the socialization of styles, the controversial issue of style
value has also been hotly debated in the field. Are some styles more adaptive
or desirable than others? Some scholars (e.g., Kwang & Rodrigues, 2002)
have argued that style value depends on the different contexts where styles
are used. Other scholars (e.g., Zhang & Sternberg, 2009) believe that some
styles are laden with more positive value than others. In this section, studies
contributing to the issue of style value are reviewed along with four bodies
of literature. The first one is about studies that examine the relationships
between styles and individual variables related to cognitive development.
The second is about research that explores the relationships between styles
and individual variables related to social and emotional development. The
third is related to teacher-student interaction in educational settings. The
fourth is about evidence from style training programs.
The first line: The relationships between styles and cognitive

development
In Zhang and Sternberg’s (2006, 2009) comprehensive review of the studies

related to the value of styles, the results from the research on the
relationships of styles to abilities, creativity, cognitive development,
giftedness, and academic performance all support the idea that styles are
value laden. For example, students with higher scores in self-rated abilities,
higher levels of cognitive development, better performance in creativity and
critical thinking measures, or who are seen as gifted tended to have higher
scores in Type I styles (i.e., styles that are characterized by low degrees of
structure, cognitive complexity, nonconformity, and autonomy, such as the
legislative thinking style, deep learning approach, field independent, and the
holistic style). Meanwhile, students with lower scores in self-rated abilities,
lower levels of cognitive development, and lower performance in creativity
and critical thinking measures, or students who are seen as non-gifted tend
to have higher scores in Type II styles (i.e., styles that are characterized by
high degrees of structure, cognitive simplicity, conformity, and authority,
16 Chapter One
such as the local thinking style; the surface learning approach; field
dependence; and the analytic style) (e.g., Alborzi & Ostovar, 2007;
Dollinger, Palaskonis, & Pearson, 2004; Kim & Michael, 1995; A. L.
Miller, 2007; S.-K. Park, Park, & Choe, 2005; Torrance & Reynolds, 1978;
Yang & Lin, 2004; Zhang, 2002b, 2003b, 2004c, 2005a).
However, most of the studies that examined the relationships between styles
and cognitive outcomes directly focused on academic achievement and task
performance. The results from these studies seem more complicated than
the ones from studies on abilities, creativity, cognitive development, and
gifted students mentioned above. Some studies found that students with
Type I styles outperformed students with Type II styles. For example, field-
independent (Type I style) students were found to have better levels of
academic achievement (e.g., Bagley & Mallick, 1998; Cameron & Dwyer,
2005; Hite, 2004), problem-solving skills (Williams, 2001), and programming
performances (Johnson & Kane, 1992; D. J. Wilson, Mundy-Castle, &
Sibanda, 1990). Huang and Chao’s (2000) study also showed that students
with learning disabilities had higher scores in field dependence than those
without learning disabilities. The results from some studies on other style
constructs, such as Biggs’ (1978) learning approaches, Myers and
McCaulley’s (1988) personality types, Kagan and his colleagues’ (1964)
reflective-impulsive styles, and Sternberg’s (1997) thinking styles also
support the belief that Type I styles have more positive value in terms of
academic achievement and task performance, which resonates with the
findings on field-dependence/independence (Chamorro-Premuzic &
Furnham, 2008; Fisher, 1994; Furnham, Christopher, Garwood, & Martin,
2007; Nachmias & Shany, 2002; Stahl, Erickson, & Rayman, 1986;
Sternberg & Grigorenko, 1993). Even in non-academic settings, the results
on the positive relationship between Type I styles and task performance
have been also found among reflective-impulsive styles (Sayles-Folks &
Harrison, 1989), field-dependence/independence (Wegner, 1980), and
adaption-innovation styles (Foxall & Hackett, 1992; Gelade, 1995).
However, other studies found inconsistent results. Sometimes, students with

Type II styles outperform those with Type I styles. Thinking style studies in
Hong Kong, the Philippines, and Spain have found that some Type II
thinking styles (e.g., the conservative, executive, and local styles) are
positively related to academic achievement (Bernardo, Zhang, & Callueng,
2002; Cano-Garcia & Hughes, 2000; Cheung, 2002; Zhang, 2001b, 2001c,
2004d; Zhang & Sternberg, 1998). Some scholars also found that, in some
disciplines (e.g., social sciences, literature, chemistry, and geography),
students with Type II styles (e.g., field-dependent, surface learning
approach) had higher scores in academic tests (Varma & Thakur, 1992;
Zhang, 2000). Kern and Matta (1988) found that when students were tested
on their performance on software use, students with a sensing (Type II)
preference did better than students with an intuition (Type I) preference. In
a traditional learning mode, the analytic style (Type II) was also found to be
positively related to academic performance (Liao & Chuang, 2007).
Zhang and Sternberg (2006) argued that these inconsistent results were due
to the different cultures, disciplines, learning materials, tasks and
assessments, and school levels involved in the studies. In spite of this, all of
them have still proved that styles are value laden rather than value free.
The second line: The relationship of styles to social and emotional

development
Besides cognitive development, social and emotional development (e.g.,

personality, emotion management, self-esteem, identity, and well-being) is
also an important aspect of student development. Unlike research on the
relationships between styles and academic achievement, the research related
to the relationships between styles and social and emotional development is
limited, but the existing findings seem to be more consistent.
Most studies in this field focus on the relationship between styles and
personality. Type I styles have been found to be more related to adaptive
personality traits, such as openness and conscientiousness (Busato et al.,
1999; Chamorro-Premuzic et al., 2007; Shokri, Kadivar, Farzad, & Sangari,
2007; Zhang, 2002a, 2002d; Zhang & Huang, 2001). As for other aspects
of personality in a much wider sense (such as moral, emotional, and
psychosocial aspects), Type I styles have also been proved to be more
related to certain developmental outcomes that always are seen as positive
attributes. For example, field independence (Type I), reflectivity (Type I)
and Type I thinking styles were associated with higher levels of moral
maturity (S.-K. Park et al., 2005; Schleifer & Douglas, 1973). Field
independent people also exhibit better control of their behaviors (Leventhal
& Sisco, 1996). Moreover, the innovative style (Type I) was found to be
negatively related to pessimism and positively related to tolerance of
ambiguity (R. T. Keller & Holland, 1978b; Wunderley, Reddy, & Dember,
1998). People with the innovative style and Type I thinking styles also had
more positive attitudes towards new technologies (Chakraborty, Hu, & Cui,
2008; Zhang, 2007c; Zhang & He, 2003). Meanwhile, Type I thinking styles
were also found to be positively related to a more adaptive type of academic
motivation (i.e., motivation to approach success) (W. Fan & Zhang, 2009).
18 Chapter One
Zhang (2004a, 2008b, 2008c, 2010) conducted a series of studies that

examined the role of thinking styles in individual psychosocial development
and also found that students with Type I styles had better psychosocial
development.
Another important field in individual social and emotional development is

one’s concept or image of themselves, which is also often seen as an aspect
of personality, such as self-confidence, self-esteem, and self-efficacy.
Research on the relationship between styles and the self-related variables
mentioned above all supported the fact that Type I styles (e.g., field
independence, the deep learning approach, innovation, the legislative style,
and the liberal style) are laden with positive value in terms of these self-
related constructs (e.g., Bosacki, Innerd, & Towson, 1997; Phan, 2007;
Tierney, 1997; Watkins & Dahlin, 1997; Zhang, 2001d; Zhang & Postiglione,
2001).
However, personality and the self-concepts mentioned above have

limitations as indicators for the style value because they can be easily seen
as antecedents of styles rather than outcomes. Compared with personality
and self-concepts, well-being can be seen as a more conclusive outcome and
provides more powerful evidence on the value issue of intellectual styles.
The first aspect related to well-being is mental health. Chen and Zhang
(2008) found that Type I thinking styles (such as the hierarchical and liberal
styles) are positively related to mental health while Type II thinking styles
(such as the local and monarchic styles) are negatively related to it. Zhang’s
(2009) study also found that students with Type I thinking styles have lower
scores on anxiety. Research conducted among patients with mental diseases
also echoed the above results. For example, patients with depression were
found to be more field-dependent (Type II) (Calamari, Pini, & Puleggio,
2000), those with borderline personality disorder tended to favor the impulsive
style (Type II), and those with Attention Deficit and Hyperactivity Disorder
were found more impulsive and more field-dependent (López-Villalobos et
al., 2003). The second aspect related to well-being is life satisfaction (i.e.,
satisfaction with one’s general life) and domain satisfaction (i.e., satisfaction
with specific domains of one’s life), which also seem to be positively related
to Type I styles. For example, field-independent people have higher levels
of job satisfaction (Hageman, 1990), while teachers with Type I teaching
styles and adults with Type I thinking styles in other workplaces have more
positive perceptions of work environments (Lee, 2002; Zhang, 2005b;
Zhang & Higgins, 2008).
It has to be admitted that there are still a few studies on social and emotional
development that have inconsistent results, though they are much less
common than the studies on academic achievement. For example, the
holistic style (Type I) was found to be positively related to anxiety in Jong,
Merckelbach, and Nijman’s (1995) study, and the analytic style (Type II)
was found to be associated with better self-control in Gadzella’s (1990)
study. However, according to the above review, most studies on the
relationship of styles with social and emotional development support the
argument that styles are value-laden and Type I styles are more adaptive
than Type II.
The third line: Teacher-student interaction in educational settings
The value issue of intellectual styles can also be inferred from the research
on styles and teacher-student interactions in educational settings. First,
teachers with Type I styles are more likely to have effective teaching
behaviors. For example, it was found that teachers with Type I thinking
styles tended to exhibit more student-centered teaching behaviors and use
more beneficial humor styles in teaching, while teachers with Type II
thinking styles tended to exhibit more teacher-centered teaching behaviors
and use more detrimental humor styles (G. H. Chen, 2007; Lee, 2002; D.
M. Yu, 2008; Zhang, 2001a). Similar results were also found in research on
other style constructs. For instance, more field independent (Type I) and
more reflective (Type I) teachers have more positive attitudes to teaching
and students, and also display teaching behaviors which are normally
considered to be beneficial for students, such as more encouragement, more
nurturing, more democracy, and less criticism (see the review in Zhang &
Sternberg, 2006). Meanwhile, effective teachers’ MBTI profiles always
included intuition (Type I) rather than sensing (Type II) (Mills, 2003;
Rushton, Morgan, & Richard, 2007). Second, teachers prefer students with
Type I styles (Zhang, Fu, & Jiao, 2008). Third, students also prefer teachers
with Type I styles (Zhang, 2004e, 2006a, 2008a; Zhang, Huang, & Zhang,
2005) and tend to have better course satisfaction if their teachers use Type
I styles (Betoret, 2007; G. H. Chen, 2007). All of the evidence indicates that
Type I styles are more valued than Type II styles in educational settings,
although in some situations Type II styles are probably encouraged by
academic assessments in a practical sense (as previously discussed).
The fourth line: Style training programs
It is noticeable that most style training programs aim to help people develop
certain types of styles rather than others. Type I styles, such as field
20 Chapter One
independence, reflective styles, deep learning approaches, divergent

thinking, and a holistic mode of thinking, are always the styles that these
programs aim to foster (see the review in Zhang & Sternberg, 2006). It is
logical that when people think some types of styles are more adaptive than
others, then they will make efforts to cultivate them. With respect to
thinking styles, there are also two studies that had the purpose of training
Type I styles. The first is Fan’s (2006) study that aimed to facilitate Type I
styles through a hypermedia instructional environment relative to a
traditional instructional environment. Although the results did not completely
support this prediction, it at least showed that Type I thinking styles were
the ones that people prefer to cultivate. The other is Zhang’s (2005) study
where teachers were trained to teach in the legislative condition, the
executive condition, the judicial condition, and the combined condition (a
combination of aforementioned three styles). It was found that teaching in
the combined condition or the legislative condition (Type I styles) benefited
student achievement more than the others, which partially supported the
idea that styles are value laden.
In summary, though some styles are value-differentiated depending on

different contexts, most intellectual styles seem value-laden in almost all
situations, with Type I styles being more adaptive and desirable than Type
II. However, there are still some limitations. First, these inconsistent results
indicate that the value of some styles varies across certain specific contexts.
Second, most variables that have been examined with styles to infer their
value are not persuasive enough to be treated as developmental outcomes
(e.g., personality and self-esteem). Third, the developmental outcomes
examined to explore the issue of style value are still within a narrow range.
Although most research to date has focused on the role of styles in students’
academic achievement, it is believed that paying attention to styles’
contribution to students’ emotional and career development is equally
important (Zhang, 2013). Therefore, in the research depicted in this book,
career decision self-efficacy as well as subjective well-being were examined
in order to expand the range of developmental outcomes studied in style
research and provide more powerful evidence to further clarify the issue of
the value of intellectual styles.
About This Book

This book is aimed at addressing the major three controversial issues (i.e.,
style identity, style malleability, and style value) in the field of styles
through the lens of thinking styles. It is rooted in the empirical findings from
the author’s PhD research project, which utilizes a quantitatively-driven

mixed research methods design. In doing so, this book provides some
answers to current research gaps in literature (as previously discussed). For
example, no research has systematically investigated the relationships of
students’ perceived parenting styles and perceived learning environments to
their thinking styles. More importantly, the present research has a new
design which simultaneously examines these two environmental factors
with personality traits to explore their competing influences on students’
thinking styles using a longitudinal approach. In addition, there has been no
research conducted that explores the relationships of student thinking styles
to their career decision self-efficacy and subjective well-being, and which
also examines the possible mediating function of thinking styles in the
relationships of parenting styles, learning environments, and personality
traits to particular developmental outcomes for students (i.e., career decision
self-efficacy and subjective well-being). Moreover, no research has directly
compared the malleability of thinking styles and that of personality traits.
Through filling the above research gaps, this book offers valuable
information on these controversial issues and facilitates a better
understanding of the nature of thinking styles. In addition to the theoretical
significance, the most important practical application of this present
research lies in its implications for parenting and education. If the research
findings reveal that some styles have more positive adaptive value than
others in terms of student development, and also indicate that some
characteristics in parenting styles and learning environments can promote
specific thinking styles, then reasonable suggestions can be provided to
parents, teachers, school administrators, and educators about effective
strategies that optimize the development of student thinking styles.
In the following three chapters (i.e., Chapter 2, 3, and 4), each of the three
controversial issues will be discussed with a review of previous empirical
evidence in literature, as well as the major findings from the present research
project. In Chapter 5, all of the findings relevant to the three issues are
summarized and further discussed within the broader context of intellectual
styles. The final chapter highlights the conclusions, as well as discusses
their contributions and implications for theories, research, and educational
practice.
In order to help readers better understand the research findings presented in

the subsequent chapters, I would like to introduce the project’s research
methods before they move on to next chapters. The whole research
comprises three phases. The first phase was the pilot study. The second was
the main study (a quantitative longitudinal study). Finally, the third study
22 Chapter One
was a follow-up qualitative study. The pilot study mainly aimed at

validating the inventories that would be used in the main study. For the
purpose of brevity, this part is not included in this book. The research
methods adopted in the main study and the follow-up study are presented
below.
Main study: The quantitative, longitudinal study

The purpose of the main study was to examine the following questions:
x What is the relationship between personality traits and thinking styles?
x Are thinking styles changeable?
x What are the roles of students’ perceived parenting styles, perceived
learning environments, and personality traits in predicting their thinking
styles? Which one is the strongest predictor for the students’ thinking
styles?
x What are the roles of thinking styles in predicting the students’ specific
developmental outcomes (i.e., career decision self-efficacy and
subjective well-being)?
x Do students’ thinking styles mediate the relationships of perceived
parenting styles, perceived learning environments, and personality traits
to their developmental outcomes (i.e., career decision self-efficacy and
subjective well-being)?
This study is a longitudinal study. Participants were asked to respond twice

to the questionnaires that assess the key variables in the present research
with a one-year interval. All of the above questions were examined in the
control of relevant demographic factors, such as gender, grade, academic
major, hometown, and socioeconomic status.
Participants
926 students from a comprehensive university in Shanghai, China (the same

institution in the pilot study) participated in the first wave of data collection
(Time 1). In the second wave of data collection (Time 2), 743 out of the 926
students participated. The distribution of the sample is shown in Table 1.2.
The age of the participants at the beginning of the main study ranged from
18 to 22 (Mean = 18.86, SD = 1.21). The results from the t-tests on the
differences between students who participated in both data collections and
the ones who dropped out showed that the students who participated in both
data collections had higher scores on agreeableness (t = 3.11, p<.01) and
conscientiousness (t = 2.15, p<.05), and lower scores on negative affect (t =
-1.99, p<.05) and self-efficacy on career goal selection (t = -2.29, p<.05)
than the dropouts. The results of the main study should be understood in the
light of these group differences.
Table 1.2: The Distribution of the Sample in the Main Study
Time 1 Time 2
Grade freshmen 609 504
juniors 317 239
missing 0 0
Major humanities and social sciences 282 224
science and technology 643 518
missing 1 1
Gender male 520 414
female 405 328
missing 1 1
Total 926 743
Measures
Seven inventories were adopted in this study: a) the Parenting Style Index
(Steinberg, Lamborn, Darling, Mounts, & Dornbusch, 1994) for assessing
perceived parenting styles; b) the Inventory of Students’ Perceived Learning
Environment (self-designed) for assessing the perceived learning
environment; c) the NEO Five-Factor Inventory-3 (McCrae & Costa, 2007)
for assessing students’ personality traits; d) the Thinking Style Inventory-
Revised II (Sternberg, Wagner, & Zhang, 2007) for assessing thinking
styles; e) Career Decision-Making Self-Efficacy Scale––Short Form (Betz,
Klein, & Taylor, 1996) for assessing career decision self-efficacy; f) the
Satisfaction With Life Scale (E. Diener, Emmons, Larsen, & Griffin, 1985)
and g) the Positive and Negative Affect Schedule (Watson, Clark, &
Tellegen, 1988) for assessing subjective well-being. All of the measures
have been validated in the research. More details about each measure can
be found in Appendix 2 and Appendix 3.
Procedures
At the beginning of the academic year, all the participants were asked to
respond to the questionnaire, which include all the inventories mentioned
above and some necessary demographic information (including gender, age,
grade, major, hometown, and socioeconomic status) in class in the presence
24 Chapter One
of the researcher and their teachers. It should be noted that at the time of the
data collection, the freshmen were asked to recall their experience in high
school when they responded to the Inventory of Students’ Perceived
Learning Environment, while the juniors were asked to recall their
experience in university. In addition, all the participants were asked to recall
their experiences with their parents during the first 16 years of their lives
when they responded to the Parenting Style Index.
One year after the first data collection, the participants were asked to
respond again to the same questionnaire, which include all above
inventories and some necessary demographic information. It should be
noted that, at the time of the data collection, all of them were asked to
respond to the Inventory of Students’ Perceived Learning Environment
according to their experience in university during the period between the
two data collections. They were also asked to respond to the Parenting Style
Index according to their experience with their parents in the period between
the two data collections.
Follow-up study: The qualitative study

The qualitative study was conducted based on the findings from the
quantitative study. If the main study found that thinking styles were
changeable and could be influenced by environmental factors, then the
purpose of the qualitative study would be to further explore how thinking
styles change, especially how environmental factors (i.e., perceived
parenting styles and learning environments) together with personal
characteristics (i.e., personality traits) influence students’ thinking styles
and how students’ thinking styles influence their career decision self-
efficacy and subjective well-being.
Participants
Twenty-nine participants were selected from students who had taken part in
both of the main study’s data collections and had been found to have
significant changes in their thinking styles over the course of a year
according to the results of the Reliable Change Index, which is a statistical
technique that can detect significant individual change by excluding
measure errors.
Interview Protocol
The interview protocol consisted of three parts. The first part included
questions for ice-breaking, where the interviewer attempted to obtain the
interviewee’ trust and provide feedback on the interviewee’s thinking styles

based on the main study’s results. The second part included questions about
the socialization of thinking styles, where the issue of how and why the
interviewee’s thinking styles changed was discussed. The third part
included questions concerning the value of thinking styles, where the issue
of how thinking styles influence career decision self-efficacy and subjective
well-being was discussed. The details about the interview protocol are
presented in Appendix 4.
Procedures
Individual interviews were conducted in this study. Each interview lasted

about 60 minutes and was audio-recorded. At the beginning of each
interview, the researcher provided the feedback report on interviewees’
thinking styles based on results from the main study, and discussed the
concept of “style” with them in order to make sure that it was correctly
understood. Following this, the researcher asked questions to stimulate a
discussion of the relevant issues. All of the discussions were conducted in
Mandarin, which is the participants and the interviewer’s (researcher)
mother language.
CHAPTER TWO
PERSONALITY AND ITS RELATIONSHIP

WITH THINKING STYLES 1
As von Wittich and Antonakis (2011) point out, “constructs must be

thoroughly tested before they can assume their place in the nomological
network of individual differences and used in practice.” (p. 1047). Although
the field of styles has been in existence for eight decades, the identity of
styles as a unique psychological construct is still hotly debated. One
important issue relevant to the construct validity of styles is their
relationship with personality: whether intellectual styles are distinct from or
a part of personality traits.
Some scholars argue that styles are subordinate to personality (e.g., Jackson
& Lawty-Jones, 1996; von Wittich & Antonakis, 2011), whereas other
scholars insist that personality and styles are distinct constructs and they
both make unique contributions to the understanding of individual
differences (e.g., Li & Armstrong, 2015; Zhang, 2006b). From an empirical
perspective, the relationship between personality and styles can be
understood from the following three lines of research: (1) research that
examines the extent of the overlap between personality and styles; (2)
research that reveals the unique contributions of styles to other outcome
variables beyond personality; and (3) research that compares the difference
in changeability between personality and thinking styles. This chapter
discusses the association between personality and thinking in three sections,
which focus on the aforementioned three lines of research. In each section,
I will review the relevant literature in the field of intellectual styles and
present the empirical evidence from my research project.
1The present chapter draws heavily on the work by Zhang, L. F., Chen, C., and
myself, titled “Thinking styles: Distinct from personality?”, published in Personality
and Individual Differences (Fan, Zhang, & Chen, 2018).
Personality and its Relationship with Thinking Styles 27
The Overlap Between Personality and Thinking Styles

The first line of the research examined the correlation between styles and
personality, as well as the percentages of styles explained by personality.
These results showed the discriminant validity of styles. For example,
Chamorro-Premuzic and Furnham (2009) summarized seven studies and
found several consistent associations between learning approaches and
personality. The extent of these overlaps varied from low (r = 0.1–0.3) to
moderate (r = 0.3–0.5). As another example, only the extraversion
personality trait was found to be significantly (to a moderate degree)
associated with Kolb’s (1976) learning styles (Li & Armstrong, 2015). In a
more recent study, field-dependence/independence (FDI) was not related to
personality traits at all (Xie, 2015). However, other studies showed more
overlaps between styles and personality traits. For example, Von Wittich
and Antonakis (2011) found that personality traits accounted for 67% of the
variance in the adaption-innovation (KAI) styles specified by Kirton (1961,
1976). The associations between personality types (Myers & McCaulley,
1988; which is also considered to be a style model) and personality traits
were also strong, especially between the extraversion-introversion dimension
in Myers and McCaulley’s (1988) personality types and extraversion
personality trait (r = .65–.71) (Furnham, 1996; Furnham, Dissou, Sloan, &
Chamorro-Premuzic, 2007; Tobacyk, Livingston, & Robbins, 2008).
The inconsistent findings are partially due to the different style constructs
examined in various studies (Chamorro-Premuzic & Furnham, 2009).
Grigorenko and Sternberg (1995) classified various style models into three
traditions: cognition-centered, personality-centered, and activity-centered.
From this conceptual perspective, personality-centered styles are supposed
to be closer to personality than cognition-centered and activity-centered
styles. To some extent, the previous findings supported this assumption.
Activity-centered styles, such as Biggs’ (1978) learning approaches and
Kolb’s (1976) learning styles, only had a low to moderate overlap with
personality. Cognition-centered styles, such as Witkin's (1962) field
dependence/independence, had no or a low overlap with personality, while
the strongest overlaps between personality and styles were found in
personality-centered styles, such as the personality types defined by Myers
and McCaulley (1988) and the KAI styles specified by Kirton (1961, 1976).
Studies based on a more comprehensive style model that includes all of the
aforementioned three traditions would deepen our understanding of the
relationship between styles and personality. Unlike binary models of styles
(e.g., Witkin’s field-dependence/independence; Kirton’s adaption-innovation
styles), the model of thinking styles (also known as the theory of mental
28 Chapter Two
self-government; Sternberg, 1997) allows for a comprehensive profile of

individual differences based on how people vary in their relative preference
for each of the 13 thinking styles. It concerns all of Grigorenko and
Sternberg’s (1995) three traditions (cognition-centered, personality-centered,
and activity-centered).
Zhang and her colleagues (Zhang, 2002a, 2002d, 2006b; Zhang & Huang,
2001) conducted a series of studies showing that personality traits had a
weak to moderate correlation with thinking styles, and the variance in
thinking styles that could be explained by personality traits was less than
35%. Most of these studies adopted correlations or ordinary least squares
(OLS) regressions to analyze the relationships between thinking styles and
personality traits. However, Von Wittich and Antonakis (2011) argue that
correlations and OLS regressions tend to underestimate the overlap between
variables because these methods of data analysis do not take measurement
errors into consideration. More rigorous data analysis methods, such as
errors-in variables (EIV) least squares regression models, should be used to
avoid attenuating coefficient estimation.
Therefore, in the present research, EIV regressions were conducted with

control variables to estimate the overlap between thinking styles and
personality traits. This part of the analysis utilized data from the first wave
where 926 students responded to measures, which included the inventories
of thinking styles (Thinking Styles Inventory-Revised II, Sternberg,
Wagner, & Zhang, 2007) and personality traits (The NEO Five-Factor
Inventory-3, McCrae & Costa, 2007).
MANOVA (multivariate analysis of variance) and correlations were

conducted to examine the relationships of multiple demographic factors
(e.g., gender, grade, major, hometown, and socioeconomic status) to key
variables before the analysis of the relationships between thinking styles
and personality traits was performed. The multivariate indices showed that
gender (Wilks' ȁ = .97, F = 1.91, df = 13, p<.05 Ș2 = .03), grade (Wilks' ȁ =
.96, F = 3.08, df = 13, p<.001, Ș2 = .04), and hometown (Wilks' ȁ = .94, F =
1.44, df = 39, p<.05, Ș2 = .02) made statistically significant differences in
thinking styles. Gender (Wilks' ȁ = .98, F = 2.98, df = 5, p<.05, Ș2 = .02)
and hometown (Wilks' ȁ = .96, F = 2.71, df = 15, p<.001, Ș2 = .02) made
significant differences in personality traits. Furthermore, the results from
the correlations showed that socioeconomic status (SES) was significantly
associated with the five dimensions of thinking styles (r = .07–.16) and the
three dimensions of personality traits (r = .12–.16). Therefore, in the
subsequent analyses, gender, grade, hometown, and SES were controlled.
Table 2.1: Partial Correlations Between Thinking Styles and Personality Traits
Personality Thinking Styles

Traits
Legislative Executive Judicial Global Local Liberal Conservative Hierarchical Monarchic Oligarchic Anarchic Internal External
Neuroticism -.10** .12*** -.04 -.06 .03 -.14*** .21*** -.13*** .10** .03 .08 .04 -.17***
Extraversion .17*** .10** .23*** .16*** .06 .21*** -.04 .27*** -.03 .19*** .10** - .53***
.14***
Openness .40*** .06 .37*** .13*** .09** .36*** -.12*** .20*** .10** .06 .03 .19*** .26***
Agreeableness -.02 .00 -.01 - .05 .01 -.08 .03 -.06 .10** -.06 - .22***
.13*** .18***
Conscient- .29*** .22*** .24*** .05 .21*** .21*** .05 .54*** .19*** .16*** .03 .16*** .26***
iousness
Notes: Control variables: gender, grade, hometown, and SES; and **p<.01, ***p<.001.
30 Chapter Two
Table 2.2: Regressions of Thinking Styles on Personality Traits (EIV)

Thinking
Legislative Executive Judicial Global Local Liberal Conservative Hierarchical Monarchic Oligarchic Anarchic Internal External
Styles
R2Total 0.33 0.14 0.27 0.12 0.09 0.26 0.14 0.39 0.12 0.12 0.06 0.28 0.44
R2demo 0.05 0.00 0.04 0.02 0.02 0.05 0.02 0.01 0.02 0.01 0.02 0.03 0.02
R2Personality 0.28 0.14 0.23 0.11 0.07 0.21 0.12 0.38 0.10 0.11 0.04 0.26 0.41
F 35.95*** 10.96*** 28.13*** 9.47*** 7.70*** 27.19*** 11.35*** 51.33*** 10.53*** 8.93*** 5.24*** 24.33*** 59.12***
df 915 915 915 915 915 915 915 915 915 915 915 915 915
ȕneuroticism -.12* .38*** .17** -.16** .39*** .11* .15** .34*** .20*** -.17*** .26***
ȕextraversion -.11* .16** .12* .20*** -.15* .33*** .26*** -.48*** .75***
ȕopenness .58*** .49*** .21*** .52*** .29*** -.12* .41***
ȕagreeableness -.26*** -.19*** -.35*** -.21*** -.10* .21*** -.43*** .26***
ȕconscientiousness .22*** .38*** .13** .32*** .25*** .72*** .36*** .17** .31***
Notes: R2Total = contribution of gender, grade, hometown, SES, and personality to thinking styles; R2demo = contribution of gender,
grade, hometown, and SES to thinking styles; R2Personality = unique contribution of personality to thinking styles; and *p<.05,
**p<.01, ***p<.001. For greater clarity, insignificant regression coefficients are not shown.
Table 2.1 shows the correlations between thinking styles and personality
with the above control demographics. In order to balance type I and type II
errors, .01 was used as the cut-off value for the level of statistical
significance in partial correlations between thinking styles and personality
traits. 41 of the 65 correlations were significant (r = .09–.54).
The results from the EIV models (Table 2.2) showed that the percentages of
the variance in thinking styles explained by personality traits ranged from
4%–41%, with only two exceeding 30%: the variances in the external and
hierarchical thinking styles accounted for by personality traits. Among the
44 significant regression coefficients, almost two-thirds of the coefficients
were under .30, while four coefficients exceeded .50. The largest regression
coefficients were the ones for the extraversion personality trait for the
external thinking style and the conscientiousness personality trait for the
hierarchical thinking style.
The examination of the overlap between personality traits and thinking

styles (discriminant validity) revealed that, although the coefficients
estimated by EIV regression models in this study were slightly larger than
those obtained in previous studies (Zhang, 2002a, 2006b) based on OLS
regression models, the overlaps between personality traits and most of the
thinking styles were only low to moderate. In reference to the overlaps
between Kirton’s adaption-innovation (KAI) styles and personality traits,
which were based on the same method (EIV: R2=.67, von Wittich &
Antonakis, 2011), thinking styles seem to be more distinctive from
personality than KAI styles. It was also noted that there were two
exceptions: the overlaps between extraversion and the external thinking
style, as well as between conscientiousness and the hierarchical thinking
style, were much larger than the overlaps between personality traits and
other thinking styles. These findings suggest that, when researchers
examine personality traits and thinking styles simultaneously, the
hierarchical and external subscales may be removed if the length of
questionnaires and the accuracy of the measures need to be balanced.
However, these findings do not suggest that the whole inventory of thinking
styles can be dropped when researchers measure personality traits because
a large portion of the variance in thinking styles cannot be explained by
personality traits.
32 Chapter Two
The Unique Contribution of Thinking Styles Beyond

Personality
The second line of research indicates the incremental validity of styles by
showing the unique explanatory power of styles for other outcome variables
beyond personality. For example, it was found that personality styles
assessed using the Myers-Briggs Type Indicator (MBTI, Myers &
McCaulley, 1988) explained about an additional 5% of the variance in
divergent thinking after nearly 10% of the variance was explained by the
Big Five personality traits (Furnham et al., 2009). Some studies also found
that learning approaches uniquely contributed to academic performance
beyond personality traits and intelligence (Chamorro-Premuzic & Furnham,
2008; Rosander & Bäckström, 2012). However, there are also inconsistent
results. For example, von Wittich and Antonakis (2011) found that the KAI
styles did not significantly contribute to leadership beyond personality traits.
Regarding thinking styles, there is only one study (W. Fan, Zhang, &
Watkins, 2010) that examined the predictive power of thinking styles for
academic achievement after controlling for personality traits and academic
motive. The results showed that thinking styles explained an additional 10–
32% of the variance in academic performance beyond personality and
academic motive.
Compared with the amount of research that examines the discriminant

validity of styles, there has been much less work that focuses on
investigating its incremental validity. Furthermore, the findings in these
limited studies are also inconsistent. Therefore, further studies should be
conducted to examine the incremental validity of thinking styles, especially
those with non-academic outcomes; this could provide a more
comprehensive understanding of thinking styles’ unique contributions to
individual differences in performance. In the present study, the incremental
validity of styles was examined by analyzing their unique statistical
contributions to career decision self-efficacy (CDSE) and subjective well-
being (SWB) beyond personality traits.
CDSE is defined as one’s beliefs about how well he/she can perform in
career choice tasks (Taylor & Betz, 1983). CDSE has been proven to be an
important predictor for one’s career performance (Choi et al., 2012).
Meanwhile, career development is a major issue that university students
have to deal with because they are just about to begin their working life (J.
Fan, 2016). Furthermore, studies have found a close relationship between
thinking styles and CDSE (J. Fan, 2016), as well as a significant relationship
between personality and CDSE (Hartman & Betz, 2007). It is reasonable to
put thinking styles and personality into one regression model in order to
identify the amount of variance in CDSE which can be explained by styles
beyond personality.
The most accepted concept of subjective well-being is currently the one that
defines subjective well-being as “a person’s cognitive and affective
evaluations of his or her life as a whole” (E. Diener, Oishi, & Lucas, 2009,
p.187). According to this definition, subjective well-being includes a
cognitive judgment of life that is always indicated by life satisfaction and
emotional reactions to life events that are always indicated by positive affect
and negative affect. Furthermore, this definition also implies the subjective
nature of well-being, which refers to one’s personal perception of his/her
own well-being. Top-down theories of SWB contend that people’s
predispositions influence their experiences and the perceptions of their lives.
Among all of the personal predispositions, personality is the most
extensively studied variable. For example, following a meta-analysis of 137
studies, DeNeve and Cooper (1998) found that personality variables, such
as repressive-defensiveness, trust, emotional stability, locus of control,
desire for control, hardiness, self-esteem, and tension, are all associated with
subjective well-being. Specific to the Big Five personality traits, extraversion
(as a positive predictor) and neuroticism (as a negative predictor) are the
most repeatedly reported traits that are closely related to subjective well-
being. In addition, agreeableness and conscientiousness were also found to
have positive correlations with subjective well-being in some other studies
(Lucas & Diener, 2008). In contrast, intellectual styles have received little
attention in the field of subjective well-being. Despite this, a few studies on
the relationships between intellectual styles and mental health (e.g., Wise et
al., 1978; Chen & Zhang, 2010) have indicated that styles probably also play
a functional role in subjective well-being. Therefore, it is also reasonable to
put thinking styles and personality into one regression model to identify the
amount of variance in SWB which can be explained by styles beyond
personality.
The results from hierarchical liner regressions of CDSE on personality traits

and thinking styles with control variables (gender, grade, hometown, and
SES) are presented in Table 2.3. The results showed that personality traits
explained 18–26% of the variance in CDSE, while thinking styles
contributed an additional 10–11% after personality traits had been taken into
account. This result verified the incremental validity of thinking styles.
34 Chapter Two
However, is this finding transferable to other outcome variables, such as

SWB? Table 2.4 presents the results from hierarchical liner regressions of
SWB on personality traits and thinking styles with control variables (gender,
grade, hometown, and SES). The results showed that personality traits
explained 26–41% of the variance in SWB, while thinking styles
contributed an additional 5–11% after personality traits had been taken into
account. Personality traits seem to have a more important role in SWB than
CDSE. However, thinking styles still have additional explanatory power for
SWB beyond personality traits. Again, this result verified the incremental
validity of thinking styles.
Table 2.3: Regressions of CDSE on Personality Traits and Thinking Styles
Occupational Self- Goal

CDSE Planning Problem Solving
Information Appraisal Selection
R2Total 0.32 0.40 0.34 0.39 0.38

R2demo 0.04 0.05 0.04 0.03 0.04
R2Personality 0.18 0.24 0.21 0.26 0.24
R2TS 0.11 0.11 0.10 0.11 0.11
F 11.60*** 24.79*** 19.22*** 23.93*** 22.75***
df 914 914 914 914 914
Personality ȕneuroticism -.09** -.13*** -.15*** -.07* -.15***
ȕextraversion .11**
ȕopenness .08* .17*** .16*** .14***
ȕconcientiousness .10** .12*** .12*** .20*** .12***
Thinking Styles ȕlegislative .13** .09* .11**
ȕglobal .08* .13***
ȕlocal .09* .13***
ȕliberal .08* .09* .12**
ȕhierarchical .17*** .09* .23* .12***
ȕmonarchic -.08*
ȕoligarchic .09* .07*
ȕinternal .11** .09*
ȕexternal .14*** .10* .09* .14***
Notes: CDSE = career decision self-efficacy; R2Total = contribution of gender, grade, hometown, SES, personality, and thinking styles
to CDSE; R2demo = contribution of gender, grade, hometown, and SES to CDSE; R2Personality = unique contribution of personality to
CDSE; R2TS = unique contribution of thinking styles to CDSE; and *p<.05, **p<.01, ***p<.001.
36 Chapter Two
Table 2.4: Regressions of SWB on Personality Traits and Thinking Styles
Life Positive Negative

SWB(I)
Satisfaction Affect Affect
R2Total 0.35 0.55 0.41
R2demo 0.04 0.03 0.02
R2Personality 0.26 0.41 0.31
R2TS 0.05 0.11 0.08
F(24,891) 20.25*** 46.03*** 25.62***
df 915 915 915
Personality ȕneuro -.36*** -.19*** .49***
ȕextra .12*** .30***
ȕopen .07* .07**
ȕagree -.05* -.07*
ȕconcien .09* .14*** -.13***
Thinking
ȕleg .10** -.10*
Styles
ȕexe -.10*
ȕglo .07* .12***
ȕloc .08* .12*** .09*
ȕlib .16***
ȕhie .09* .08*
ȕmon .11**
ȕana .09**
ȕext .08* -.15***
Notes: SWB = subjective well-being; R2Total = the contribution of gender, grade,
hometown, SES, personality, thinking styles to SWB; R2demo = contribution of
gender, grade, hometown, and SES to SWB; R2Personality = unique contribution of
personality to SWB; R2TS = unique contribution of thinking styles to SWB; and
*p<.05, **p<.01, ***p<.001.
Comparing the Malleability of Personality

and Thinking Styles
Some scholars believe that styles and personality differ in the nature
because styles are more changeable than personality (e.g., Chamorro-
Premuzic & Furnham, 2009; Zhang & Sternberg, 2005). So, the third line
of the research that contributes to the distinction between personality and
styles contrasts personality’s malleability level with that of styles. After an
extensive review, Zhang (2013) found that, although only a few style
models were involved in existing longitudinal studies of style malleability,
the findings essentially suggested that styles are changeable. In terms of
thinking styles, Sternberg (1997) has argued that thinking styles are stable,
but can be partially socialized. Four experiments provided empirical
evidence for the malleability of thinking styles by exploring if teachers’
interpersonal styles (T. M. Yu, 2012), instructional modes (W. Fan, 2012),
or teaching styles (Lau, 2014; Tai, 2012) made a difference in students’
thinking styles. Although the directions of change in some thinking styles
were unexpected, all of the four studies showed that thinking style changed
to some extent over 13–32 weeks. Nevertheless, it is hard to determine
whether styles and personality traits differ in malleability without directly
comparing them.
In the present research, participants responded to the inventories of thinking

styles and personality twice with a one-year interval between the responses.
MANOVAs for repeated measures were conducted to detect changes in
thinking styles and personality traits over a year, with gender, grade,
hometown, and SES in control. The results from the multivariate test
showed that time generally had a statistically significant effect on thinking
styles (Wilks' ȁ = .96, F = 2.58, df = 13, p<.01, Ș = .05). Specifically, nine
of the 13 thinking styles changed significantly over the year (the statistics
for each thinking style are shown in Table 2.5). The estimates of effect size
ranged from .01 to .03, which were small (J. Cohen, 1988). In contrast, the
multivariate analysis showed that time had no significant effect on
personality traits (Wilks' ȁ = .99, F = 1.27, df = 5, p>.05, Ș = .01) (the
statistics for each personality trait are shown in Table 2.5).
38 Chapter Two
Table 2.5: Repeated Measures on Thinking Styles and Personality Traits
Round 1 Round 2 Partial

Eta
Constructs Subscales F(1,716)
Mean SD Mean SD Squared
Thinking Legislative 5.19 0.94 5.20 0.85 8.76** 0.01

Styles Executive 5.01 0.94 5.04 0.82 0.61 0.00
Judicial 4.70 0.97 4.81 0.94 18.06*** 0.03
Global 4.21 0.88 4.34 0.85 0.60 0.00
Local 4.10 1.04 4.21 1.00 7.72** 0.01
Liberal 4.44 1.15 4.55 1.06 6.97** 0.01
Conservative 4.47 1.02 4.62 0.93 5.40* 0.01
Hierarchical 4.90 1.00 4.94 0.92 2.44 0.00
Monarchic 4.45 1.06 4.58 0.99 1.57 0.00
Oligarchic 4.59 0.96 4.75 0.96 5.29* 0.01
Anarchic 3.92 0.94 4.09 0.97 8.48** 0.01
Internal 4.44 1.04 4.57 0.94 7.24** 0.01
External 4.85 1.10 4.88 1.02 7.55** 0.01
Personality Neuroticism 3.06 0.95 2.92 0.90 1.96 0.00
Traits Extraversion 3.41 0.81 3.40 0.73 0.23 0.00
Openness 4.11 0.75 3.94 0.75 0.20 0.00
Agreeableness 3.69 0.67 3.54 0.70 2.21 0.00
Conscientiousness 3.72 0.76 3.62 0.70 0.17 0.00
Notes: *p<.05, **p<.01, ***p<.001
The comparison of the malleability of personality traits and thinking styles

showed that personality traits stayed stable over a one-year period but
thinking styles significantly changed statistically over the same period of
time, although the estimates of the effect sizes were small. This finding
supported Zhang and Sternberg’s (2005) argument that styles are stable
states rather than permanent traits, which implies that thinking styles are
relatively stable, but still more changeable than personality traits.
In summary, the above analyses demonstrated that personality traits only

explained limited variance in thinking styles, that thinking styles uniquely
contributed to a third variable (i.e., career decision self-efficacy and
subjective well-being) beyond personality traits, and that the two constructs
differ regarding their degrees of malleability. All of these findings lead to
the conclusion that thinking styles are distinct from, rather than subordinate
to, personality traits. This concurs with the view of some researchers, who
state that style is not a redundant construct to personality (e.g., Busato et al.,
1999; Fjell & Walhovd, 2004; Zhang & Sternberg, 2006). The findings of
this study help to clarify the position of thinking styles in the field of
individual differences. It suggests that investigating thinking styles beyond
personality is worthwhile in order to achieve a comprehensive understanding
of individual differences in human performance and development.
CHAPTER THREE
THE SOCIALIZATION OF THINKING STYLES
As one of the most controversial issues, style malleability has been hotly
debated for a long time (Zhang, 2013; Zhang & Sternberg, 2006). Some
scholars believe that styles are inborn characteristics and that they are
relatively static (e.g., Riding & Rayner, 1998), while others argue that they
are a joint product of nature and nurture, which means they can be cultivated.
For example, Sternberg (1994; Sternberg, 1997) pointed out in his mental
self-government theory that thinking styles are at least in part socialized.
Although some scholars tend to stress the genetic effect on the development
of styles (e.g., Riding & Rayner, 1998), Zhang and Sternberg (2005) claim
that styles represent states rather than traits based on the accumulating
findings from abundant studies. This means, to some extent, that styles can
be socialized and modified by the environment in which people reside.
However, it has to be admitted that research on the socialization of thinking
styles is still in its infancy.
In order to address the issue of the socialization of styles, two questions

need to be answered. First, are styles changeable? Second, do environmental
factors contribute to styles and, if yes, to what extent? This chapter presents
research evidence on the change of thinking styles as well as the
contribution of family factors and learning environments to thinking styles.
The change of thinking styles was been roughly compared with changes in
personality traits in the last chapter, and now, in the first section, this chapter
will explore multiple types of style change. The second and third sections
of this chapter focus on the contribution of familial factors and learning
environments to thinking styles. The reason why family and learning
envrionments were selected from the abundant environmental factors is
rooted in Bronfenbrenner’s bioecological theory (Bronfenbrenner, 1979;
Bronfenbrenner & Ceci, 1994). According to this theory, among the
environment’s hierarchical systems that exert influence on individual
development, family and school are in the most proximal system and viewed
as the primary engines of individual development. Furthermore, inspired by
Kurt Lewin’s (1936) classical formula of B = f (P × E), Bronfenbrenner and
Ceci (1994) pointed out that human development takes place during the
The Socialization of Thinking Styles 41
process of person and environment interaction. That is to say, studying

student development is ineffective without considering personal characteristics
besides environmental factors. Therefore, this research includes both
environmental factors and personal characteristics, which makes it possible
to compare the competing predictive power of factors in families, factors in
schools/universities, and personal characteristics for students’ thinking
styles. In doing so, we can reach a more comprehensive understanding of
the socialization of thinking styles. This part of the examination is presented
in the final section of this chapter.
The Change of Thinking Styles 2

Most of the studies that detected the change of styles were based on
several particular style models (e.g., reflectivity-impulsivity, field
dependence/independence, deep/surface learning approaches, and career
personality types) (Zhang, 2013). For example, after a systematic and in-depth
review, Zhang (2013) found that many intervention studies on style constructs
focusing on reflectivity-impulsivity and field dependence/independence were
conducted between the 1960s and the 1980s, especially among populations
with behavioral problems or special needs. In addition, most of them
revealed the expectant change of styles after the intervention. In contrast, a
handful of studies on learning approaches had inconsistent findings. Some
studies showed that learning approaches changed as expected after
particular interventions (e.g., Gordon & Debus, 2002; K. Wilson & Fowler,
2005), while others found no significant change or even opposite changes
(e.g., Gijbels et al., 2008; Groves, 2005) (see more details in Zhang, 2013).
Longitudinal studies based on career personality types were mostly without
intervention and these studies showed that one’s career interest types were
stable but the extent of the stability depended on certain contingent factors
such as age, gender, and different methods of data analysis (Zhang, 2013).
However, as Zhang (2013) mentioned, the existing research has several
limitations, such as the lack of effect sizes and the fact that its focus is
mainly on particular populations. Furthermore, besides the style models in
the previous studies, other style models, including thinking styles, have not
been extensively examined in terms of their change using longitudinal
studies. To date, there are only two published studies on the change of
thinking styles. Fan (2012) examined changes of thinking styles in students
2The present section draws heavily on the work by Zhang, L. F., Hong, Y., and
myself, titled “The Malleability of Thinking Styles Over One Year”, published in
Educational Psychology (Fan, Zhang, & Hong, 2020).
42 Chapter Three
using two different instructional conditions over one semester. He found

that some thinking styles (e.g., hierarchical, executive, conservative)
changed under the traditional instruction condition, while other thinking
styles (e.g., judicial, liberal, legislative) changed under the hypermedia
instruction condition. Fan, Zhang, and Chen (2018) also found that nine of
the 13 thinking styles changed significantly over one year among university
students. Besides the two published studies, there are also three unpublished
doctoral dissertations that conducted experiments to explore if teachers’
interpersonal styles (T. M. Yu, 2012) or teaching styles (Lau, 2014; Tai,
2012) would cause changes in students’ thinking styles over 13–32 weeks.
Although students’ thinking styles did not change in the researchers’
expected directions, the research findings confirmed that thinking styles
were changeable (Lau, 2014; Tai, 2012; T. M. Yu, 2012). However, all of
the above studies only focused on the mean-level change of thinking styles.
It has been realized that changes at the individual level could become
blurred if researchers only focus on changes at the population level (Loo,
1997; Low, Yoon, Roberts, & Rounds, 2005). Multiple types of changes
should be examined in order to attain a more comprehensive picture of style
change. For example, does one’s relative rank status of thinking styles in
the population change over time? How about individual-level changes?
What is the proportion of people who experience true change? Would one’s
configuration of the 13 thinking styles change over time? In the field of
personality, four analytic approaches have been extensively used to
systematically study change (Roberts, Wood, & Caspi, 2008). These
analytic approaches can be used to help us answer the above questions,
enabling us to reach a more comprehensive understanding of style change.
At the population level (i.e., the group level), there are two types of changes
that are frequently examined in longitudinal studies. One is mean-level
change, and the other is rank-order consistency. Mean-level change (also
called normative change) refers to people’s general trends of change over
time (Roberts, Wood, & Caspi, 2008). This type of change is always
analyzed by comparing the difference between mean scores of certain
characteristics across two points in time. Rank-order consistency (also
called differential stability) is used to explore whether people remain stable
in the level of certain characteristics relative to their age cohorts. This
consistency is always indicated by correlation coefficients between two
occasions with the higher correlation coefficients meaning less changes to
one’s relative status in the level of certain characteristics (Donnellan,
Conger, & Burzette, 2007; Roberts et al., 2008).
However, only focusing on the population-level change might obscure the

changes happening at the individual-level (Donnellan et al., 2007). For
example, if a great number of individuals has a substantial increase in the
level of certain characteristics and, at the same time, an almost equal number
of individuals has a substantial decrease in the level of the same
characteristics, there would be no mean-level change detected because they
would cancel each other out when the mean scores were computed (Roberts,
Caspi, & Moffitt, 2001). At the individual level, there are also two types of
change: individual stability and ipsative consistency. Individual stability
(also called individual differences in change) indicates a person’s unique
trajectory of changes regardless of the general trend of change found in the
population. The Reliable Change Index (N. S. Jacobson & Truax, 1991) is
the most widely used statistics for indicating this type of change because of
its strength in ruling out score changes caused by measurement errors.
Unlike individual stability, which focuses on the changes in one particular

dimension of characteristics within each comparison, ipsative consistency
(also called ipsative stability) reflects the changes in one’s profile with
multiple dimensions (Roberts et al., 2001; Roberts et al., 2008). In other
words, it examines whether the configuration of ones’ characteristics is
maintained over time. Cronbach and Gleser’s (1953) D2 statistic was often
used to detect ipsative consistency. However, since Cronbach and Gleser’s
(1953) D2 only indicates the raw distance between two profiles, it has
limitations in terms of determining the extent of the disparity between two
profiles. Therefore, further normalization of distance-based coefficients is
encouraged (Barrett, 2003).
This section presents the results from the examination of style change using
the above four analytic approaches (i.e., mean-level change, rank-order
consistency, individual stability, and ipsative consistency). The data was
extracted from the two waves of the present research with a one-year
interval. 743 of the 926 university students responded to the same
questionnaire at the second wave. The attrition analysis conducted by the t-
tests showed that there was no significant difference in any dimension of
thinking styles between the students who participated in both data
collections and those who dropped out. Therefore, the analyses of the
change of thinking styles were based on the sample of 743 participants who
completed both waves of data collection. The reasons that the present
research was conducted among Chinese university students were based on
the following concerns. When students enter university, they experience a
transition in both family and school life. With regard to family life, they
become more distant from home and learn to be more independent from
44 Chapter Three
their parents. In terms of school life, learning environments in universities

are more diverse than the examination-oriented environments in high
schools. Studying students during this transition period can provide a more
heuristic value for the issue of style malleability.
Mean-level change of thinking styles

Paired samples T-tests were conducted to examine the mean-level change
of 13 thinking styles over one year. Meanwhile, Cohen’s ds were generated
to indicate the effect size of the change. The results (Table 3.1) showed that
scores in nine of the 13 thinking styles (i.e., judicial, global, local, liberal,
conservative, monarchic, oligarchic, anarchic, and internal) significantly
increased after one year. The Cohen’s ds of these nine thinking styles ranged
from .09–.18. The |d| values were all under .2, so the effect size was
considered to be small (McCartney & Rosenthal, 2000).
Repeated measures were conducted to further examine if individual factors,

such as gender, grade, and major, made differences in the change of thinking
styles over one year. No gender or major differences were found in the
change of thinking styles, but the interaction effect of grade and time was
significant (Wilks’ ೷ = .95, F = 3.13, p<.001, Partial Eta Squared = .05).
This means that freshmen and juniors had different changes in their thinking
styles over one year. Specifically, grade made significant differences in the
change of six thinking styles: judicial (F = 4.45, p<.05), global (F = 3.93,
p<.05), liberal (F = 10.91, p=.001), oligarchic (F = 8.30, p<.01), internal (F
= 5.39, p<.05), and external (F = 4.79, p<.05) (see Figure 1). The results
showed that, apart from the internal thinking style, juniors had a
significantly larger increase in the other five thinking styles, while freshmen
did not change or even had a slight decrease.
Table 3.1: Mean-Level Change and Rank-Order Consistency of Thinking Styles
Time 1 Time 2 Bayesian Estimation Effect Size

TS r'
Mean SD Mean SD 澔 95%HCI Pr(diff>0) Pr(diff<0) 澔 Mean 95%HDI
legislative 5.19 0.94 5.20 0.85 [-0.067, 0.061] 0.462 0.538 0.00 [-.079, 0.071] .62***
executive 5.01 0.94 5.04 0.82 [-0.095, 0.041] 0.208 0.792 -0.03 [-.103, .047] .55***
judicial 4.70 0.97 4.81 0.94 [-0.180,-0.035] 0.001 0.999 -0.11 [-.185, -.037] .56***
global 4.21 0.88 4.34 0.85 [-0.200,-0.065] <0.001 >0.999 -0.16 [-.235, -.078] .61***
local 4.10 1.04 4.21 1.00 [-0.190,-0.034] 0.003 0.997 -0.11 [-.191, -.033] .53***
liberal 4.44 1.15 4.55 1.06 [-0.170,-0.016] 0.009 0.991 -0.10 [-.173, -.015] .59***
conservative 4.47 1.02 4.62 0.93 [-0.250,-0.100] <0.001 >0.999 -0.19 [-.273, -.108] .54***
hierarchical 4.90 1.00 4.94 0.92 [-0.100, 0.030] 0.143 0.857 -0.04 [-.118, .033] .65***
monarchic 4.45 1.06 4.58 0.99 [-0.210,-0.050] <0.001 >0.999 -0.12 [-.199, -.049] .54***
oligarchic 4.59 0.96 4.75 0.96 [-0.240,-0.093] <0.001 >0.999 -0.17 [-.242, -.089] .51***
anarchic 3.92 0.94 4.09 0.97 [-0.240,-0.087] <0.001 >0.999 -0.16 [-.237, -.088] .59***
internal 4.44 1.04 4.57 0.94 [-0.200,-0.052] <0.001 >0.999 -0.13 [-.209, -.056] .62***
external 4.85 1.10 4.88 1.02 澳 [-0.100, 0.041] 0.193 0.807 澳 -0.04 [-.115, .046] .61***
Notes. TS = thinking styles; r’ are the correlation coefficients corrected for attenuation. *** p < .001.
46 Chapter Three
Figure 3.1 Grade Differences in the Change of Six Thinking Styles Freshmen
Juniors
The results showed that nine of the 13 thinking styles had an increase at the
mean-level. Given that the participants were experiencing the transition
from adolescence to young adulthood, stronger preferences for most of the
thinking styles after a year could help students to be more flexible to deal
with different types of tasks. Furthermore, the change of thinking styles
showed significant grade differences. On the one hand, the results showed
that freshmen had a larger increase in the internal thinking style than juniors
over one year. During the first year in Chinese universities, the majority of
the courses in which freshmen were enrolled were in a traditional lecture
format and required significant self-study efforts. This might enhance the
freshmen’s preferences for working independently (i.e., internal). On the
other hand, compared with the freshmen, the juniors had a significantly
larger increase in the judicial, global, liberal, oligarchic, and external
thinking styles. This suggests that juniors had become more creative when
working on tasks (i.e., liberal), thought more critically when evaluating or
judging something (i.e., judicial), paid more attention to the overall picture
(i.e., global), had become more flexible dealing with multiple tasks (i.e.,
oligarchic), and had increased their preference to work with people (i.e.,
external) over the duration of a year. These results make sense because when
transitioning from juniors to seniors over the course of a year, students begin
to access more advanced courses in their disciplines. Teachers may require
students to possess more critical thinking skills and ask them to apply what
they have learned to manage small projects. During this period, students
also need to prepare themselves for their coming careers, so they may have
more diverse experiences rather than just academic learning. All of these
environments and experiences could be reinforcing their use of the above
thinking styles.
Although thinking styles showed significant changes at the mean level, it

should be noted that the effect size of the change was small. This suggests
that, in general, thinking styles are changeable, but only to a small extent.
Rank-order consistency of thinking styles

Correlations were conducted between scores in the 13 thinking styles at
Time 1 and Time 2. The results are also shown in Table 3.1. It was found
that the correlation coefficients of three thinking styles (i.e., liberal,
hierarchical, and external) exceeded .5, which could be seen as strong
(McCartney & Rosenthal, 2000). The correlation coefficients of the rest of
the thinking styles were between .39 to .48, which indicated that the
correlations were moderate (McCartney & Rosenthal, 2000). These results
mean that students’ rankings in the majority of thinking styles stayed stable
48 Chapter Three
to moderate after one year, and students’ relative status in the levels of the
liberal, hierarchical, and external thinking styles were even more stable.
The results reveal that thinking styles had moderate to strong rank-order
stability, which means that students’ relative status among their counterparts
in the level of each thinking style was retained over one year. It is interesting
that the liberal thinking style had a significant change at the mean level
while, at the same time, it showed a strong rank-order consistency. This
pattern indicated that the liberal thinking styles may have a normative
increase over one year, but most students may have similar amounts of
increase, which leads to their relative rank status staying stable.
In addition, studies on personality change showed that personality had

moderate to strong rank-order stability (r = .40–.67) over eight to nine years
in the period from adolescence to young adulthood (Donnellan et al., 2007;
Roberts et al., 2001). Compared with these results, thinking styles had
slightly smaller correlation coefficients (r = .39–.52) within a much shorter
interval (i.e., one year), which suggests that one’s relative status in thinking
styles is less stable than that of personality.
Individual stability of thinking styles

To explore the individual-level change of thinking styles, the Reliable
Change Index (RCI) was calculated for its strength in excluding the effect
of measurement errors. If an individual’s RCI score was below -1.96 or
above 1.96, the change can be considered to be reliable (N. S. Jacobson &
Truax, 1991). If changes were random in a population, there should be 2.5%
of individuals whose RCI scores are lower than -1.96, 95% of individuals
whose RCI scores are between -1.96 and 1.96, and 2.5% of individuals
whose RCI scores are higher than -1.96. If this is not the case, it indicates
that the changes detected are true changes (N. S. Jacobson & Truax, 1991).
The observed distribution of the RCI in 13 thinking styles is presented in
Table 3.2. Chi-square tests showed that the distributions of RCI in the
oligarchic, anarchic, and internal thinking styles were significantly different
from the random-change pattern.
Table 3.2: The Distribution of the Reliable Change Index
Chi-square
TS Decreased Stayed stable Increased (df = 2, N =
743)
legislative 1.70% 96.00% 2.30% 1.83
executive 2.40% 94.60% 3.00% 0.69
judicial 1.50% 95.40% 3.10% 4.15
global 1.90% 94.80% 3.40% 3.38
local 2.20% 95.70% 2.20% 0.71
liberal 1.60% 95.40% 3.00% 2.96
conservative 2.00% 94.80% 3.20% 2.30
hierarchical 1.90% 95.70% 2.40% 1.14
monarchic 2.00% 95.80% 2.20% 1.05
oligarchic 2.00% 93.30% 4.70% 15.62***
anarchic 1.10% 93.50% 5.40% 31.12***
internal 1.10% 95.40% 3.50% 9.01*
external 2.40% 95.30% 2.30% 0.14
Notes. TS = thinking styles; Decreased = Reliable Chang Index below -1.96; Stayed
stable = Reliable Change Index between -1.96 and 1.96; Increased = Reliable
Change Index above 1.96. The Chi-square tests were used to examine if the observed
distribution of Reliable Change Index was significantly different from 2.5%
(decreasing), 95% (stayed stable), and 2.5% (increasing). * p<.05. ** p < .01. *** p
< .001.
MANOVAs were conducted to further examine if gender, grade, and major

made differences in students’ RCI scores. No group difference was found
based on gender and major, but the results showed that the grade made
significant differences in the students’ RCI scores of thinking styles (Wilks’
೷ = .95, F = 3.13, p<.001). Therefore, a series of the Reliable Change Index
was generated among freshmen (Time 1) and juniors (Time 2) separately.
Table 3.3 shows the observed distributions of the RCI in thinking styles
among two separate groups based on grade. Chi-square tests showed that,
in the group of freshmen, the distributions of the RCI of two thinking styles
(i.e., anarchic and internal) were significantly different from the random-
change pattern, while in the group of juniors, the distributions of the RCI of
four thinking styles (i.e., judicial, liberal, oligarchic, and anarchic) were
significantly different from the random-change pattern.
50 Chapter Three
Table 3.3: The Distribution of the Reliable Change Index Among Two Grades
Freshmen at Time 1 (Sophomores at Time 2) Juniors at Time 1 (Seniors at Time 2)

Chi-
Chi-square
TS Stayed Stayed square (df
Decreased Increased (df = 2, N = Decreased Increased
stable stable = 2, N =
504)
239)
legislative 3.00% 94.00% 3.00% 1.03 2.90% 94.10% 2.90% 0.35
executive 2.00% 95.60% 2.40% 0.55 2.50% 94.10% 3.30% 0.68
judicial 1.80% 95.60% 2.60% 1.03 0.80% 94.60% 4.60% 6.84*
global 2.60% 94.40% 3.00% 0.53 1.70% 93.70% 4.60% 4.87
local 2.40% 95.40% 2.20% 0.22 1.70% 96.20% 2.10% 0.87
liberal 1.80% 96.00% 2.20% 1.23 2.10% 92.50% 5.40% 8.49*
conservative 2.20% 94.20% 3.60% 2.61 1.70% 95.80% 2.50% 0.68
hierarchical 2.20% 95.20% 2.60% 0.21 1.30% 96.70% 2.10% 1.74
monarchic 2.20% 95.80% 2.00% 0.73 1.70% 95.80% 2.50% 0.68
oligarchic 2.40% 93.70% 4.00% 4.58 1.30% 92.50% 6.30% 15.16***
anarchic 1.40% 92.90% 5.80% 24.36*** 1.30% 93.70% 5.00% 7.54*
internal 1.00% 95.40% 3.60% 6.92* 1.30% 95.40% 3.30% 2.17
external 1.60% 96.80% 1.60% 3.44 2.50% 94.10% 3.30% 0.68
Notes. TS = thinking styles; Decreased = Reliable Chang Index below -1.96; Stayed stable = Reliable Change Index between -1.96 to
1.96; Increased = Reliable Change Index above 1.96. The Chi-square tests were used to examine if the observed distribution of the
Reliable Change Index was significantly different from 2.5% (decreasing), 95% (stayed stable), and 2.5% (increasing). * p<.05. ** p
< .01. *** p < .001.
The above examination at the individual-level change based on RCI showed

that the majority of participants stayed stable over the year. Reliable
changes were only found in a small number of thinking styles, which was
much less than the ones found at the mean-level change. This is because
RCI is a more conservative evaluation of change and it excluded the change
caused by measurement errors. Although a high degree of individual
stability was detected, it was undeniable that reliable changes did occur.
Despite the fact that the portion of participants who had reliable changes
was small, the statistics revealed that these detected changes cannot be
attributed to measurement errors.
Furthermore, with several exceptions, the portion of the participants who

had an increase was larger than those who had a decrease in thinking styles.
This trend corresponded with what was found at the mean-level change,
thereby consistently indicating that students tend to have stronger
preferences for diverse thinking styles over time. In addition, the juniors
were found to have true changes in more thinking styles (e.g., judicial and
liberal) than freshmen, which also resonates with the grade differences
found in the mean-level change. The results once again imply the possible
influence of the different environments that the freshmen and juniors
experienced in the past year on their thinking styles.
Ipsative consistency of thinking styles

The Cronbach and Gleser’s (1953) D2 was calculated to examine the change
in the profiles of thinking styles over one year. The D2s ranged from .00 to
82.76, with a mean of D2 =13.86 (SD=11.83). If one’s thinking styles profile
did not change at all, the change detected should be entirely due to
measurement errors. A D2 value solely based on measurement errors was
calculated (Derror2 = 14.51), and 46.10% of the participants had greater D2
values than what would be expected by measurement errors alone. This
result indicated that nearly half of the students had a true change in their
profiles of thinking styles. To further explore the extent of change,
normalized D values (dn) were computed by dividing D values by the
maximum distance that could be observed between two profiles. The closer
to 1.00 dn is, the more different the profile becomes (Barrett, 2003). The
results showed that the dns ranged from .00 to .33, with a mean of dn =0.09
(SD=0.06). This result indicated that the extent of the profile change over
one year was small.
52 Chapter Three
Further examination was conducted to see if gender, grade, and major made
differences in the ipsative consistency of thinking styles. Because D2 scores
were not normally distributed, Mann-Whitney U tests were adopted. The
results showed that there was no significant group difference in D2 based on
gender, grade, and major, suggesting that the configuration of thinking
styles changed in similar patterns across gender, grade, and major.
The above analysis of the ipsative consistency of thinking styles revealed

that the change of the configuration of thinking styles within subjects was
detected among nearly half of the participants, after excluding the
dissimilarity of profiles caused by measurement errors. However, the extent
of the profile change was small. This indicated that, although small changes
were detected, the ipsative consistency of thinking styles over one year was
remarkable.
In summary, given the results from the four types of changes examined,
thinking styles have a moderate to high degree of stability, but this does not
mean that they are fixed characteristics because, even though it is at a small
magnitude, thinking styles are still changeable. The research findings
largely supported Zhang and Sternberg's (2005) view on style malleability:
styles are stable states rather than permanent traits.
Furthermore, the individual differences in style changes found in this study

legitimate the need for further exploration on why some of participants had
a true change in thinking styles while others did not. Were some people
more susceptible to change? Or did some specific experiences or
environmental factors trigger such changes?
Qualitative findings of style change

In my research’s follow-up qualitative study, 29 university students who
had significant changes in their thinking styles over one year were selected
to be the interviewees in order to further explore how these changes
happened. When the interviewees were asked to discuss why and how their
thinking styles changed during the year between the pretest and the posttest,
most of them attributed this change to the nature of the university learning
environment. They stated that, in general, the teachers encouraged them to
think creatively and critically, and that they obtained more autonomy over
their own life in university compared with their experiences in high school.
The following are some of the comments from the interviewees:
During high school, you just did whatever teachers told you to do. You did
not need to think independently. […] In university, the atmosphere
encourages you, voluntarily or by force, to become more open and

extroverted. It is good that you can have your own opinions on things and
that you are encouraged to express them in class discussions. (JXX)
High school teachers gave you homework every day and told you exactly
what you needed to do step-by-step. In university, there is no homework at
all! Teachers also do not provide any instruction on what you need to do. It
makes me feel like my life has no direction […] but it also allows me to read
whatever I like. (WHN)
In high school, everyone did the same thing and you always had teachers
keep an eye on you. In university, we all have different goals. There are
various modes of life open for you to choose. […] No one guides you. You
have great freedom and you have to consider everything for yourself. You
need to choose your own goal and your own way. You have to be global.
(MX)
It seems that, unlike the experience in high school where students were
restricted by an organized schedule set by teachers, as well as by the
pressure of the National College Entrance Examination, in university,
students become the master of their own lives; this autonomy provides many
opportunities for them to practice Type I styles. Meanwhile, university
learning activities rewarded Type I styles more than the ones in high schools.
The detailed descriptions provided by interviewees about what they had
experienced in the university environment resonate with the findings from
the quantitative study concerning the influence of learning environments on
students’ thinking styles, which will be discussed in detail in the third
section of this chapter. In general, the finding that students became more
likely to use the creativity-generating styles (Type I styles) when they
entered higher levels in the educational system echoes Fer’s (2012) review
of the studies on the correlations between educational level and intellectual
styles, which also indicates that the higher levels encourage the
development of Type I styles.
Furthermore, in terms of style change trends, the results from the main study
also showed that the increase of Type I thinking styles among juniors was
significantly greater than among freshmen. Based on the interviewees’
description, this is probably because of the differences between the
experiences that freshmen and juniors had during the research interval.
Although the university environment generally encouraged Type I thinking
styles, juniors seemed to experience more opportunities that promote Type
I thinking styles. It is interesting that, based on the quantitative study, the
learning environment that freshmen experienced and the one that the juniors
had experienced in the year between the pretest and the posttest did not
54 Chapter Three
differ significantly, but the interviewees still claimed some differences that
may not have been clearly detected by the inventory of learning
environment in the quantitative study.
First, even though both the learning environments for freshmen and juniors
encouraged constructivist learning, juniors become more deeply involved in
this type of learning than freshmen, which means that they were somewhat
more influenced by it than the freshmen. Some interviewees mentioned that,
in the first year of university life, freshmen had just left the harsh levels of
studying in high schools behind and, therefore, they were attracted to the
variety of university activities on offer. During this period, academic study
lost its core status in the students’ lives. Although the learning environment
seemed to create a climate that facilitates students’ Type I thinking styles,
the freshmen did not spend much time and energy becoming involved in it.
As one of the interviewees mentioned,
I spent most of my time in student activities and associations during the first
year [in university]. I only worked hard at learning at the end of the semester,
in order to deal with the exams. I know a lot of students like me. We all
embrace “Buddha's feet” at the last moment [we would just be cramming for
exams] (QBK).
On the contrary, the juniors were already over the period of being curious
about the novelty of university life and they had, instead, become gradually
aware of the importance of academic performance in their later career
development. For example,
My GPA (Grade Point Average) in the first year after I entered the university
was embarrassing. […] Last year, I decided to go abroad after my graduation
and I realized that GPA is a key element when applying for good overseas
universities. I am now working much harder than before and my GPA has
improved. (QBK)
Therefore, when the juniors adjusted their time and energy allocation, and
became more involved in academic learning, they increased their likelihood
of being influenced by learning environments.
Second, some of the freshmen’s experiences had buffered the influence of

learning environments. As I have just mentioned, freshmen have spent a
significant portion of their lives in student organizations and associations.
Often, the positions and responsibilities that they held were quite entry-level.
They usually just followed orders and instructions from the leaders and
performed quite practical and executive tasks. For example, Interviewee
FQC stated the following:
During the first year of my university life, I did a lot of detailed things. For
example, when I was a junior officer in student association, I usually
delivered posters and persuaded students to participate in our activities. I
never designed an activity myself. This kind of general planning thing is
usually done by senior officers or managers.
The type of job roles that freshmen played in student associations and
organizations probably enabled them to practice their Type II styles instead
of encouraging Type I styles. In addition, the learning environment that
freshmen experienced may not be as constructivist as the one that the juniors
experienced, although the differences were not explicitly revealed in the
quantitative data. It may be because the students have used their previous
experiences as a baseline when responding to the questionnaires, in which
case, freshmen and juniors perceive their learning environments as almost
equally constructivist when the extent of the discrepancy between the
current learning environment and the one they experienced previously (high
school for freshmen and lower grades in university for juniors) were
perceived to be similar. Therefore, the quantitative data did not reveal the
differences between the learning environment perceived by freshmen and
the one perceived by juniors. However, the results from the qualitative study
did show that, although the learning environment that the freshmen
experienced in the first year of the university life was more constructivist
than the one in high schools, it was not as constructivist as the one
experienced by the juniors. For example, during the first year of university
life, although the students were impressed by certain courses taught in a
constructivist way, they still had a great number of elementary courses that
largely pertained to basic information on particular subject matters, whereas
the number of constructivist-oriented courses increased among juniors.
Meanwhile, the freshmen’s assessments included open-ended questions as
well as information recitation. In contrast, the portion of assessments that
completely valued deep thinking increased for juniors. For example,
Generally speaking, teachers of specialized courses tend to implement class
discussions more and encourage students to think critically and deeply. In
the first two years of university life, we did not have too many specialized
courses. At that time, we also had many basic courses. In basic courses, it
was usually just the teacher who did the talking and almost 100 students in
the same classroom were just listening. (ZJL)
For example, courses like “basic experiments in physics” just required

students to strictly follow the prescribed procedures. Once I obtained a result
that was inconsistent with the teacher’s, even though I stuck to the right
procedures. There might have been something wrong with the equipment.
However, the teacher insisted that I repeat the experiment until I got the
56 Chapter Three
“right” result. (YMY)
In university, we have different forms of assessments. Sometimes essays,

sometimes exams […] I don’t like exams. It seems like a knowledge
competition in a limited time. In contrast, there is no limited time for writing
essays. You can really think through something deeply and present your
ideas thoroughly in essays. Writing essays is more common when you enter
the higher grades. I like it. (XZY)
Therefore, the mixed nature of the courses and assessments that freshmen
experienced may have also confounded the trend of developing Type I
thinking styles to some extent.
Third, there are probably some factors outside the campus that also exert an
impact on the students’ development of thinking styles. During the
interviews, many juniors mentioned that they undertook internships during
the third year of their university life. As they were brand-new experiences,
the internships provided new ways of viewing themselves and the world
around them, which also significantly affected their thinking styles. For
example,
I think the internship had a great impact on me. I worked for a liberal
magazine. The atmosphere was liberal and creative. You could bring out
every idea that came into your mind. In addition, the fact that you need to
handle multiple tasks simultaneously helped you to learn how to manage
stress and become increasingly better at multitasking. (ZJL)
During the internship, I was independently in charge of a lot of tasks. You

learned from these experiences a lot, whether they were successes or failures.
Gradually, it changed me and I became more confident in the belief that I
can do anything I want to do. (CCC)
Besides offering more opportunities to practice Type I styles, it seems that

the internship experience triggered an identity transition from a student to
an independent adult, who needs to be responsible for themselves and for
their actions. This conceptual autonomy in identity might be a good
facilitator for the development of Type I thinking styles.
In summary, the results indicated that, although it was not to an extensive

degree, thinking styles did change significantly over one year, which
supported the argument that style is a stable but still changeable state (Zhang,
2014; Zhang & Sternberg, 2005). The interview data indicated that the
change of styles was due to the life circumstances that students experienced.
The next two sections discuss how two particular environmental factors
(familial factors and learning environments) influence the development of
thinking styles based on both quantitative results and qualitative results.
Family Factors and Thinking Styles

It is widely recognized that parents play a critical role in child development
(Collins, Maccoby, Steinberg, Hetherington, & Bornstein, 2000; Steinberg
& Morris, 2001). Although family is always seen as a major socialization
agent, little research focuses on the influence of family settings in shaping
students’ thinking styles. Only two factors in family environments have
been examined with thinking styles. The first one is socioeconomic status
(SES). It was found that higher socioeconomic status (SES) was positively
related to higher cognitive complexity and more creative styles (Ho, 1998;
Robert J. Sternberg & Grigorenko, 1995; Tse, 2003; Yang & Lin, 2004;
Zhang & Postiglione, 2001). Zhang and Sternberg (2006) explained that
students from higher SES families have more opportunities to experience
different issues and situations, and so their previous cognitive structures
have more opportunities to be challenged, which could enable them to
develop more complex and creative styles. The second familial factor,
which was examined with students’ thinking styles, was parents’ thinking
styles. Zhang’s (2003) study found that there was a congruence between
parents’ thinking styles and children’s thinking styles to some extent.
However, as Zhang (2003) mentioned, this study failed to clarify whether
the correlation between parents' and children's styles was due to genetic
factors or socialization effects (Zhang, 2003). So, further studies that
examine more direct environmental factors (e.g., parenting styles) should
be conducted to explore the issue of the socialization of intellectual styles
in family settings. In this section, the findings from previous research on
parenting styles and intellectual styles are summarized. After this, the
theoretical foundation of the present research for parenting styles is briefly
introduced, which is followed by the major findings regarding the
relationships between familial factors and thinking styles.
Previous findings: Parenting styles and intellectual styles

Parenting styles are broadly defined as “a constellation of attitudes towards
the child that are communicated to the child and that, taken together, create
an emotional climate in which the parent’s behaviors are expressed”
(Darling & Steinberg, 1993, p. 488). They are one of the most important
factors in the family context. They are a widely studied familial variable
that has been repeatedly found to play an important role in the individual
socialization process and various developmental outcomes (Collins et al.,
2000; Steinberg & Morris, 2001), such as self-esteem (Milevsky, Schlechter,
58 Chapter Three
Netter, & Keehn, 2007), mental health and well-being (Fletcher, Walls,
Cook, Madison, & Bridges, 2008), personality (Reti et al., 2002), academic
achievement (Spera, 2005), and social behaviors (Steinberg, 2001).
However, in the broader context of intellectual styles, it also has to be

admitted that there is a lack of research that systematically and directly
examines the relationships between parenting styles and students’ intellectual
styles. Relatively speaking, field-dependence/independence (FDI) is a style
construct that has been paid a little more attention than the other intellectual
style constructs in terms of the influence of parenting. Some studies indicate
that paternal involvement and the level of autonomy children receive from
their parents had positive effect on their development of field independence
(Páramo & Tinajero, 1998), whereas punishment and coercion in parenting
tended to encourage field dependence (Witkin & Goodenough, 1981).
However, there are also some inconsistent findings. For example,
Moskowitz, Dreyer, and Kronsberg (1981) found that parenting styles had
no significant impact on children’s FDI. Some scholars (e.g., Kogan &
Block, 1991; van Meel, 1991) have suggested that more valid instruments
should be used to examine the different characteristics of parenting. In
addition, all of these studies on field-dependence/independence examined
parenting use the dimensional approach rather than the typological approach.
Unlike the studies on FDI, one study (Stansbury & Coll, 1998) on Myers
and McCaulley’s (1988) personality types (MBTI) adopted the typological
approach rather than the dimensional one to examine parenting styles.
Stansbury and Coll (1998) found that the authoritarian parenting style led to
higher scores on the introversion and judging scales. However, in Stansbury
and Coll’s (1998) study, the types of parenting styles were assessed using
just one question. The reliability and validity of this measurement of
parenting styles is questionable. Another study (Penn, 2004) also examined
the types of parenting styles with the desired structure levels within one’s
learning style according to Dunn’s (1978) model. In this study, it was found
that fathers’ authoritarian parenting style was positively related to higher
levels of desired structure in learning. However, the learning styles
examined in this study are based on Dunn’s (1978) model, which has been
criticized for its poor reliability and validity (Coffield et al., 2004) and is
not included in intellectual styles according to Zhang and Sternberg’s (2005)
classification.
Theoretical foundation: Baumrind’s parenting styles

Before I examined the relationships between parenting styles and students’
thinking styles, I had carefully considered the potential theories and
measures of parenting styles to use in this research. The theories and
research into parenting styles are usually based on two approaches:
dimensions and typologies. The dimensional approach means that parenting
styles are described by the fit level on the dimensional continuum proposed
by the relevant theories. The typological approach means that parenting
styles are classified into different types according to the combination of
characteristics of different parenting dimensions. By now, Baumrind’s
threefold typology and later fourfold typology are widely used in many
studies on parenting styles. Compared with the dimensional framework, the
typological approach contributes to a better understanding of the interaction
of different parenting dimensions on children (Fletcher et al., 2008).
However, the theories of parenting styles which use the typological
approach cannot be simply be generalized to different cultures because there
are different cultural interpretations; therefore, the dimensional approach
still has strengths in research. For example, the dimensional approach can
provide a much closer look at different aspects of parenting compared with
the typological approach (Parke, 2001). Therefore, a measure that can assess
both the main dimensions of parenting and the types of parenting styles is
more favorable. Steinberg et al.’s (1994) instrument is superior because it
has its basis in influential theories and its strength lies in examining both
dimensions and types of parenting styles. Therefore, the present research
adopted this instrument for assessing parenting styles.
It also needs to be noted that perceived parenting styles instead of actual

parenting styles were measured in the present research. Bronfenbrenner and
Ceci (1994) have suggested that, in addition to the environment’s objective
properties, its subjective properties, as perceived by people who reside in
that environment, also have an important influence on human development.
This means that it is unreasonable to examine the environmental influence
on students’ development without their interpretation. There is also
evidence indicating that individuals’ perceptions of environments are
probably more influential than actual environments (Bolkan, Sano, De
Costa, Acock, & Day, 2010; Nijhuis, Segers, & Gijselaers, 2008). Therefore,
it is necessary to investigate the influence of perceived parenting styles in
families on students’ thinking styles. For the sake of brevity, “perceived
parenting styles” is referred to as “parenting styles” in the following
chapters.
60 Chapter Three
Current findings
In the first round of data collection, participants were asked to recall their
experiences with their parents during the first 16 years of their lives when
they responded to the Parenting Style Index. In the second round of data
collection, which took place a year later, they responded to the Parenting
Style Index according to their experience with their parents in the period
between the two occasions of data collection. This enables us to see both
remote and recent influences of parenting styles on thinking styles. As in
other analyses, I controlled the relevant demographic factors in the
subsequent analyses. In the first wave of data, gender, grade, and hometown
made significant differences in thinking styles, while gender and major made
significant differences in parenting styles. In the second wave of data, thinking
styles differed based on gender and grade, while parenting styles differed
based on gender and hometown. Furthermore, socioeconomic status (SES)
was significantly associated with both parenting styles and thinking styles in
both waves of data. As SES is one of the familial factors, the relationships
between SES and thinking styles were highlighted (see Table 3.4).
Table 3.4: Zero-Order Correlations Between Socioeconomic Status (SES)

and Thinking Styles
Constructs Dimensions Time 1 Time 2

Thinking styles legislative .15*** .08*
executive
judicial .16*** .08*
global .10**
local -.09*
liberal .09**
conservative
hierarchical .07* .10**
monarchic
oligarchic
anarchic
internal
external .14***
Notes: *p<.05; **p<.01, ***p<.001; non-significant correlation coefficients are not
reported.
The legislative, judicial, global, liberal, and hierarchical thinking styles that
were found to have significant positive relationships with SES all belong to
Type I thinking styles, which means that students from families with a
higher socioeconomic status were more likely to have styles characterized
by higher cognitive complexity and creativity. This finding was consistent
with earlier studies (Ho, 1998; Sternberg & Grigorenko, 1995; Tse, 2003;
Yang & Lin, 2004; Zhang & Postiglione, 2001).
The relationships between students’ perceived parenting styles (parenting styles

hereafter) and their thinking styles were explored based on both the dimensional
and the typological approaches. The relevant demographic factors were
controlled when hierarchical regressions were conducted to predict thinking
styles from three dimensions of parenting styles (acceptance/involvement,
psychological autonomy granting, and strictness/supervision), while
MANCOVA (Multivariate Analysis of Covariance) was carried out to compare
differences in thinking styles among four types of parenting styles (authoritative,
authoritarian, indulgent, and neglectful). The means of two dimensions (i.e.,
acceptance/involvement and strictness/supervision) were used to partition
parenting styles. Parenting styles with scores both higher than the mean in
acceptance/involvement and the mean in strictness/supervision were classified
as the authoritative parenting style. Parenting styles with scores lower than the
mean in acceptance/involvement and scores higher than the mean in
strictness/supervision were classified as the authoritarian parenting style.
Parenting styles with scores higher than the mean in acceptance/involvement
and scores lower than the mean in strictness/supervision were classified as
the indulgent parenting style. Parenting styles with scores both lower than
the mean in acceptance/involvement and the mean in strictness/supervision
were classified as the neglectful parenting style.
62 Chapter Three
Table 3.5: Predicting Thinking Styles from the Three Dimensions of Parenting Styles (Time 1)
TS Type I Type II Type III
Leg Jud Glo Lib Hie Exe Loc Con Mon Oli Ana Int Ext
2
R Total 0.07 0.07 0.04 0.06 0.07 0.08 0.04 0.08 0.06 0.09 0.05 0.03 0.12
R2demo 0.04 0.03 0.02 0.05 0.01 0.00 0.01 0.02 0.02 0.01 0.02 0.02 0.02
R2PS 0.03 0.04 0.02 0.01 0.06 0.08 0.03 0.08 0.04 0.08 0.03 0.01 0.10
F 12.30*** 11.22*** 5.77*** 10.26*** 12.10*** 12.65*** 5.91*** 13.60*** 9.52*** 15.31*** 8.72*** 5.30*** 20.81***
df 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918 6,918
*** *** *** ** *** *** *** *** *** *** ***
ȕinvol .19 .20 .14 .11 .26 .24 .13 .17 .16 .29 .16 .34***
ȕpsy_au -.10** -.09** -.13*** -.09** -.16*** -.12*** -.14*** -.18*** -.10**
ȕsuper .09** .09**
Notes: TS = thinking styles; Leg = Legislative; Jud = Judicial; Glo = Global; Lib = Liberal; Hie = Hierarchical; Exe = Executive; Loc
= Local; Con = Conservative; Mon = Monarchic; Oli = Oligarchic; Ana = Anarchic; Int = Internal; Ext = External; R2Total = the
contribution of gender, grade, SES, and parenting styles to thinking styles; R2demo = the contribution of gender, grade, and SES to
thinking styles; R2PS = the unique contribution of parenting styles to thinking styles; invol = acceptance/involvement; psy_au =
psychological autonomy-granting; super = strictness/supervision; and *p<.05, **p<.01, ***p<.001.
Table 3.6: Predicting Thinking Styles from the Three Dimensions of Parenting Styles (Time 2)
Type I Type II Type III

TS
2
R Total 0.09 0.09 0.08 0.12 0.11 0.09 0.08 0.10 0.08 0.16 0.08 0.06 0.13
2
R demo 0.03 0.03 0.05 0.07 0.01 0.01 0.02 0.02 0.03 0.02 0.04 0.04 0.01
R2PS 0.06 0.06 0.03 0.05 0.09 0.08 0.06 0.08 0.06 0.14 0.04 0.02 0.13
F 8.37*** 8.03*** 6.62*** 10.76*** 9.55*** 7.88*** 6.87*** 8.94*** 7.39*** 15.08*** 7.25*** 5.14*** 12.45***
df 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724 9,724
*** *** *** *** *** *** *** *** *** *** **
ȕinvol .29 .28 .14 .27 .32 .27 .23 .21 .24 .36 .13 .40***
ȕpsy_au -.15*** -.15** -.09* -.11** -.18** -.20*** -.17*** -.21*** -.21*** -.15*** -.08*
ȕsuper -.10* -.09* .08*
contribution of gender, grade, hometown, SES, and parenting styles to thinking styles; R2demo = the contribution of gender, grade,
hometown, and SES to thinking styles; R2PS = the unique contribution of parenting styles to thinking styles; invol =
acceptance/involvement; psy_au = psychological autonomy-granting; super=strictness/supervision; and *p<.05, **p<.01, ***p<.001.
64 Chapter Three
Table 3.7: Style Differences Based on the Types of Parenting Styles (MANCOVA) (Time 1)
Thinking Styles Sum of Squares df Mean Square F Sig. Partial Eta Squared
Type I Legislative 19.95 3 6.65 8.20 .000 .03
Judicial 21.92 3 7.31 8.32 .000 .03
Hierarchical 40.82 3 13.61 14.50 .000 .05
Type II Executive 37.53 3 12.51 14.70 .000 .05
Local 13.70 3 4.57 4.35 .005 .01
Conservative 20.40 3 6.80 6.54 .000 .02
Monarchic 15.66 3 5.22 5.00 .002 .02
Type III Oligarchic 40.10 3 13.37 14.76 .000 .05
External 78.36 3 26.12 24.27 .000 .07
Table 3.8: Style Differences Based on the Types of Parenting Styles (MANCOVA) (Time 2)
Thinking Styles Sum of Squares df Mean Square F Sig. Partial Eta Squared
Type I Legislative 36.53 3 12.18 18.68 .000 .072
Judicial 21.35 3 7.12 8.48 .000 .034
Liberal 36.65 3 12.22 12.01 .000 .047
Hierarchical 52.07 3 17.36 22.11 .000 .084
Type II Executive 29.45 3 9.82 15.59 .000 .060
Local 17.42 3 5.81 5.97 .001 .024
Conservative 11.90 3 3.97 4.73 .003 .019
Monarchic 14.92 3 4.97 5.30 .001 .021
Type III Oligarchic 47.93 3 15.98 18.90 .000 .072
External 75.45 3 25.15 26.36 .000 .098
66 Chapter Three
The results based on the data from Time 1 and Time 2 were consistent,
which means that the contribution of parenting styles in either the first 16
years of the students’ lives or in the past year to thinking styles was quite
similar. In terms of the dimensional approach, it was found that parenting
styles could explain 1–10% and 2–14% of the variance in thinking styles at
Time 1 (Table 3.5) and at Time 2 (Table 3.6), respectively. At both Time 1
and Time 2, it was found that parental acceptance/involvement was positively
associated with all but the internal style, which partially supported the
hypothesis. Meanwhile, also partially in line with expectations,
psychological autonomy-granting was negatively related to some Type I
styles, Type II styles, and Type III styles. In addition, perceived parents’
supervision positively predicted some Type II styles at both Time 1 and
Time 2, and it also negatively contributed to two Type I styles at Time 2,
which was totally consistent with expectations.
In terms of the typological approach, the results from the MANCOVA

confirmed that parenting styles made statistically significant differences in
students’ thinking styles (for Time 1, Wilks' ೷ = .861, F = 3.558, p<.001;
for Time 2, Wilks' ೷ = .791, F = 4.479, p<.001). Further examinations on
each thinking style found that nine of the 13 thinking styles varied based on
types of parenting styles at Time 1 (Table 3.7) and ten thinking styles varied
based on types of parenting styles at Time 2 (Table 3.8).
Post-hoc tests (Scheffé) were conducted on these thinking styles to further

identify the mean differences of the thinking styles between every two types
of parenting styles. The cutoff score for the level of statistical significance
(the alpha value) was amended using the Bonferroni correction, which made
it .008. Eventually, the significant differences of seven thinking styles based
on types of parenting styles at Time 1 were confirmed (Table 3.9) and the
significant differences of six thinking styles based on types of parenting
styles at Time 2 were confirmed (Table 3.10). Most of the relationship
patterns were as the same at Time 1 as at Time 2. For example, at both times,
students who perceived their parents as using the authoritative and indulgent
parenting styles significantly scored higher on two Type I (legislative and
hierarchical) thinking styles and two Type III (oligarchic and external)
thinking styles than students who perceived their parents as using the
neglectful parenting style. Also, at both times, students from authoritative
families scored significantly higher in the executive (Type II) thinking
styles than those from neglectful families. Nevertheless, at both times, there
were also differences in the hierarchical thinking style between the
authoritarian and the authoritative parenting styles. In the external thinking
style, students who perceived their parents as using the authoritative and
indulgent parenting styles also scored higher than those who perceived their
parents as using the authoritarian parenting style. The only differences
between the results at Time 1 and Time 2 were that, at Time 1, style
differences based on types of parenting styles were also found in the judicial
style and the conservative style, while at Time 2, there were also differences
in the liberal style based on types of parenting styles.
The above results revealed that parenting styles had a significant influence
on students’ thinking styles. Specifically, parental involvement positively
contributed to the majority of thinking styles while the psychological
autonomy (from parents) that students perceived negatively contributed to
the majority of thinking styles, which partially supported the specific
research hypotheses. In addition, it was found that, exactly as hypothesized,
parental strictness/supervision negatively contributed to Type I thinking
styles and positively contributed to Type II styles. The results regarding
types of parenting styles also repeatedly found that students raised under the
authoritative and the indulgent parenting styles are more likely to develop
Type I thinking styles and some of Type III thinking styles than those raised
by parents using the authoritarian and the neglectful parenting styles.
68 Chapter Three
Table 3.9: Multiple Comparisons for Thinking Styles Based on Types of Parenting Styles (Time 1)
95% CI
Thinking Styles PS(I) PS(J) Mean Difference(I-J) Std. Error Sig.
Lower Bound Upper Bound
Type I Legislative 1 4 0.35 0.08 .000 0.13 0.56
3 4 0.32 0.09 .008 0.06 0.58
Judicial 1 4 0.36 0.08 .000 0.14 0.58
3 4 0.42 0.10 .000 0.15 0.69
Hierarchical 1 2 0.33 0.09 .003 0.08 0.59
1 4 0.53 0.08 .000 0.31 0.76
3 4 0.48 0.10 .000 0.21 0.75
Type II Executive 1 4 0.48 0.08 .000 0.26 0.69
Conservative 1 4 0.33 0.09 .002 0.09 0.58
Type III Oligarchic 1 4 0.47 0.08 .000 0.25 0.70
3 4 0.36 0.10 .003 0.09 0.63
External 1 2 0.57 0.10 .000 0.30 0.84
1 4 0.63 0.09 .000 0.39 0.88
3 2 -0.58 0.11 .000 0.27 0.90
3 4 0.65 0.10 .000 0.36 0.94
Notes: PS = parenting styles; 1 = authoritative; 2 = authoritarian; 3 = indulgent; 4 = neglectful.
Table 3.10: Multiple Comparisons for Thinking Styles Based on Types of Parenting Styles (Time 2)
95% CI
Thinking Styles PS(I) PS(J) Mean Difference(I-J) Std. Error Sig.
Lower Bound Upper Bound
Type I Legislative 1 2 0.33 0.09 .006 0.07 0.59
1 4 0.41 0.07 .000 0.21 0.62
3 4 0.42 0.09 .000 0.17 0.68
Liberal 1 2 0.43 0.12 .004 0.10 0.76
3 2 0.52 0.14 .002 0.14 0.90
Hierarchical 1 2 0.51 0.10 .000 0.23 0.79
1 4 0.61 0.08 .000 0.39 0.83
3 4 0.46 0.10 .000 0.19 0.74
Type II Executive 1 4 0.47 0.07 .000 0.27 0.66
Type III Oligarchic 1 4 0.49 0.08 .000 0.26 0.73
3 4 0.47 0.10 .000 0.18 0.76
External 1 2 0.59 0.11 .000 0.28 0.91
1 4 0.56 0.09 .000 0.32 0.81
3 2 0.68 0.13 .000 0.32 1.04
3 4 0.64 0.11 .000 0.34 0.95
Notes: PS = parenting styles; 1 = authoritative; 2 = authoritarian; 3 = indulgent; 4 = neglectful.
70 Chapter Three
Data from the follow-up interviews showed that students were not fully
aware of the influence of parenting styles on thinking styles. When they
were asked to respond to “what factors caused the change of your thinking
styles”, they talked a lot about learning environments but did not voluntarily
mention parenting. However, when they were asked the question “what
about the influence from the parenting you experienced”, they all responded
to the question without any hesitation: “of course it has an influence;
actually, a very important influence on my thinking styles”. But when it
came to the question of “how”, they all failed to address it specifically. This
is probably because, first, after entering university, students spent most of
their time in learning environments and their interactions with their parents
only filled a minimal portion of their life, which made them less likely to
instantly recall the factor of parenting. After all, the memories about their
learning environments were fresher. Second, parents seem to influence
children through osmosis. It is subtle and gradually formed. Furthermore,
people tend to not be as sensitive to the influence of a relatively constant
environment (e.g., family environment) than to the influence of a
changeable environment (e.g., learning environment).
However, when the interviewees were asked to just describe how their
parents parented them and how they responded to this parenting, the
connection between parenting and thinking styles emerged. This finding
further explains the results obtained from the main study.
First, students who perceived high acceptance/involvement from their

parents said that this supportive climate made them feel safe and
comfortable and that they were able to try anything without any stress or
emotional burden. Therefore, these students may be more likely to
experience a greater variety of environments and tasks than their peers,
which enables them to have more opportunities to practice a broad range of
thinking styles. This may explain the quantitative result that students with
higher parental acceptance/involvement scored higher in the majority of
thinking styles than their counterparts.
Second, students who perceived that their parents had given them high
psychological autonomy had different responses to this kind of parenting.
On the one hand, some students said that, because their parents always
respected them and encouraged them to make decisions by themselves, they
had never thought that it was necessary to obey authorities. Therefore, these
students were less likely to develop Type II styles. On the other hand, some
students also considered psychological autonomy to be an indicator of their
parents’ trust and that this should be reciprocal. Therefore, they were more
willing to be a “standard good” child and they usually tried to avoid doing
anything that might make their parents worried or anxious. The concept of
“not being rebellious as a repayment for parents’ trust” (MSJ) probably led
these students to suppress the development of Type I styles, which are
creativity-oriented and fond of novelty. The ambivalence in the ways the
students treated the psychological autonomy obtained from their parents
may partially explain the results from the main study, which indicated that
students who perceived higher levels of psychological autonomy from their
parents had lower scores on both Type I and Type II thinking styles.
Actually, some studies (e.g., Greenfield, Keller, Fuligni, & Maynard, 2003;
Qin, Pomerantz, & Wang, 2009) have suggested that psychological
autonomy granted by parents may not benefit child development in China
as much as it does in Western countries due to the cultural differences in
attitudes to the value of independence and interdependence. The students’
complicated emotions and feelings when they responded to the
psychological autonomy granted by their parents may be ascribed to the
mixed influence of their increasing need for autonomy during the process
of growth and the cultural beliefs about interdependence and harmony
within the family.
Third, the results on the relationship between parental strictness/supervision

and thinking styles completely supported the hypothesis. Generally speaking,
interference, control, and discipline from parents encouraged students’ Type
II thinking styles and discouraged their Type I thinking styles. For example,
students who perceived high strictness/supervision in the first 16 years of their
lives tended to be more compliant, as well as less creative and autonomous
when performing tasks (the executive style). This is probably because too
much interference and strict supervision from parents kept students in a
relatively simple and limited environment where they may not have enough
opportunities to make plans and decisions about their own lives. In addition,
students who perceived high strictness/supervision during the research
interval were also less likely to form their own judgments and have critical
attitudes (the opposite of the judicial style), and less likely to pursue novelty
(the opposite of the liberal style). This may be because parents who had
strict “family rules” for their children’s lives were usually committed to
maintaining parental authority. In this context, students may be less likely
to question and challenge parents or other people and conventions, and they
may tend to be less critical and liberal.
However, despite the role that parental strictness/supervision played in

students’ thinking styles, it should be noted that this dimension was related
to a much smaller number of thinking styles than the other two parental
72 Chapter Three
dimensions. Furthermore, the results on the types of parenting found that

students raised by the authoritative and the indulgent parenting styles are
more likely to develop Type I thinking styles and some of the Type III
thinking styles than students raised by parents with the authoritarian and the
neglectful parenting styles. As the authoritative parenting style and the
indulgent parenting style are similar in terms of the high level of parental
acceptance/involvement but differ in the level of strictness/supervision, this
result also indicates that parental strictness/supervision may play a less
important role in shaping students’ thinking styles than parental
acceptance/involvement.
Learning Environments and Thinking Styles 3

Besides parental influence mentioned in the previous section, learning
environments in schools, colleges, and universities are also within the
proximal environmental system where student development happens. As
mentioned in Chapter 1, some previous studies supported the relationships
of schools, academic disciplines, and courses to students’ thinking styles
(e.g., Cheung, 2002; Verma, 2001; Wu & Zhang, 1999; Zhang & Sachs,
1997), but they did not clarify what particular factors in schools, academic
disciplines or courses influence the development of students’ thinking styles,
because they did not systematically examine the specific dimensions of
learning environments (e.g., the organization of courses, the types of
assessments, the supportive climate around students, and learning facilities
outside the classroom) to see their influence on students’ thinking styles. In
this section, we will look into previous studies that indicate the possible
impact of particular dimensions of learning environments on intellectual
styles. Furthermore, I will also discuss why and how my new inventory of
learning environments was designed before discussing the current
research’s specific findings on the relationships between learning
environments and thinking styles.
Learning environments and intellectual styles

There are a handful of studies that have examined the influence of certain
aspects of learning environments on students’ intellectual styles. Among
these studies, most of them assessed intellectual styles based on Biggs’
(1978) theory of learning approaches (a model of intellectual styles) and
3 This section partially draws on the work by Zhang, L. F. and myself, titled “The
role of learning environments in thinking styles”, published in Educational
Psychology (Fan & Zhang, 2014).
several other studies were based on Witkin’s (1962) theory of field-

dependence/independence (another model of intellectual styles). The
findings in this area are inconsistent.
In terms of the relationships between learning environments and learning

approaches, the influence of assessment types and the effect of course
process that are relevant to constructivist characteristics have been
examined in previous studies. With regard to assessment types, some
researchers found that assessments and assignments that were authentic,
contextualized, or essay-based led to students’ deep learning approach
(Gordon & Debus, 2002; Gulikers et al., 2008; Seddon, 2008; Tian, 2007),
while other researchers failed to find similar results (Reid, Duvall, & Evans,
2005; Segers et al., 2008; S. N. Smith & Miller, 2005). On the contrary,
some of them even found that assessment types, such as portfolio
assessment and formative assessment, facilitated the surface learning
approach among students (Baeten et al., 2008; Gijbels & Dochy, 2006).
Some scholars argue that it is the way that assessments are implemented
during the learning process rather than how the assessments are designed
that influences the development of different learning approaches (Segers et
al., 2008). In addition to assessment types, course processes with
constructivist characteristics were also examined. For example, Wilson and
Fowler’s (2005) study explored the effectiveness of a constructivist learning
environment by comparing a conventional course and an action course (with
project work and learning groups). They found that, in the action course,
students who were deep learners at the beginning of the course maintained
their learning approach to the end of the course, while students who were
initially surface learners changed their learning styles to the deep learning
approach. Gordon and Debus (2002) also found similar changes when
teaching methods were modified by cooperative group problem-based
learning pedagogy. Nijhuis and his colleagues (2008) also demonstrated that
clear goals, independent learning, and proper workload were positively
related to the deep learning approach. However, other studies failed to prove
the effectiveness of constructivist learning environment when cultivating
students’ deep learning approach (Klinger, 2006; McParland et al., 2004).
Some studies even found that courses designed according to constructivist
principles increased the surface learning approach (Gijbels et al., 2008;
Groves, 2005).
There are also several studies that assess intellectual styles using Witkin’s
(1962) theory of field dependence/independence to examine the influence
of a constructivist learning environment on intellectual styles. Luk (1998)
found that in a self-directed learning environment created by a distance-
74 Chapter Three
education program, students became more field independent. Cathcart

(1990) also found that LOGO instruction based on constructivist principles
had a positive relationship with students’ field independence. However,
another study (Azadi, 2009) that compared the problem solving method and
lecturing failed to prove the unique contribution of a constructivist learning
environment to the development of field independence.
In addition to learning approaches and field dependence/independence, the

intellectual styles defined in other models have been paid little attention by
researchers who study the effectiveness of learning environments. As the
style models of learning approaches and field dependence/independence are
both based on only one bipolar dimension, a more general style model that
describes a repertoire of styles (i.e., thinking styles) is needed to conduct a
more comprehensive examination of the effects of learning environments
on intellectual styles. Therefore, the present research examined the
development of students’ thinking styles by directly exploring the influence
of various specific dimensions of learning environments.
Learning environments assessed in this research

Fraser (1998) identified a learning environment as “the social, psychological,
and pedagogical contexts in which learning occurs and which affect student
achievement and attitudes” (p. 3). Because of rapid progress in science,
especially the development of information technology (Jonassen, 1996), the
traditional vision of teaching and learning that emphasized information
transmission as learning has become incompatible with the modern world
(Sawyer, 2006). Changes, such as from learning as content to learning as
process, from rot learning to reflective learning, from knowledge as truth to
knowledge as relative, and from teacher-centered to student-centered, have
gradually risen (Jarvis, Holford, & Griffin, 1998). Along with changes in
the understanding of learning and teaching, the view of what make the most
effective high-quality learning environment is also changing.
What type of learning environment is viewed as the most effective one

generally depends on the learning theory that is utilized. Behaviorism
assumes that learning can be seen as changes in behavior, which happens
through one’s responses to external stimuli (B. F. Skinner, 1968). From the
perspective of behaviorism, human learning totally depends on how
teachers and educators facilitate or restrict students’ behaviors or responses
through conditioning principles, which means that students play a passive
role in the learning process (Wong, 2003). Furthermore, behaviorism sees
the mechanism inside human beings as a “black box” that cannot be
understood. Such a view ignores the cognitive process in human learning

(Siemens, 2004). That’s why behaviorism has gradually fallen into disfavor
with educators. Based on memory and information processing models
developed in cognitive psychology, cognitivism views learning as a mind-
based process instead of overt behaviors. Similar to a computer, human
learning is about information processing, storage, and retrieval, which
means that someone can be seen as a good learner if he/she can memorize
and retrieve information efficiently when it is required (Kirschner, Sweller,
& Clark, 2006).
According to this view, effective teaching is characterized by successfully

attracting students’ attention and teaching in a way that learning materials
can be easily remembered. Cognitivism is the polar opposite to behaviorism,
which focuses on organism rather than environment, mind rather than
behavior, and thinking rather than practice (Bredo, 1997). However,
behaviorism and cognitivism still have some common assumptions and
limitations. For example, they both believe that knowledge is objective and
independent of learners. A teacher’s role is as a knowledge transmitter and
the transmission is unidirectional from teacher to student. Both of these
theories ignore the interaction between teachers and students, the active role
that students perform in the learning process, and the social contexts where
the learning process takes place (Tobias & Duffy, 2009; Wong, 2003).
Constructivism has been influential in education since the 1980s (Phillips,

2000; Tobin & Tippins, 1993). Generally speaking, constructivism views
learning as a process of active knowledge construction (Brooks & Brooks,
1993; Loyens & Gijbels, 2008; Steffe & Gale, 1995). That means, first, that
learners play an active role in learning processes. Second, knowledge is
constructed by learners with their own interpretations of contradictory
situations and by integrating new knowledge with prior experiences, which
suggests that knowledge is relative and subjective. According to this view,
one can be seen as knowledgeable if one’s knowledge is constructed from
experiences that are richer and full of variety (Jonassen, Peck, & Wilson,
1999). There are some variations of constructivism, such as cognitive
constructivism, radical constructivism, and social constructivism.
Nevertheless, these variations still have the consensus that learning is a
process of active knowledge construction. From the perspective of
constructivism, teachers are no longer authoritative, and nor is knowledge.
Learning environments are seen to be effective when teachers pay more
attention to the process of knowledge construction rather than the direct
transmission of particular information, and when students are encouraged
to be responsible for their own learning. Learning environments are also
76 Chapter Three
considered to be effective when interaction and cooperation during the

learning process are supported.
Due to the rapid development of information technology which allows

knowledge to be easily transmitted outside educational settings, students
have new learning needs. This means that the educational principles in
constructivism are definitely significant (Duffy & Jonassen, 1992a). There
have been a considerable number of instructional models and teaching
strategies which are underpinned by basic constructivist ideas, such as the
student/learner-centered approach, cooperative learning, problem-based
learning, situated learning, portfolio-based learning, experiential learning,
and inquiry-based teaching (Baeten et al., 2008; Lave & Wenger, 1991;
Loyens, Rikers, & Schmidt, 2009; Savery & Duffy, 1996; Slavin, 2003).
Even instructional technology that was initially primarily based on
cognitivism has begun to be combined with some constructivist principles
(Duffy & Jonassen, 1992b). However, despite its popularity today,
constructivism is still criticized due to the lack of practical and testable
instructional models/teaching methods with solid empirical support (De
Corte, 2000; Loyens & Gijbels, 2008; Tobias & Duffy, 2009). This is
especially the case because of the dearth of appropriate testable learning
outcomes to indicate the effectiveness of a constructivist learning
environment (O'Connor, 1998).
Some scholars argue that traditional learning theories cannot be totally

abandoned. The effectiveness of different learning theories depends on the
different learning goals and learning subjects involved (e.g., Jonassen, 1999;
Loyens, Rikers, & Schmidt, 2008). For example, cognitivism is more
applicable to novices, while constructivism is more applicable to experts or
students who have an abundant knowledge base in a particular field
(Dinsmore, Alexander, & Loughlin, 2008; Kirschner, 2009; Loyens et al.,
2008). In addition, a constructivist learning environment is believed to be
more appropriate if the learning goals are relevant to problem solving or
thinking skills (Jonassen, 1999). Given the nature of the research
participants (i.e., university students) and the dependent variables (i.e.,
thinking styles, career decision self-efficacy, and subjective well-being)
involved in the present research, it is appropriate to summarize the features
of an effective learning environment mainly based on constructivism’s
educational principles because, first, university students already have a
knowledge base to process high-quality thinking, and second, the dependent
variables are close to the thinking process and problem-solving skills rather
than traditional cognitive assessment.
Based on the educational principles derived from constructivism, as well as

a review of several existing inventories that were designed for assessing the
dimensions of an effective learning environment, such as the Student
Engagement Questionnaire (Kember & Leung, 2009b), the Experiences of
Teaching and Learning Questionnaire (Entwistle, McCune, & Hounsell,
2003), the College Student Experience Questionnaire (Pace & Kuh, 2007),
and the Inventory of Perceived Study Environment (Wierstra, Kanselaar,
Linden, & Lodewijks, 1999), four common characteristics were found to be
the features of an effective learning environment.
1) Constructivist learning
As mentioned earlier, one of the major contentions in constructivism is that

learning is a process of knowledge construction rather than the simply
transmission or reproduction of knowledge. Based on this contention, De
Corte (1995, 2000) emphasized that teachers should create a powerful
environment that could encourage students’ high-quality thinking. Collis
and Winnips (2002) also pointed out that it is important to make sense of
the subject matter and help students construct mental models that could be
applicable to new fields and situations. Similarly, Moreno and Mayer (1999)
stressed the significance of helping students to understand the deep structure
and the process underlying the subject matter. Therefore, teachers and
educators’ major focus should be making students understand how to
construct knowledge instead of merely teaching them to memorize
information. Meanwhile, students should be provided with abundant
opportunities to experience the process of knowledge construction in a
variety of helpful activities, such as case studies, interest groups, internships,
and seminars.
2) Student autonomy
The second major contention of constructivism is that learners play an

active role in the learning process. As Vermunt (2003) stated, making
students responsible for the learning processes could facilitate the
development of self-directed learning and independent thinking. De Corte
(1995, 2000) also supported the view that encouraging students’ self-
regulation in studying is an important factor in an effective learning
environment. Therefore, it is important for teachers and educators to allow
students to have autonomy in their learning, to make them participate in the
teaching process, and to help them to cultivate self-study habits.
78 Chapter Three
3) Interaction and cooperation
The third major contention of constructivism is that teachers are no longer

sages. They are just facilitators, and so are peers. According to Vygotsky’s
(1978) theory on “scaffolding”, which is also one of the foundations of
constructivism, appropriate help from teachers and peers is beneficial to
student development. This type of appropriate help can be largely facilitated
by interactions between teachers and students and cooperation among
students (Pontecorvo, 1993). Furthermore, interaction and cooperation are
also conducive to cultivating student initiative in the learning processes and
encourage their critical reflection on knowledge (De Corte, 1995, 2000; Van
Merriënboer & Paas, 2003). Therefore, an effective learning environment
should encourage interactions with teachers and cooperation with peers.
4) Clear goals and coherence of curricula
Many criticisms on the efficacy of constructivism are about its minimal

instruction strategy. However, this understanding is a misinterpretation of
constructivism (Tobias & Duffy, 2009). Actually, learning goals are
considered to be a driver for student learning and they are also helpful in the
process of knowledge construction (Tobias & Duffy, 2009). Furthermore, if
courses proceed in a way that is oriented by the learning goals and if they
make sense to students, they can be seen to be a good demonstration of the
process of knowledge construction. As Broekkamp and his colleagues (2002)
pointed out, clear goals and coherence of curricula have a significant
influence on students’ learning process, such as the learning strategies they
adopt. In fact, the factor concerning clear goals and coherence of curricula
has been frequently included in learning environment inventories (e.g.,
Entwistle et al., 2003; Kember & Leung, 2009b; Ramsden, 1991; Wierstra
et al., 1999). Therefore, it is reasonable to see clear goals and coherence
within a curriculum as essential characteristics of effective learning
environments.
The measure that assessed the learning environments in the present research
aimed to reflect the above four characteristics. Meanwhile, this measure was
supposed to be able to assess students’ perception of the general learning
environment rather than just a specific course. However, to the best of my
knowledge, most of the existing inventories on learning environments were
designed to assess learning environments within particular courses.
Furthermore, many inventories focus on a limited range of dimensions in
teaching, such as course design, assessments, and teaching methods, which
have been criticized for being too teacher-centered (e.g., Centra, 1993;
Kolitch & Dean, 1999). Fortunately, some inventories that have been
recently developed have noticed other dimensions in learning environments.
For example, Pace and Kuh (2007) designed a questionnaire to assess
students’ experience of activities on campus, while Wierstra, Kanselaar, van
der Linden, & Lodewijks (1999) added the dimension of student
participation in their inventory. Others included teacher-student and
student-student relationships (e.g., Entwistle, McCune, & Hounsell, 2003;
Kember & Leung, 2009). However, although these inventories succeeded
in expanding the breadth of the dimensions included in the questionnaires
that assess the learning environment to some extent, none of them was
comprehensive enough to provide a systematic examination of the general
learning environment. Therefore, given that the existing inventories on
learning environments are either too class-specific or only include a limited
range of dimensions, it became meaningful to develop an encompassing
learning environment inventory that can assess students’ general perception
of various learning dimensions both inside and outside classroom.
Therefore, I developed a comprehensive inventory (Inventory of Students’

Perceived Learning Environment, ISPLE) to meet the needs of the present
research. The contents of the inventory concern all of the four common
features of a constructivist learning environment (as summarized above).
Furthermore, the dimensions of the inventory were based on Entwistle,
McCune, and Hounsell’s (2003) conceptual model of teaching-learning
environments and on several existing inventories, including the Student
Engagement Questionnaire (Kember & Leung, 2009b), the Experiences of
Teaching and Learning Questionnaire (Entwistle, et al., 2003), the College
Student Experience Questionnaire (Pace & Kuh, 2007), and the Inventory
of Perceived Study Environment (Wierstra et al., 1999). Dimensions that
were frequently found in the factor analysis of these inventories and which
fit into Entwistle, McCune, and Hounsell’s (2003) conceptual model were
selected. In addition, extra dimensions focusing on learning environments
outside the classroom were added. Eventually, eight dimensions were
identified: 1) constructivist-oriented teaching; 2) clear goals and coherence
of curricula; 3) student autonomy; 4) assessments and assignments; 5)
teacher-student interaction; 6) student-student cooperation; 7) peer morale
and identities; and 8) learning facilities. The inventory was validated in a
pilot study. Further details about this inventory can be seen in Appendix 1
and Appendix 2.
80 Chapter Three
Current findings
In the literature it has been shown that a constructivist learning environment
usually aims to encourage students to experience the process of knowledge
construction through exploration, as well as to be autonomous in their
learning. These objectives are in line with the characteristics of Type I styles
that manifest autonomy, initiative, and creativity, but they are contrary to
the characteristics of Type II styles that are instruction-conforming and
conservative. Therefore, it was hypothesized that, statistically, constructivist
learning environments would positively contribute to Type I thinking styles
(Hypothesis 1) and negatively contribute to Type II thinking styles
(Hypothesis 2). As the nature of Type III styles is context-dependent, no
specific hypothesis concerning Type III styles was made. The research
findings revealed that some of the above hypotheses were supported while
others were not.
Hierarchical regressions were conducted to explore the predictive power of

learning environments for students’ thinking styles at both Time 1 and Time
2 with relevant demographic factors being taken into consideration (Table
3.11 and Table 3.12). The unique contribution of learning environments to
thinking styles ranged from 3–30% at Time 1 and 6–31% at Time 2. As
expected, the dimensions of the learning environments that manifested
constructivist characteristics positively predicted Type I thinking styles.
However, contrary to our expectations, some of the dimensions also
positively contributed to Type II styles. In both sets of data, the dimension
of constructivist-oriented teaching and the dimension of peer morale and
identities were the main contributors to Type I thinking styles, while the
dimension of clear goals and coherence of curricula and the dimension of
assessment and assignment were the major contributors to Type II thinking
styles. In addition, five dimensions of learning environments were involved
in predicting particular Type III thinking styles at both Time 1 and Time 2,
where student-student cooperation was the major contributor. The results of
Time 1 and those of Time 2 mainly differed in two ways. First, besides its
contribution to Type II styles, the dimension of assessment and assignment
also contributed to four Type I styles at Time 1, whereas it only contributed
to one Type I style at Time 2. Second, the dimension of learning facilities
contributed to more thinking styles (including styles from all three types) at
Time 2 than it did at Time 1.
It should be noted that during the data collection of Time 1 in the main study,
freshmen were asked to recall their experiences in high schools while
juniors were asked about their university experiences. Separate analyses of
the relationships between perceived learning environments and thinking

styles at Time 1 for freshmen and juniors showed that the patterns of
relationships were similar to the ones found in the analyses that considered
freshmen and juniors as a whole sample. Given the limited space, detailed
information about specific relationships between learning environments and
thinking styles for freshmen and juniors has been omitted.
82 Chapter Three
Table 3.11: Predicting Thinking Styles from Learning Environments (Time 1)

TS
2
R Total 0.15 0.12 0.05 0.15 0.15 0.09 0.11 0.06 0.07 0.13 0.07 0.05 0.32
R2demo 0.04 0.03 0.02 0.05 0.01 0.00 0.02 0.03 0.02 0.01 0.02 0.03 0.02
R2LE 0.11 0.09 0.03 0.10 0.14 0.09 0.10 0.03 0.05 0.12 0.05 0.03 0.30
F 12.20*** 9.80*** 3.43*** 12.33*** 11.97*** 6.95*** 8.83*** 4.32*** 5.49*** 9.96*** 5.50*** 3.98*** 32.29***
df 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899 15,899
*
ȕcons .10
ȕgoal .16*** .16*** .11* .13**
ȕauto .11* .15***
ȕasse .12** .14*** .22*** .14*** .16*** .09* .11** .19*** .09**
ȕtsin -.10* .12* .10*
ȕssco .10* .14** -.19*** .37***
ȕpeer .09* .10* .12** .12** .15***
ȕfaci .10*
2 2
Notes: TS = thinking styles; R Total = the contribution of gender, grade, major, hometown, SES, and learning environments to thinking styles; R demo = the contribution of
gender, grade, major, hometown, and SES to thinking styles; R2LE = the unique contribution of learning environments to thinking styles; cons = constructivist-oriented
teaching; goal = clear goals and coherence of curricula; auto = student autonomy; asse = assessment and assignment; tsin = teacher-student interaction; ssco = student-student
cooperation; peer = peer morale and identities; faci = learning facilities; and *p<.05, **p<.01, ***p<.001.
Table 3.12: Predicting Thinking Styles from Learning Environments (Time 2)
TS
2
R Total 0.25 0.20 0.12 0.19 0.22 0.17 0.17 0.10 0.13 0.22 0.11 0.10 0.33
2
R demo 0.04 0.03 0.05 0.07 0.01 0.01 0.02 0.02 0.02 0.03 0.04 0.04 0.02
2
R LE 0.21 0.18 0.07 0.12 0.21 0.16 0.15 0.07 0.11 0.19 0.07 0.06 0.31
*** *** *** ** *** *** *** *** *** *** *** ***
F 19.72 15.42 8.20 13.95 17.42 12.04 12.28 6.63 9.08 16.65 7.27 6.61 29.74***
df 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728 12,728
ȕcons .20*** .18*** .12* .14*
ȕgoal .17*** .12* .15**
ȕauto
ȕasse .12** .23*** .10* .17***
ȕtsin
ȕssco .16** .16*** -.19*** .34***
ȕpeer .09* .13** .12** .10* .10*
ȕfaci .22*** .12** .09* .10* .26*** .14** .19*** .12**
Notes: TS = thinking styles; R2Total = the contribution of gender, grade, major, SES, and learning environments to thinking styles; R2demo = the contribution of gender, grade,
major, and SES to thinking styles; R2LE = the unique contribution of learning environments to thinking styles; cons = constructivist-oriented teaching; goal = clear goals and
coherence of curricula; auto = student autonomy; asse = assessment and assignment; tsin = teacher-student interaction; ssco = student-student cooperation; peer = peer morale
and identities; faci = learning facilities; and *p<.05, **p<.01, ***p<.001.
84 Chapter Three
Generally speaking, the results indicated that learning environments were

significantly related to students’ thinking styles. Specifically, the hypothesis
about the positive relationships between the constructivist learning
environment and Type I thinking styles was supported. However, contrary
to expectations, the constructivist learning environment was also positively
related to Type II thinking styles.
Some dimensions of the constructivist learning environment were found to

be statistical contributors to Type I thinking styles, such as constructivist-
oriented teaching and peer morale and identities. These results indicated
that, first, if students perceived their teachers to be stressing the process of
knowledge building, rather than knowledge accumulation, and helping them
think critically about what they have learned (constructivist-oriented
teaching), they were more comfortable when dealing with unstructured
tasks (legislative style), preferred to evaluate people and things around them
(judicial style), and tended to manage multiple tasks in an orderly way
(hierarchical style). It is possible that constructivist-oriented teaching
encourages students to practice Type I styles and thus facilitates their
development. In the qualitative data, the feature of constructivist-oriented
teaching was also frequently addressed by the interviewees when they
compared the learning environment in universities with the one in high
schools. They were impressed by teachers who teach courses in a
constructivist way. For example, as indicated by two of the interviewees:
Teaching is more flexible in university than in high school. University
teachers expand the range of knowledge beyond textbooks, which really
enriches our knowledge […] they link one topic to another naturally, which
made us interested. (ZYY)
University teachers point out an argument and explore it by analyzing the

situation nowadays, which is more interesting and insightful. (DJ)
Second, the results also indicated that students whose classmates and friends
were proactive in learning and interested in knowledge connection and
application (i.e. had higher scores in peer morale and identities) tended to
think critically (judicial style) and globally (global style), tended to
demonstrate open-mindedness (liberal style), and preferred unstructured
thinking (legislative style). Compared with peer influence, the contribution
of teacher-student interaction was tenuous, especially in the university
environment. Interviewees in the follow-up qualitative study claimed that
the frequency and the intensity of teacher-student interaction decreased
dramatically from high school to university. Instead, the time spent with
peers in learning occasions or daily life significantly increased. For example,
Most teachers just showed up during lectures. If you are not proactive
enough, it is then hard to approach them out of the class. The undergraduate
courses seem too simple and basic for teachers, and they are busy doing their
cutting-edge research. If we encounter difficulties in studying, we usually
need to seek help through the internet or discuss it with our peers. We seldom
look to teachers. (MSJ)
The communication among peers in high school mostly focused on things

like examination skills. This type of communication does not influence you
as a person. Even though sometimes the topic pertained to ideas about life
and things like that, it would not develop into a deeper level because of the
limited time. It is different in university. You can talk about everything with
your friends. It has a very free style. You are easily influenced by them. (MX)
Therefore, it is no wonder that students were influenced by peer morale and

learning identities. If their peers engage with proactive learning activities,
students are more likely to be inspired by them and perform similar
academic activities, which enable them to practice Type I styles.
Another dimension that was found to contribute to Type I thinking styles

was assessments and assignments. However, it is noted that this dimension
is also positively associated with Type II thinking styles. High scores in the
dimension of assessments and assignments mean that the evaluation of
performance requires a deep understanding of the subjects rather than
simply memorizing facts. It is understandable that this dimension was
positively related to Type I thinking styles that require more complex,
unstructured, and creative thinking. However, it was also positively related
to two Type II styles: the preference for focusing on detail (local) and
dealing with one task at a time (monarchic style). This result may indicate
that memorizing facts is a necessary step in the process of deep
understanding to some extent (Hess & Azuma, 1991). These Type II
thinking styles seem to be helpful for memorizing tasks. Therefore,
assessments and assignments that require for deep understanding of certain
subject matters seem to provide opportunities for students to practice both
Type I and Type II styles. In addition, as some of the interviewees
mentioned, although students needed to have a deep understanding of the
subject to receive a good mark in certain assessments, they often also
included some items that assess basic information, which just required
memory skills. In this type of assessment, if the students are not motivated
to get excellent marks, they can still get a reasonable score if they answer
the information recitation questions successfully.
86 Chapter Three
The dimension of clear goals and coherence of curricula contributed more

to Type II thinking styles. It was anticipated that clear goals and coherence
of curricula would provide a good demonstration for students about the
process of knowledge building. The results did find this demonstration
helped students to develop a tendency to manage their lives in a more
organized manner (i.e., the hierarchical style, one of the Type I styles).
However, the predictive power of clear goals and coherence of curricula for
Type II styles (e.g., local and conservative) also suggests that students in a
learning environment where teachers clearly explain the course aims and
made sense of the teaching process to achieve these objectives tended to pay
attention to detail (local style) and to adhere to existing rules or use familiar
ways to deal with tasks (conservative style). It is possible that students using
the local or the conservative styles prefer to choose courses that have clear
goals and coherent sessions, but it is also possible that clear instruction
providing the big picture for the course means that they do not bother
focusing on general issues. Instead, they focus on details. Furthermore,
students tend to follow the process of learning that teachers present rather
than attempting novel ways. If this is the case, teachers should provide
appropriate levels of support because too much may lead students to step
back and become “lazy”, which stops them from exploring by themselves.
Future research should closely examine the detailed process (e.g., the
amount, the timing, and the context) of how course instruction should be
implemented in the classroom, in order to achieve a deep understanding of
how to construct clear goals and present coherence in the curricula.
Apart from the above dimensions, it was also found that the dimension of
student-student cooperation played a more salient role in their Type III
styles. The results indicated that students who perceived more support and
cooperation from classmates tended to select their tasks based on others’
suggestions (oligarchic style) and to work with others (external style) rather
than to do things alone (internal style). This finding makes sense because
students who experienced more student cooperation would value others’
suggestions to a greater degree and would be more willing to work with
others. Nonetheless, an alternative explanation could be that students with
the oligarchic and external styles rely more on others, which leads them to
interact more with others; this then enables them to perceive more student-
student cooperation than their counterparts. Students with the internal
thinking style are more independent, so they tend to interact with others less,
which would make them perceive less student-student cooperation than
their counterparts. Although the two explanations both seem tempting, the
interview data provided plausible evidence to support the former
explanation. For example, as one interviewee stated,
I was quite introverted before, mostly because I was not confident enough, I
think. During university life, after experiencing many student cooperation
projects and class discussions, I found that even if some students could not
express themselves very well or their arguments did not make sense, they
were still able to confidently give their opinion and other people listened to
them with great tolerance. I gradually realized that I can do this too. (ZJL)
One dimension that was found to contribute to all types of thinking styles
(but only at Time 2) was learning facilities. In terms of learning facilities,
both quantitative data and qualitative data indicated that learning facilities
in universities were significantly more abundant than in high schools.
Therefore, it is not surprising that the associations between learning
facilities and thinking styles were weak based on the analysis of the first
wave of the data set, which included the data of high school environments.
However, the analysis of the second wave of data set, where the research
totally focused on one-year experience in university, found that learning
facilities played a salient role in all three types of thinking styles. This may
be because the nature and features of the available resources and
opportunities on campus are diverse. The diversity of learning facilities and
the different ways that students used them provided them with opportunities
to develop diverse thinking styles. For example, some students may use
library resources to download information, while some others may use them
to design a project. Some laboratories may provide students with
opportunities to apply their knowledge as research assistants and invite them
to participate in the process of knowledge construction, while others may
only ask them to maintain equipment and clean. In order to achieve a better
understanding of how specific learning facilities influence students’
thinking styles, further research is needed to refine the exploration of
different learning facilities and interrogate the ways in which students use
resources and opportunities, rather than merely examining their availability.
In summary, the above findings demonstrated that the general constructivist

learning environment played an important role in students’ thinking styles.
Specifically, constructivist-oriented teaching, as well as peer morale and
identities in learning, were positively associated with thinking styles that
are characterized by cognitive complexity, nonconformity, autonomy, and
low degrees of structure (characteristics of Type I styles), while assessments
and assignments oriented to deep understanding, as well as learning
facilities, had positive associations with both Type I (creativity-generating
styles) and Type II styles (norm-conforming styles). Clear goals and
coherence of curricula had positive relationships with Type II styles while
student-student cooperation played a statistically significant role in Type III
88 Chapter Three
styles. Although some of the associations were in unexpected directions, the

above findings illustrated the important role of learning environments in
students’ socialization of thinking styles.
It is necessary to note that, the results from the follow-up interviews also
indicated that there were other factors in university life beyond the range of
dimensions examined in the quantitative study which also exerted an
influence on their thinking styles. These include experiences in student
associations, internships outside campus, and the management of daily life
(see “The change of thinking styles” section in this chapter for more details).
The Competitive Influence of Parenting, Learning

Environments, and Personality on the Development
of Thinking Styles
In addition to parenting styles and learning environments, personality traits
were also found to play an important role in thinking styles based on the
results on the regressions of thinking styles on personality traits. The next
question is that, among parenting styles, learning environments, and
personality traits, which group of variables is the major predictor for
students’ thinking styles? There is a clear need to study the competing
predictive power of familial environments, learning environments, and
personal characteristics for two reasons.
First, nature versus nurture is a classic controversy in the field of

developmental psychology. It also seems to be an unavoidable issue when
the development of intellectual styles is examined. Although the present
research cannot directly clarify the issue of nature versus nurture, the
present results should provide some insight into the extent to which thinking
styles can be nurtured by comparing the predictive power of environmental
factors and relatively stable personality characteristics (i.e., personality
traits).
Second, during adolescence, students experience a transition with an

increasing demand for independence and autonomy (Wentzel & Battle,
2001), especially when they enter university and are physically away from
home. During this process, the importance of parenting in student
development is likely to be gradually challenged by schools and universities’
influence (B. K. Keller & Whiston, 2008). This consideration suggests the
importance of examining the competitive influence of familial environments
and learning environments on student development.
To explore their competitive predictive power, hierarchical regressions of

thinking styles on parenting styles, learning environments, and personality
traits were conducted, with the relevant demographic factors in control
(Tables 3.13 and 3.14). When the predictive power of the three key variables
was examined simultaneously, it was found that the unique contribution of
parenting styles to thinking styles ranged from 1% to 10% (Time 1) and
from 2% to 14% (Time 2); the unique contribution of learning environments
to thinking styles ranged from 2% to 20% (Time 1) and from 4% to 21%
(Time 2); and the unique contribution of personality traits to thinking styles
ranged from 2% to 19% (Time 1) and from 3% to 17% (Time 2). Parenting
styles were the strongest predictor for two thinking styles (executive and
oligarchic at Time 1; conservative and oligarchic at Time 2) in both data
sets. Learning environments had the strongest predictive power for three
thinking styles (local, anarchic, and external) at Time 1 and for eight
thinking styles (legislative, judicial, global, executive, local, monarchic,
anarchic, and external) at Time 2. In addition, personality traits were the
strongest predictor for eight thinking styles at Time 1 and for three thinking
styles at Time 2. Therefore, among the three key predictors, personality
traits were the major contributor to the majority of thinking styles at Time
1, whereas learning environments were the major contributor to the majority
of thinking styles at Time 2. The specific relationships of various
dimensions of parenting styles and learning environments to thinking styles
were similar to the ones identified earlier in the separate regression models
of each predictor.
90 Chapter Three
Table 3.13: Predicting Thinking Styles from Parenting Styles, Learning Environments, and Personality Traits (Time
1)
2
R Total 0.29 0.24 0.11 0.26 0.35 0.18 0.15 0.18 0.14 0.19 0.12 0.20 0.46
R2demo 0.04 0.03 0.02 0.05 0.01 0.00 0.02 0.03 0.03 0.01 0.02 0.03 0.02
2
R PS 0.03 0.04 0.02 0.01 0.07 0.08 0.03 0.06 0.04 0.09 0.03 0.01 0.10
R2LE 0.08 0.06 0.02 0.09 0.08 0.05 0.08 0.02 0.03 0.07 0.05 0.03 0.20
R2Pers 0.13 0.11 0.05 0.10 0.19 0.06 0.03 0.08 0.05 0.03 0.02 0.14 0.13
F 15.43*** 12.26*** 4.79*** 13.55*** 20.77*** 8.72*** 6.93*** 8.55*** 6.54*** 8.84*** 5.07*** 9.75*** 33.34***
df 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891 23,891
ȕinvol .07* .15*** .16*** .17*** .09* .08**
ȕpsy_au -.08* -.07* -.08** -.11** -.10** -.12*** -.10** -.12*** -.14*** -.08*
ȕsuper -.07* -.08* -.07*
ȕcons
ȕgoal .12** .13** .09* .10*
ȕauto .11* .18***
ȕasse .08* .09* .20*** .08* .13*** .08* .17*** .07*
ȕtsin -.11* .10*
ȕssco .10* -.14** .29***

ȕpeer .10* .10* .13**
ȕfaci .09*
ȕneuro -.09** .23*** .08* .23*** .10** .16*** .10** .09***
ȕextra .08* .12** -.13*** .13*** .11** -.27*** .41***
ȕopen .34*** .30*** .09** .31*** -.19*** -.09* .23***
ȕagree -.13*** -.10** -.21*** -.09** -.07* -.08* -.20*** .07*
ȕcient .15*** .10** 澳澳 .47*** .19*** .17*** .09* .19*** 澳澳 .23*** 澳
contribution of gender, grade, major, hometown, SES, parenting styles, learning environments, and personality to thinking styles;
R2demo = the contribution of gender, grade, major, hometown, and SES to thinking styles; R2PS = the unique contribution of parenting
styles to thinking styles; R2LE = the unique contribution of learning environments to thinking styles; R2Pers = the unique contribution of
personality to thinking styles; invol = acceptance/involvement; psy_au = psychological autonomy-granting; super =
strictness/supervision; cons = constructivist-oriented teaching; goal = clear goals and coherence of curricula; auto = student autonomy;
asse = assessment and assignment; tsin = teacher-student interaction; ssco = student-student cooperation; peer = peer morale and
identities; faci = learning facilities; neuro = neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness;
cient = conscientiousness; and *p<.05, **p<.01, ***p<.001.
92 Chapter Three
Table 3.14: Predicting Thinking Styles from Parenting Styles, Learning Environments, and Personality Traits (Time
2)
2
R Total 0.39 0.33 0.20 0.34 0.42 0.25 0.22 0.20 0.21 0.30 0.17 0.26 0.47
R2demo 0.04 0.03 0.05 0.07 0.01 0.02 0.02 0.03 0.03 0.03 0.04 0.04 0.02
2
R PS 0.06 0.06 0.03 0.05 0.09 0.08 0.06 0.08 0.06 0.14 0.04 0.02 0.13
R2LE 0.16 0.13 0.06 0.09 0.14 0.10 0.11 0.04 0.08 0.10 0.05 0.05 0.21
2
R Pers 0.13 0.10 0.06 0.13 0.17 0.05 0.03 0.06 0.05 0.03 0.03 0.15 0.12
F 19.32*** 14.94*** 7.45*** 15.57*** 22.18*** 10.02*** 8.51*** 7.78*** 8.39*** 13.35*** 6.27*** 10.84*** 27.69***
df 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709 23,709
ȕinvol .14** .10* .16*** .13** .18*** .10**
ȕpsy_au -.10** -.08* -.07* -.11** -.12** -.10** -.17*** -.12** -.13***
ȕsuper -.11** -.09*
ȕcons .12*
ȕgoal .13** .13*
ȕauto .10* .10*
ȕasse .10* .23*** .14**
ȕtsin
ȕssco .10* .17*** .14** -.13** .28***

ȕpeer .08* .13** .13*** .08*
ȕfaci .16*** .10* .25*** .15*** .09* .16*** .12**
ȕneuro .11** .08* .21*** .11** .23*** .10* .16*** .11** .13***
ȕextra .10* .10* .12** .08* -.09* .16*** .13** -.30*** .41***
ȕopen .37*** .30*** .17*** .33*** .11*** -.17*** .25***
ȕagree -.11*** -.13*** -.16*** -.17*** -.07* -.13*** .12*** -.13*** -.25*** .09**
ȕcient .12*** -.18*** .08* .42*** .17*** .15*** .19*** .13*** -.07*
contribution of gender, grade, major, hometown, SES, parenting styles, learning environments, and personality to thinking styles;
R2demo = the contribution of gender, grade, major, hometown, and SES to thinking styles; R2PS = the unique contribution of parenting
styles to thinking styles; R2LE = the unique contribution of learning environments to thinking styles; R2Pers = the unique contribution of
personality to thinking styles; invol = acceptance/involvement; psy_au = psychological autonomy-granting; super =
strictness/supervision; cons = constructivist-oriented teaching; goal = clear goals and coherence of curricula; auto = student autonomy;
asse = assessment and assignment; tsin = teacher-student interaction; ssco = student-student cooperation; peer = peer morale and
identities; faci = learning facilities; neuro = neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness;
cient = conscientiousness; and *p<.05, **p<.01, ***p<.001.
94 Chapter Three
The above analyses were based on cross-sectional data sets. In order to

provide more solid evidence for the causal relationships between
environmental factors and thinking styles, further analyses based on
longitudinal data were conducted to see how parenting styles, learning
environments, and personality traits influence the change of thinking styles
over one year.
There are two types of change models. One is called the unconditional
change model where the change of the dependent variable is predicted by
the independent variable at Time 1. This change model assumes that the
change of the dependent variable is irrelevant to the initial score of the
dependent variable, which is seldom true in reality. Therefore, the
conditional change score model that controls for the initial score of the
dependent variable is always preferred in research to capture the factors that
really cause the change.
It has to be noted that, although in longitudinal change models, the

independent variable assessed at the pretest (i.e., at Time 1) is usually used
as the predictor, this research used personality traits at Time 1 but parenting
styles and learning environments at Time 2 as predictors. This is because
the research question was whether the change in students’ thinking styles
was caused by parenting styles and their learning environments in the year
between the pretest and the posttest, or their initial personality traits. In the
present research, the parenting styles and learning environments that
students experienced in the year between the pretest and the posttest were
assessed by retrospective data, which means that, although the data for the
parenting styles and learning environments was collected at Time 2, it
actually represented students’ experience in the research interval from the
pretest to the posttest.
Therefore, to explore whether the change of thinking styles over the

research interval was because of the environmental factors that students
experience in the year between the pretest and the posttest or whether it was
due to their initial personality traits, regressions for the change of thinking
styles were conducted with parenting styles (Time 2), learning environments
(Time 2), and personality traits (Time 1) as the predictors (Table 3.15).
Among these regressions, thinking styles (Time 1) and relevant
demographic factors were controlled. The results showed that thinking
styles at Time 1 were negatively correlated with the change of thinking
styles, which means there was a “regression toward the mean” effect. This
effect was often found in longitudinal studies. It means high scorers at
pretest tend to have lower scores at posttest while low scorers at pretest tend
to have higher scores at posttest. (Taris, 2000). Although there was the
“regression toward the mean” effect, the parenting styles and the learning
environments that students experienced in the year between the pretest and
the posttest still had made unique contributions to the change of thinking
styles. Specifically, parental involvement contributed to the increase of
most thinking styles, while psychological autonomy from parents
contributed to a decrease in most thinking styles, which partially supported
the hypotheses. Parental supervision mainly decreased Type I thinking
styles, which was in line with expectations. To further explore the
interactions between parenting dimensions, the role of parenting styles in
the change of thinking styles was further examined based on the typology
of parenting styles by conducting MANOVAs. It was found that the types
of parenting styles that students experienced during the year between the
pretest and the posttest did not make any significant difference in the change
of thinking styles.
The results from the conditional change models (Table 3.15) about the
contribution of learning environments to the change of thinking styles
partially supported the relevant hypotheses. Specifically, constructivist-
oriented teaching and peer morale mainly promoted Type I styles, while
clear goals and coherence of curricula primarily increased Type II styles. In
addition, student-student cooperation mainly contributed to the change of
Type III styles, and learning facilities had an effect on all three Types of
thinking styles. This pattern was quite similar to the one found in the cross-
sectional data.
Furthermore, based on the percentages of the variance in thinking style

change explained by the predictors (i.e., R2), it was indicated that, among
parenting styles, learning environments, and personality, personality was
the major predictor for eight thinking styles while learning environments
was the major predictor for five thinking styles and contributed almost the
same as personality to another two styles (legislative and liberal). Although
parenting styles also made unique contributions to the change of thinking
styles, their predictive power was subtle. Therefore, the ranking of the
contributions of the three predictors for the change of thinking styles over
one year was personality traits at the initial stage, learning environments
experienced in the year between Time 1 and Time 2, and then parenting
styles experienced in the year between Time 1 and Time 2.
In general, when parenting styles, learning environments, and personality

traits were examined simultaneously, it was found that the three predictors
all had unique contributions to students’ thinking styles, but the predictive
96 Chapter Three
power differed. In the first wave of data collection, the predictive power of
parenting styles that students experienced in the first 16 years in their lives
was examined with previous experience in learning environments and
personality traits together. The results from this set of data indicated that the
parenting styles that students experienced in the first 16 years of their lives
were the major predictors for two of the 13 thinking styles (executive and
oligarchic). However, in the longitudinal data analysis where the influence
of parenting styles that students experienced in the year between the pretest
and the protest was examined with learning environments in the same
research interval and personality traits at the initial stage of the research, it
was found that parenting styles were not a major predictor for any of the
thinking styles. These findings indicated that, although parenting styles had
made unique contributions to students’ thinking styles, they were not
powerful predictors for thinking styles; in addition, the influence of
parenting styles on thinking styles decreased over one year. It is
understandable that the influence of parenting styles on students’
development decreased when students went into university education
because, during this period, they lived away from their parents and had
limited communication with them.
In terms of the predictive power of learning environments, it was shown that

the learning environment that students had experienced before the pretest
served as a major predictor for three of the 13 thinking styles, while the
learning environment that students experienced during the year between the
pretest and the posttest was the most responsible for the change of five
thinking styles over one year and was almost equally responsible as
personality traits for the change of another two thinking styles. The increase
in the predictive power of learning environments is probably because the
one that students had experienced before the pretest was relatively constant
and they were used to it, but the ones that they experienced during the year
between the pretest and the posttest contained many different features; this
was especially true for freshmen. In a changeable environment, students
may have to adapt themselves to meet the new demands of the environment
and, during this process, students may be more likely to make changes in
their thinking styles accordingly. In addition, the finding that learning
environments were more of a major predictor for thinking styles than
parenting styles is in accordance with the results that emerged from the
interview study that students attributed their change of thinking styles more
to learning environments than to parenting styles.
However, compared with parenting styles and learning environments,

personality traits had the strongest predictive power for the majority of
thinking styles, because personality traits were found to be the major

contributor for eight of the 13 thinking styles at Time 1 and another eight at
Time 2 after controlling for the initial scores of thinking styles. During the
interview study, when students were asked why their thinking styles
changed over the research interval, they mostly referred to factors in the
learning environment; however, when they were asked how their thinking
styles had formed in the first place, they mostly mentioned personality. For
example, two interviewees commented as follows:
It is difficult to imagine that an infant has thinking styles, but I think that we
have personalities at very early stage, which determine what we value.
During the process of growth, the surrounding circumstances provide
various tasks for us to complete. Personality matters when dealing with these
tasks. We respond to tasks based on our personality, which is sometimes
successful and sometimes fails. Gradually, based on our experiences, we
formed a preference for dealing with tasks. (XZY)
I think personality is the foundation for one’s thinking styles. If one is

extroverted, then they would be more likely to be external, proactive, and
liberal. However, if someone is introverted, they are more likely to be
conservative and to prefer doing things alone. (QBK)
Considering all of the above findings, I would like to address two points.
First, the specific relationships of the various dimensions of parenting styles
and learning environments to thinking styles found in longitudinal models
were similar to the ones identified earlier in the separate regression models
of each predictor based on the cross-sectional data. This means that the
relationships of parenting styles and learning environments to thinking
styles are more likely to be causal than occasional. Second, even though
personality traits served as the major predictors for the majority of thinking
styles, the two environmental factors (parenting styles and learning
environments) still made unique contributions beyond personality traits to
thinking styles, although the magnitude of the predictive power was not
large. It seems that personality traits provide a baseline of thinking styles
and the environmental factors then adjust them. These adjustments may not
be dramatic, but they do happen. This finding once again supported the
argument that styles are stable status rather than permanent traits.
98 Chapter Three
Table 3.15: Conditional Change Model of Thinking Styles
C- Type I Type II Type III

TS Leg Jud Glo Lib Hie Exe Loc Con Mon Oli Ana Int Ext
R2Total 0.49 0.43 0.41 0.42 0.44 0.51 0.45 0.44 0.44 0.47 0.37 0.41 0.51
R2demo 0.01 0.03 0.01 0.02 0.01 0.01 0.01 0.01 0.01 0.02 0.01 0.02 0.02
R2PS 0.01 0.01 0.01 0.02 0.02 0.01 0.02 0.01 0.01 0.02 0.01 0.01 0.01
R2LE 0.07 0.04 0.03 0.03 0.04 0.03 0.04 0.02 0.02 0.03 0.01 0.03 0.06
R2Pers 0.07 0.05 0.01 0.03 0.08 0.04 0.02 0.01 0.03 0.01 0.01 0.05 0.09
R2TS1 0.33 0.30 0.35 0.32 0.29 0.42 0.36 0.38 0.38 0.39 0.32 0.31 0.33
27.91 22.58 20.20 21.08 23.37 30.70 24.03 22.87 23.11 26.21 18.29 20.88 30.78
F *** *** *** *** *** *** *** *** *** *** *** *** ***
df 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708 24,708
Parenting Styles
ȕinvol .09** .11** .13*** .13*** .10** .10** .10** .11** .15*** .14***
ȕpsy_au -.09** -.09** -.08* -.06* -.11*** -.12*** -.11*** -.14*** -.13*** -.12***
ȕsuper -.07* -.11*** -.07* -.09** -.09**
Learning Environments
ȕcons .14** .17*** .11* .10*
ȕgoal .10* .09* .11*
ȕauto .09* .11**
ȕasse .16*** .11**
ȕtsin .11*
ȕssco .09* .11** .12** -.13** .27***
ȕpeer .11** .09*
ȕfaci .16*** .10* .09* .18*** .10* .14*** .09*
Personality
ȕneuro .08* .07*
ȕextra -.07* .07* -.08* .11**
ȕopen -.08** -.09** -.08** .07* -.07*
ȕagree -.07* -.07*
ȕcient .08* .07* .08*
TS
ȕtime1 -.66*** -.63*** -.62*** -.65*** -.67*** -.70*** -.64*** -.66*** -.67*** -.67*** -.59*** -.63*** -.73***
Notes: C-TS = Change of thinking styles; Leg = Legislative; Jud = Judicial; Glo = Global; Lib = Liberal; Hie = Hierarchical; Exe =
Executive; Loc = Local; Con = Conservative; Mon = Monarchic; Oli = Oligarchic; Ana = Anarchic; Int = Internal; Ext = External;
R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles at Time 2, learning environments at Time 2,
personality at Time 1, and thinking styles at Time 1 to the change of thinking styles from Time 1 to Time 2; R2demo = the contribution
of gender, grade, major, hometown, and SES to the change of thinking styles; R2PS = the unique contribution of parenting styles during
the year between the pretest and the posttest to the change of thinking styles; R2LE = the unique contribution of learning environments
during the year between the pretest and the posttest to the change of thinking styles; R2Pers = the unique contribution of personality at
Time 1 to the change of thinking styles; R2TS1 = the unique contribution of thinking styles at Time 1 to the change of thinking styles;
invol = acceptance/involvement; psy_au = psychological autonomy-granting; super = strictness/supervision; cons = constructivist-
oriented teaching; goal = clear goals and coherence of curricula; auto = student autonomy; asse = assessment and assignment; tsin =
teacher-student interaction; ssco = student-student cooperation; peer = peer morale and identities; faci = learning facilities; neuro =
neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness; cient = conscientiousness; and *p<.05,
**p<.01, ***p<.001. The number of R2 in bold means the F change was significant after the predictor was added into the regression
model.
CHAPTER FOUR
THE ROLE OF THINKING STYLES

IN STUDENT DEVELOPMENT
Another major controversial issue in the field of intellectual styles is

whether styles are value-laden (Zhang & Sternberg, 2005). In other words,
are some styles more adaptive than others? This issue is unavoidable when
researchers explore the socialization of styles, because findings on the ways
in which particular styles can be developed would only be meaningful when
there are particular styles that we should aim to develop. The issue of style
value has been hotly debated. Some scholars (e.g., Kwang & Rodrigues,
2002; Messick, 1994; A. Miller, 1987; Riding, 1997; Eugene Sadler-Smith,
2009) have argued that the value of a certain style can be either positive or
negative, depending on the specific contexts where it is used. Other scholars
(e.g., Kogan, 1989; Messer, 1976; Zhang, 2012; Zhang & Sternberg, 2009)
have claimed that some styles manifest more positive value than others in
most situations.
The value of styles can be reflected through the line of research that
examines the relationships between styles and individual developmental
outcomes. If some specific styles have been very often found to play more
positive roles than others in individual developmental outcomes, then styles
can be regarded as value-laden. If styles have been found to play a diverse
role in different contexts, they can be considered to be value-differentiated.
In this chapter, evidence emerging from research on the relationships

between styles and academic development is reviewed in the first section.
The second section reviews previous studies examining the role of styles in
individuals’ non-academic development, which is followed by the present
research’s findings on the role of thinking styles in students’ career decision
self-efficacy and subjective well-being.
The Role of Thinking Styles in Student Development 101
Thinking Styles and Academic Development

Previous empirical studies that addressed the issue of style value mainly
focused on the relationships between styles and students’ cognitive
developmental outcomes, especially academic achievement. However, the
findings are complex. Some studies found that students with Type I
intellectual styles outperform students with Type II intellectual styles. For
example, field-independent (Type I) students were found to have better
academic performances (e.g., Bagley & Mallick, 1998; Cameron & Dwyer,
2005; Hite, 2004), better problem-solving performances (Williams, 2001),
and better programming performance s(Johnson & Kane, 1992; D. J. Wilson
et al., 1990) than their field-dependent (Type II) counterparts. Similar
results were also found regarding Biggs’s (1978) learning approaches and
Kagan and his colleagues’ (1964) reflective-impulsive styles (Chamorro-
Premuzic & Furnham, 2008; Fisher, 1994; Furnham, Christopher, et al.,
2007; Stahl et al., 1986), where Type I styles (i.e., the deep learning
approach and the reflective style) positively contributed to academic
outcomes. In contrast, some other studies found that students with Type II
intellectual styles outperform students with Type I intellectual styles. For
example, some scholars found that, in some disciplines (e.g., social sciences,
literature, chemistry, and geography), students with Type II styles (e.g.,
field-dependent, surface learning approach) had higher scores in academic
tests (Varma & Thakur, 1992; Zhang, 2000). Fan and He (2012) reviewed
eight style models and their relationships with academic achievement. They
found that, although previous studies confirmed the unique contributory role
of intellectual styles in academic achievement even beyond abilities and
personality, the specific relationships between styles and academic
achievement were complicated. Inconsistent findings were often found
among some multidimensional models, such as Kolb’s (1976) learning
styles, Gregorc’s (1979) mind styles, and Riding and Cheema’s (1991)
cognitive styles. For example, O’Brien (1994) found that concrete
sequential types were superior to other types in Gregorc’s (1979) mind
styles in terms of students’ learning performance, while in Miller’s (2005)
study, it was found that concrete sequential students actually perform worse
than concrete random students.
The research findings are also complex in terms of thinking styles. On the
one hand, Type I thinking styles (e.g., legislative, judicial, and hierarchical)
were found to play more positive roles in academic achievement in contrast
to Type II thinking styles (e.g., executive and conservative) in some studies
(W. Fan et al., 2010; Grigorenko & Sternberg, 1997; Zhang, 2004d). On the
other hand, some studies conducted in Hong Kong, the Philippines, and
102 Chapter Four
Spain found that some Type II thinking styles (e.g., the conservative,
executive, and local styles) are positively related to academic achievement
(Bernardo et al., 2002; Cano-Garcia & Hughes, 2000; Varma & Thakur,
1992; Zhang, 2002e, 2004b, 2007b).
These inconsistent results are probably due to the different cultures,

different disciplines, different learning materials, and different school levels
involved in these studies (Zhang & Sternberg, 2006). For example,
Grigorenko and Sternberg (1997) found that the legislative thinking style
and the judicial thinking style (both Type I styles) were positively
associated with academic performance while the executive thinking style
(Type II) was negatively associated with academic performance in western
samples. However, the findings came to the opposite conclusion in Zhang
and Sternberg’s (2000) research, which was conducted among Chinese
populations. In this research, Type I thinking styles were found to play a
negative role in academic performance, while Type II thinking styles (e.g.,
executive and conservative) were found to play a positive one. These
inconsistent findings may have something to do with the different cultures
in western and eastern societies. Nevertheless, it needs to be noted that this
inconsistency could also be attributed to the different requirements of
learning in different stages of education, because the western samples in
Grigorenko and Sternberg’s (1997) research were college students, while
the Chinese samples in Zhang and Sternberg’s (2000) research were
secondary school students. This speculation was partially supported by Fan
and his colleagues’ (2010) study. They found that Chinese university
students with Type I thinking styles also tend to outperform students with
Type II thinking styles, which is similar to their counterparts in western
colleges.
However, it is still difficult to directly compare the findings from different

studies on the relationships between styles and academic performance. First,
this is because most of these studies directly used GPA or course grades as
indicators of academic performance, and the reliability of these indicators
largely depended on the quality of the specific assessments adopted in these
schools or colleges. Second, the diversity of the assessment design, even
within the same subject, may influence the relationships between styles and
academic performance found in relevant studies. For example, Fan and his
colleagues (2010) found that Type II thinking styles (e.g., executive, local,
conservative, and monarchic) were more likely to be negatively related to
achievements if the tests used for assessing achievements consisted of
analysis and problem-solving tasks. Therefore, it is inappropriate to directly
compare the relevant findings across different studies on the relationships
between styles and academic performance if the assessments for academic

performance are not standardized tests.
Compared with the evaluation of academic performance, the evaluation of

non-academic developmental outcomes (e.g., self-esteem, emotional
competence, social competence, and psychological well-being) seem to be
relatively more comparable, because they are usually assessed by unified
inventories and their reliability and validity are explicitly depicted in the
research. However, studies addressing the relationships of styles to non-
academic outcomes are not as abundant as those examining the relationships
between styles and academic outcomes.
Thinking Styles and Non-academic Development 4

Although academic outcomes are dominant concerns in education,
increasing numbers of educators are arguing that we should also place
importance to non-academic outcomes. However, studies examining the
role of styles in individuals’ non-academic development are still far from
adequate (Gebbia & Honigsfeld, 2012). This prevents a comprehensive
understanding of the value of intellectual styles in student development.
This section begins with the general review of the existing research on the
relationships between intellectual styles and non-academic outcomes. After
that, career self-efficacy and its relationship with styles are highlighted,
which is followed by relevant findings from the present research. I will then
discuss subjective well-being and its relationship with styles based on
previous empirical evidence before presenting the relevant findings from
the present research. At the end of this section, the research results on the
mediating role of thinking styles in these particular student developmental
outcomes are presented.
Intellectual styles and non-academic outcomes

While they are limited in number, most of the existing studies on the
relationships between intellectual styles and non-academic outcomes
consistently indicated that Type I styles manifest positive value. For
example, Type I styles (e.g., field independence, deep learning approach,
innovation, the legislative style, and the liberal style) were consistently
found to be associated with more positive self-concept or higher self-esteem
4The section draws partially on my work titled, “The role of thinking styles in career
decision-making self-efficacy among university students”, published in Thinking
Skills and Creativity (Fan, 2016).
104 Chapter Four
(e.g., Bosacki et al., 1997; Phan, 2007; Tierney, 1997; Watkins & Dahlin,
1997). Moreover, field-independent (Type I) people were found to have
higher levels of job satisfaction (Hageman, 1990). In contrast, some studies
revealed that Type II styles played a negative role in certain social and
emotional outcomes. For example, Type II styles (e.g., field-dependent,
impulsive) were also easily found among patients with depressive symptoms,
borderline personality disorder, and Attention Deficit and Hyperactivity
Disorder (Calamari et al., 2000; López-Villalobos et al., 2003).
Research specifically on thinking styles also reveals the same pattern. On

the one hand, Type I thinking styles (e.g., the legislative and liberal styles)
were found to be positively related to self-esteem (Zhang, 2001d; Zhang &
Postiglione, 2001). Furthermore, Type I thinking styles were also correlated
with low levels of anxiety (Zhang, 2009), proper emotion management
(Zhang, 2008b), good mental health (G.-H. Chen & L. F. Zhang, 2010), and
positive psychosocial development (Zhang, 2008c, 2010; Zhang & He,
2011). On the other hand, Zhang (2009) found that the conservative style
(one of the Type II thinking styles) was associated with high levels of
anxiety. Zhang (2010, Zhang & He, 2011) also found that Type II thinking
styles (e.g., conservative and monarchic) were negatively related to
psychosocial development. Moreover, Type II thinking styles (e.g., local
and monarchic) were also found to be associated with psychological
problems (Chen & Zhang, 2010).
In terms of Type III styles, previous studies showed that Type III thinking
styles play a positive role in some developmental outcomes sometimes but
a negative role in others (e.g., Zhang, 2001d, 2008b; Zhang, 2010). These
findings resonate with Zhang and Sternberg’s (2006) contention that Type
III styles are value-differentiated, because they manifest the nature of either
Type I or Type II styles depending on the specific context.
Generally speaking, compared with studies on the relationships between

intellectual styles and academic outcomes, the existing studies on the
relationships of intellectual styles to non-academic outcomes have obtained
more consistent results, which show that Type I styles tend to play a positive
role whereas Type II styles tend to play a negative one. In addition, Type
III tend to be value-differentiated. However, given the limited numbers of
studies on the role of intellectual styles in individuals’ non-academic
development, the prior findings are insufficient to draw a definitive
conclusion concerning the value of styles. The coverage of non-academic
outcomes examined in this type of research needs to be expanded in order
to provide additional knowledge and deepen our understanding of the style
value issue. In addition, it is claimed that thinking styles can be applied to

non-academic settings as well as academic settings (Zhang & Sternberg,
2005): that is, thinking styles can influence not only one’s performance in
educational settings but also one’s performance in other settings, such as
work and daily life. This theoretical argument also warrants the further
examination of the role of thinking styles in one’s career development and
quality of life. As university students need to be prepared for the future
world of work, developing an understanding of their career development
has a significant role in decreasing career-relevant problems later on in their
lives (B. K. Keller & Whiston, 2008). In addition, students’ psychological
well-being and mental health, which has a profound influence on their lives,
is also being given more value than previously in education (Noddings,
2005). Therefore, the present research selected career decision self-efficacy
and subjective well-being as the outcome variables in order to shed light on
how to optimize student development in terms of career readiness and
psychological well-being, as well as to achieve a better understanding of the
style value issue.
Career decision self-efficacy and its relationship

with intellectual styles
As Bandura (2006) stated, career decision self-efficacy has been demonstrated
to be a salient variable in the field of career development since it was first
proposed by Taylor and Betz (1983). Taylor and Betz’s (1983) theory is an
integration of Crites’s (1961, 1965) career maturity model and Bandura’s
(1977) self-efficacy theory. Taylor and Betz (1983) applied the concept of
self-efficacy to career choice competencies and defined career decision self-
efficacy as individuals’ beliefs about how well they can perform career
choice tasks. At the same time, they borrowed the structure of career choice
competencies from Crites’s (1961, 1965) career maturity model and
designed an inventory to assess career decision self-efficacy according to
its five aspects: a) gathering occupation information, b) accurate self-
appraisal, c) goal selection, d) making plans, and e) problem solving (Crites,
1961, 1965). In the field of career research, it is believed that career decision
self-efficacy is a good indicator of career maturity (Long, Fang, & Ling,
2000).
The theory of career decision self-efficacy, which was derived from the
integration of Bandura’s (1977) self-efficacy theory and Crites’ (1961, 1965)
career maturity theory, and its subsequent research have contributed to the
development of the social cognitive career theory (SCCT), which was
proposed by Lent, Brown, and Hackett (1994). As one of the integral social
106 Chapter Four
cognitive variables in SCCT, self-efficacy, together with outcome

expectations, interests, and choice goals, are seen as powerful predictors for
career outcomes, such as choice actions and performance. As a
comprehensive model of career development, SCCT has stimulated
abundant studies on self-efficacy within the realm of career research (Lent,
2005), including ones exploring the personal antecedents of career decision
self-efficacy.
Many studies have found that personal factors, such as gender, ethnicity,
social status, personality, identity, and self-reliance, play a role in individual
differences in career decision self-efficacy (e.g., Gianakos, 2001; Gloria &
Hird, 1999; Hartman & Betz, 2007; Nauta & Kahn, 2007; Thompson &
Subich, 2006). For example, in terms of one of the widely studied personal
inputs––personality traits––Hartman and Betz (2007) found that
conscientiousness and extraversion were positively related to career
decision self-efficacy, while neuroticism was negatively related. Similar
results were also found in Page, Bruch, and Haase’s (2008) study and Jin,
Watkins, and Yuen’s (2009) research. However, although intellectual styles
are seen as the interface between personality and ability (Zhang & Sternberg,
2005), their role in career decision self-efficacy has been under-researched.
Although little research has directly examined the relationships between

intellectual styles and career decision self-efficacy, there are still a few
studies that have considered the relationships between styles and some types
of career content self-efficacy. Among these studies, Holland’s (1973)
theory of career personality types (a model of intellectual styles) was the
most frequently used; this is probably because the six career personality
types proposed in it have been treated as representatives for career interests
in the field of career development. These studies found moderate
correlations between career interests in the six domains proposed in
Holland’s theory (i.e., realistic, investigative, artistic, social, enterprising,
and conventional) and content/task-specific self-efficacy, such as science
self-efficacy, math self-efficacy, or specific self-efficacy in the same six
domains (Rottinghaus, Larson, & Borgen, 2003). The types of self-efficacy
mentioned in the above study were classified as belonging to the content
domain of career self-efficacy by Hacket and Betz (1981). Although this
study shed light on the role of intellectual styles in career decision self-
efficacy, it did not provide much information on the value of intellectual
styles because the variables (i.e., career domains, such as science, math,
artistic, and enterprising domains) involved in these studies cannot be easily
judged as one being more adaptive than another, thereby making them poor
indicators for the value of intellectual styles. Therefore, process self-
efficacy, such as career decision self-efficacy, needs to be examined with

intellectual styles in order to achieve a better understanding of the value of
intellectual styles as well as the development of career maturity. In the
process domain of career self-efficacy, higher career decision self-efficacy
implies better levels of career maturity, which could be a good indicator of
style value. Therefore, in the present research, career decision self-efficacy
was examined with thinking styles.
Current findings: Thinking styles and career decision self-

efficacy
Given that Type I thinking styles were consistently found to be positively
related to other social and emotional developmental outcomes (e.g., self-
esteem, psychological well-being, and psychosocial development) and Type
II thinking styles negatively related, the present research hypothesized that
Type I thinking styles would be positively related to career decision self-
efficacy, while Type II thinking styles would be negatively related. No
hypothesis was made in terms of Type III styles because, according to
previous studies, their value might differ based on specific contexts.
I analyzed the relationship between thinking styles and career decision self-
efficacy based on two sets of data (Time 1 and Time 2) to see if the
relationship between the two variables remained the same across time.
Furthermore, to exclude the possible influence of confounding variables, the
relationship between thinking styles and career decision self-efficacy was
examined with parenting styles, learning environments, personality traits,
and relevant demographic factors (i.e., gender, grade, major, hometown, and
SES) as controls. The results based on hierarchical regressions showed that
thinking styles had made unique statistical contributions to CDSE beyond
parenting styles, learning environments, and personality traits. Specifically,
thinking styles uniquely explain 7% to 9% of the variance in CDSE at Time
1 (Table 4.1) and 2% to 4% of the variance in CDSE at Time 2 (Table 4.2).
108 Chapter Four
Table 4.1: Predicting Career Decision Self-Efficacy from Thinking Styles with Parenting Styles, Learning
Environments, Personality Traits, and Relevant Demographic Factors in Control (Time 1)
Occupational Self- Goal Problem

CDSE Planning
Information Appraisal Selection Solving
R2Total 0.34 0.42 0.36 0.40 0.39
2
R demo 0.04 0.05 0.04 0.03 0.04
2
R PS 0.08 0.06 0.05 0.06 0.06
2
R LE 0.09 0.09 0.08 0.08 0.09
2
R Personality 0.07 0.14 0.12 0.15 0.13
2
R TS 0.07 0.09 0.07 0.09 0.08
F 12.79*** 17.96*** 13.58*** 16.23*** 15.27***
df 36,878 36,878 36,878 36,878 36,878
Type I ȕleg .12** .10*
ȕjud
ȕglo .12***
ȕlib .10* .09* .10* .13**
ȕhie .17*** .08* .22*** .12**
Type II ȕexe
ȕloc .08* .11**
ȕcon
ȕmon -.08*
Type III ȕoli
ȕana
ȕint .12** .11*
ȕext .11* .09* .10*
Notes: CDSE = career decision self-efficacy; Total = the contribution of gender, grade, major, hometown, SES, parenting styles,
R2
learning environments, personality traits, and thinking styles to career decision self-efficacy; R2demo = the contribution of gender, grade,
major, hometown, and SES to career decision self-efficacy; R2PS = the unique contribution of parenting styles to career decision self-
efficacy; R2LE = the unique contribution of learning environments to career decision self-efficacy; R2Personality = the unique contribution
of personality to career decision self-efficacy; R2TS = the unique contribution of thinking styles to career decision self-efficacy; leg =
legislative; jud = judicial; glo = global; lib = liberal; hie = hierarchical; exe=executive; loc = local; con = conservative; mon = monarchic;
oli = oligarchic; ana = anarchic; int = internal; ext = external; *p<.05; **p<.01; and ***p<.001.
110 Chapter Four
Table 4.2: Predicting Career Decision Self-Efficacy from Thinking Styles with Parenting Styles, Learning
Environments, Personality Traits, and Relevant Demographic Factors in Control (Time 2)
Occupational Goal Problem

CDSE Self-Appraisal Planning
Information Selection Solving
R2Total 0.38 0.47 0.40 0.47 0.51

R2demo 0.03 0.03 0.02 0.02 0.03
R2PS 0.11 0.12 0.10 0.13 0.14
2
R LE 0.13 0.13 0.10 0.12 0.18
2
R Personality 0.10 0.15 0.14 0.18 0.13
2
R TS 0.02 0.04 0.02 0.02 0.03
*** *** *** ***
F 11.60 17.10 12.60 16.92 20.09***
df 36,696 36,696 36,696 36,696 36,696
Type I ȕleg .15***

ȕjud .09*
ȕglo
ȕlib .12**
ȕhie .09*
Type II ȕexe
ȕloc
ȕcon
ȕmon
Type III ȕoli
ȕana .12***
ȕint
ȕext -.10* -.09*
Notes: CDSE = career decision self-efficacy; R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles,
learning environments, personality traits, and thinking styles to career decision self-efficacy; R2demo = the contribution of gender, grade,
major, hometown, and SES to career decision self-efficacy; R2PS = the unique contribution of parenting styles to career decision self-
efficacy; R2LE = the unique contribution of learning environments to career decision self-efficacy; R2Personality = the unique contribution
of personality to career decision self-efficacy; R2TS = the unique contribution of thinking styles to career decision self-efficacy; leg =
legislative; jud = judicial; glo = global; lib = liberal; hie = hierarchical; exe = executive; loc = local; con = conservative; mon =
monarchic; oli = oligarchic; ana = anarchic; int = internal; ext = external; *p<.05; **p<.01; and ***p<.001.
112 Chapter Four
Generally speaking, it was found at both Time 1 and Time 2 that, as

expected, some Type I thinking styles (i.e., legislative, liberal, and
hierarchical styles) positively contributed to CDSE. In addition, also as
expected, the global style positively predicted self-efficacy on career
planning at Time 1, while the judicial style positively predicted self-efficacy
on career problem solving. In terms of Type II thinking styles, it was found
that at Time 1, unexpectedly, the local style was positively associated with
CDSE, while, as expected, the monarchic style was negatively associated
with CDSE. At Time 2, Type II thinking styles showed no statistically
significant contribution to CDSE. In terms of Type III thinking styles, it was
found that both the internal style and the external style positively
contributed to CDSE at Time 1. At Time 2, the anarchic style was positively
associated with self-efficacy on goal selection, while the external style was
negatively related to self-appraisal and problem solving.
The conditional change models were also adopted to explore the

longitudinal contribution of thinking styles to the change of career decision
self-efficacy over the course of a year. Regressions in the change of CDSE
on parenting styles experienced in the year between the pretest and the
posttest, learning environments in the year between the pretest and the
posttest, personality traits at Time 1, thinking styles at Time 1, and the initial
scores of CDSE were conducted with the relevant demographic factors
being in control (Table 4.3). The effect of “regression toward the mean” was
also found for the change of CDSE. But still, students’ thinking styles at
Time 1 exclusively explained 3–6% of the variance in the change of CDSE.
Specifically, it was shown that two Type I thinking styles (legislative and
hierarchical) were the major contributors to the increase of CDSE over one
year.
Table 4.3: Predicting the Change of Career Decision Self-Efficacy from Thinking Styles (Time 1) in the Presence of
Parenting Styles (Time 2), Learning Environments (Time 2), and Personality Traits (Time 1)

Change of CDSE Self-Appraisal Planning
R2Total 0.45 0.42 0.41 0.46 0.47

R2demo 0.02 0.03 0.03 0.03 0.02
2
R PS 0.01 0.02 0.02 0.03 0.03
2
R LE 0.01 0.02 0.01 0.02 0.03
R2Personality 0.04 0.03 0.04 0.05 0.06
R2TS 0.06 0.05 0.05 0.06 0.06
R2Time 1 0.30 0.27 0.26 0.28 0.27
F 15.34*** 13.60*** 12.95*** 16.28*** 16.50***
df 37,695 37,695 37,695 37,695 37,695
Parenting Styles ȕinvol .14*** .19*** .16*** .19*** .19***
ȕpsyc_auto -.08*
Learning ȕconstru .14**
Environments
ȕclear goal .10*
114 Chapter Four
ȕautonomy .10*
ȕassessment .08*
ȕs-s cooper .17*** .10* .18*** .15*** .16***
ȕpeer -.11** -.10** -.08* -.18***
ȕfacilities .08* .11**
Personality ȕneuro .07*
ȕextra .09* .09*
Thinking Styles
Type I ȕleg .10* .10* .09*
ȕhie .09* .12** .08*
Type II ȕmon -.08*
CDSE ȕtime1 -.68*** -.68*** -.63*** -.70*** -.69***
2
Notes: CDSE = career decision self-efficacy; R Total = the contribution of gender, grade, major, hometown, SES, parenting styles at Time 2, learning environments at Time 2,
personality at Time 1, thinking styles at Time 1, and career decision self-efficacy at Time 1 to the change of career decision self-efficacy from Time 1 to Time 2; R2demo = the
contribution of gender, grade, major, hometown, and SES to the change of career decision self-efficacy; R2PS = the unique contribution of parenting styles during the year
between the pretest and the posttest to the change of career decision self-efficacy; R2LE = the unique contribution of learning environments during the year between the pretest
and the posttest to the change of career decision self-efficacy; R2Personality = the unique contribution of personality at Time 1 to the change of career decision self-efficacy; R2TS
= the unique contribution of thinking styles at Time 1 to the change of career decision self-efficacy; R2Time1 = the unique contribution of career decision self-efficacy at Time
1 to the change of career decision self-efficacy; invol = acceptance/involvement; psyc_auto = psychological autonomy-granting; super = strictness/supervision; constru =
constructivist-oriented teaching; clear goal = clear goals and coherence of curricula; autonomy = student autonomy; assessment = assessment and assignment; t-s inter =
teacher-student interaction; s-s cooper = student-student cooperation; peer = peer morale and identities; facilities = learning facilities; neuro = neuroticism; extra = extraversion;
open = openness to experience; agree = agreeableness; conscien = conscientiousness; leg = legislative; jud = judicial; glo = global; lib = liberal; hie = hierarchical; exe =
executive; loc = local; con = conservative; mon = monarchic; oli = oligarchic; ana = anarchic; int = internal; ext = external; *p<.05, **p<.01, ***p<.001; and the number of
R2 in bold means that the F change was significant after the predictor was added into the regression model. Due to space limitations, the dimensions of the predictors that
were not significantly related to any of the change of CDSE were omitted.
The results from the analysis of two sets of cross-sectional data and that of
the longitudinal data confirmed the important role that thinking styles play
in the students’ career maturity, as represented by career decision self-
efficacy. Actually, the contributions of thinking styles to career decision
self-efficacy (CDSE) were even beyond those contributed by parenting
styles, learning environments, and personality traits. Specifically speaking,
the results from both the cross-sectional data analysis and the longitudinal
data analysis demonstrated that Type I thinking styles were positively
associated with CDSE, which was completely in line with the hypothesis.
For example, the legislative style and the hierarchical style were found to
be significant predictors for CDSE. It is understandable that if students
prefer to deal with unstructured tasks and arrange multiple tasks in an
efficient way, they would be more confident in coping with career-related
tasks, such as collecting occupational information, making career plans, and
solving problems during career development.
During interviews, although the interviewees were not consciously aware of

the role that thinking styles played in career maturity when they were asked
to talk about it, the connection emerged that students who scored high on
Type I styles were found to have frequently considered issues that relate to
career development, while students who scored low on Type I styles
claimed that they had not thought about the career issue. This discrepancy
in terms of the proactive consideration of career development was especially
obvious among freshmen with different styles, as career decisions are not
as urgent for them as they are for juniors. All of the aforementioned results
support that Type I thinking styles seem to be more adaptive than others in
terms of career maturity.
In terms of Type II styles, although the local style was found to be positively
related to career decision self-efficacy (CDSE) based on the analyses of data
at Time 1, this relationship was not confirmed by the results from the
longitudinal data analyses, thereby indicating that the relationship found at
Time 1 was more likely to be covariant than causal. However, the negative
relationship between the monarchic style and CDSE found in the cross-
sectional data analysis (Time 1) was confirmed by the results from the
longitudinal data analyses, thereby indicating that students who prefer to
deal with one task at one time (monarchic) did not have enough confidence
when dealing with tasks relevant to their career development, which may
require multiple foci. This result supports the argument that Type II styles
are less adaptive.
116 Chapter Four
In terms of Type III styles, although some of the relationships between Type
III thinking styles and CDSE were found based on the cross-sectional data
analyses, none of them were confirmed based on the longitudinal data
analyses. This indicates that Type III styles did not predict students’ CDSE
in certain way, which suggests that Type III styles’ value is not explicit and
it may, instead, depend on specific situations.
Subjective well-being and its relationship with intellectual styles

Happiness and well-being have been popular topics among philosophers
and religious leaders for centuries (McMahon, 2006). However, subjective
well-being was not systematically studied until the early twentieth century
(E. Diener et al., 2009). Subsequently, Diener’s influential work in 1984
largely urged made the realm of subjective well-being into a science (E.
Diener et al., 2009). Historically, subjective well-being has been defined
from different perspectives and using different terms (E. Diener, 1984).
Currently, the most extensively accepted concept of subjective well-being
defines it as “a person’s cognitive and affective evaluations of his or her life
as a whole” (E. Diener et al., 2009, p.187). According to this definition,
subjective well-being includes cognitive judgment of life, which is always
indicated by life satisfaction and emotional reactions to life events that
include positive affect and negative affect. Furthermore, this definition also
implies the subjective nature of well-being, which means that it is one’s own
perception of his/her well-being that is meaningful to him/her.
Theories of subjective well-being can be distinguished by two major

approaches: top-down and bottom up (E. Diener, 1984). Top-down theories
contend that people’s predispositions (e.g., gene, temperament, and
personality) influence their experiences and perceptions of their lives. This
approach has always been supported by studies on genetic influence,
personality influence, and the stability of subjective well-being across life
situations (e.g., E. Diener & Larsen, 1984; Lykken & Tellegen, 1996; Steel,
Schmidt, & Shultz, 2008). Bottom-up theories suggest that one’s perception
of well-being is derived from the summation of momentary pleasure
experienced in different domains of life. According to this approach, life
events and circumstances influence people’s judgments of domain
satisfaction and emotional experience, and, in turn, influence people’s
global judgments of well-being. Bottom-up theories have always been
supported by studies on the influence of domain satisfaction on general life
satisfaction, as well as those on the influence of life events and
circumstances on subjective well-being (E. Diener, Lucas, Oishi, & Suh,
2002; Paykel, 2003).
Both approaches have been supported by empirical evidence. On the one

hand, many personal correlates have been proved to play important roles in
one’s subjective well-being, such as genes, temperament, personality, self-
esteem, attributional styles, coping styles, and optimism (e.g., DeNeve &
Cooper, 1998; Lykken & Tellegen, 1996; C. L. Park & Adler, 2003; Rigby
& Huebner, 2005; Schimmack & Diener, 2003). For example, after a meta-
analysis of 137 studies, DeNeve and Cooper (1998) found that personality
variables such as repressive-defensiveness, trust, emotional stability, locus
of control, desire for control, hardiness, self-esteem, and tension were all
associated with subjective well-being. Specific to the Big Five personality
traits, extraversion (as a positive predictor) and neuroticism (as a negative
predictor) are the most repeatedly reported traits that are closely related to
subjective well-being. In addition, agreeableness and conscientiousness
were also found to have positive correlations with subjective well-being in
some other studies (Lucas & Diener, 2008). There has also been increasing
attention paid to the effect of cognitive factors (e.g., attention biases,
categorization tendencies, and self-regulation abilities) on subjective well-
being (see Robinson & Compton, 2008, for a review). On the other hand,
situational factors such as ethnicity, culture, income, familial relationships,
marriage, employment, and other life events have also been frequently
studied in terms of their correlations with subjective well-being (e.g., E.
Diener & Biswas-Diener, 2002; M. L. Diener & Diener McGavran, 2008;
Lucas, 2005; Lucas, Clark, Georgellis, & Diener, 2004; Oishi, 2001;
Veenhoven, Ehrhardt, Ho, & de Vries, 1993). The majority of these studies
found weak to moderate correlations between these situational factors and
subjective well-being. For example, studies in this field of literature
consistently showed that parental warmth, support, involvement, and care
were positively associated with life satisfaction, happiness, and mental
health, while parental control, rejection, and indifference were negatively
associated with these outcomes (e.g., Cheng & Furnham, 2004; Flouri, 2004;
Furnham & Cheng, 2000; Huppert, Abbott, Ploubidis, Richards, & Kuh,
2010; Maynard & Harding, 2010; Shek, 2007; Suldo & Huebner, 2004;
Taris & Bok, 1996; Zheng & Tao, 2001). Learning environments with
meaningful activities that permit student autonomy and engaged students
were also proved to lead to enjoyment and satisfaction (Maton, 1990;
Shernoff, Csikszentmihalyi, Shneider, & Shernoff, 2003).
Although the influence of life circumstance and life events on subjective

well-being has been supported by empirical studies, the influential power of
these situational factors is weaker than personal predispositions, when
results from studies on genes and personality are compared (Lucas & Diener,
2008). However, it does not mean the effect of situational factors can be
118 Chapter Four
overlooked when studying subjective well-being because they do have an

effect. Some scholars have suggested that people evaluate their lives based
on their life experiences, but how these life experiences influence their
evaluation depends on the ways they view and interpret them, which are
both relevant to their personal predispositions (Kim-Prieto, Diener, Tamir,
Scollon, & Diener, 2005; Lucas & Diener, 2008). As personal preferences
for managing activities and processing information in life, intellectual styles
quite possibly play a role in subjective well-being. Furthermore, as
mentioned above, personality and cognitive factors have been found to be
associated with subjective well-being. It is also reasonable to speculate that
intellectual styles may also contribute to subjective well-being to some
extent, as styles are considered to be the interface between personality and
cognition.
However, intellectual styles have received little attention in the field of

subjective well-being. Only two studies on Myers and McCaulley’s (1988)
personality types were found in literature on subjective well-being and
happiness (Francis & Jones, 2000; Shewchuk & O'Connor, 1995). Both of
them revealed that extraverted participants showed higher levels of well-
being and happiness than introverted ones. In addition, there are also a few
studies on the relationships between intellectual styles and mental health.
For example, Wise and his colleagues (1978) found that field dependence
(Type II style) was positively related to depression. Zhang (2009) also found
that Type I styles (e.g., legislative, judicial, liberal, and hierarchical) were
negatively related to anxiety. In addition, Chen and Zhang (2010) found that
thinking styles had contributions to mental health. These studies all indicate
that intellectual styles may play an important role in subjective well-being.
Another study from Updegraff and Suh (2007) also provided some insight.
The researchers asked their participants to describe important aspects of
themselves and their lives, and found that people who saw themselves or
their lives in an abstract way showed higher levels of life satisfaction than
people who saw themselves or their lives in a concrete way. Therefore, as
an indicator of the ways people adapt to deal with information and manage
their lives, intellectual styles, such as thinking styles probably also play a
functional role in subjective well-being.
Current findings: Thinking styles and subjective well-being

Given that Type I thinking styles were consistently found to be positively
related to other social and emotional developmental outcomes (e.g., self-
esteem, psychological well-being, and psychosocial development) and Type
II thinking styles negatively related, it was also hypothesized in the present
research that Type I thinking styles would be positively related to subjective

well-being, while Type II thinking styles would be negatively related to
subjective well-being. No hypothesis was made in terms of Type III styles
because, according to earlier studies, their value might differ based on
specific contexts.
Similar to the analysis procedure I adopted for career decision self-efficacy,

the analyses of the relationship between thinking styles and subjective well-
being were also first based on two sets of data (Time 1 and Time 2) to see
if the relationship between the two variables stayed the same across time.
Furthermore, to exclude the possible influence of confounding variables, the
relationship between thinking styles and subjective well-being was
examined with parenting styles, learning environments, personality traits,
and relevant demographic factors (i.e., gender, grade, major, hometown, and
SES) as controls. The results based on hierarchical regressions showed that
thinking styles made unique contributions to SWB beyond parenting styles,
learning environments, and personality traits. Specifically, thinking styles
explained 3% to 8% of the variance in SWB at Time 1 (Table 4.4) and 1%
to 5% of the variance in SWB at Time 2 (Table 4.5). It was found that, as
expected, some Type I thinking styles (legislative, global, and liberal at
Time 1; liberal and hierarchical at Time 2) predicted SWB positively.
However, unexpectedly, the global style was also found to be positively
related to negative affect at Time 1, while the judicial style was found to be
negatively related to positive affect at Time 2. Another unexpected result
was that Type II thinking styles (i.e., executive, local, conservative, and
monarchic) were found to be positively associated with SWB at Time 1. In
terms of Type III styles, it was found that, at both Time 1 and Time 2, the
anarchic style positively predicted negative affect, while the external style
did so negatively.
120 Chapter Four
Table 4.4: Predicting Subjective Well-being from Thinking Styles with Parenting Styles, Learning Environments,
Personality Traits, and Relevant Demographic Factors in Control (Time 1)
SWB Life Satisfaction Positive Affect Negative Affect

R2Total 0.40 0.57 0.42
2
R demo 0.04 0.03 0.03
2
R PS 0.15 0.10 0.05
2
R LE 0.05 0.12 0.03
R2TS 0.03 0.08 0.05
*** ***
F 16.31 31.99 17.46***
df 36,878 36,878 36,878
Type I ȕleg .10** -.09*
ȕjud
ȕglo .07* .11***
ȕlib .17***
ȕhie
Type II ȕexe -.10*

ȕloc .08* .11*** .07*
ȕcon .08*
ȕmon .09**
Type III ȕoli
ȕana .08*
ȕint
ȕext -.14***
Notes: SWB = subjective well-being; R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles, learning
environments, personality traits, and thinking styles to subjective well-being; R2demo = the contribution of gender, grade, major,
hometown, and SES to subjective well-being; R2PS = the unique contribution of parenting styles to subjective well-being; R2LE = the
unique contribution of learning environments to subjective well-being; R2Personality = the unique contribution of personality to subjective
well-being; R2TS = the unique contribution of thinking styles to subjective well-being; leg = legislative; jud = judicial; glo = global; lib
= liberal; hie = hierarchical; exe = executive; loc = local; con = conservative; mon = monarchic; oli = oligarchic; ana = anarchic; int =
internal; ext = external; *p<.05; **p<.01; and ***p<.001.
122 Chapter Four
Table 4.5: Predicting Subjective Well-being from Thinking Styles with Parenting Styles, Learning Environments,
Personality Traits, and Relevant Demographic Factors in Control (Time 2)
SWB Life Satisfaction Positive Affect Negative Affect

R2Total 0.47 0.59 0.51
2
R demo 0.05 0.03 0.03
2
R PS 0.16 0.15 0.16
2
R LE 0.10 0.12 0.05
R2TS 0.01 0.05 0.02
*** ***
F 17.02 27.67 20.30***
df 36,696 36,696 36,696
Type I ȕleg
ȕjud -.09*
ȕglo
ȕlib .18***
ȕhie .12***
Type II ȕexe
ȕloc
ȕcon
ȕmon
Type III ȕoli
ȕana .09** .07*
ȕint
ȕext -.09*
Notes: SWB = subjective well-being; R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles, learning
environments, personality traits, and thinking styles to subjective well-being; R2demo = the contribution of gender, grade, major,
hometown, and SES to subjective well-being; R2PS = the unique contribution of parenting styles to subjective well-being; R2LE = the
unique contribution of learning environments to subjective well-being; R2Personality = the unique contribution of personality to subjective
well-being; R2TS = the unique contribution of thinking styles to subjective well-being; leg = legislative; jud = judicial; glo = global; lib
= liberal; hie = hierarchical; exe = executive; loc=local; con = conservative; mon = monarchic; oli = oligarchic; ana = anarchic; int =
internal; ext = external; *p<.05; **p<.01; and ***p<.001.
124 Chapter Four
Table 4.6: Predicting the Change of Subjective Well-being from Thinking Styles (Time 1) in the Presence of Parenting
Styles (Time 2), Learning Environments (Time 2), and Personality Traits (Time 1)
Change of SWB life satisfaction positive affect negative affect

R2Total 0.44 0.49 0.44
2
R demo 0.01 0.02 0.00
2
R PS 0.01 0.02 0.03
2
R LE 0.02 0.03 0.01
R2TS 0.04 0.06 0.04
2
R Time 1 0.30 0.28 0.29
F 14.45*** 17.89*** 14.90***
df 37,695 37,695 37,695
Parenting Styles ȕinvol .18*** .23***
ȕpsyc_auto -.28***
ȕsuper -.07*
Learning ȕconstru .10*
Environments
ȕclear goal .11*
ȕassessment .08*
ȕs-s cooper .18***

ȕpeer .08*
Personality ȕneuro .12**
ȕextra .12**
Thinking Styles
Type I ȕlib .09*
ȕhie .12**
Type II ȕexe -.09*
ȕloc -.08*
ȕmon .11**
SWB ȕtime1 -.71*** -.83*** -.72***
2
Notes: SWB = subjective well-being; R Total = the contribution of gender, grade, major, hometown, SES, parenting styles at Time 2, learning environments at
Time 2, personality at Time 1, thinking styles at Time 1, and subjective well-being at Time 1 to the change of subjective well-being from Time 1 to Time 2; R2demo
= the contribution of gender, grade, major, hometown, and SES to the change of subjective well-being; R2PS = the unique contribution of parenting styles during
the year between the pretest and the posttest to the change of subjective well-being; R2LE = the unique contribution of learning environments during the year
between the pretest and the posttest to the change of subjective well-being; R2Personality = the unique contribution of personality at Time 1 to the change of subjective
well-being; R2TS = the unique contribution of thinking styles at Time 1 to the change of subjective well-being; R2Time1 = the unique contribution of subjective well-
being at Time 1 to the change of subjective well-being; invol = acceptance/involvement; psyc_auto = psychological autonomy-granting; super =
strictness/supervision; constru = constructivist-oriented teaching; clear goal=clear goal and coherence of curricula; autonomy = student autonomy; assessment =
assessment and assignment; t-s inter = teacher-student interaction; s-s cooper = student-student cooperation; peer = peer morale and identities; facilities = learning
facilities; neuro=neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness; conscien = conscientiousness; leg = legislative; jud =
judicial; glo = global; lib = liberal; hie = hierarchical; exe = executive; loc = local; con = conservative; mon = monarchic; oli = oligarchic; ana = anarchic; int =
internal; ext = external; *p<.05, **p<.01, ***p<.001; and the number of R2 in bold means the F change was significant after the predictor was added into the
regression model. Due to space limitations, the dimensions of the predictors that were not significantly related to any of the change of SWB were omitted.
126 Chapter Four
The conditional change models were also adopted to explore the

longitudinal contribution of thinking styles to the change of subjective well-
being over one year. Regressions of the change of SWB on parenting styles
experienced in the year between the pretest and the posttest, learning
environments in the year between the pretest and the posttest, personality
traits at Time 1, thinking styles at Time 1, and the initial scores of SWB
were conducted with the relevant demographic factors as controls (Table
4.6). The effect of the “regression toward the mean” was also found for the
change of SWB. Nevertheless, students’ thinking styles at Time 1 still
exclusively explained 4–6% of the variance in the change of SWB.
Specifically, it was shown that two Type I thinking styles (liberal and
hierarchical) were the major contributors to the increase of positive affect
over one year. As for Type II styles, the executive style contributed to the
decrease of life satisfaction, the local style contributed to the decrease of
negative affect, and the monarchic style contributed to the increase of
negative affect. All of these results were consistent with hypotheses, apart
from the role of the local style (Type II) in subjective well-being.
The analyses of the quantitative data from the main study demonstrated that
thinking styles play an important role in subjective well-being (SWB).
However, just as they were not aware of the relationships between thinking
styles and career maturity when they were asked about the relationships
between the two, the interviewees were also not fully aware of the
relationships between thinking styles and SWB. The connection only
emerged when they were encouraged to freely discuss anything about their
subjective well-being, which further confirmed the results from the
quantitative data, as well as providing possible explanations for them.
Specifically, as expected, Type I thinking styles positively contributed to

students’ subjective well-being. For example, the longitudinal data analyses
showed that students with higher scores on the liberal style at the initial
stage were found to experience more positive affect after one year. This is
probably because students who prefer to try novel things (i.e., using the
liberal style) would be more comfortable dealing with the changes in
environment. Additionally, students with the liberal style tend to be
rewarded by the education orientation within university environments,
which makes them happy with university life. The interview data also
indicated that students who scored low in Type I styles admired peers who
manifested the characteristics of Type I styles and adapted well to the
university environment. For example, as one of the interviewees
commented,
Some of my classmates seemingly know a lot. They have read beyond the
things taught in class, even before they enter university. They have had
abundant experiences. During the class discussions, they present their own
ideas confidently and clearly. This makes me feel inferior and stressed. (MX)
Furthermore, the longitudinal data also indicated that the hierarchical style
was a positive predictor for students’ positive affect after one year. As some
interviewees described, the university environment is a multitasking
environment. For instance,
In high school, I only needed to focus on one thing because my parents had
done other things for me. In university, I have to take care of a lot of things
myself, such as how to allocate money, when to buy new clothes, and other
detailed things in daily life […] Furthermore, besides dealing with difficult
courses, I also have to prepare for TOFEL and GRE. In addition, a lot of
stuff needs to be done in the student associations I participate in. I feel upset
that my energy is so limited but there are too many things to do. (QBC)
Therefore, in this type of multitasking environment, it would be easier for

students who prefer to perform multiple tasks in an orderly way to adapt. In
addition, it was shown that the global and judicial styles had negative
relationships with SWB based on the cross-sectional data analyses.
However, these relationships were insignificant in the longitudinal data
analyses, thereby indicating that the global and judicial styles did not predict
SWB in a causal sense.
Although positive relationships between some Type II styles and SWB were
found in the cross-sectional data analyses at Time 1, the same positive
relationships were not confirmed in the longitudinal data analyses. In fact,
in the longitudinal data analyses, just like the hypothesis, some Type II
styles (e.g., the executive style and the monarchic style) were found to
negatively contribute to subjective well-being. It is probably because in the
university environment where direct instructions from teachers and parental
supervision decrease, students who prefer to follow others’ instructions
(executive) do not have many resources to depend on, which probably leads
them to sometimes feel helpless, which would influence their subjective
well-being. During interviews, it was also found that students who scored
higher in the executive style expressed more confusion during their adaption
to university life. When there was no explicit instruction and step-by-step
guidance in university, they felt “lost (㏞ⲉ)” (WHN) for a while until they
managed to successfully adapt. In addition, as previously mentioned, the
university is a multi-tasking environment where students who prefer to do
only one task at a time (monarchic) are likely to feel difficulties during the
128 Chapter Four
process of adaption, which probably leads to the decrease of subjective well-

being. The only exception that was not in line with the anticipation was that
the local style (Type II) was found to positively predict students’ subjective
well-being. This relationship probably has something to do with academic
achievement. In previous literature, Type II styles were sometimes found to
be beneficial for academic achievement (Zhang & Sternberg, 2009) and, in
the present research, it was also found that university assessments
encouraged the local style. Furthermore, the interviewees did mention that
academic achievement was one of the important factors that influenced their
emotional state. Therefore, it is understandable that students who use the
local style would have a positive mood when they achieve a satisfying
academic performance.
In terms of Type III styles, although some relationships between Type III
thinking styles and subjective well-being were found during the cross-
sectional analyses, none of them were confirmed from the longitudinal data
analyses. It indicated that Type III styles did not predict students’ subjective
well-being in a certain way, which suggests that Type III styles’ value
probably varies depending on the particular context.
In summary, the examination of the role of thinking styles in career decision

self-efficacy and subjective well-being revealed that Type I thinking styles
were conducive to students’ career decision self-efficacy and subjective
well-being. This finding resonates with Zhang’s studies, which examined
the relationships between thinking styles and a series of students’ social and
emotional developmental outcomes, such as anxiety (Zhang, 2009), emotion
management (Zhang, 2008b), psychosocial development (Zhang, 2010),
and self-esteem (Zhang, 2001d); these studies also found that Type I
thinking styles were more adaptive than others in terms of these social and
emotional outcomes. In addition, in the present research, it was also found
that, in general, Type II thinking styles were less adaptive in terms of the
students’ career decision self-efficacy and subjective well-being, although
the value of Type II thinking styles was sometimes blurred when academic
achievement was probably involved. The present study did not find any
orientation of value in Type III thinking styles. Generally speaking, the
research findings lent support to Zhang and Sternberg’s (2005) judgment of
the value of styles.
The mediating role of thinking styles in student development

Another limitation in the research of this field is the lack of attention to the
mediating role that intellectual styles may play in the relationships between
potential antecedents and student developmental outcomes. Enlightened by

Biggs’ (1990) 3P model, the possible mediating role of thinking styles in
student development is also taken into account in this research. In Biggs’
3P model, it is contended that process factors (e.g., learning approach-a
construct of styles) could mediate the influence of presage factors (e.g.,
student characteristics and teaching context) on product factors (e.g.,
learning outcomes), which implies the mediating role that styles may play
in the effect of personal characteristics and environmental factors on student
developmental outcomes.
Actually, in the field of psychopathology, cognitive vulnerability (which is

sometimes called dysfunctional cognitive style), attributional/ explanatory
style, and locus of control have been found to mediate the relationship
between parenting and certain psychological symptoms, such as depression
and anxiety (Alloy et al., 2001; McGinn, Cukor, & Sanderson, 2005;
O'Donnell, 2009; Schapman, 2003). Based on these results, it can be
inferred that intellectual styles probably also mediate the relationship
between parenting and student developmental outcomes, which is worth
further exploration. As we already know, the contribution of parenting
styles, learning environment, and personality to thinking styles, as well as
the contribution of thinking styles to career decision self-efficacy and
subjective well-being, have been confirmed in this research. Therefore, it is
tempting to explore the mediating role that thinking styles may play in the
relationships between these antecedents and students’ developmental
outcomes.
Therefore, the present research examined the mediating role that thinking
styles may play in the relationships of parenting styles, learning
environment, and personality to students’ career decision self-efficacy and
subjective well-being. The analyses were based on two-wave longitudinal
data.
There are three approaches for the mediation analysis with two-wave
longitudinal data (Roth & MacKinnon, 2012). The first one is raw change
score analysis. The path of this approach is from the predictor(s) to the
change of mediator(s) to the change of outcome(s). As mentioned before,
this approach overlooks the effect of the initial score on the mediator(s) and
the outcome(s). Without considering the initial scores of the mediator(s) and
the outcome(s), it would be insufficient to infer causal relationships (Cole
& Maxwell, 2003). The second approach, which is called baseline-adjusted
change score analysis, atones for this shortcoming. It controls the
relationships due to the initial differences of the mediator(s) and the
130 Chapter Four
outcome(s) by putting the initial scores of the mediator(s) and the outcome(s)
into the analysis model. The third approach is autoregressive model analysis,
where the path is from the predictor(s) to the mediator(s) at posttest to the
outcome(s) at posttest with the initial scores of the mediator(s) and the
outcome(s) being controlled. As for the total effect, the second approach
and the third approach are identical, but the mediation effect achieved by
the third approach is usually bigger than the one from the second approach
because, in autoregressive model analysis, the calculation of the unique
contribution of the mediator(s) to the outcome(s) includes both within-
subject covariation and between-subjects covariation, while the baseline-
adjusted change score analysis restricts itself to the within-subjective
covariation. The second approach is recommended when the research focus
is pure change within individuals (Roth & MacKinnon, 2012). Therefore,
the baseline-adjusted change score analysis was adopted in the present
research.
There are four prerequisites for establishing mediation models. First, the
independent variable (IV) should significantly predict the mediator. Second,
the IV should significantly predict the dependent variable (DV) without the
mediator. Third, the mediator should have made a unique contribution to
the DV with the presence of the IV. Fourth, compared with the contribution
of the IV to the DV in the absence of the mediator, the contribution of the
IV to the DV after the mediator is added would decrease or even disappear.
A series of regressions (see Table 3.15 and Appendices 5–8) were
conducted to determine the relationships that met the four conditions.
Eventually, 13 potential mediating relationships for career decision self-
efficacy (CDSE) and six potential mediating relationships for subjective
well-being (SWB) were found (Table 4.7); however, no statistical mediating
effect of thinking styles in the relationships between personality traits and
the outcome variables was found. The 19 possible mediating relationships
were put into two path models (one for CDSE and the other for SWB) to
demonstrate the mediating effect through the baseline-adjusted change
score analysis (Figure 4.1 and Figure 4.2).
The model fit indices of the CDSE path model (Figure 4.1) are as follows:
Chi-square (df = 236, N = 743) = 416.80, p<.001, RMSEA = .032, SRMR
= .069, NFI = .98, NNFI = .99, and CFI = .99. Three Type I thinking styles
played significant roles as mediators in the relationships of parenting styles
and learning environments to CDSE. Specifically, the judicial style
significantly mediated the influence of parental acceptance/involvement on
self-efficacy of occupational information collection, the influence of
psychological autonomy-granting from parents on career planning, the
influence of constructivist-oriented teaching on problem solving, and the

influence of learning facilities on self-efficacy of occupational information
collection and problem solving. Furthermore, the liberal style significantly
mediated the relationships of acceptance/involvement and peer morale to
self-appraisal, while the hierarchical style significantly mediated the
relationships of one dimension of parenting styles (i.e., parental
acceptance/involvement) and one dimension of the learning environment
(i.e., clear goals and coherence of curricula) to students’ self-efficacy of
career goal selection.
132 Chapter Four
Table 4.7: Testing Longitudinal Relationships that are Potentially Mediated by Thinking Styles
Mediat
IV DV Path Model Testing
or
indirect effect of direct effect of IV
PS TS CDSE a b a*b
IV on DV on DV
occupational
acceptance/involvement jud .07* .08*** .01
information
.01* .20***
occupational
acceptance/involvement ext .11*** .02 .00
information
acceptance/involvement jud planning .07* .08*** .01 .01 .27***
acceptance/involvement jud problem solving .07* .08*** .01 .01 .25***
psychological autonomy-
jud planning -.06* .08*** -.01 -.01* -.06***
granting
acceptance/involvement lib self-appraisal .08** .09*** .01 .01* .26***
acceptance/involvement hie goal selection .16*** .05* .01 .01* .23***
LE TS CDSE a b a*b
IV on DV on DV
constructivist-oriented teaching jud problem solving .15*** .08*** .01 .01** .09***
occupational
learning facilities jud .07* .08*** .01 .01* .07**
information
learning facilities jud problem solving .07* .08*** .01 .01* .08***
peer morale and identities lib self-appraisal .13*** .09*** .01 .01** .03
clear goal and coherence of
hie goal selection .22*** .05* .01 .01* .01
curricula
occupational
student-student cooperation ext .34*** .02 .01 .01 .03
information
PS TS SWB a b a*b
IV on DV on DV
acceptance/involvement lib positive affect .11*** .16*** .02
.03*** .19***
acceptance/involvement hie positive affect .14*** .08** .01
strictness/supervision lib positive affect -.06 .16*** -.01 -.01 -.07*
loc negative affect -.07** .03 -.00
granting
-.02** -.29***
ana negative affect -.12*** .11*** -.01
granting
LE TS SWB a b a*b
IV on DV on DV
constructivist-oriented teaching hie positive affect .20*** .08** .02 .02** .20***
Notes: PS = parenting styles; LE = learning environments; TS = thinking styles; SWB = subjective well-being; CDSE = career decision
self-efficacy; jud = judicial; lib = liberal; hie = hierarchical; loc = local; ana = anarchic; ext = external.
134 Chapter Four
Figure 4.1 Baseline-adjusted Path Model of Career Decision Self-efficacy

Notes: invol = acceptance/involvement; psyc_auto = psychological autonomy-granting; constru = constructivist-oriented teaching; goal
= clear goals and coherence of curricula; s_s coop = student-student cooperation; peer = peer morale and identities; facilities = learning
facilities; jud = judicial; lib=liberal; hie = hierarchical; ext = external; OI = occupational information; SA = self-appraisal; GS = goal
selection; PL = planning; PS = problem solving; _1 = scores at Time 1; and ᶭ = change of scores from Time1 to Time 2.
The model fit indices of SWB’s path model (Figure 4.2) were as follows:
Chi-square (df = 90, N = 743) = 262.07, p<.001, RMSEA = .051, SRMR
= .044, NFI = .93, NNFI = .93, and CFI = .95. Three thinking styles
significantly mediated the influence of parenting styles and learning
environments on SWB. Specifically, two Type I styles, the liberal style and
the hierarchical style, partially mediated the influence of parental
acceptance/involvement on positive affect. The hierarchical style also
partially mediated the influence of constructivist-oriented teaching on
positive affect. In addition, the mediating effect of one Type III style, the
anarchic style, in the relationship between psychological autonomy-
granting from parents and negative affect was also significant.
Overall, thinking styles, especially Type I thinking styles, were partial

mediators in the influence of parenting styles and learning environments on
CDSE and SWB, while the influence of personality traits on CDSE and
SWB were direct.
136 Chapter Four
Figure 4.2 Baseline-adjusted Path Model of Subjective Well-being

Notes: invol = acceptance/involvement; psyc_auto = psychological autonomy-granting; super = strictness/supervision; constru =
constructivist-oriented teaching; lib = liberal; hie = hierarchical; loc = local; ana = anarchic; _1 = scores at Time 1; and ᶭ = change of
scores from Time1 to Time 2.
The results based on the longitudinal data analysis showed that thinking
styles played the role of partial mediators in the relationships of two
environmental predictors (parenting styles and learning environments) to
career decision self-efficacy (CDSE) and subjective well-being (SWB),
where the mediating function was mostly found among Type I and Type III
thinking styles. Although the indirect effect of parenting styles and learning
environments on CDSE and SWB through thinking styles was not very
strong, the mediating function of thinking styles is still statistically
significant.
For example, three Type I styles partially mediated the influence of one
dimension of parenting styles and two dimensions of learning environments
on the students’ development of CDSE. These results indicated that, if the
students perceived their parents as caring and involved (parental
acceptance/involvement), they would be more likely to develop Type I
thinking styles that are characterized as thinking critically (judicial),
favoring novel tasks or issues (liberal), and handling multiple tasks in an
orderly way (hierarchical), which accordingly promoted their self-efficacy
in dealing with tasks relevant to career development. In addition, the results
also indicated that, when students experienced a learning environment
where teachers taught in a constructivist-oriented way, or when sufficient
learning facilities could be accessed, they also tended to develop the judicial
style (Type I), which might have increased their self-efficacy in their career
development. In addition, Type I thinking styles also completely mediated
the influence of another two dimensions of learning environments on CDSE.
It was indicated that when peers were proactive in learning, or when
curricula were designed with clear goals, students would be more likely to
develop the liberal style (Type I) or the hierarchical style (Type I), which
subsequently increased their self-efficacy in dealing with career-related
tasks.
With regard to subjective well-being (SWB), the longitudinal data analyses

also revealed that two Type I thinking styles and one Type III style partially
mediated the influence of parenting styles and learning environments on
SWB. The results indicated that if students perceived more acceptance and
warmth from their parents, they would prefer novel things (liberal, Type I)
and tend to deal with multiple tasks in an orderly manner (hierarchical, Type
I), which then might have enhanced their experience of positive affect.
Furthermore, the results also indicated that the psychological autonomy
granted by parents was likely to have decreased the students’ tendency to
handle tasks in a fragmented way (anarchic, Type III), which then might
have reduced the students’ experience of negative affect. In addition,
138 Chapter Four
constructivist-oriented teaching in the learning environment was also likely

to have benefited the students’ development of the hierarchical style, which
would probably be helpful when they are dealing with multiple tasks in
university, thereby increasing their experience of positive affect.
Generally speaking, the finding that thinking styles function as mediators in

the relationships of parenting styles and learning environments to students’
career decision self-efficacy and subjective well-being lent empirical
support to Biggs’ (1979) 3P model of student performance. This finding also
suggests that the influence of parenting styles and learning environments on
students’ thinking styles would accordingly influence other development
outcomes in student life (e.g., CDSE and SWB), albeit mildly. Furthermore,
this finding demonstrated that shaping Type I thinking styles had an
adaptive value for student development.
CHAPTER FIVE
THE RESPONSE TO THREE

CONTROVERSIAL ISSUES
After Allport (1937) first introduced the notion of style to psychology,

various style constructs and theories proliferated between the late 1950s and
the early 1970s (Zhang & Sternberg, 2006). However, because of the lack
of internal dialogue, the plethora of style constructs became a major obstacle
to the advancement of this field (Peterson, Rayner, & Armstrong, 2009).
Some constructs labeled as styles may not actually be styles, while some
different style labels have been used interchangeably. The notion of styles
has been confusing not only to the public but also to academics during the
early research stages, as it has too many theories surrounding it. In Coffield
and his colleagues’ (2004) review, they found over 71 theories of styles.
The lack of both an internal dialogue and a common conceptual framework
has provided a major obstacle to the advancement of the field. Fortunately,
in the previous three decades, considerable endeavors have been made to
integrate various style constructs and clarify controversial issues, which
have significantly helped to move the field forward. Recently, “intellectual
styles” has been used as a superordinate concept that encompasses all style
constructs (e.g., cognitive styles, learning styles, and thinking styles).
However, there are still some important but unsolved issues, such as those
surrounding identity, style malleability, and style value. The applicability of
styles into educational practice has been very often questioned and criticized
due to these controversial issues. However, it has also motivated researchers
to conduct empirical studies to clarify these confusions. In this chapter, I
will summarize the present research’s major findings regarding each
controversial issue, combined with a discussion of these controversial issues
in the larger context of intellectual styles using evidence from other style
models.
Style Identity
For decades, the field of intellectual styles has suffered from a blurred
identity in the broader context of psychology and education (Zhang, 2013).
140 Chapter Five
Such a challenge has partially arisen from the difficulty in distinguishing

styles from abilities and personality. Although it has been declared that
intellectual styles are at the interface between abilities and personality
(Robert J. Sternberg, 1997), many debates on this issue remain mostly at the
conceptual level, especially with regard to the relationship between styles
and personality. Some researchers consider style to be a subordinate
variable of personality, while others believe these two variables to be
distinct constructs (Roodenburg, Roodenburg, & Rayner, 2012; Zhang &
Sternberg, 2006). The present research examined the relationship between
personality traits and thinking styles in order to contribute to the discussions
surrounding this issue. The relationship between personality traits and
thinking styles can be discussed from three perspectives: 1) the overlap
between personality traits and thinking styles; 2) the role of personality
traits and thinking styles in explaining a third variable; and 3) a comparison
of the malleability of personality traits and thinking styles.
First, the regressions of thinking styles on personality traits presented in

Chapter 2 showed that personality traits could explain less than 31% of the
variance in thinking styles. This result parallels the findings in previous
studies conducted by Zhang (2002a, 2006b). However, it is necessary to
note that when parenting styles, learning environments, and personality
traits were simultaneously put into regression models to predict thinking
styles, it was found that the percentage of the variance in thinking styles
explained by personality decreased to less than 19% (see Chapter 3: Tables
3.13 and 3.14). Together, these findings suggest that the overlap between
personality traits and thinking styles is small. Second, as shown in Chapter
2, when personality traits and thinking styles were put together to predict
the two outcome variables (i.e., career decision self-efficacy and subjective
well-being), it was found that thinking styles had made unique contributions
to the two outcomes of student development beyond personality traits.
Moreover, it is also noted that, even after parenting styles and learning
environments were put into regression models with thinking styles and
personality traits to predict the two outcomes (i.e., career decision self-
efficacy and subjective well-being) (see the tables in Chapter 4), the
incremental validity of thinking styles over personality was repeatedly
demonstrated based on both the cross-sectional and the longitudinal data
analyses. These results suggest that personality traits and thinking styles
have distinct construct contents so that each of them can explain different
parts of the variance in the additional variables. Third, the comparison of
the malleability of personality traits and thinking styles found that
personality traits stayed stable over the course of a year but thinking styles
changed significantly during the same period of time, which suggests that
The Response to Three Controversial Issues 141
thinking styles are more changeable than personality traits.
In summary, the present research demonstrated that the overlap between

thinking styles and personality traits is small, that both thinking styles and
personality traits make unique contributions to developmental outcomes for
students, and that the nature of these two constructs differs in terms of
malleability. All of these findings converge to form the conclusion that
thinking styles and personality traits are two distinct constructs rather than
one being subordinate to the other. This concurs with the opinions of certain
researchers who also insist that style is not a redundant construct in the field
of individual differences (e.g., Busato et al., 1999; Fjell & Walhovd, 2004;
Zhang & Sternberg, 2006).
If we look closer at particular dimensions of thinking styles, we will find

that the overlap between different dimensions of thinking styles and
personality traits varies. Although most of the dimensions of thinking styles
only have low to moderate associations with personality traits, there are two
outliers: the external style and the hierarchical style. At the conceptual level,
it is understandable that the external style overlaps extraversion to a great
extent. The tendency to work with people is more likely to happen among
those who are outgoing and enjoy social gatherings. It was also found that
the hierarchical style was profoundly related to conscientiousness. People
with the hierarchical style prefer to prioritize tasks and deal with them in a
reasonable order. This type of person is more likely to be organized and
self-disciplined. A high level of self-discipline is actually a major
characteristic of conscientiousness. The different levels of overlap between
the particular dimensions of thinking styles and personality traits can also
be understood using Grigorenko and Sternberg’s (1995) model. As I
mentioned in Chapter 2, Grigorenko and Sternberg (1995) classified various
style models into three traditions: cognition-centered, personality-centered,
and activity-centered. From this conceptual perspective, personality-
centered styles would be closer to personality than cognition-centered and
activity-centered styles. In previous research on other style models, the
preliminary evidence has shown that cognition-centered styles, such as field
dependence-independence (Witkin, 1962), are related less to personality but
more to abilities, whereas personality-centered styles, such as the theory of
psychological types from Myers and McCaulley (1988), are related more to
personality but less to abilities (see Chapter 2). Unlike other style models,
the theory of thinking styles includes all of the three field of styles traditions.
Based on the findings shown in this research, it has been confirmed that the
hierarchical and external thinking styles are more personality-centered than
the other dimensions.
142 Chapter Five
However, does the close relationship of the hierarchical and external

thinking styles to personality traits mean that it is not worthwhile to study
them? I do not think so. Zhang (2017) argued that the correlation between
styles and other constructs in individuality (e.g., ability, intelligence, and
personality) did not warrant the collapse of the identity of styles in the field
of psychology and education. The high correlation found between the two
variables does not necessarily mean that they are identical, as it could be
also possible that a third variable may simultaneously influence them. It is
just like the high correlation that one may find between ice-cream sales and
the number of drowning accidents. The correlation is actually attributable
to the third variable—temperature. When the weather gets hotter, more
people buy ice-creams and, at the same time, more people go swimming,
which increases the probability of drowning. In this case, we would not say
that ice-cream sales and the incidents of drowning have identical variables.
By the same token, the correlations found between styles and personality is
probably due to an underlying common genetic mechanism. Furnham (2012)
also pointed out that it would be more desirable to examine the incremental
validity of styles than the correlation between constructs. The evidence that
styles have a unique explanatory power for student performance beyond that
of other constructs in individuality (e.g., personality and intelligence) could
consolidate the identity of intellectual styles as a valuable field. However,
previous research has rarely addressed this issue. The present study is one
of the very few studies that have examined the incremental validity of styles
with personality and its findings are inspiring. As expected, it was found
that thinking styles do uniquely contribute to the two developmental
outcomes (i.e., career decision self-efficacy and subjective well-being)
beyond personality traits. Apart from the above general finding, it was noted
that the hierarchical and external styles, which are the two thinking styles
that are highly correlated with personality traits, still provide an additional
explanation for the two student developmental outcomes. Based on the
regression coefficients, the predictive power of the hierarchical style and the
conscientiousness element for career decision self-efficacy were comparable.
Moreover, the predictive power of the external style for career decision self-
efficacy was even larger than for extraversion. Similar results were also
found in terms of subjective well-being. These findings specifically support
Zhang (2017) and Furnham’s (2012) arguments and suggest that high
correlations between particular styles and personality do not simply cancel
out their incremental validity. Another strength of the present research is
that the malleability of styles and personality were compared directly, which
is rarely found in previous studies. The findings provide evidence that the
nature of styles and personality differ. It was also noted that the external
style changed over one year while extraversion in personality traits did not
change, although the two dimensions highly overlapped according to the
analysis of the bivariate relationship. Again, this result suggests that the
examination of bivariate relationships between styles and other constructs
of individual differences based on simple correlations or regressions is
insufficient to clarify the identity issue. Unfortunately, this is exactly the
case in most of the existing research that has tried to distinguish styles from
personality or abilities. Therefore, more studies anchored in other style
models are required to both examine the incremental validity of styles and
compare their nature and the targeted constructs based on longitudinal data.
This will hopefully allow us to achieve a clear picture of the identity of
styles in the broader context of psychology and education.
For now, based on the current research’s findings on thinking styles and
personality traits, a tentative agreement can be reached that personality traits
and thinking styles are two distinct constructs and that it is worthwhile
studying thinking styles in order to have a comprehensive understanding of
student performance. However, is it possible to apply this conclusion to
other style models? Further studies of other style models that focus on the
incremental validity of styles, as well as comparing the malleability of styles
and personality, are also required.
Style Malleability
Another major piece of puzzle in the field of styles is the controversial issue
of style malleability. Can styles be socialized or modified? It is especially
vital to clarify this issue for educators, because it is meaningless to apply
the notion of styles in practice if they are fixed traits. Like the other
controversial issues mentioned in this book, this matter has been strongly
debated for decades. Many discourses on the issue were at the conceptual
level and the research evidence was not scientifically sound (Zhang, 2013).
Based on research into intelligence and personality, Mandelman and

Grigorenko (2012) speculated that styles are determined by genetic factors
to a lesser extent as they are the interface between intelligence and
personality. It implies that styles can be socialized to a greater extent than
intelligence and personality. However, the hypothesis regarding the genetic
effect has not yet been fully examined. As mentioned in Chapter 1, many
studies addressing the issue of style malleability have examined the
relationships between style and demographic factors, such as age, gender,
grade, academic disciplines, and hometown. In the present research, gender,
grade, major, and hometown were also found to make a difference in terms
144 Chapter Five
of students’ thinking styles. For example, in both rounds of data, it was

found that males scored significantly higher in both the liberal and the
internal thinking styles. This could be explained by gender-role
socialization to some extent because, according to our society’s traditional
gender stereotypes, males are encouraged to be more novelty-seeking and
less sociable. Zhang (2013) also performed a systematic review of the
research on style malleability. In her book, she delineated the research on
style and demographic factors, such as gender, nationality/geographical
locations, academic discipline, and occupation. She concluded that, in most
of the existing studies, intellectual styles, regardless of the theories that they
are anchored in, were found to vary based on the above demographic factors,
which alludes to the possible socialization effect.
However, as Zhang (2013) acknowledged in her book, most of these studies

are cross-sectional, which means that there are always alternative explanations
for these associations beyond style socialization. For example, the
relationship between age and styles could be due to the cohort effect or the
process of biological maturity. If the process of biological maturity (e.g.,
brain growth) leads to stronger or weaker preferences for certain specific
styles among older populations, the style differences detected in studies
across age groups would not mean that styles are shapeable or malleable.
By the same token, the gender differences in thinking styles could be
attributed to genetic differences between males and females rather than
the effect of gender socialization. Moreover, it is possible that the various
learning/working environments created by different academic
disciplines/occupations shape individuals’ styles, but there is also the
possibility that individuals with specific intellectual styles tend to seek out
learning activities, disciplines, and occupational settings that accommodate
them. Therefore, in order to clarify whether thinking styles are changeable,
longitudinal studies are needed to provide more rigorous evidence on style
malleability.
Nevertheless, longitudinal studies are rare in the field of intellectual styles.

Although Zhang (2013) identified quite a few longitudinal studies with and
without interventions in her book, she also pointed out their obvious
limitations. For example, most of the longitudinal studies were carried out
in the 1970s and 1980s, and their sample sizes were small with severe
attrition. The studies anchored in some style models (e.g., the reflectivity-
impulsivity construct) were mainly conducted among special populations
(e.g., children with special learning needs, juvenile delinquents, and
hyperactive children). Other limitations include the lack of effect sizes and
the lack of controlling for confounding variables.
The present research contributes to the discussions of the issue of style

malleability through a longitudinal research design with methodological
rigor. First, the research traced the change of styles in a large sample
selected from university students. Second, the effect sizes of the style
change were reported. Third, this research examined not only mean-level
change but also other indicators of change/stability (e.g., rank-order
consistency, individual stability, and ipsative consistency), which provides
a more comprehensive view of style change. Fourth, the research explored
the role of environmental factors in the socialization of styles with relevant
demographic factors as controls and personality traits included for
comparison. Consequently, this research has provided solid empirical
evidence that contributes to our understanding of the dynamic nature of
thinking styles.
My research’s major findings on the issue of style malleability can be

summarized in the answers to the following two questions: Are thinking
styles changeable? If they are, what factors cause these changes?
Are thinking styles changeable?

The present research examined four types of style change over one year.
The results showed that, at the mean-level, nine out of the thirteen thinking
styles had an increase, but the size of the change was small. This suggests
that, in general, thinking styles are changeable, but only to a small extent.
Furthermore, the results revealed that thinking styles had moderate to strong
rank-order stability, which means that students’ relative status among their
counterparts in terms of their level of each thinking style was almost
retained over the course of a year. At the individual level, after excluding
the influence of measurement errors, the research found that a small portion
of participants did have reliable changes in each dimension of thinking
styles over the course of a year. Meanwhile, the change in the configuration
of thinking styles within subjects was detected among nearly half of the
participants, although the extent of profile change was small. Given the
above results from the four types of changes examined, it can be confidently
concluded that thinking styles have a moderate to high degree of stability,
but this does not mean that they are fixed characteristics. Thinking styles
are still changeable, even if only to a small degree.
What factors cause the changes in thinking styles?

A small but interesting finding of the present research that I would like to
note is that hometowns made a difference in thinking styles in freshmen but
146 Chapter Five
not in juniors. This suggests that the influence of the cultural climate in the
students’ hometowns on their thinking styles may vanish after experiencing
university life for a number of years. However, it is also probably due to the
cohort effect. But which speculation is true? The results from the analysis
of the longitudinal data supported the former speculation. The effect of the
students’ hometowns on thinking styles became insignificant after a year.
This finding suggests that university experiences and environments may
play a role in shaping students’ thinking styles. The follow-up interview
study further confirmed this speculation. In the interviews, students mainly
attributed these style changes to the change of learning environment,
personal involvement in the learning environment, changes in daily life, and
other personal experiences (e.g., involvement in student organizations and
internships). It revealed that students’ thinking styles changed when they
made an effort to adapt to their changeable environments. Moreover, the
findings from the quantitative, longitudinal study on the role of learning
environments in students’ thinking styles provided more evidence that
consolidate this conclusion. Despite the unexpected results on some specific
relationships, all of the findings demonstrated that learning environments
significantly contributed to the development of thinking styles. The
quantitative findings on the hometown’s disappearing effect on thinking
styles after the students entered university, the quantitative analyses of the
role of various dimensions of learning environments in students’ thinking
styles, and the qualitative findings from interviews construct a triangulation.
This makes the conclusion that learning environments play a significant role
in shaping students’ thinking styles more solid and compelling. Based on
all of the above evidence, the dynamic nature of thinking styles is
indisputable.
Besides learning environments and experiences in university, the present

research also examined the role of familial environments in the socialization
of thinking styles. In line with the findings from earlier studies (Ho, 1998;
Robert J. Sternberg & Grigorenko, 1995; Tse, 2003; Yang & Lin, 2004;
Zhang & Postiglione, 2001), the present research found that students from
families with a higher socioeconomic status tended to have Type I thinking
styles that were characterized by higher cognitive complexity and creativity.
However, the result for the association between familial socioeconomic
status and thinking styles has limited implications for educational practice
because there is little we can do to change this situation even if we want to
promote children’s Type I thinking styles. It would be relatively realistic to
provide interventions targeted at parenting styles if one wants to cultivate a
particular thinking style. Therefore, the identification of the relationships
between parenting styles and thinking styles is more meaningful for
educational practice. As expected, the present research found that parenting

styles significantly contributed to thinking styles, albeit to a small degree.
This finding confirms the role of parenting styles in the socialization of
thinking styles. Interestingly, it was also found that the influence of
parenting styles on thinking styles decreased after a year at university,
which again alludes to the influence of their learning environments.
Although there was a decreasingly small influence found in the relationship
between parenting styles and thinking styles, it does not necessarily mean
that the influence of parenting styles should be ignored for two reasons. First,
during the transition to universities and colleges, it was the first time that
most of the students were away from their families for a long period of time.
During this period, learning environments gradually replaced family
environments to become proximal environments that exert more influence
on students. Despite this, the contribution of parenting styles to thinking
styles was still significant, which suggests that its influence is long-standing.
Second, we are not sure whether we would find a stronger influence from
parenting styles on thinking styles if we examined the relationships between
these two constructs in younger students (e.g., during adolescence or middle
childhood).
Although the development of thinking styles can be at least partially

attributable to the influence of two environmental factors (parenting styles
and learning environments), it has to be admitted that personality traits are
the major contributors to most of the thirteen thinking styles among the three
predictors. Given the findings on the competitive influence of family
environments, learning environments, and personality, it seems that
personality traits provide a baseline of thinking styles and then environmental
factors adjust them. Learning environments and experiences play an
important role in shaping students’ thinking styles, whereas the influence of
parenting styles on thinking styles is small but long-standing. By and large,
all of the findings lead to the conclusion that thinking styles are states rather
than traits. In other words, thinking styles are relatively stable, but they are
still changeable.
Can this conclusion be generalized for other models of intellectual styles?

Given the large number of style models that exist in the field of intellectual
styles, the answer to this question is complicated. In Zhang’s book (2013),
following an extensive review of style research over the past seven decades,
she found that the vast majority of studies indicated that the development of
styles was contingent upon gender, academic driplines, vocations, and other
environments. In addition, there were also a handful of longitudinal studies
which confirmed style change. Although she also pointed out that the
148 Chapter Five
methodology of most studies could have been more rigorous, the findings
indicating the dynamic nature of styles based on various style models cannot
just be coincidence. Therefore, she asserted that styles are malleable.
However, intellectual styles based on different style models may be
malleable to varying degrees. This could be discussed within the context of
Curry’s “onion” model and Grigorenko and Sternberg’s (1995) three-
tradition model. Taking the “onion” as a metaphor, Curry’s (1983) model
claimed that there are three layers of styles. The innermost layer involves
personality dimensions, the middle layer involves information processing,
and the outermost layer involves instructional preferences. Theoretically, it
is expected that styles which fall into the innermost layer are less changeable
than the ones in the middle layer, while styles that fall into the outermost
layer are the most changeable among the three “onion” layers. According to
Grigorenko and Sternberg’s (1995) model, all of the existing style
constructs fall into three traditions: cognition-centered, personality-centered,
and activity-centered. It is also speculated that personality-centered styles
are the least changeable while activity-centered styles are the most
changeable. The above hypotheses are partially supported by a few studies.
For example, Low and his colleagues (2005) found that the rank-order
stability of vocational interests (also called career personality types, which
is a style model that falls in the innermost layer of the “onion” and in the
personality-centered tradition) were even stronger than personality traits. In
contrast to other style models, learning approaches (styles falling in the
outermost layer of the “onion” and in the activity-centered tradition) are the
most widely examined style constructs regarding the issue of style
malleability in longitudinal studies. Although the direction of the change of
learning approaches is inconsistent across studies, most of the existing
research supported the idea that learning approaches are changeable.
However, as Zhang (2013) pointed out in her book, the existing literature is
insufficient to make a reliable comparison between different style models
in terms of their degree of malleability due to the paucity of longitudinal
studies in contrast to the large number of various style models and the lack
of methodological rigor in most studies. Therefore, it is recommended that
further research should be undertaken based on other style models with the
following features: 1) longitudinally tracing style change; 2) excluding the
effect of measurement errors; 3) involving a large sample and keeping the
rate of attrition low; 4) providing a fuller account of confounding factors
while detecting the effect of specific environmental factors; and 5) reporting
effect sizes. It will only be possible to compare the degree of style change
and ascertain if the malleability of styles is contingent upon their distance
from the center of Curry’s “onion” and their positions in Grigorenko and
Sternberg’s (1995) traditions after more rigorous studies have been

conducted across different style models.
Style Value
Just like other controversial matters in the field of intellectual styles, the
issue of style value has gone through many conceptual arguments. Some
scholars contend that there are no good or bad styles (e.g., Kwang &
Rodrigues, 2002; Messick, 1994; Riding, 1997). The value of styles depends
on the different contexts where they are used. When Sternberg (1997)
proposed the concept of thinking styles, he also stated that they are value-
free. However, he gradually changed his opinion based on the accumulating
evidence. In his work with Zhang, he supported the argument that styles are
value-laden rather than value-free in today’s society (Zhang & Sternberg,
2009). In Zhang’s (2017) recent book, she explicitly stated that Type I styles
that manifest cognitive complexity and creativity are more adaptive than
Type II styles that manifest norm-conformity. However, as I mentioned in
the preceding chapter, the range of developmental outcomes examined with
intellectual styles is very narrow, and it is particularly restricted to academic
outcomes. Studies examining the role of styles in non-academic
development are still limited (Gebbia & Honigsfeld, 2012). This prevents a
comprehensive understanding of the value of intellectual styles in student
development.
In the field of thinking styles, most studies on the relationships between

styles and non-academic outcomes were conducted by Zhang and her
colleagues. For example, Zhang’s studies (e.g., 2001d, 2002c; 2002d, 2002e,
2008c, 2010) have found that some thinking styles are more closely
correlated with certain developmental outcomes that are commonly viewed
as positive human attributes, such as higher levels of self-esteem, cognitive
development, and psychosocial development. The present research
expanded the range of developmental outcomes to include career decision
self-efficacy and subjective well-being. Compared with earlier studies, the
present research has three methodological strengths. First, it was a
longitudinal study. Second, the research put confounding variables in
control when examining the concerned relationships. Third, the research
consisted of a large sample. Therefore, the present research’s findings can
be considered to be more convincing.
In general, the present research’s findings are in line with Zhang’s (2017)
judgment about the value of intellectual styles. As expected, Type I thinking
styles positively contribute to students’ career decision self-efficacy and
150 Chapter Five
subjective well-being, while Type II thinking styles did so negatively with

only one exception (the local style to subjective well-being). Interviews also
revealed that students with Type I thinking styles had strengths when
considering career-relevant issues, as well as adapting to university life,
which led to higher levels of satisfaction and more positive emotions.
Together, these findings suggest that Type I thinking styles are more
adaptive than other styles in terms of these social and emotional outcomes.
In addition, Type III thinking styles do not manifest explicit value in terms
of these two developmental outcomes, thereby indicating that Type III
thinking styles are much more likely to be value-differentiated.
The findings of the present research in terms of career decision self-efficacy

and subjective well-being resonate with Zhang’s studies, which examined
the relationships between thinking styles and a series of students’ social and
emotional developmental outcomes, such as anxiety (Zhang, 2009),
emotion management (Zhang, 2008b), psychosocial development (Zhang,
2010), and self-esteem (Zhang, 2001d). These studies also found that Type
I thinking styles were more adaptive than others in terms of these social and
emotional outcomes. Although there were inconsistent findings regarding
the role of styles in academic outcomes, the findings for non-academic
outcomes are quite consistent. Therefore, one can confidently declare that
Type I thinking styles, such as legislative, liberal, and hierarchical, are
superior to other ones with regard to many of students’ social and emotional
outcomes.
In terms of Type II thinking styles, the findings are slightly ambivalent. In

the cross-sectional data at Time 1, it was found that the local thinking style
was positively related to career decision self-efficacy; in addition,
conservative and monarchic styles were positively related to subjective
well-being. Based on these findings, it is tempting to conclude that Type II
thinking styles are positively associated with desirable attributes like Type
I thinking styles, which would mean that they are value-free. However,
these associations were not detected in the analyses of either the cross-
sectional data at Time 2 or the longitudinal data. In fact, in the analyses of
the longitudinal data, some Type II styles (e.g., the monarchic thinking style)
were found to be negatively related to career decision self-efficacy and
subjective well-being. It indicated that the aforementioned positive
relationships between Type II thinking styles and the social and emotional
outcomes might be occasional and that, furthermore, they were definitely
not causal. These unstable correlations remind researchers that it is
insufficient to allude to the value of intellectual styles based on just one
cross-sectional study.
Nevertheless, there was one exception. The local style was still found to
positively contribute to subjective well-being in the analyses of the
longitudinal data. In the preceding chapter, I mentioned that it might be
relevant to the association between the local style and academic
performance. The assessment and evaluation in Chinese universities might
reward students using the local style, whereas a good academic performance
is probably one of the important origins of subjective well-being for
students. Besides this, there might be an alternative explanation from a
much broader perspective. Although success could be positively related to
happiness, the attributes that are desirable for success and happiness might
be slightly different. People who like paying attention to detail (the local
style) might not achieve great career success, but they could still have
positive emotions in daily life. However, as only one weak link was found
between the local style and subjective well-being, more studies are needed
to further examine this speculation.
Despite this small exception, the findings of the present research generally
suggest that Type I thinking styles have a greater positive value in student
development. One may naturally wonder if the superiority of Type I
thinking styles is culturally-specific or universal.
Zhang and Sternberg (2005) have pointed out that intellectual styles are
sociological, which means that “the use of a style is affected by the
preferences of the society in which one lives for various ways of thinking
(p. 2)”. At the conceptual level, it is understandable that culture has an
influence on shaping people’s styles. Based on this theoretical consideration
and Hofstede’s (1980) theory of cultural dimensions, Zhang (2013)
proposed a hypothesis that countries characterized by low power distance
(i.e., the unequal distribution of power is less acceptable), low uncertainty
avoidance (i.e. society is more tolerant of ambiguity), individualism (i.e.,
individual thoughts and behaviors are more respected), and masculinity (i.e.,
traditional masculine attributes, such as assertiveness and decisiveness, are
more valued) would encourage people to develop Type I styles, whereas
countries characterized by high power distance (i.e., the unequal distribution
of power is more acceptable), high uncertainty avoidance (i.e. society is less
tolerant of ambiguity), collectivism (i.e. conformity with other members of
the society is more appreciated), and femininity (i.e., traditional feminine
attributes, such as rule-following and obedience, are more valued) would
encourage people to develop Type II styles. After reviewing existing
research on the basis of several style constructs, such as field
dependence/independence, learning approaches, and career personality type,
Zhang (2013) concluded that, in spite of a few exceptions, the research
152 Chapter Five
hypothesis on cultural differences was generally supported. Nonetheless,

the findings of the present research, which was conducted among Chinese
participants, did not lend support to Zhang’s (2013) hypothesis. In cultural
classification, China is often considered to be a society that features high
power distance, high uncertainty avoidance, collectivism, and femininity.
According to Zhang’s (2013) hypothesis, Type II styles would be more
rewarded in this type of society. However, the present research found that
Type I thinking styles are more rewarded in contemporary Chinese society.
There are two possible reasons that could help to explain this finding.
First, even in a society that emphasizes hierarchies and conformance, the

attributes that are conducive to a successful career (e.g., being leaders)
might be similar to societies that emphasize equality and freedom. They all
require cognitive complexity, independent thinking, and creativity.
Therefore, although, at the collective level, people in a society with high
power distance, high uncertainty avoidance, collectivism, and femininity
are more likely to have Type II styles, people who have higher levels of
career maturity and the potential to achieve a better career still tend to use
Type I styles more.
Second, social norms and culture are gradually changing along with
economic evolution and the process of modernization (Matsumoto, 2002).
Although traditional Chinese societies manifested characteristics which
indicated high power distance, high uncertainty avoidance, collectivism,
and femininity, many things have been changing in the process of rapid
economic growth and accelerated modernization in Chinese society in
recent decades. Social norms and culture in contemporary Chinese society
is in the process moving from one polar to the other in Hofstede’s (1980)
culture dimensions. Accordingly, society’s attitudes to some individual
attributes are also changing. For example, in a series of studies conducted
by Chen and his colleagues in 1990, they found that shyness in Chinese
children was associated with a good adjustment, but this association had
become weaker in 1998. In 2002, a reverse correlation was even found
between shyness and child adjustment, which means that shy Chinese
children appear to be poorly adjusted (X. Chen, Cen, Li, & He, 2005). The
researchers claimed that this trend could be attributed to the rapid expansion
of a competitive market-oriented economy in recent decades. Under this
economic and social background, assertiveness and sociability are required
for success and therefore become more valued and encouraged by society.
By the same token, the change of culture and social norms may be also
responsible for the superiority of the Type I thinking styles found in the
present research. In contemporary Chinese society, characteristics linked
with Type I thinking styles seem to be more socially desirable.
In summary, given the findings generated from previous studies on other

styles models (see Chapter 1) and the results from the present research, it
can be concluded, at least in the current era, that Type I styles are more
adaptive than Type II styles, especially in terms of social and emotional
development. However, one must keep in mind that, from a historical
perspective, we are not sure if Type I styles would maintain their superiority
in student development in the far future (e.g., a hundred years later), as
social norms will evolve along with economic and social development.
Despite this uncertainty, we can now confidently claim that styles are value-
laden rather than value-free.
CHAPTER SIX
CONCLUDING REMARKS
Compared with the majority of studies in the field of styles, the present
research has advantages in various aspects, such as its longitudinal design
and its full consideration of the potential confounding variables. This has
led to solid and sound findings. Based on the findings of the present research
and the discussion in preceding chapters, some of the conclusions are well
grounded. First of all, this chapter will summarize the conclusions of the
research undertaken. In the second section, the theoretical contributions, as
well as its implications for both practice and research, will be elucidated.
This will be followed by a discussion of possible future directions in the
field of intellectual styles.
Conclusions
The present research addressed three major controversial issues in the field
of styles. It explored the socialization of students’ thinking styles and traced
their changes over the course of a year. It also examined the role of thinking
styles in terms of developmental outcomes with regard to career decision
self-efficacy and subjective well-being. This led to the following four major
conclusions.
1) Personality traits and thinking styles are two distinct constructs.
This conclusion is based on the following three findings. First, personality

traits only explain the limited variance in thinking styles. Second, thinking
styles contributed to the two developmental outcomes (i.e., career decision
self-efficacy and subjective well-being) beyond personality traits. Third,
over the course of a year, thinking styles were more malleable than
personality traits. These results suggest that thinking styles are distinct from,
rather than subordinate to, personality traits.
2) Thinking styles are stable rather than permanent traits.
Although styles are relatively stable, they are still changeable. Moreover,
Concluding Remarks 155
the development of thinking styles is a joint function of personality traits

and environmental factors, such as parenting styles and learning
environments. Among the three groups of predictors, personality traits are
major antecedents for most thinking styles. In addition, parenting styles and
learning environments also make unique contributions to shaping thinking
styles.
3) Thinking styles are value-laden.
Type I thinking styles manifest more adaptive attributes in terms of students’

career decision self-efficacy and subjective well-being, while, apart from
the local style, Type II thinking styles are less adaptive. In addition, Type
III thinking styles do not manifest explicit value in terms of these two
developmental outcomes.
4) Thinking styles partially mediate the environmental influence on

students’ developmental outcomes.
Thinking styles (mainly Type I) significantly mediate the relationships of

some aspects of parenting styles and learning environments to the
developmental outcomes (self-efficacy in terms of career decisions and
subjective well-being). In other words, parenting styles and learning
environments influence students’ self-efficacy and subjective well-being,
partially because they affect the students’ thinking styles.
One may naturally wonder whether the above conclusions can be

generalized to other style models of intellectual styles. Given the
tremendous number of style constructs in the field, it is not possible to
provide a simple answer to this question. Based on a comprehensive review
of the existing evidence, I am confident to say that, by and large, intellectual
styles are distinct from personality and are also partially malleable, but a
certain intellectual style’s distance from personality and the extent of its
malleability varies. On the one hand, it is quite possible that personality-
centered styles, as well as those that fall in the center of Curry’s “onion”,
are much closer to personality than thinking styles and these styles may also
be consequently more stable. On the other hand, activity-centered styles, as
well as those that fall in the outermost layer of Curry’s “onion”, are more
likely to be distinct from personality and are, therefore, more modifiable.
More rigorous research on various style models is needed to systematically
examine the supposition. Only a full examination of existing style models
will enable the exclusion of constructs that have been mistakenly labelled
by styles, thereby revealing the true spectrum of real style constructs. In
156 Appendices
terms of the value of intellectual style, the research, as well as whatever

theories of styles it is anchored in, showed consistent findings with regard
to the role that styles play in social and emotional development. Despite the
variability in academic outcomes, it does not mean that styles are value-free
because the inconsistency could be attributed to factors irrelevant to style
value (e.g., different modalities of assessment). Based on the general picture,
it can be tentatively stated that intellectual styles are value-laden, with Type
I styles more likely to have better developmental outcomes. However, we
should always keep in mind the possible interference from cultural
influences and the possibility that social reform and economic development
could have an effect on the change of style values in the future.
Contributions and Implications

The present research’s findings on the identity, malleability, and value of
thinking styles have significant theoretical contributions, practical
implications, and insights for future studies.
Theoretical contributions
First, and most importantly, the present research has further clarified the
nature of styles by addressing three major controversial issues. Bewildering
information and the lack of consensus have been seen as the major obstacles
that impede the advancement of the theory of styles (Peterson et al., 2009).
This research has provided empirical evidence to address three of the major
controversial issues in the field: how personality traits and styles are related
to each other; whether styles are malleable; and whether styles are value-
laden. The empirical evidence provided has profound significance for the
clarification of these controversial issues due to its three major strengths.
First, most of the relevant studies in the existing literature are based on
cross-sectional data, which limits the interpretation of the relationships
within covariations. The longitudinal design provides greater credibility
when interpreting the possible directions of the examined relationships.
Second, the present research examined personal and environmental factors
simultaneously and across time, which provides a comprehensive and
dynamic picture of the relationships involved. Third, the present research
adopted mixed methods where the qualitative data was used to triangulate
and further interpret the findings of quantitative studies, which makes its
conclusions more convincing. This compelling evidence significantly
contributes to the clarification of three of the controversial issues in the field
of styles and facilitates a better understanding of the nature of intellectual
styles, thereby advancing the field.
Second, the present research provides empirical support for certain theories.
For instance, the research discovered the combined functions of personality
traits, parenting styles, and learning environments on the development of
students’ thinking styles. This supports Bronfenbrenner’s bioecological
theory (Bronfenbrenner, 1979; Bronfenbrenner & Ceci, 1994), which
claims that human development is a joint function of person and context.
The research also revealed the mediating function of thinking styles in
student development; this supports Biggs’ 3P model (1990), which states
that presage factors (experiences) influence product factors (performance)
through their effect on process factors (styles). Furthermore, as the 3P
model was initially proposed to predict students’ academic performance, the
present research not only supports this model but also extends it to non-
academic performance in various areas (e.g., career maturity and subjective
well-being).
Implications for educational practice

In addition to making contributions to the above theories, the present
research also has profound implications for educational practice. The
research mainly provides insights into the following two practical concerns:
“which thinking styles should be cultivated to benefit student development?”
and “how can these thinking styles be cultivated?” In addition, based on the
above findings and implications, the present research also sheds light on the
discussions surrounding the “matching hypothesis” in education.
First, the value of thinking styles identified in the present research provides
information for parents, teachers, administrators in educational institutions,
and students on what styles they should encourage and cultivate. Based on
the findings of the present research, Type I thinking styles (which are
characterized as creativity-generating and having higher levels of cognitive
complexity) seem to be more beneficial to student development. Keeping
this point in mind during daily educational practice has profound
implications, especially when traditional Chinese culture treats norm-
conformity positively. Parents and teachers should transfer the emphasis
from obeying authority to encouraging creativity and autonomy.
Educational administrators could conduct some workshops or courses that
aim at cultivating students’ Type I styles. In addition, students should be
aware of practicing their Type I styles in daily life.
Second, while the findings on the value of thinking styles provide insights
into the type that should be cultivated by parents and educational
practitioners, the findings on the malleability of thinking styles indicate that
158 Appendices
the cultivation of thinking styles is actually possible, as well as providing

potential ways to do this. For example, parents are encouraged to provide a
secure climate by being responsive and caring, as well as minimizing their
strict attitudes to their children’s lives in order to promote the development
of Type I thinking styles. If teachers aim to cultivate Type I styles among
students, then they need to teach in a constructivist way, thereby focusing
on the process of knowledge construction rather than merely on knowledge
transmission. Teachers are also advised to create an academic atmosphere
that facilitates proactive learning and provides students with opportunities
to inspire one another. In addition, it is recommended that teachers ensure
their students’ involvement during student-student cooperative activities
and also provide proper supervision during this process. University
administrators are advised to create a supportive environment for teachers
to practice constructivist methods of education and also to provide abundant
learning facilities so that their students can always find resources and
opportunities when they want to practice particular styles. In addition, the
findings also help to enlighten students. The findings suggest that students
should choose courses that explore the construction of knowledge rather
than those that rely on memorizing material if they want to promote their
Type I thinking styles. Students should also be encouraged to spend time
with friends and classmates who proactively enrich their knowledge and
enthusiastically put this knowledge into practice. It is also better for students
to be more involved in collaborative activities and to interact with their
peers in order to experience the whole process of communication, conflict,
and integration. Nevertheless, it should be noted that, given the degree of
style stability, it is unlikely that practitioners will be able to quickly change
their students’ thinking styles. Moreover, given the significant individual
differences in style changes, it is suggested that practitioners should take
individual differences into consideration when designing training programs.
Third, the present research provides insights to the “matching hypothesis”

of styles. The “matching hypothesis” assumes that individuals perform best
when the stylistic demands of environments match their own styles. Based
on this hypothesis, one may claim that teachers should deploy different
teaching styles to accommodate their students’ various styles in order to
improve learning. However, individualized instruction is incredibly difficult,
if not impossible. This is also one of the major critiques that the field of
intellectual styles frequently receives. Some scholars have claimed that it is
unrealistic to ask teachers to routinely change their instruction styles to meet
the needs of every single student and so, for this reason, they have hastily
dismissed the relevance of styles to educational practice. However, the fact
that Type I styles are generally more adaptive in contemporary societies and
are, therefore, malleable must lead us to rethink the “matching hypothesis”.

Would it be more desirable to ask teachers to cultivate and promote Type I
styles instead of accommodating and consolidating their students’ original
styles? The answer is likely to be yes. One study investigated students’
preferences for teaching styles; it found that Type I teaching styles were
generally preferred by students, regardless of their own style (Zhang, 2007a).
This finding implies that all students, whatever their style, may benefit from
teachers who adopt Type I teaching styles. Given this fact and the present
research’s findings on style value and malleability, it is necessary to reflect
on and reconstruct the “matching hypothesis” in the field of intellectual
styles.
Implications for research

There are also some issues emerging from the present research’s findings
that are worth pursuing. For example, the bewildering attitudes of students
to the psychological autonomy granted by parents need to be further
explored through a comparison with samples from western countries to
deepen the understanding of the cultural differences behind the phenomenon.
The type of assessments on students’ thinking styles requires further
exploration with a close look at their content and student motivation.
Furthermore, it is better to examine exactly what happens in the process of
student-student cooperation instead of merely focusing on the existence of
cooperative activities. Although data from the qualitative part of the present
research implies that the different roles students play during cooperation, as
well as their varying levels of involvement, may have an impact on different
thinking styles, a systematic examination of this issue is still required. For
this purpose, items in the Inventory of Students’ Perceived Learning
Environment (ISPLE), which provides a scale for student-student cooperation,
could be updated to assess the process of cooperative activities instead of
merely assessing the availability of cooperative opportunities. Using an
updated scale in future quantitative studies would help to confirm if the
differential influence of factors during the cooperation process on thinking
styles is a common phenomenon among university students. By the same
token, it is also recommended to further examine the nature of different
learning facilities and the ways in which students use them rather than just
confirming their availability. The relevant ISPLE items could then be
updated in order to get a systematic understanding of how various learning
facilities make a difference in the development of students’ thinking styles.
In addition, the positive role of a local style with regard to subjective well-
being requires further studies. For example, a study that simultaneously
explores thinking styles, subjective well-being, and academic achievement
160 Appendices
needs to be carried out so that the possible mediating role of academic

achievement in the relationship between the local style and subjective well-
being can be fully examined.
Future Directions
The field of styles has suffered from the bewildering array of style
constructs for decades. Although efforts to integrate these theories have
been made in recent decades, the criticisms of the field are still ongoing.
One of the reasons for this is that some of the fundamental issues that
threaten the identity and practical value of style in education remain
unsolved. This book reviewed the existing literature and discussed the
findings of a recent longitudinal study on thinking styles. It demonstrated
that 1) thinking styles are distinct from personality rather than subordinate
to it; 2) thinking styles are stable to some extent but still changeable; and 3)
thinking styles are value-laden with Type I thinking styles being more
adaptive. As the model of thinking styles is one of the most recent style
theories and because it has several strengths compared with many other
style constructs (as discussed in Chapter 1), the findings related to it provide
considerable insights, which enable us to understand the nature of
intellectual styles. I believe that the research findings disclosed in this book
help fortify the credibility of the field of styles and, at the same time, instill
more confidence in educational practitioners.
Although the evidence shown in this book is compelling, it has to be

acknowledged that the current research findings only uncovered a small part
of the picture. For example, the interval used in this study was only a year
long. It is possible that more substantial changes would be detected when
this interval is extended. It is also worthwhile to examine whether thinking
styles are more changeable in younger groups. Therefore, it is suggested
that in the future longitudinal studies with longer intervals using different
age groups should be undertaken. In addition, research should also be
conducted among people in other cultural contexts to see if the current
findings could be further generalized. Moreover, a longitudinal research
design with three or more time points is also recommended because it is
advantageous for resolving both the random error and the time precedence
issues when examining mediating relationships. This would help to
facilitate a better understanding of both the dynamic and the possible causal
relationships involved.
More importantly, although the current findings provide valuable

information about the nature of thinking styles, the character of many other
styles anchored in various theories is still unclear. More longitudinal

research based on other style models needs to be designed and conducted
with methodological rigor. This will promote the integration of massive
style constructs, facilitate internal dialogue among the different theories, as
well as responses to critiques, which will help to fundamentally advance the
field of intellectual styles.
In the concluding remarks of the final chapter as well as the whole book, I
would like to remind readers that, although the present research only
provides a piece of the puzzle for intellectual styles, the findings elucidated
are compelling and intriguing. Given the well-established facts that thinking
styles explain students’ performance beyond personality and that they are
more changeable than personality, educational practitioners should attempt
to enhance student performance by encouraging them to develop particular
thinking styles. It is also exciting news because it does not mean that
teachers need to tailor their instructions to meet every student’s
individualized stylistic needs in order to optimize their performance.
Research suggests that teachers should make concerted efforts with parents
to foster students’ Type I styles, which would benefit their long-term
development.
APPENDICES
Appendix 1: The Threefold Model of Intellectual Styles
Style Construct TYPE I TYPE II TYPE III

Learning Approach
Deep Surface Achieving
(Biggs, 1978)
Realistic,
Career Personality Type Investigative,
Artistic Conventional
(Holland, 1973) Social,
Enterprising
Mode of Thinking
Holistic Analytic Integrative
(Torrance, 1988)
Personality Type Thinking, Feeling,
Intuitive,
(Myers & McCaulley, Sensing, Judging Introversion,
Perceiving
1988) Extraversion
Abstract Random,
Mind Style Concrete Concrete
Abstract
(Gregorc, 1985) Random Sequential
Sequential
Decision-Making Style
Innovation Adaptation
(Kirton, 1976)
Conceptual Tempo
Reflectivity Impulsivity
(Kagan, 1966)
Structure of Intellect Divergent Convergent
(Guilford, 1967) Thinking Thinking
Perceptual Style Field
Field Dependent
(Witkin, 1962) Independent
Legislative,
Executive, Local, Oligarchic,
Thinking Style Judicial, Global,
Monarchic, Anarchic,
(Sternberg, 1988) Hierarchical,
Conservative Internal, External
Judicial
Note. Adapted from Zhang and Sternberg, 2009, p. 65.
Thinking Styles: Identity, Value, and Malleability 163
Appendix 2: The Inventories in the Research

The Parenting Style Index
The Parenting Style Index (Steinberg et al., 1994) is a measure of perceived
parenting styles (parenting styles hereafter for brevity), which is designed
based on Baumrind’s (1967) framework and subsequent research (Maccoby
& Martin, 1983). It has three dimensions: acceptance/involvement,
strictness/supervision, and psychological autonomy granting. There are nine
items each for the acceptance/involvement dimension and the psychological
autonomy granting dimension with a 4-point scale (1 = strongly disagree, 4
= strongly agree). There are eight items for the strictness/supervision
dimension with two items to indicate how late they can stay up on school
nights and weekend nights (7-point scale), three items to indicate to what
extent parents try to understand their children’s lives, and another three to
indicate to what extent they really understand their children’s lives (3-point
scale). In this research, the scales of the items in all of the dimensions were
changed into 7-point Likert scale to achieve consistency within the whole
questionnaire. The means of the dimensions for acceptance/involvement
and strictness/supervision were used to partition parenting styles. Parenting
styles with scores higher than both the mean in acceptance/involvement and
the mean in strictness/supervision were classified as the authoritative
parenting style. Parenting styles with scores lower than the mean in
acceptance/involvement and scores higher than the mean in
strictness/supervision were classified as the authoritarian parenting style.
Parenting styles with scores higher than the mean in acceptance/involvement
and scores lower than the mean in strictness/supervision were classified as
the indulgent parenting style. Parenting styles with scores both lower than
the mean in acceptance/involvement and the mean in strictness/supervision
were classified as the neglectful parenting style. This inventory has been
used in different cultures and has good reliability and validity (e.g., Chao,
2001; Milevsky et al., 2007; Steinberg et al., 1994). This inventory has been
recently utilized among Chinese high school students (C. Chen & Fan, 2010)
and demonstrated acceptable reliability and validity.
It is noted that, due to its poor performance in terms of reliability and the
validity in the pilot study, item 12 was modified in the main study. Besides,
it should be also noted that, at Time 1 (pretest) in the main study, students
were asked to respond to the PSI according to their experience during the
first 16 years of their lives, while at Time 2 (posttest) students were asked
to respond to the PSI according to their experience in the year between the
pretest and the posttest. Because all the students resided on campus instead
164 Appendices
of at home after they entered university, some items in the PSI, especially
the items in the scale of strictness/supervision were no longer suitable for
their situation. Therefore, relevant items were modified slightly in terms of
their wording for investigation at Time 2. Because the test instructions and
some items were different between Time 1 and Time 2, the psychometric
properties of the PSI for both Time 1 and Time 2 were analyzed.
Factor Structure
The result of the EFA for the PSI is favorable. Principal Component
Analysis with Varimax Rotation was carried out. Three factors were
forcibly extracted according to the theoretical structure. At Time 1, three
factors explained 39.72% of the variance in the data. Apart from item 12,
which cross-loaded on the dimensions of psychological autonomy granting
and acceptance/involvement, all items loaded on the theoretically expected
factors. At Time 2, three factors explained 46.62% of the variance in the
data. All items loaded on the theoretically expected factors and none of them
were cross-loaded.
Reliability
The Cronbach’s alpha coefficients of the PSI’s three scales for Time 1
were .73 (acceptance/involvement), .75 (psychological autonomy-granting),
and .81 (strictness/supervision). The Cronbach’s alpha coefficients of the
PSI’s three scales for Time 2 were .82 (acceptance/involvement), .81
(psychological autonomy-granting), and .89 (strictness/supervision). These
results indicated that the reliability of the PSI was respectable. The current
internal consistency of the PSI was comparable with the results in Chen’s
(2010) study of Chinese adolescents.
Inventory of Students’ Perceived Learning Environment

This inventory is self-designed to meet the needs of the present research; it
assesses student perception of the general learning environment. Most of the
existing inventories on learning environments were designed for course
evaluation, which means they only examine variables happening inside
classroom and are usually aimed at assessing the learning environment in
one specific course. Furthermore, many inventories focus on variables in
teaching, such as course design, assessments, and teaching methods, which
have been criticized for being too teacher-centered (e.g., Centra, 1993;
Kolitch & Dean, 1999). Fortunately, some recently developed inventories
have noticed some other dimensions or levels in learning environments. For
example, Pace and Kuh (2007) designed a questionnaire to assess students’
experience of activities on campus and Wierstra, Kanselaar, van der Linden,

and Lodewijks (1999) added the dimension of student participation in their
inventory. Others included teacher-student and student-student relationships in
their inventories (e.g., Entwistle et al., 2003; Kember & Leung, 2009a).
However, though these inventories succeeded expanding the breadth of the
dimensions included in questionnaires that assess learning environments to
some extent, none of them are comprehensive enough to provide a
systematic examination of the general learning environment. Therefore,
given that existing inventories on learning environments are either too class-
specific or include a limited range of dimensions, it is meaningful to develop
an encompassing learning environment inventory that can assess students’
general perception of various learning dimensions both inside and outside
the classroom. The design and the utilization of this inventory can facilitate
an exploration of the general learning environment’s influence on students’
developmental outcomes within one study, rather than merely focusing on
the influence of one specific course or limited environmental dimensions.
As illuminated in the Chapter 3, four common characteristics of an effective

learning environment have been identified based on learning theories from
constructivism and corresponding studies: a) constructivist learning; b)
student autonomy; c) interaction and cooperation; and d) clear goals and
coherence of curricula. The items selected and designed for this inventory
aim to reflect these four characteristics.
The dimensions of this inventory were constructed based on the major

dimensions described in Entwistle, McCune, and Hounsell’s (2003)
conceptual model and the dimensions frequently found in factor analysis
results of corresponding inventories, such as the Student Engagement
Questionnaire (Kember & Leung, 2009a), the Experiences of Teaching and
Learning Questionnaire (Entwistle, McCune, & Hounsell, 2003), the
College Student Experience Questionnaire (Pace & Kuh, 2007), and the
Inventory of Perceived Study Environment (Wierstra et al., 1999). Efforts
were made to integrate these dimensions and, finally, eight were identified:
1) constructivist-oriented teaching; 2) clear goals and coherence of curricula;
3) student autonomy; 4) assessment and assignment; 5) teacher-student
interaction; 6) student-student cooperation; 7) peer morale and identities;
and 8) learning facilities.
There are 32 items in this inventory in total, with four items for each
dimension. Some items were borrowed and modified from the four
inventories mentioned above (i.e., Student Engagement Questionnaire,
Experiences of Teaching and Learning Questionnaire, College Student
166 Appendices
Experience Questionnaire, and Inventory of Perceived Study Environment),

while some items were self-designed according to the characteristics that
are theoretically supposed to be in the particular dimensions. The procedure
of translation and back-translation was executed for all of the items. A 7-
point scale was utilized to ask students to indicate how accurately each
statement described their perceived learning environment (1=extremely
inaccurate, 7=extremely accurate). The contents of this inventory have been
reviewed by experts and a few students responded to it as a preliminary trial
before it was used in a pilot study. According to the results of the pilot study
on the analysis of ISPLE’s psychometric properties, two problematic items
were modified for the main study. The psychometric properties of the final
inventory are presented below.
Factor Structure
The results from CFA for the ISPLE preliminarily supported the eight-factor
model. According to the three criteria of good indices proposed by Hu and
Bentler (1999) (irrelevant to sample size; punish complex models; and
sensitive to misfit models), RMSEA, NNFI, and CFI are considered to be
good indicators of a model fit (Hu & Bentler, 1999; Wen, Hau, & Marsh,
2004). Hu and Bentler identified the cutoff values for RMSEA, NNFI, and
CFI as .06, .95, and .95, respectively. However, Marsh, Hau, and Wen (2004)
criticized Hu and Bentler’s (1999) decision, as they felt that the rules were
too stringent and may increase Type I errors. They recommended that
RMSEA <.08 is acceptable, and so are NNFI and CFI >.90. Marsh, Hau,
and Wen (2004) also claimed that, in large samples, RMSEA, NNFI, and
CFI are better indices than the Chi-square. The model fit indices for the
ISPLE eight-factor model were as follows: Chi-square (df = 435, N = 926)
= 2459.22, p<.001, RMSEA = .071, NNFI = .94, and CFI = .95. According
to this, the ISPLE’s eight-factor model was acceptable.
Reliability
The Cronbach’s alpha coefficients of the ISPLE’s eight scales at Time 1

were .71 (constructivist-oriented teaching), .62 (clear goals and coherence
of curricula), .66 (student autonomy), .60 (assessment and assignment), .75
(teacher-student interaction), .79 (student-student cooperation), .72 (peer
morale and identities), and .75 (learning facilities). At Time 2, the
estimation of the reliability of eight scales were .75 (constructivist-oriented
teaching), .69 (clear goals and coherence of curricula), .64 (student
autonomy), .67 (assessment and assignment), .75 (teacher-student
interaction), .79 (student-student cooperation), .76 (peer morale and
identities), and .63 (learning facilities). Although some of the dimensions

still have room for improvement, the ISPLE’s reliability could be
considered to be generally acceptable.
NEO Five-Factor Inventory-3

The NEO Five-Factor Inventory-3 (NEO-FFI-3, McCrae & Costa, 2007) is
the second revision of the NEO Five-Factor Inventory (NEO-FFI, Costa &
McCrae, 1992), which is a widely used short version of the NEO Personality
Inventory (NEO-PI, Costa & McCrae, 1985). It assesses five dimensions of
personality: Neuroticism, Extraversion, Openness, Agreeableness, and
Conscientiousness. For each dimension, there are 12 items with a 5-point
Likert scale. In this research, it was changed into a 7-point Likert scale (1 =
strongly disagree, 7 = strongly agree) to achieve consistency within the
whole questionnaire. The reliability and validity of NEO-FFI have already
been confirmed by ample research (e.g., Costa & McCrae, 1992; Courneya
& Hellsten, 1998; Saucier, 1998). After two revisions, NEO-FFI-3 showed
further improved psychometric properties and became more readable
(McCrae & Costa, 2007). The present research updated a Chinese version
of the NEO-FFI (Ye, 2008; Zhang & Huang, 2001) by replacing 15 old
items in NEO-FFI with 15 corresponding new items from NEO-FFI-3. The
15 items were translated from English to Chinese especially for this research.
All 15 items have gone through the translation and back-translation
procedures.
Factor Structure
The results from the CFA for the NEO-FFI-3 showed that the data did not
fit the five-factor model very well. The model fit indices were as follows:
Chi-square (df = 1700, N = 926) = 11557.48, p<.001, RMSEA = .079, NNFI
= .84, and CFI = .84. The misfit of the five-factor model has been
demonstrated in previous studies using NEO-FFI (e.g., Egan, Deary, &
Austin, 2000; Holden & Fekken, 1994). The current CFA results from the
NEO-FFI-3 were consistent with these previous findings. McCrae et al.
(1996) argued that the model misfit was because of a problem in Structural
Equation Modeling, while Gignac et al. (2007) claimed that the problem
was more likely to be in the five-factor structure than in the method. The
issue of the misfit of the NEO-FFI’s five-factor model calls for further
research.
168 Appendices
Reliability
The Cronbach’s alpha coefficients of the NEO-FFI’s five scales

were .82/.82 (Time 1/Time 2, Neuroticism), .78/.75 (Time 1/Time 2,
Extraversion), .71/.74 (Time 1/Time 2, Openness to experience), .63/.68
(Time 1/Time 2, Agreeableness), and .79/.77 (Time 1/Time 2,
Conscientiousness). Generally speaking, these results indicate that the
reliability data of the NEO-FFI scales was acceptable. The internal
consistency of the NEO-FFI-3 found in Western samples ranged from .72
(Agreeableness) to .88 (Conscientiousness) (McCrae & Costa, 2007).
Compared with the original English version, the current Chinese version
needs to make an effort to enhance the reliability of the Agreeableness scale.
However, compared with the Chinese version of the NEO-FFI (Costa &
McCrae, 1992), the internal consistency of the Openness scale has been
improved.
Thinking Style Inventory-Revised II

The Thinking Style Inventory-Revised II (TSI-R2, Sternberg, Wagner &
Zhang, 2007) is the second revised version of the Thinking Style Inventory
(TSI, Sternberg & Wagner, 1992) based on Sternberg’s theory of mental
self-government. It is a self-report test in which respondents rate themselves
on a 7-point scale, with 1 indicating that the statement does not describe
them at all and 7 indicating that the statement characterizes them extremely
well. There are 65 items, which assess the 13 different style scales.
Both the TSI (Sternberg & Wagner, 1992) and the Thinking Style
Inventory-Revised (TSI-R, Sternberg, Wagner, & Zhang, 2003) have been
found to have acceptable reliability and validity in previous studies in
several cultures, including Hong Kong, mainland China, the United
Kingdom, and the United States (Zhang, 2004a, 2004d, 2005b, 2008b,
2008c; Zhang & Higgins, 2008), apart from its relatively low reliability for
the Anarchic scale. As the latest version, TSI-R2 has been applied in the
Chinese context (e.g., Zhang, 2009, 2010), and it was found that the internal
consistency for the Anarchic scale had improved. The psychometric
properties of the TSI-R2 in the present research are described below.
Factor Structure
The results from the CFA for the TSI-R2 preliminarily supported the 13-
factor model. The model fit indices were as follows: Chi-square (df = 1937,
N = 926) = 9021.39, p<.001, RMSEA = .063, NNFI = .93, and CFI = .94.
Reliability
The Cronbach’s alpha coefficients of the TSI-R2’s 13 scales were .81/.80

(Time 1/Time 2, legislative), .75/.72 (Time 1/Time 2, executive), .76/.79
(Time 1/Time 2, judicial), .69/.68 (Time 1/Time 2, global), .72/.76 (Time
1/Time 2, local), .86/.87 (Time 1/Time 2, liberal), .79/.80 (Time 1/Time 2,
conservative), .79/.82 (Time 1/Time 2, hierarchical), .74/.75 (Time 1/Time
2, monarchic), .75/.80 (Time 1/Time 2, oligarchic), .63/.71 (Time 1/Time 2,
anarchic), .77/.76 (Time 1/Time 2, internal), and .83/.83 (Time 1/Time 2,
external). The results were comparable with previous findings in other
Chinese samples, where the internal consistency of the 13 scales ranged
from .62 to .85 (e.g., Zhang & He, 2011; Zhang & Wong, 2011). The
Cronbach’s alpha coefficient of the anarchic scale was the lowest, but it has
been improved compared with the magnitude found in previous versions of
the TSI (e.g., Fan, 2006; Zhang & Higgins, 2008).
Career Decision Self-Efficacy Scale––Short Form

The Career Decision Self-Efficacy Scale––Short Form (CDSE-SF)
developed by Betz, Klein, and Taylor (1996) aims to assess career decision
self-efficacy. It is the short version of the Career Decision-Making Self-
efficacy Scale (CDMSE, Taylor & Betz, 1983). The CDSE-SF includes 25
items, with five items for each subscale (self-appraisal, occupational
information, goal selection, planning, problem-solving). Originally, it
required participants to respond to each item on a 5-point scale (1 = no
confidence at all; 5 = complete confidence). However, in this research, it
was changed into a 7-point Likert scale to achieve consistency within the
whole questionnaire. The coefficients of the reliability range from .73 to .83
for the subscales and .94 for the total score (Betz et al., 1996). The
convergent validity has been also supported by its relationship to vocational
indecision, vocational decidedness, occupational self-efficacy, and general
self-efficacy (Betz et al., 1996; Hampton, 2006). This inventory has been
applied to Chinese graduate students in one study (Jin, Watkins, & Yuen,
2009) and demonstrated acceptable reliability and validity. Two items
(items 13 and 17) had poor psychometric properties in the pilot study and
were, therefore, slightly modified for the main study.
Factor Structure
The results from the CFA for the CDSES-SF indicated that the 5-factor
model was acceptable. The model fit indices were as follows: Chi-square
(df = 264, N = 926) = 1604.27, p<.001, RMSEA = .074, NNFI = .96, and
170 Appendices
CFI = .97. The values of RMSEA were similar to those (RMSEA = .07)
found in another study conducted among Chinese postgraduates (Jin et al.,
2009). The value of CFI was higher in the present study than those (CFI
= .86) found in Jin et al.'s (2009) study.
Reliability
The Cronbach’s alpha coefficients of the CDSES-SF’s five scales at Time

1 were .76 (occupational information), .76 (self-appraisal), .77 (goal
selection), .77 (planning), and .72 (problem solving). The estimates of the
reliability of the five scales at Time 2 were .76 (occupational information), .78
(self-appraisal), .80 (goal selection), .80 (planning), and .79 (problem
solving). Generally speaking, the reliability of the CDSES-SF was favorable
in the main study and better than the estimates of reliability found in Jin et
al.’s (2009) study among Chinese postgraduates.
The Satisfaction with Life Scale and The Positive and Negative
Affect Schedule
The Satisfaction with Life Scale (SWLS; E. Diener et al., 1985) is a widely
used inventory that assesses the cognitive component of subjective well-
being and life satisfaction. The SWLS contains five items in total with a 7-
point scale (1 = not at all true of me; 7 = exactly true of me).
This inventory has been applied in different countries besides English-

speaking ones, such as the Arab world, the Netherlands, Spain, and France
(e.g., Abdallah, 1998; Arrindell, Heesink, & Feij, 1999; Atienza, Balaguer,
& García-Merita, 2003; Blais, Vallerand, Pelletier, & Brière, 1989). All of
these studies show the inventories’ favorable psychometric properties with
high reliability and great internal and external validity.
The Positive and Negative Affect Schedule (PANAS; Watson et al., 1988)
has also been widely used to assess the affective component of subjective
well-being. The inventory consists of two 10-item subscales for positive
affect and negative affect. Participants were asked to respond to a 7-point
scale (1 = not at all true of me; 7 = exactly true of me) to indicate their
affective experience in the past month. Satisfactory psychometric properties
have been supported by various studies in different cultures (e.g., Krohne,
Egloff, Kohlmann, & Tausch, 1996; Melvin & Molloy, 2000; Paradowski,
2001).
The SWLS and the PANAS were used together in the present research as
the indicators for subjective well-being (SWB). Thus, the two inventories
were put together for factor analysis. The psychometric properties of the
two inventories are shown below.
Factor Structure
The results from CFA supported the three-factor model as theoretically

expected. The model fit indices were as follows: Chi-square (df = 271, N =
926) = 1591.76, p<.001, RMSEA = .073, NNFI = .96, and CFI = .96.
Reliability
The Cronbach’s alpha coefficients were .83/.83 (Time 1/Time 2) for the
scale of life satisfaction, .89/.90 (Time 1/Time 2) for positive affect,
and .91/.91 (Time 1/Time 2) for the scale of negative affect. These results
indicated that the reliability of the measures for SWB were very good. The
results were very consistent with those in Ye’s (2008) study among Chinese
samples, indicating that the two measures for SWB were stable and reliable.
172 Appendices
Appendix 3: Sample Inventory Items

Inven-
Dimension Sample Item
tory
PSI acceptance/involvement I can count on my parents to help me out, if I have some kind of problem.
*My parents say that you shouldn't argue with adults.
granting
strictness/supervision How much do your parents try to know what you do with your free time?
ISPLE Constructivist-oriented
The teaching helps me to think about the evidence underpinning different views.
teaching
clear goal and coherence of
It is clear to me what I am supposed to learn
curricula
student autonomy We are allowed some choices over what aspects of the subjects to concentrate on.
Assessment and assignment To do well in courses, I have to think critically about the topics.
There is sufficient feedback on activities and assignments to ensure that we learn from
teacher-student interaction
the work we do.
student-student cooperation We always have opportunities to work together with others on learning.
The students around me always search for further knowledge other than what have been
peer morale and identities
taught in class.
I can always find materials other than textbooks in library, database, or other places on
learning facilities
campus to do further reading.
NEO-FFI- neuroticism At times I have felt bitter and resentful.
3
extroversion *I prefer jobs that let me work alone without being bothered by other people.
I enjoy concentrating on a fantasy or daydream and exploring all its possibilities, letting it grow
openness
and develop.
agreeableness When I’ve been insulted, I just try to forgive and forget.
conscientious-
I keep my belonging clean and neat.
ness
TSI-R2 legislative When faced with a problem, I use my own ideas and strategies to solve it.
executive I like to figure out how to solve a problem following certain rules.
judicial I like situations where I can compare and rate different ways of doing things.
global I tend to pay little attention to details.
local I prefer to deal with problems that require me to attend to a lot of details.
liberal I like to challenge old ideas or ways of doing things and to seek better ones.
conservative I stick to standard rules or ways of doing things.
hierarchical I like to set priorities for the things I need to do before I start doing them.
monarchic When talking or writing about ideas, I prefer to focus on one idea at a time.
When discussing or writing about a topic, I stick to the points of view accepted by my
oligarchic
colleagues.
I can switch from one task to another easily, because all tasks seem to me to be equally
anarchic
important.
internal I like to control all phases of a project, without having to consult with others.
external When starting a task, I like to brainstorm ideas with friends or peers.
174 Appendices
CDSE-SF occupational information Find out the employment trends for an occupation over the next five years
self-appraisal Determine what your ideal job would be
goal selection Choose a career that will fit your preferred lifestyle
planning Make a plan of your goals for the next five years
problem solving Identify some reasonable career alternatives if you are unable to get your first choice
SWLS life satisfaction In most ways my life is close to my ideal.
PANAS positive affect enthusiastic
negative affect nervous
Notes. *reverse scored item.
Appendix 4: Interview Protocol

PART I: Ice-breaking
(Self-introduction and report delivery)
Q: Do you have any questions about your thinking styles report? Is anything
unclear? Or is there any information you would like?
Q: In your opinion, what are thinking styles? (Description of thinking styles

and further clarification of major differences between ability, personality,
and thinking styles based on the discussion.)
Q: Do you think the description of thinking styles in your feedback report

is consistent with your own appraisal of them? (If yes, could you provide
some examples for the way in which the report reflects your actual thinking
styles? If no, why?)
PART II: The socialization of thinking styles
Q: Do you think that your thinking styles have changed? When did these
changes happen? (Did you have any changes in your thinking styles last
year?) What kind of changes? Could you provide some examples about your
changes in thinking styles? How have these changes happened? (Followed
by further enquiries about the factors that the research focuses on.)
Q: Among family environment, learning environment, and personality,

which one influences your thinking styles the most? How? Do the other two
factors have any influence on your thinking styles? If yes, how? If no, why?
–– Are your parents strict? Or do they treat you with a lot of care and
make few demands? Do your parents insist that you adopt their
opinions, or do they understand that you have different ones? Did
the parenting you experienced, such as the above aspects, influence
your development of thinking styles? Could you provide some
examples on how your thinking styles were influenced by your
parenting?
–– Did your parents’ attitude to you and their parenting change after
you went to university? Did these changes in parenting styles
contribute to the change in your thinking styles? How?
–– (Give a description of dimensions of learning environments here.)

Is there any difference between the learning environment in your
176 Appendices
high school and the one in university? What are the differences?
Did these differences contribute to the change of your thinking
styles? Which aspects of learning environments affect your
thinking styles? How?
–– Among the different courses (university, or high school and

university, dependent on the participant; highlight but do not limit
it to the past year) are there any differences in teaching methods,
the extent to which you can decide your learning content and
process, assessment types, available resources, morale, and other
aspects of the learning environment? Could you provide some
examples? What kind of teaching methods, assessment types,
learning morale, and other aspects of the learning environment do
you find most often in your courses? What kind of teaching
methods, assessment types, learning morale, and other aspects of
the learning environment do you like best? Why? (Remember, to
obtain more details about assessments, teacher-student interactions,
student autonomy, and student cooperation.)
–– Are there one or two courses that have influenced your thinking
styles? What are the characteristics of these courses?
–– What do you think of the relationship between your personality and

thinking styles? Which one of these two characteristics is more
changeable? Why? Could you provide some examples?
PART III: The role of thinking styles in student development
Q: Have you considered career development issues, such as goal selection,

information collection, self-appraisal, and dealing with frustrations and
problems? What do you think of your ability with regard to these aspects of
career development?
Q: What factors influence your ability to handle your career development?

Could you provide some examples?
Q: Among these factors, how do you think that your thinking styles affect
your career maturity?
––Is any thinking style beneficial when you are dealing with tasks
related to career development (e.g., career planning, self-appraisal,
and goal selection)? What type of thinking styles? Could you provide
some examples?
–– Is there any thinking style that you believe has prevented you from
effectively dealing with tasks related to career development? What
type of thinking styles? Could you provide some examples?
Q: Are you satisfied with your life? Are you happy? Why?
Q: What factors influence your happiness? Could you provide some

examples?
Q: Among these factors, how do you believe your thinking styles have
affected your subjective well-being?
––Is any thinking style beneficial to your well-being/happiness (e.g., life

satisfaction and positive affect)? What type of thinking styles? Could
you provide some examples?
––Does any thinking style hinder your well-being/happiness? What type

of thinking styles? Could you provide some examples?
178 Appendices
Appendix 5: Regressions of the Change of Career Decision Self-efficacy on Parenting Styles (Time 2), Learning
Environment (Time 2), and Personality (Time 1)

R2Total 0.43 0.40 0.39 0.45 0.45
Adj R2Total 0.41 0.38 0.37 0.43 0.43
R2demo 0.02 0.03 0.03 0.03 0.02
R2PS 0.01 0.02 0.02 0.03 0.03
R2LE 0.01 0.02 0.01 0.02 0.03
R2Personality 0.04 0.03 0.04 0.05 0.06
R2Time 1 0.35 0.29 0.29 0.32 0.31
F 22.55*** 19.32*** 19.11*** 24.20*** 23.97***
df 732 732 732 732 732
Parenting Styles ȕinvol .13*** .17*** .15*** .18*** .17***
ȕpsyc_auto -.07*
ȕsuper
Learning ȕconstru .10* .15**
Environment ȕclear goal .10*
ȕautonomy .09*
ȕassessment .08*
ȕt-s inter
ȕs-s cooper .16*** .17*** .14** .14***
ȕpeer -.11** -.09* -.08* -.16***
ȕfacilities .08* .11**
Personality ȕneuro
ȕextra .07*
ȕopen
ȕagree -.07*
ȕconscien
CDSE ȕtime1 -.69*** -.66*** -.64*** -.69*** -.68***
Notes: CDSE = Career decision self-efficacy; R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles at
Time 2, learning environment at Time 2, personality at Time 1, thinking styles at Time 1, and career decision self-efficacy at Time 1
to the change of career decision self-efficacy from Time 1 to Time 2; R2demo = the contribution of gender, grade, major, hometown, and
SES to the change of career decision self-efficacy; R2PS = the unique contribution of parenting styles during the past year to the change
of career decision self-efficacy; R2LE = the unique contribution of learning environment during the past year to the change of career
decision self-efficacy; R2Personality = the unique contribution of personality at Time 1 to the change of career decision self-efficacy; R2TS
= the unique contribution of thinking styles at Time 1 to the change of career decision self-efficacy; R2Time1 = the unique contribution
of career decision self-efficacy at Time 1 to the change of career decision self-efficacy; invol = acceptance/involvement; psyc_auto =
psychological autonomy-granting; super = strictness/supervision; constru = constructivist-oriented teaching; clear goal = clear goals
and coherence of curricula; autonomy = student autonomy; assessment = assessment and assignment; t-s inter = teacher-student
interaction; s-s cooper = student-student cooperation; peer = peer morale and identities; facilities = facilities for learning; neuro =
neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness; conscien = conscientiousness; *p<.05,
**p<.01, ***p<.001; and the number of R2 in bold means the F change was significant after the predictor was added into the regression
model.
180 Appendices
Appendix 6: Regressions of the Change of Subjective Well-being on

Parenting Styles (Time 2), Learning Environment (Time 2), and
Personality (Time 1)

Change of SWB
Satisfaction Effect Effect
R2Total 0.43 0.47 0.43
Adj R2Total 0.41 0.45 0.41
R2demo 0.01 0.02 0.00
2
R PS 0.01 0.02 0.03
2
R LE 0.02 0.02 0.01
2
R Personality 0.06 0.09 0.08
R2Time 1 0.33 0.32 0.31
F 22.79*** 26.82*** 22.76***
df 732 732 732
Parenting ȕinvol .18*** .22***
Styles
ȕpsy_au .06* -.28***
ȕsuper -.08*
Learning ȕcons .10*
Environment
ȕgoal .11*
ȕauto
ȕasse .08*
ȕtsin
ȕssco .17***
ȕpeer .08*
ȕfaci -.09*
Personality ȕneuro .11**

ȕextra .12***
ȕopen
ȕagree -.07*
ȕconscien
SWB ȕtime1 -.72*** -.79*** -.71***
R2
Notes: SWB = subjective well-being; Total = the contribution of gender, major,
hometown, SES, parenting styles at Time 2, learning environment at Time 2,
personality at Time 1, thinking styles at Time 1, and subjective well-being at Time
1 to the change of subjective well-being from Time 1 to Time 2; R2demo = the
contribution of gender, major, hometown, and SES to the change of subjective well-
being; R2PS = the unique contribution of parenting styles during the past year to the
change of subjective well-being; R2LE = the unique contribution of learning
environment during the past year to the change of subjective well-being; R2Personality
= the unique contribution of personality at Time 1 to the change of subjective well-
being; R2TS = the unique contribution of thinking styles at Time 1 to the change of
subjective well-being; R2Time1 = the unique contribution of subjective well-being at
Time 1 to the change of subjective well-being; invol = acceptance/involvement;
psy_au = psychological autonomy-granting; super = strictness/supervision; cons =
constructivist-oriented teaching; goal = clear goals and coherence of curricula; auto
= student autonomy; asse = assessment and assignment; tsin = teacher-student
interaction; ssco = student-student cooperation; peer = peer morale and identities;
faci = facilities for learning; neuro = neuroticism; extra = extraversion; open =
openness to experience; agree = agreeableness; conscien = conscientiousness;
*p<.05, **p<.01, ***p<.001; and the number of R2 in bold means the F change was
significant after the predictor was added into the regression model.
182 Appendices
Appendix 7: Regressions of the Change of Career Decision Self-efficacy on Parenting Styles (Time 2), Learning
Environment (Time 2), Personality (Time 1), and the Change of Thinking Styles

R2Total 0.47 0.43 0.43 0.49 0.48
Adj R2Total 0.44 0.40 0.40 0.46 0.45
R2demo 0.02 0.03 0.03 0.03 0.02
R2PS 0.01 0.02 0.02 0.03 0.03
R2LE 0.01 0.02 0.01 0.02 0.03
R2Personality 0.04 0.03 0.04 0.05 0.06
R2ƸTS 0.10 0.09 0.09 0.09 0.09
R2Time 1 0.29 0.24 0.24 0.27 0.26
F 16.32*** 14.41*** 14.20*** 17.74*** 17.30***
df 732 732 732 732 732
Parenting Styles ȕinvol .10** .14*** .12** .15*** .14***
Learning ȕconstru .12**
Environment ȕclear goal .10*
ȕassessment .08*
ȕs-s cooper .14*** .09* .17*** .13** .13**
ȕpeer -.11** -.09* -.08* -.15***
ȕfacilities .08*
Personality ȕextra .08* .09*
ȕagree -.07* -.07*
ȕconscien .09*
Change of Thinking Styles
Type I ȕjud .11** .08* .10**
ȕlib .11***
ȕhie .09*
Type III ȕana .10**
ȕext .07*
CDSE ȕtime1 -.64*** -.61*** -.59*** -.65*** -.63***
Notes: CDSE = career decision self-efficacy; R2Total = the contribution of gender, grade, major, hometown, SES, parenting styles at
Time 2, learning environment at Time 2, personality at Time 1, thinking styles at Time 1, and career decision self-efficacy at Time 1
to the change of career decision self-efficacy from Time 1 to Time 2; R2demo = the contribution of gender, grade, major, hometown, and
SES to the change of career decision self-efficacy; R2PS = the unique contribution of parenting styles during the past year to the change
of career decision self-efficacy; R2LE = the unique contribution of the learning environment during the past year to the change of career
decision self-efficacy; R2Personality = the unique contribution of personality at Time 1 to the change of career decision self-efficacy; R2TS
= the unique contribution of thinking styles at Time 1 to the change of career decision self-efficacy; R2Time1 = the unique contribution
of career decision self-efficacy at Time 1 to the change of career decision self-efficacy; invol = acceptance/involvement; psyc_auto =
psychological autonomy-granting; super = strictness/supervision; constru = constructivist-oriented teaching; clear goal = clear goals
and coherence of curricula; autonomy = student autonomy; assessment= assessment and assignment; t-s inter = teacher-student
interaction; s-s cooper = student-student cooperation; peer = peer morale and identities; facilities = facilities for learning; neuro =
neuroticism; extra = extraversion; open = openness to experience; agree = agreeableness; conscien = conscientiousness; leg = legislative;
jud = judicial; glo = global; lib = liberal; hie = hierarchical; exe = executive; loc = local; con = conservative; mon = monarchic; oli =
oligarchic; ana = anarchic; int = internal; ext = external; *p<.05, **p<.01, ***p<.001; and the number of R2 in bold means the F change
was significant after the predictor was added into the regression model. Due to space limitations, the dimensions of the predictors that
were not significantly related to any change of CDSE were omitted.
184 Appendices
Appendix 8: Regressions of the Change of Subjective Well-being on

Parenting Styles (Time 2), Learning Environment (Time 2), Personality
(Time 1), and the Change of Thinking Styles

Change of SWB
Satisfaction Effect Effect
R2Total 0.45 0.51 0.45
Adj R2Total 0.42 0.48 0.42
R2demo 0.01 0.02 0.00
2
R PS 0.01 0.02 0.03
2
R LE 0.02 0.03 0.01
2
R Personality 0.06 0.08 0.08
R2ƸTS 0.06 0.13 0.05
R2Time 1 0.29 0.23 0.28
F 15.92*** 19.18*** 15.51***
df 732 732 732
Parenting ȕinvol .16*** .19***
Styles
ȕpsy_au -.26***
Learning ȕcons .09*
Environment
ȕgoal .10*
ȕasse .08*
ȕssco .17***
ȕpeer .08*
ȕfaci -.08*
Personality ȕneuro .12***
ȕextra .12***
ȕopen .08*
ȕagree -.08*
Change of Thinking Styles

Type I ȕlib .12***
ȕhie .07*
Type II ȕloc .10**
Type III ȕana .08*
SWB ȕtime1 -.69*** -.71*** -.70***
R2
Notes: SWB = subjective well-being; Total = the contribution of gender, grade,
major, hometown, SES, parenting styles at Time 2, learning environment at Time 2,
personality at Time 1, thinking styles at Time 1, and subjective well-being at Time
1 to the change of subjective well-being from Time 1 to Time 2; R2demo = the
contribution of gender, grade, major, hometown, and SES to the change of subjective
well-being; R2PS = the unique contribution of parenting styles during the past year to
the change of subjective well-being; R2LE = the unique contribution of learning
environment during the past year to the change of subjective well-being; R2Personality
= the unique contribution of personality at Time 1 to the change of subjective well-
being; R2TS = the unique contribution of thinking styles at Time 1 to the change of
subjective well-being; R2Time1 = the unique contribution of subjective well-being at
Time 1 to the change of subjective well-being; invol = acceptance/involvement;
psy_au = psychological autonomy-granting; cons = constructivist-oriented teaching;
goal = clear goals and coherence of curricula; assessment = assessment and
assignment; s-s cooper = student-student cooperation; peer = peer morale and
identities; faci = facilities for learning; neuro = neuroticism; extra = extraversion;
open = openness to experience; agree = agreeableness; lib = liberal; hie =
hierarchical; loc = local; ana = anarchic; *p<.05, **p<.01, ***p<.001; and the
number of R2 in bold means the F change was significant after the predictor was
added into the regression model. Due to space limitations, the dimensions of the
predictors which were not significantly related to any change of SWB were omitted.
REFERENCES
Abdallah, T. (1998). The Satisfaction with Life Scale (SWLS):

Psychometric properties in an Arabic-speaking sample. International
Journal of Adolescence &Youth, 7(2), 113-119.
Alborzi, S., & Ostovar, S. (2007). Thinking styles of gifted and nongifted
students in Iran. Psychological Reports, 100, 1076-1082.
Alloy, L. B., Abramson, L. Y., Tashman, N. A., Berrebbi, D. S., Hogan, M.
E., Whitehouse, W. G., . . . Moroco, A. (2001). Developmental origins
of cognitive vulnerability to depression: Parenting, cognitive, and
inferential feedback styles of the parents of individuals at high and low
cognitive risk for depression. Cognitive Therapy and Research, 25(4),
397-423.
Allport, G. W. (1937). Structure et dévelopment de la personnalité.
Neuchâtel: Delachaux-Niestlé.
Armstrong, S. J., Peterson, E. R., & Rayner, S. G. (2012). Understanding
and defining cognitive style and learning style: A Delphi study in the
context of educational psychology. Educational Studies, 38(4), 449-455.
doi: http://dx.doi.org/10.1080/03055698.2011.643110
Arrindell, W. A., Heesink, J., & Feij, J. A. (1999). The Satisfaction With
Life Scale (SWLS): Appraisal with 1700 health young adults in The
Netherlands. Personality and Individual Differences, 26(5), 815-826.
Atienza, F. L., Balaguer, I., & García-Merita, M. L. (2003). Satisfaction
with Life Scale: analysis of factorial invariance across sexes.
Personality and Individual Differences, 35(6), 1255-1260.
Azadi, M. (2009). Academic achievement in students with field
dependent/independent cognitive styles. Journal of Iranian Psychologists,
5(18), 119-126.
Baeten, M., Dochy, F., & Struyven, K. (2008). Students' approaches to
learning and assessment preferences in a portfolio-based learning
environment. Instructional Science, 36(5-6), 359-374.
Bagley, C., & Mallick, K. (1998). Field independence, cultural context and
academic achievement: A commentary. British Journal of Educational
Psychology, 68, 581-587.
Balasooriya, C. D., Hughes, C., & Toohey, S. (2009). Impact of a new
integrated medicine program on students' approaches to learning.
Higher Education Research & Development, 28(3), 289-302.
Bandura, A. (1977). Self-efficacy: Toward a unifying theory of behavioral

change. Psychological Review, 84, 191-215.
Bandura, A. (2006). Adolescent development from an agentic perspective.
In F. Pajares & T. Urdan (Eds.), Self-efficacy beliefs of adolescents (pp.
1-43). Greenwich, CT: Information Age.
Barnhart, R. K. (2003). Adaptation-Innovation Theory of Cognitive Style in
Aviation. International Journal of Applied Aviation Studies, 3(1), 105-
115.
Barrett, P. (2003). Person-Target Profiling for Selection and Recruitment
Part 2: The Coefficient Distribution Analyzer. Retrieved from
http://www.pbarrett.net/presentations/workshop2_NZ2003.pdf
Baumrind, D. (1967). Child care practices anteceding three patterns of
preschool behavior. Genetic Psychology Monographs, 75(1), 43-88.
Bernardo, A. B., Zhang, L. F., & Callueng, C. M. (2002). Thinking styles
and academic achievement among Filipino students. Journal of Genetic
Psychology, 163(2), 149-163.
Betoret, F. D. (2007). The influence of students' and teachers' thinking styles
on student course satisfaction and on their learning process. Educational
Psychology, 27(2), 219-234.
Betz, N. E., Klein, K. L., & Taylor, K. M. (1996). Evaluation of a short form
of the Career Decision-Making Self-Efficacy Scale. Journal of Career
Assessment, 4, 47-57. doi:
http://dx.doi.org/10.1177/106907279600400103
Biggs, J. B. (1978). Individual and group differences in study processes.
British Journal of Educational Psychology, 48, 266-279. doi:
http://dx.doi.org/10.1111/j.2044-8279.1978.tb03013.x
Biggs, J. B. (1979). Individual differences in study process and the quality
of learning outcomes. Higher Education, 8, 381-394.
Biggs, J. B. (1990). Teaching for desired learning outcomes. In N. J.
Entwistle (Ed.), Handbook of educational ideas and practice.
Beckenham: Croom Helm.
Bin, L. (2009). A research on the trait of vocational/technical college
students' learning motivation. Psychological Science (China), 32(3),
724-726.
Blais, M. R., Vallerand, R. J., Pelletier, L. G., & Brière, N. M. (1989). The
satisfaction scale: Canadian-French validation of the Satisfaction with
Life Scale. Canadian Journal of Behavioural Science, 21(2), 210-223.
Bolkan, C., Sano, Y., De Costa, J., Acock, A. C., & Day, R. D. (2010). Early
Adolescents' Perceptions of Mothers' and Fathers' Parenting Styles and
Problem Behavior. Marriage & Family Review, 46(8), 563-579.
doi:10.1080/01494929.2010.543040
188 References
Bosacki, S., Innerd, W., & Towson, S. (1997). Field independence-

dependence and self-esteem in preadolescents: Does gender make a
difference? Journal of Youth and Adolescence, 26(6), 691-703.
Bredo, E. (1997). The social construction of learning. In G. D. Phye (Ed.),
Handbook of academic learning: construction of knowledge. San Diego,
Calif: Academic Press.
Broekkamp, H., van Hout-Wolters, B. H. A. M., Rijlaarsdam, G., & van den
Bergh, H. (2002). Importance in instructional text: Teachers' and
students' perceptions of task demands. Journal of Educational
Bronfenbrenner, U. (1979). The ecological of human development.
Cambridge, MA: Harvard University Press.
Bronfenbrenner, U., & Ceci, S. J. (1994). Nature-nurture reconceptualized
in developmental perspective: A bioecological model. Psychological
Review, 101(4), 568-586.
Brooks, J. G., & Brooks, M. G. (1993). In search of understanding: the case
for constructivist classrooms. Alexandria, VA: Association for
Supervision and Curriculum Development.
Brown, R. T. (1980). Impulsivity and psychoeducational intervention in
hyperactive children. Journal of Learning Disabilities, 13(5), 249-254.
Brown, R. T. (1986). Methylphenidate and cognitive therapy in children
with attention deficit disorder: A double-blind trial. Journal of
Developmental & Behavioral Pediatrics, 7(3), 163-170.
Busato, V. V., Prins, F. J., Elshout, J. J., & Hamaker, C. (1999). The relation
between learning styles, the Big Five personality traits and achievement
motivation in higher education. Personality and Individual Differences,
26(1), 129-140. doi: http://dx.doi.org/10.1016/S0191-8869(98)00112-3
Cakan, M. (2003). Cross-cultural Aspect of the Group Embedded Figures
Test: Norms For Turkish Eighth Graders. Perceptual and Motor Skills,
97(2), 499-509.
Calamari, E., Pini, M., & Puleggio, A. (2000). Field dependence and
verbalized strategies on the portable rod-and-frame test by depressed
outpatients and normal controls. Perceptual and Motor Skills, 91(3),
1221-1229.
Cameron, B., & Dwyer, F. (2005). The effect of online gaming, cognition
and feedback type in facilitating delayed achievement of different
learning objectives. Journal of Interactive Learning Research, 16(3),
243-258.
Campbell, B. J. (1991). Planning for a student learning style. journal of
Education for Business, 66(6), 356-360.
Cano-Garcia, F., & Hughes, E. H. (2000). Learning and thinking styles: An

analysis of their relationship and influence on academic achievement.
Educational Psychology, 20(4), 413-430.
Cathcart, W. G. (1990). Effects of Logo instruction on cognitive style.
Journal of Educational Computing Research, 6(2), 231-242.
Centra, J. (1993). Reflective faculty evaluation. San Francisco: Jossey Bass.
Chakraborty, I., Hu, P. J.-H., & Cui, D. (2008). Examining the effects of
cognitive style in individuals' technology use decision making. Decision
Support Systems. Special Issue: I.T. and value creation, 45(2), 228-241.
Chamorro-Premuzic, T., & Furnham, A. (2008). Personality, intelligence
and approaches to learning as predictors of academic performance.
Personality and Individual Differences, 44(7), 1596-1603. doi:
http://dx.doi.org/10.1016/j.paid.2008.01.003
Chamorro-Premuzic, T., & Furnham, A. (2009). Mainly Openness: The
relationship between the Big Five personality traits and learning
approaches. Learning and Individual Differences, 19(4), 524-529. doi:
http://dx.doi.org/10.1016/j.lindif.2009.06.004
Chamorro-Premuzic, T., Furnham, A., & Lewis, M. (2007). Personality and
approaches to learning predict preference for different teaching
methods. Learning and Individual Differences, 17(3), 241-250.
Retrieved from http://www.sciencedirect.com/science/article/B6W5P-
4MP5JGW-1/2/3fba87c6697bb299af8a082100d9f659
Chao, R. K. (2001). Extending research on the consequences of parenting
style for Chinese Americans and European Americans. Child
Development, 72(6), 1832-1843.
Chen, C., & Fan, J. (2010). The applicability of the Parenting Style Index
to Nanjing and Shanghai adolescents. Journal of Nanjing Normal
University (Social Science), 6, 107-112.
Chen, G.-H., & Zhang, L.-F. (2010). Mental health and thinking styles in
Sternberg's theory: An explanatory study. Psychological Reports,
107(3), 784-794. doi:
http://dx.doi.org/10.2466/02.04.09.PR0.107.6.784-794
Chen, G.-H., & Zhang, L. F. (2010). Mental health and thinking styles in
Sternberg's theory: An explanatory study. Psychological Reports,
107(3), 784-794. doi:
http://dx.doi.org/10.2466/02.04.09.PR0.107.6.784-794
Chen, G. H. (2007). University teachers' humor production in the classroom
and students' ratings of teaching effectiveness. Unpublished
dissertation. The University of Hong Kong.
Chen, G. H., & Zhang, L. F. (2008). The role of thinking styles in mental
health. Unpublished manuscript.
190 References
Chen, X., Cen, G., Li, D., & He, Y. (2005). Social Functioning and
Adjustment in Chinese Children: The Imprint of Historical Time. Child
Development, 76(1), 182-195. doi:10.1111/j.1467-8624.2005.00838.x
Cheng, H., & Furnham, A. (2004). Perceived parental rearing style, self-
esteem and self-criticism as predictors of happiness. Journal of
Happiness Studies, 5(1), 1-21.
Cheung, F. (2002). Thinking styles and achievement in mathematics and
language learning. Unpublished manuscript. The University of Hong
Kong.
Choi, B. Y., Park, H., Yang, E., Lee, S. K., Lee, Y., & Lee, S. M. (2012).
Understanding career decision self-efficacy: A meta-analytic approach.
Journal of Career Development, 39(5), 443-460. doi:
http://dx.doi.org/10.1177/0894845311398042
Coffield, F., Moseley, D., Hall, E., & Ecclestone, K. (2004). Learning styles
and pedagogy in post-16 learning: A systematic and critical review.
London: Learning & Skills Research Centre.
Cohen, J. (1988). Statistical power analysis for the behavioral science.
Hillsdale, NJ: Erlbaum.
Cohen, N. J. (1981). valuation of the relative effectiveness of
methylphenidate and cognitive behavior modification in the treatment
of kindergarten-aged hyperactive children. Journal of Abnormal Child
Psychology: An official publication of the International Society for
Research in Child and Adolescent Psychopathology, 9(1), 43-54.
Collings, J. N. (1985). Scientific thinking through the development of
formal operations: Training in the cognitive restructuring aspect of field-
independence. Research in Science & Technological Education, 3(2),
145-152.
Collins, W. A., Maccoby, E., Steinberg, L., Hetherington, E. M., &
Bornstein, M. H. (2000). Contemporary research on parenting: The case
for nature and nurture. American Psychologist, 55, 218-232.
Collis, B., & Winnips, K. (2002). Two scenarios for productive learning
environments in the workplace. British Journal of Educational
Technology, 33, 133-148.
Costa, P. T., Jr., & McCrae, R. R. (1985). The NEO Personality Inventory
manual. Odessa, FL: Psychological Assessment Resources.
Costa, P. T., Jr., & McCrae, R. R. (1992). NEO Personality Inventory-
Revised(NEO-PI-R) and NEO Five-Factor Inventory (NEO-FFI)
professional manual. Odessa, FL: Psychological Assessment Resources
Inc.
Courneya, K. S., & Hellsten, L.-A. M. (1998). Personality correlates of
exercise behavior, motives, barriers and preferences: An application of
the five-factor model. Personality and Individual Differences, 24(5),

625-633.
Crites, J. O. (1961). A model for the measurement of vocational maturity.
Journal of Counseling Psychology, 8(255-259).
Crites, J. O. (1965). Measurement of vocational maturity in adolescence: I:
Attitude scale of the Vocational Development Inventory. Psychological
Monographs, 79(2, Whole No. 595).
Cronbach, L. J., & Gleser, G. C. (1953). Assessing similarity between
profiles. Psychological Bulletin, 50(6), 456-473. doi:
http://dx.doi.org/10.1037/h0057173
Curry, L. (1983). An organization of learning styles theory and constructs.
ERIC Document 235, 185
Daniel von Wittich, J. A. (2011). The KAI cognitive style inventory: Was it
personality all along? Personality and Individual Differences, 50(7),
1044-1049. doi: http://dx.doi.org/10.1016/j.paid.2011.01.022
Darling, N., & Steinberg, L. (1993). Parenting style as context: An
integrative model. Psychological Bulletin, 113, 487-496.
De Corte, E. (1995). Fostering cognitive growth: A perspective from
research on methemetics. Educational Psychologist, 30, 37-46.
De Corte, E. (2000). Marrying theory building and the improvement of
school practice. Learning and Instruction, 10, 249-266.
DeNeve, K. M., & Cooper, H. (1998). The happy personality: A meta-
analysis of 137 personality traits and subjective well-being.
Psychological Bulletin, 124(2), 197-229.
Diener, E. (1984). Subjective well-being. Psychological Bulletin, 95(3),
542-575.
Diener, E., & Biswas-Diener, R. (2002). Will money increase subjective
well-being? Social Indicators Research, 57(2), 119-169. Retrieved from
http://dx.doi.org/10.1023/A:1014411319119
Diener, E., Emmons, R. A., Larsen, R. J., & Griffin, S. (1985). The
Satisfaction With Life Scale. Journal of Personality Assessment, 49(1),
71-75.
Diener, E., & Larsen, R. J. (1984). Temporal stability and cross-situational
consistency of affective, behavioral, and cognitive responses. Journal of
Personality and Social Psychology, 47(4), 871-883.
Diener, E., Lucas, R. E., Oishi, S., & Suh, E. M. (2002). Looking up and
down: Weighting good and bad information in life satisfaction
judgments. Personality & Social Psychology Bulletin, 28(4), 437-445.
Diener, E., Oishi, S., & Lucas, R. E. (2009). Subjective well-being: The
science of happiness and life satisfaction. In C. R. Snyder & S. J. Lopez
192 References
(Eds.), Oxford handbook of positive psychology (2nd ed., pp. 187-194).

Oxford, New York: Oxford University Press.
Diener, M. L., & Diener McGavran, M. B. (2008). What makes people
happy?: A developmental approach to the literature on family
relationships and well-being. In M. Eid & R. J. Larsen (Eds.), The
science of subjective well-being (pp. 347-375). New York, NY, US:
Guilford Press.
Dillbeck, M. C., Assimakis, P. D., Raimondi, D., & Orme-Johnson, D. W.
(1986). Longitudinal effects on the Transcendental Meditation and TM-
Sidhi program on cognitive ability and cognitive style. Perceptual and
Motor Skills, 62(3), 731-738.
Dinsmore, D. L., Alexander, P. A., & Loughlin, S. M. (2008). The impact
of new learning environments in an engineering design course.
Instructional Science, 36(5-6), 375-393.
Dollinger, S. J., Palaskonis, D. G., & Pearson, J. L. (2004). Creativity and
intuition revisited. Journal of Creative Behavior, 38(4), 244-259.
Donnellan, M. B., Conger, R. D., & Burzette, R. G. (2007). Personality
development from late adolescence to young adulthood: Differential
stability, normative maturity, and evidence for the maturity-stability
hypothesis. Journal of Personality, 75(2), 237-264. doi:10.1111/j.1467-
6494.2007.00438.x
Duffy, T. M., & Jonassen, D. H. (1992a). Constructivism: New implications
for instructional technology. In T. M. Duffy & D. H. Jonassen (Eds.),
Constructivism and the technology of instruction: A conversation.
Hillsdale, New Jersey: Lawrence Erlbaum Associates.
Duffy, T. M., & Jonassen, D. H. (Eds.). (1992b). Constructivism and the
technology of instruction: A conversation. Hillsdale, New Jersey:
Lawrence Erlbaum Associates.
Dunn, R. S., & Dunn, K. J. (1978). Teaching students through their
individual learning styles: A practical approach. Reston, Va.: Reston
Pub. Co. .
Egan, V., Deary, I., & Austin, E. (2000). The NEO-FFI: emerging British
norms and an item-level analysis suggest N, A and C are more reliable
than O and E. Personality and Individual Differences, 29(5), 907-920.
Retrieved from
http://www.sciencedirect.com/science/article/pii/S0191886999002421
Entwistle, N., McCune, V., & Hounsell, J. (2003). Investigating ways of
enhancing university teaching-learning environments: Measuring
students' approaches to studying and perceptions of teaching. In E. de
Corte, L. Verschaffel, N. Entwistle, & J. Van Merriënboer (Eds.),
Powerful learning environments: Unravelling basic components and
dimensions (pp. 89-107). Oxford, England: Pergamon/Elsevier Science

Ltd.
Entwistle, N., & Ramsden, P. (1983). Understanding student learning.
London: Croom Helm.
Fan, J. (2016). The role of thinking styles in career decision-making self-
efficacy among university students. Thinking Skills and Creativity, 20,
63-73. doi: http://dx.doi.org/10.1016/j.tsc.2016.03.001
Fan, J., Zhang, L. F., & Chen, C. (2018). Thinking styles: Distinct from
personality? Personality and Individual Differences, 125, 50-55. doi:
https://doi.org/10.1016/j.paid.2017.12.026
Fan, W. (2012). An experimental comparison of the flexibility in the use of
thinking styles in traditional and hypermedia learning environments.
Thinking Skills and Creativity, 7(3), 224-233. doi:
http://dx.doi.org/10.1016/j.tsc.2012.04.007
Fan, W., & He, Y. (2012). Academic achievement and intellectual styles. In
L. F. Zhang, R. J. Sternberg, & S. Rayner (Eds.), Handbook of
intellectual styles: Preferences in cognition, learning, and thinking. (pp.
233-249). New York, NY, US: Springer Publishing Co, New York, NY.
Fan, W., & Zhang, L. F. (2009). Are achievement motivation and thinking
styles related? A visit among Chinese university students. Learning and
Individual Difference, 19, 299-303.
Fan, W., Zhang, L. F., & Watkins, D. (2010). Incremental validity of
thinking styles in predicting academic achievements: An experimental
study in hypermedia learning environments. Educational Psychology,
30(5), 605-623. doi: http://dx.doi.org/10.1080/01443410.2010.496899
Feindler, E. L., Ecton, R. B., Kingsley, D., & Dubey, D. R. (1986). Group
anger-control training for institutionalized psychiatric male adolescents.
Behavior Therapy, 17(2), 109-123.
Fer, S. (2007). What are the thinking styles of Turkish student teachers?
Teachers College Record, 109(6), 1488-1516.
Fer, S. (2012). Demographic characteristics and intellectual styles. In L. F.
Zhang, R. J. Sternberg, & S. Rayner (Eds.), Handbook of intellectual
styles: Preferences in cognition, learning, and thinking. (pp. 109-130).
New York, NY, US: Springer Publishing Co, New York, NY.
Fisher, N. M. (1994). MBTI type and writing-across-the-curriculum.
Journal of Psychological Type, 28, 43-47.
Fjell, A. D., & Walhovd, K. B. (2004). Thinking styles in relation to
personality traits: An investigation of the Thinking Styles Inventory and
NEO-PI-R. Scandinavian Journal of Psychology, 45(4), 293-300. doi:
http://dx.doi.org/10.1111/j.1467-9450.2004.00408.x
194 References
Fletcher, A. C., Walls, J. K., Cook, E. C., Madison, K. J., & Bridges, T. H.
(2008). Parenting style as a moderator of associations between maternal
disciplinary strategies and child well-being. Journal of Family Issues,
29(12), 1724-1744.
Flouri, E. (2004). Subjective well-being in midlife: The role of involvement
of and closeness to parents in childhood. Journal of Happiness Studies,
5(4), 335-358.
Foxall, G. R. (1990). An empirical analysis of mid-career managers'
adaptive-innovative cognitive styles and task orientations in three
countries. Psychological Reports, 66, 1115-1124.
Foxall, G. R., & Hackett, P. M. W. (1992). Cognitive style and extent of
computer use in organizations: Relevance of sufficiency of originality,
efficiency and rule-conformity. Perceptual and Motor Skills, 74, 491-
497.
Foxall, G. R., Payne, A. F., & Walters, D. A. (1992). Adaptive-innovative
cognitive styles of Australian managers. Australian Psychologist, 27(2),
118-122.
Francis, L. J., & Jones, S. H. (2000). Psychological type and happiness: A
study among adult churchgoers. Journal of Psychological Type, 54, 36-
41.
Fraser, B. J. (1998). The birth of a new journal: Editor's introduction.
Learning Environments Research, 1, 1-5.
Friedman, C. P., & Stritter, F. T. (1976). An empirical inventory comparing
instructional preferences of medical and other professional students.
Research in Medical Education Processing, 15th Annual conference,
November, San Francisco, 85-90.
Furnham, A. (1996). The big five versus the big four: the relationship
between the Myers-Briggs Type Indicator (MBTI) and NEO-PI five
factor model of personality. Personality and Individual Differences,
21(2), 303-307. doi: http://dx.doi.org/10.1016/0191-8869(96)00033-5
Furnham, A. (2012). Intelligence and intellectual styles. In L. F. Zhang, R.
J. Sternberg, & S. Rayner (Eds.), Handbook of intellectual styles:
Preferences in cognition, learning, and thinking. (pp. 173-192). New
York, NY, US: Springer Publishing Co, New York, NY.
Furnham, A., & Cheng, H. (2000). Perceived parental behaviour, self-
esteem and happiness. Social Psychiatry and Psychiatric Epidemiology,
35(10), 463-470.
Furnham, A., Christopher, A. N., Garwood, J., & Martin, G. N. (2007).
Approaches to learning and the acquisition of general knowledge.
Personality and Individual Differences, 43(6), 1563-1571.
Furnham, A., Crump, J., Batey, M., & Chamorro-Premuzic, T. (2009).

Personality and ability predictors of the "Consequences" Test of
Divergent Thinking in a large non-student sample. Personality and
Individual Differences, 46(4), 536-540. doi:
Furnham, A., Dissou, G., Sloan, P., & Chamorro-Premuzic, T. (2007).
Personality and intelligence in business people: A study of two
personality and two intelligence measures. Journal of Business and
Psychology, 22(1), 99-109. doi: http://dx.doi.org/10.1007/s10869-007-
9051-z
Furnham, A., Jackson, C. J., & Miller, T. (1999). Personality, learning style
and work performance. Personality and Individual Differences, 27(6),
1113-1122. Retrieved from
http://www.sciencedirect.com/science/article/B6V9F-3X52WPF-
8/2/2652006a0e77725f311bed19984610f0
Gadzella, B. M. (1990). Differences among cognitive-processing styles
groups on personality traits. Journal of Instructional Psychology, 26(3),
161-166.
Gebbia, M. I., & Honigsfeld, A. (2012). Learner developmental outcomes
and intellectual styles. In L. F. Zhang, R. J. Sternberg, & S. Rayner
(Eds.), Handbook of intellectual styles: Preferences in cognition,
learning, and thinking. (pp. 251-271). New York, NY, US: Springer
Publishing Co, New York, NY.
Gelade, G. (1995). Creative style and divergent production. Journal of
Creative Behavior, 29(1), 36-53.
Gianakos, I. (2001). Predictors of career decision-making self-efficacy.
Journal of Career Assessment, 9, 101-114.
Gignac, G. E., Bates, T. C., & Jang, K. L. (2007). Implications relevant to
CFA model misfit, reliability, and the five-factor model as measured by
the NEO-FFI. Personality and Individual Differences, 43, 1051-1062.
Gijbels, D., & Dochy, F. (2006). Students’ assessment preferences and
approaches to learning: can formative assessment make a difference?
Educational Studies, 32(4), 399-409. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=aph&AN=228
97517&site=ehost-live
Gijbels, D., Segers, M., & Struyf, E. (2008). Constructivist learning
environments and the (im)possibility to change students' perceptions of
assessment demands and approaches to learning. Instructional Science,
36(5-6), 431-443.
Gloria, A. M., & Hird, J. S. (1999). Influences of Ethnic and Nonethnic
Variables on the Career Decision-Making Self-Efficacy of College
196 References
Students. Career Development Quarterly, 48(2), 157. Retrieved from

http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=264
Gordon, C., & Debus, R. (2002). Developing deep learning approaches and
personal teaching efficacy within a preservice teacher education context.
British Journal of Educational Psychology, 72(4), 483-511. Retrieved
from
Greenfield, P. M., Keller, H., Fuligni, A., & Maynard, A. (2003). Cultural
pathways through universal development. Annual Review of
Psychology, 54, 461-490. Retrieved from
http://search.proquest.com/docview/619811604?accountid=14548
Gregorc, A. F. (1979). Learning/ teaching styles: Potent forces behind them.
Educational Leadership, 36, 234-236.
Grigorenko, E. L., & Sternberg, R. J. (1995). Thinking styles. In D.
Saklofske & M. Zeidner (Eds.), International handbook of personality
and intelligence (pp. 205-229). New York: Plenum.
Grigorenko, E. L., & Sternberg, R. J. (1997). Styles of thinking, abilities,
and academic performance. Exceptional Children, 63(3), 295-312.
Groves, M. (2005). Problem-based learning and learning approach: Is there
a relationship? Advances in Health Sciences Education, 10(4), 315-326.
Guilford, J. P. (1950). Creativity research: Past, present and future.
American Psychologist, 5, 444-454.
Gulikers, J. T. M., Kester, L., Kirschner, P. A., & Bastiaens, T. J. (2008).
The effect of practical experience on perceptions of assessment
authenticity, study approach, and learning outcomes. Learning and
Instruction, 18(2), 172-186.
Hackett, G., & Betz, N. E. (1981). A self-efficacy approach to the career
development of women. Journal of Vocational Behavior, 18, 326-339.
Hageman, P. T. (1990). An investigation of the relationship between
cognitive styles, selected dimensions of organizational climate, and job
satisfaction of registered professional staff nurses. Dissertation
Abstracts International, 51(7), 3323-B.
Hampton, N. Z. (2006). A Psychometric Evaluation of the Career Decision
Self-Efficacy Scale-Short Form in Chinese High School Students.
Journal of Career Development, 33(2), 142-155.
Hartman, R. O., & Betz, N. E. (2007). The five-factor model and career self-
efficacy: General and domain-specific relationships. Journal of Career
Assessment, 15(2), 145-161. doi:
http://dx.doi.org/10.1177/1069072706298011
Hayward, G., & Everett, C. (1983). Adaptors and innovators: Data from the
Kirton Adaptor-Innovator Inventory in a local authority setting. Journal
of Occupational Psychology, 56, 339-342.
Helwig, A. A., & Myrin, M. D. (1997). Ten-year stability of Holland codes
within one family. Career Development Quarterly, 46(1), 62-71.
Retrieved from
Hess, R. D., & Azuma, M. (1991). Cultural support of schooling: Contrasts
between Japan and the United States. Educational Researcher, 20, 2-8.
Hilliard, R. I. (1995). How do medical students learn: Medical student
learning styles and factors that affect these learning styles. Teaching and
Learning in Medicine, 7(4), 201-210.
Hite, C. E. (2004). Expository content area texts, cognitive style and gender:
Their effects on reading comprehension. Reading Research and
Instruction, 43(4), 41-74.
Ho, H. K. (1998). Assessing thinking styles in the theory of mental self-
government: A mini validity study in a Hong Kong secondary school.
Unpublished manuscript. The University of Hong Kong.
Hofstede, G. H. (1980). Culture’s consequences: International differences
in work-related values. Beverly Hills, CA: Sage.
Holden, R. R., & Fekken, G. C. (1994). The NEO five-factor inventory in a
Canadian context: Psychometric properties for a sample of university
women. Personality and Individual Differences, 17(3), 441-444.
Retrieved from
http://www.sciencedirect.com/science/article/pii/0191886994902917
Holland, J. L. (1973). Making vocational choices: A theory of careers.
Englewood Cliffs, NJ: Prentice-Hall.
Holland, P. A., Bowskill, I., & Bailey, A. (1991). Adaptors and innovators:
Selection versus induction. Psychological Reports, 68, 1283-1290.
Holmes, C. S. (1981). Reflective training and causal attributions in
impulsive mildly retarded children. Journal of Clinical Child
Psychology, 10(3), 194-199.
Hommerding, L. (2003). Thinking style preferences among the public
library directors of Florida. Dissertation Abstracts International
(Section B): The Sciences and Engineering, 63(11B), 5545.
Huang, J., & Chao, L. (2000). Field dependence versus field independence
of students with and without learning disabilities. Perceptual and Motor
Skills, 90(1), 343-346.
Huppert, F. A., Abbott, R. A., Ploubidis, G. B., Richards, M., & Kuh, D.
(2010). Parental practices predict psychological well-being in midlife:
198 References
Life-course associations among women in the 1946 British birth cohort.

Psychological Medicine: A Journal of Research in Psychiatry and the
Allied Sciences, 40(9), 1507-1518.
Jackson, C. J., & Lawty-Jones, M. (1996). Explaining the overlap between
personality and learning style. Personality and Individual Differences,
20(3), 293-300. doi: http://dx.doi.org/10.1016/0191-8869(95)00174-3
Jacobson, C. M. (1993). Cognitive styles of creativity: Relations of scores
on the Kirton Adaption-Innovation Inventory and the Mayers-Briggs
Type Indicator among managers in USA. Psychological Reports, 72,
1131-1138.
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical
approach to defining meaningful change in psychotherapy research.
Journal of Clinical and Consulting Psychology, 59, 12-19. doi:
http://dx.doi.org/10.1037/0022-006X.59.1.12
Jarvis, P., Holford, J., & Griffin, C. (1998). The theory and practice of
learning. London Sterling, Va: Kogan Page; Stylus Pub. Inc.
Jin, L., Watkins, D., & Yuen, M. (2009). Personality, career decision self-
efficacy and commitment to the career choices process among Chinese
graduate students. Journal of Vocational Behavior, 74(1), 47-52. doi:
http://dx.doi.org/10.1016/j.jvb.2008.10.002
Johnson, J., & Kane, K. (1992). Developmental and task factors in LOGO
programming. Journal of Educational Computing Research, 8(2), 229-
253.
Jonassen, D. H. (1996). Computers in the classroom: mindtools for critical
thinking. Englewood Cliffs, N. J.: Merrill.
Jonassen, D. H. (1999). Designing constructivist learning environments. In
C. M. Reigeluth (Ed.), Instructional-design theories and models: A new
paradigm of instructional theory (Vol. II, pp. 215-239). Mahwah, NJ,
US: Lawrence Erlbaum Associates Publishers.
Jonassen, D. H., & Grabowski, B. L. (1993). Handbook of individual
differences: Learning and instruction. Hillsdale, NJ: Lawrence Erlbaum
Associates.
Jonassen, D. H., Peck, K. L., & Wilson, B. G. (1999). Learning with
technology: a constructivist perspective. Upper Saddle River, N. J.:
Merrill.
Jones, C., Reichard, C., & Mokhtari, K. (2003). Are students' learning styles
discipline specific? Community College Journal of Research and
Practice, 27(5), 363-375.
Jong, P. J. D., Merckelbach, H., & Nijman, H. (1995). Hemisphere
preference, anxiety, and covariation bias. Personality and Individual
Differences, 18(3), 363-371.
Jung, C. (1923). Psychological types. New York: Harcourt Brace.

Kagan, J. (1965). Individual differences in the resolution of response
uncertainty. Journal of Personality and Social Psychology, 2(2), 154-
160.
Kagan, J., Rosman, B. L., Day, D., Albert, J., & Philips, W. (1964).
Information processing in the child: Significance of analytic and
reflective attitudes. Psychological Monographs, 78(1, Whole No. 578).
Kaufman, J. C. (2001). Thinking styles in creative writers and journalists.
Dissertation Abstracts International (Section B): The Sciences and
Engineering, 62(3B), 1069.
Keller, B. K., & Whiston, S. C. (2008). The role of parental influences on
young adolescents' career development. Journal of Career Assessment,
16(2), 198-217. Retrieved from
http://jca.sagepub.com/content/16/2/198.abstract
Keller, R. T., & Holland, W. E. (1978a). A cross-validation study of the
Kirton Adaption-Innovation Inventory in three research and
development organizations. Applied Psychological Measurement, 2(4),
563-570.
Keller, R. T., & Holland, W. E. (1978b). Individual characteristics of
innovativeness and communication in research and development
organizations. Journal of Applied Psychology, 63(6), 759-762.
Kember, D., & Leung, D. Y. P. (2009a). Development of a questionnaire
for assessing students' perceptions of the teaching and learning
environment and its use in quality assurance. Learning Environments
Research, 12(1), 15-29.
Kember, D., & Leung, D. Y. P. (2009b). Development of a questionnaire
for assessing students' perceptions of the teaching and learning
environment and its use in quality assurance. Learning Environments
Research, 12, 15-29.
Kendall, P. C., & Finch, A. J. (1978). A cognitive-behavioral treatment for
impulsivity: A group comparison study. Journal of Consulting and
Clinical Psychology, 46(1), 110-118.
Kern, G. M., & Matta, K. F. (1988). The influence of personality on self-
paced instruction. Journal of Computer-Based Instruction, 15(3), 104-
108.
Kim, J., & Michael, W. B. (1995). The relationship of creativity measures
to school achievement and preferred learning and thinking style in a
sample of Korean high school students. Educational and Psychological
Measurement, 55(1), 60-74.
Kim-Prieto, C., Diener, E., Tamir, M., Scollon, C., & Diener, M. (2005).
Integrating the diverse definitions of happiness: A time-sequential
200 References
framework of subjective well-being. Journal of Happiness Studies, 6(3),

261-300. Retrieved from 10.1007/s10902-005-7226-8
40&site=ehost-live
Kirschner, P. A. (2009). Epistemology or pedagogy, that is the question. In
S. Tobias & T. M. Duffy (Eds.), Constructivist Instruction: Success or
Failure? New York, London: Routledge.
Kirschner, P. A., Sweller, J., & Clark, R. E. (2006). Why minimal guidance
during instruction does not work: An analysis of the failure of
constructivist, discovery, problem-based, experiential, and inquiry-
based teaching. Educational Psychologist, 41(2), 75-86.
Kirton, M. J. (1961). Management initiative. London: Acton Society Trust.
Kirton, M. J. (1976). Adaptor and innovators: A description and measure.
Journal of Applied Psychology, 61, 622-629. doi:
http://dx.doi.org/10.1037/0021-9010.61.5.622
Kirton, M. J., & Pender, S. R. (1982). The adaption-innovation continuum,
occupational type and course selection. Psychological Reports, 51, 883-
886.
Klinger, T. H. (2006). Learning Approach, Thinking Style and Critical
Inquiry: The Online Community. Korean Journal of Thinking &
Problem Solving, 16(1), 91-113.
Kogan, N. (1989). A stylistic perspective on metaphor and aesthetic
sensitivity in children. In T. Globerson & T. Zelniker (Eds.), Cognitive
style and cognitive development (Vol. 3, pp. 192-213). Norwood, NJ:
Ablex.
Kogan, N., & Block, J. (1991). Field dependence-independence from early
childhood through adolescence: personality and socialization aspects. In
S. Wapner & J. Demick (Eds.), Field dependence-independence:
Cognitive style across the life span. Hillsdale, NJ: Erlbaum.
Kolb, D. A. (1976). The learning style inventory: Technical manual.
Boston, MA: McBer.
Kolitch, E., & Dean, A. V. (1999). Student ratings of instruction in the USA:
Hidden assumptions and missing conceptions about 'good' teaching.
Studies in Higher Education, 24(27-42).
Krohne, H. W., Egloff, B., Kohlmann, C.-W., & Tausch, A. (1996).
Investigations with a German version of the Positive and Negative
Affect Schedule (PANAS). Diagnostica, 42(2), 139-156.
Kwang, N. A., & Rodrigues, D. (2002). A Big-Five personality profile of
the adapter and innovator. The Journal of Creative Behavior, 36(4), 254-
268.
Lam, P. Y. (2000). The usefulness of thinking styles in reflecting how

individuals think and explaining school performance. Unpublished
manuscript. The University of Hong Kong.
Lapidus, L. B., Shin, S.-K., & Hutton, E. M. (2001). An Evaluation of a Six-
Week Intervention Designed to Facilitate Coping with Psychological
Stress. Journal of Clinical Psychology, 57(12), 1381-1401. Retrieved
from
Larson, L. M., Rottinghaus, P. J., & Borgen, F. H. (2002). Meta-analyses of
Big Six Interests and Big Five Personality Factors. Journal of
Vocational Behavior, 61(2), 217-239. Retrieved from
http://www.sciencedirect.com/science/article/B6WMN-46P4392-
3/2/91d13ea45951c98f771c9bc8e4232c58
Lau, C.-h. (2014). Thinking styles, motivational orientation, and academic
achievements in learning physics among Hong Kong secondary school
students. Unpublished doctoral dissertation. The University of Hong
Kong.
Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral
participation. Cambridge: Cambridge University Press.
Lee, K. L. (2002). Thinking styles and approaches in teaching among Hong
Kong kindergarten teachers. Unpublished manuscript. The University
of Hong Kong.
Lent, R. W. (2005). A social cognitive view of career development and
counseling. In S. D. Brown & R. W. Lent (Eds.), Career development
and counseling: Putting theory and research to work (pp. 101-127).
New York: John Wiley.
Lent, R. W., Brown, S. D., & Hackett, G. (1994). Toward a unifying social
cognitive theory of career and academic interest, choice, and
performance. Journal of Vocation Behavior, 45, 75-122.
Leventhal, G., & Sisco, H. (1996). Correlations among field
dependence/independence, locus of control and self-monitoring.
Perceptual and Motor Skills, 83(604-606).
Li, M., & Armstrong, S. J. (2015). The relationship between Kolb's
experiential learning styles and Big Five personality traits in
international managers. Personality and Individual Differences, 86, 422-
426. doi: http://dx.doi.org/10.1016/j.paid.2015.07.001
Liao, C., & Chuang, S.-H. (2007). Assessing the effect of cognitive styles
with different learning modes on learning outcome. Perceptual and
Motor Skills, 105(1), 184-190.
202 References
Long, L., Fang, L., & Ling, W. (2000). Advancement in the theory and
measurement of career maturity. Psychological Science (China), 23(5),
595-598.
Loo, R. (1997). Evaluating change and stability in learning style scores.
Educational Psychology, 17(1/2), 95. Retrieved from
López-Villalobos, J. A., Pintado, I. S., Sánchez-Mateos, J. D., Rodríguez, J.
M. M., Azón, M. I. S., & Sanz, F. R. (2003). Attention deficit
hypeactivity disorder: Development of reflexive-impulse, reflexive-
rigid and field dependence-independence cognitive styles. Revista de
Psiquiatría Infanto-Juvenil, 20(4), 166-175.
Low, K. S. D., Yoon, M., Roberts, B. W., & Rounds, J. (2005). The Stability
of Vocational Interests From Early Adolescence to Middle Adulthood:
A Quantitative Review of Longitudinal Studies. Psychological Bulletin,
131(5), 713-737. doi:10.1037/0033-2909.131.5.713
Loyens, S. M. M., & Gijbels, D. (2008). Understanding the effects of
constructivist learning environments: Introducing a multi-directional
approach. Instructional Science, 36(5-6), 351-357.
Loyens, S. M. M., Rikers, R. M. J. P., & Schmidt, H. G. (2008).
Relationships between students' conceptions of constructivist learning
and their regulation and processing strategies. Instructional Science,
36(5-6), 445-462.
Loyens, S. M. M., Rikers, R. M. J. P., & Schmidt, H. G. (2009). Students'
conceptions of constructivist learning in different programme years and
different learning environments. British Journal of Educational
Lucas, R. E. (2005). Time does not heal all wounds: A longitudinal study of
reaction and adaptation to divorce. Psychological Science, 16(12), 945-
950. Retrieved from http://pss.sagepub.com/content/16/12/945.abstract
Lucas, R. E., Clark, A. E., Georgellis, Y., & Diener, E. (2004).
Unemployment alters the set point for life satisfaction. Psychological
Science, 15(1), 8-13. Retrieved from
http://pss.sagepub.com/content/15/1/8.abstract
Lucas, R. E., & Diener, E. (2008). Personality and subjective well-being. In
O. P. John, R. W. Robins, & L. A. Pervin (Eds.), Handbook of
personality psychology: Theory and research (3rd ed., pp. 795-814).
New York, NY, US: Guilford Press.
Luk Suet Ching, W. (1998). The influence of a distance-learning
environment on students' field dependence/independence. Journal of
Experimental Education, 66(2), 149-160. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=ulh&AN=496
443&site=ehost-live
Lykken, D., & Tellegen, A. (1996). Happiness is a stochastic phenomenon.
Psychological Science, 7(3), 186-189.
Maccoby, E. E., & Martin, J. (1983). Socialization in the context of the
family: Parent-child interaction. In E. M. Hetherington & P. H. Mussen
(Eds.), Handbook of child psychology: Socialization, personality, and
social development (Vol. 4, pp. 1-101). New York: Wiley.
Mandelman, S. D., & Grigorenko, E. L. (2012). The etiology of intellectual
styles: Contributions from intelligence and personality. In L. F. Zhang,
R. J. Sternberg, & S. Rayner (Eds.), Handbook of intellectual styles:
Preferences in cognition, learning, and thinking. (pp. 89-107). New
York, NY, US: Springer Publishing Co, New York, NY.
Maton, K. I. (1990). Meaningful involvement in instrumental activity and
well-being: Studies of older adolescents and at risk urban teen-agers.
American Journal of Community Psychology, 18(2), 297-320. Retrieved
from http://dx.doi.org/10.1007/BF00931306
Matsumoto, D. (2002). The new Japan: debunking seven cultural
stereotypes. Yarmouth, ME: Intercultural Press.
Maynard, M. J., & Harding, S. (2010). Perceived parenting and
psychological well-being in UK ethnic minority adolescents. Child
Care, Health & Development, 36(5), 630-638.
McCartney, K., & Rosenthal, R. (2000). Effect size, practical importance,
and social policy for children. Child Development, 71, 173-180. doi:
https://doi.org/10.1111/1467-8624.00131
McCrae, R. R., & Costa, P. T., Jr. (2007). Brief versions of the NEO-PI-3.
Journal of Individual Differences, 28(3), 116-128. doi:
http://dx.doi.org/10.1027/1614-0001.28.3.116
McCrae, R. R., Zonderman, A. B., Costa Jr, P. T., Bond, M. H., &
Paunonen, S. V. (1996). Evaluating replicability of factors in the
Revised NEO Personality Inventory: Confirmatory factor analysis
versus Procrustes rotation. Journal of Personality and Social
Psychology, 70(3), 552-566. doi:10.1037/0022-3514.70.3.552
McGinn, L. K., Cukor, D., & Sanderson, W. C. (2005). The Relationship
Between Parenting Style, Cognitive Style, and Anxiety and Depression:
Does Increased Early Adversity Influence Symptom Severity Through
the Mediating Role of Cognitive Style? Cognitive Therapy & Research,
29(2), 219-242. Retrieved from
204 References
McMahon, D. M. (2006). Happiness: A history. New York: Atlantic

Monthly Press.
McParland, M., Noble, L. M., & Livingston, G. (2004). The effectiveness
of problem-based learning compared to traditional teaching in
undergraduate psychiatry. Medical Education, 38(8), 859-867.
Melvin, G., & Molloy, G. N. (2000). Some psychometric properties of the
Positive and Negative Affect Schedule among Australian youth.
Psychological Reports, 86(3), 1209-1212.
Messer, S. B. (1976). Reflection-impulsivity: A review. Psychological
Bulletin, 83(6), 1026-1052.
Messick, S. (1994). The matter of style: Manifestations of personality in
cognition, learning, and teaching. Educational Psychologist, 29, 121-
136.
Milevsky, A., Schlechter, M., Netter, S., & Keehn, D. (2007). Maternal and
paternal parenting styles in adolescents: associations with self-esteem,
depression and life-satisfaction. Journal of Child and Family Study,
16(1), 39-47.
Miller, A. (1987). Cognitive styles: An integrated model. Educational
Psychology, 7(4), 251-268.
Miller, A. L. (2007). Creativity and cognitive style: The relationship
between field-dependence-independence, expected evaluation, and
creative performance. Psychology of Aesthetics, Creativity, and the
Arts., 1(4), 243-246.
Miller, L. M. (2005). Using learning styles to evaluate computer-based
instruction. Computers in Human Behavior, 21(2), 287-306. doi:
https://doi.org/10.1016/j.chb.2004.02.011
Mills, C. J. (2003). Characteristics of effective teachers of gifted students:
Teacher background and personality styles of students. Gifted Child
Quarterly, 47(4), 272-281.
Mitchell, T., & Cahill, A. M. (2005). Cognitive style and plebe turnover at
the U.S. Naval Academy. Perceptual and Motor Skills, 101(1), 55-62.
Moreno, R., & Mayer, R. E. (1999). Multimedia-supported metaphors for
meaning making in mathematics. Cognition and Instruction, 17, 215-
248.
Moskowitz, D. S., Dreyer, A. S., & Kronsberg, S. (1981). Preschool
children's field independence: Prediction from antecedent and
concurrent maternal and child behavior. Perceptual and Motor Skills,
52(607-616).
Myers, I. B. (1962). The Myers-Briggs Type Indicator: Manual. Palo Alto,
CA: Consulting Psychologists Press.
Myers, I. B., & McCaulley, M. H. (1988). Manual: A guide to the

development and use of the Myers-Briggs Type Indicator. Palo Alto,
CA: Consulting Psychologists Press.
Nachmias, R., & Shany, N. (2002). Learning in virtual courses and its
relationship to thinking styles. Journal of Educational Computing
Research, 27(3), 315-329.
Nauta, M. M., & Kahn, J. H. (2007). Identity status, consistency and
differentiation of interests, and career decision self-efficacy. Journal of
Career Assessment, 15(1), 55-65.
Nijhuis, J., Segers, M., & Gijselaers, W. (2008). The extent of variability in
learning strategies and students' perceptions of the learning
environment. Learning and Instruction, 18(2), 121-134. Retrieved from
http://www.sciencedirect.com/science/article/B6VFW-4NBY8M8-
1/2/27025021f600a4b224568107ff17db65
Noddings, N. (2005). Happiness and education Cambridge, UK:
Cambridge University Press.
O'Brien, T. P. (1994). Cognitive learning styles and academic achievement
in secondary education. Journal of Research & Development in
Education, 28(1), 11-21.
O'Connor, M. C. (1998). Can we trace the "efficacy of social
constructivism"? Review of Research in Education, 23, 25-71.
O'Donnell, E. H. (2009). Explanatory styles, parenting, and adolescent
depression. Dissertation Abstracts International: Section B: The
Sciences and Engineering, 69(10-B), 6427.
O'Leary, M. R., Donovan, D. M., & Kasner, K. H. (1975). Shifts in
Perceptual Differentiation and Defense Mechanisms in Alcoholics.
Journal of Clinical Psychology, 31(3), 565-567. Retrieved from
Oishi, S. (2001). Culture and memory for emotional experiences: On-line
vs. retrospective judgments of subjective well-being. Dissertation
Abstracts International: Section B: The Sciences and Engineering,
61(10-B), 5625.
Pace, C. R., & Kuh, G. D. (2007). The college student experiences
questionnaire assessment program. Retrieved from
http://cseq.iub.edu/cseq_generalinfo.cfm
Page, J., Bruch, M. A., & Haase, R. F. (2008). Role of perfectionism and
Five-Factor model traits in career indecision. Personality and Individual
Differences, 45(8), 811-815. Retrieved from
http://www.sciencedirect.com/science/article/B6V9F-4TJ5YGH-
2/2/dea132dc40cfd76dcaa039f77cbd3ebb
206 References
Paradowski, J. H. (2001). Positive affectivity, negative affectivity, and job

satisfaction:A meta-analysis. Dissertation Abstracts International:
Section B: The Sciences and Engineering, 61(12-B), 6749.
Páramo, M. F., & Tinajero, C. (1998). Family process variables and field
dependence-independence: a proposal for a new conceptual framwork.
International Journal of Educational Research, 29, 205-218.
Park, C. L., & Adler, N. E. (2003). Coping style as a predictor of health and
well-being across the first year of medical school. Health Psychology,
22(6), 627-631.
Park, S.-K., Park, K.-H., & Choe, H.-S. (2005). The relationship between
thinking styles and scientific giftedness in Korea. Journal of Secondary
Gifted Education, 16(2-3), 87-97.
Parke, R. D. (2001). Comprehensive Models of Parenting: Remaining
Issues and Future Trends. In J. R. M. Gerris (Ed.), Dynamics of
Parenting (pp. 463-471). Leuven-Apeldoorn: Garant.
Pask, G., & Scott, B. C. E. (1972). Learning strategies and individual
competence. International Journal of Man-Machine Studies, 4, 217-
253.
Paykel, E. S. (2003). Life events and affective disorders. Acta Psychiatrica
Scandinavica, 108, 61-66. Retrieved from
http://dx.doi.org/10.1034/j.1600-0447.108.s418.13.x
Penn, B. A. (2004). The effects of parenting style on preferred structure
levels within individual learning style in college students at Vennard
college: A comparative study. Dissertation Abstracts International
Section A: Humanities and Social Sciences, 64(12-A), 4359.
Peterson, E. R., Rayner, S. G., & Armstrong, S. J. (2009). Researching the
psychology of cognitive style and learning style: Is there really a future?
Learning and Individual Differences, 19(4), 518-523. doi:
Phan, H. P. (2007). An examination of reflective thinking, learning
approaches, and self-efficacy beliefs at the University of the South
Pacific: A path analysis approach. Educational Psychology, 27(6), 789-
806.
Phillips, D. C. (2000). An opinionated account of the constructivist
landscape. In D. C. Philips (Ed.), Constructivism in education: Opinions
and second opinions on controversial issues (pp. 1-16). Chicago:
National Society for the Study of Education.
Pontecorvo, C. (1993). Social interaction in the acquisition of knowledge.
Educational Psychology Review, 5(3), 293-310.
Qin, L., Pomerantz, E. M., & Wang, Q. (2009). Are gains in decision-
making autonomy during early adolescence beneficial for emotional
functioning? The case of the United States and China. Child

Development, 80(6), 1705-1721. Retrieved from
http://search.proquest.com/docview/621972949?accountid=14548
Ramsden, P. (1991). A performance indicator of teaching quality in higher
education: the course experience questionnaire. Studies in Higher
Education, 16(2), 129-150.
Rani, N. J., & Rao, P. V. K. (2000). Effects of meditation on attention
processes. Journal of Indian Psychology, 18(1-2), 52-60.
Reid, W. A., Duvall, E., & Evans, P. (2005). Can we influence medical
students’ approaches to learning? Medical Teacher, 27(5), 401-407.
Retrieved from
Reti, I. M., Samuels, J. F., Eaton, W. W., Bienvenu Iii, O. J., Costa, P. T.,
Jr., & Nestadt, G. (2002). Influences of parenting on normal personality
traits. Psychiatry Research, 111(1), 55-64. Retrieved from
http://www.sciencedirect.com/science/article/B6TBV-461XR57-
1/2/02d5ee31c8af5c37bf626384a3c875bc
Rezler, A. G., & Rezmovic, V. (1974). The learning preference inventory.
Journal of Applied Health, 19(1), 28-34.
Richardson, J. T. E. (1995). Mature students in higher education: II. An
investigation of approaches to studying and academic performance.
Studies in Higher Education, 20(1), 5-17.
Riding, R. J. (1997). On the nature of cognitive style. Educational
Psychology, 17(1&2), 29-49.
Riding, R. J., & Cheema, I. (1991). Cognitive styles-an overview and
integration. Educational Psychology, 11(3&4), 193-215.
Riding, R. J., & Rayner, S. G. (1998). Cognitive Styles and Learning
Strategies: Understanding Style Differences in Learning and Behavior.
London: David Fulton.
Riding, R. J., & Wigley, S. (1997). The relationship between cognitive style
and personality in further education students. Personality and Individual
Differences, 23(3), 379-389. Retrieved from
http://www.sciencedirect.com/science/article/B6V9F-47HBTBC-
3/2/e66ff403adaebb5c4faf0ba58792062f
Rigby, B. T., & Huebner, E. S. (2005). Do causal attributions mediate the
relationship between personality characteristics and life satisfaction in
adolescence? Psychology in the Schools, 42(1), 91-99.
Roberts, B. W., Caspi, A., & Moffitt, T. E. (2001). The kids are alright:
Growth and stability in personality development from adolescence to
208 References
adulthood. Journal of Personality and Social Psychology, 81(4), 670-

683. doi:10.1037//0022-3514.81.4.670
Roberts, B. W., Wood, D., & Caspi, A. (2008). The development of
personality traits in adulthood. In O. P. John, R. W. Robins, & L. A.
Pervin (Eds.), Handbook of personality: Theory and research (3rd ed.,
pp. 375-398). New York, London: The Guilford Press.
Robinson, M. D., & Compton, R. J. (2008). The happy mind in action: The
cognitive basis of subjective well-being. In M. Eid & R. J. Larsen (Eds.),
The science of subjective well-being (pp. 220-238). New York, NY, US:
Guilford Press.
Roodenburg, J., Roodenburg, E., & Rayner, S. (2012). Personality and
intellectual styles. In L. F. Zhang, R. J. Sternberg, & S. Rayner (Eds.),
Handbook of intellectual styles: Preferences in cognition, learning, and
thinking. (pp. 209-231). New York, NY, US: Springer Publishing Co,
New York, NY.
Rosander, P., & Bäckström, M. (2012). The unique contribution of learning
approaches to academic performance, after controlling for IQ and
personality: Are there gender differences? Learning and Individual
Differences, 22(6), 820-826. doi:
Rottinghaus, P. J., Larson, L. M., & Borgen, F. H. (2003). The relation of
self-efficacy and interests: a meta-analysis of 60 samples. Journal of
Vocational Behavior, 62(2), 221-236. Retrieved from
http://www.sciencedirect.com/science/article/B6WMN-48120B9-
5/2/200cd81eb99068b36b29b718649e783b
Rushton, S., Morgan, J., & Richard, M. (2007). Teacher's Myers-Briggs
personality profiles: Identifying effective teacher personality traits.
Teaching and Teacher Education, 23(4), 432-441.
Sadler-Smith, E. (1996). Approaches to studying: age, gender and academic
performance. Educational Studies, 22(3), 367-379.
Sadler-Smith, E. (2009). A duplex model of cognitive style. In L. F. Zhang
& R. J. Sternberg (Eds.), Perspectives on the nature of intellectual styles.
(pp. 3-28): Springer Publishing Co, New York, NY.
Saucier, G. (1998). Replicable item-cluster subcomponents in the NEO
Five-Factor Inventory. Journal of Personality Assessment, 70(2), 263-
276.
Savery, J. R., & Duffy, T. M. (1996). Problem based learning: An
instructional model and its constructivist framework. In B. G. Wilson
(Ed.), Constructivist learning environments: case studies in
instructional design (pp. 135-148). Englewood Cliffs, New Jersey:
Educational Techonology Publications.
Sawyer, R. K. (2006). Introduction: The new science of learning. In R. K.

Sawyer (Ed.), The Cambridge handbook of the learning sciences. New
York: Cambridge University Press.
Sayles-Folks, S. L., & Harrison, D. K. (1989). Reflection-impulsivity and
work adjustment. Rehabilitation Counseling Bulletin, 33(2), 110-117.
Schapman, A. M. (2003). The interrelationship among perceived family
environment, control-related cognitions, and social anxiety in
adolescents: An investigation of mediational and moderational models.
Dissertation Abstracts International: Section B: The Sciences and
Engineering, 64(5-B), 2404.
Schimmack, U., & Diener, E. (2003). Predictive validity of explicit and
implicit self-esteem for subjective well being. Journal of Research in
Personality, 37(2), 100-106.
Schleifer, M., & Douglas, V. I. (1973). Moral judgments, behavior and
cognitive style in young children. Canadian Journal of Behavioural
Science, 5, 133-144.
Seddon, J. (2008). Vets and videos: Student learning from context-based
assessment in a pre-clinical science course. Assessment and Evaluation
in Higher Education, 33(5), 559-566.
Segers, M., Gijbels, D., & Thurlings, M. (2008). The relationship between
students' perceptions of portfolio assessment practice and their
approaches to learning. Educational Studies, 34(1), 35-44.
Shek, D. T. L. (2007). A longitudinal study of perceived parental
psychological control and psychological well-being in Chinese
adolescents in Hong Kong. Journal of Clinical Psychology, 63(1), 1-22.
Shernoff, D. J., Csikszentmihalyi, M., Shneider, B., & Shernoff, E. S.
(2003). Student engagement in high school classrooms from the
perspective of flow theory. School Psychology Quarterly, 18(2), 158-
176.
Shewchuk, R. M., & O'Connor, S. J. (1995). Health care executives:
Subjective well-being as a function of psychological type. Journal of
Psychological Type, 32, 23-29.
Shokri, O., Kadivar, P., Farzad, V., & Sangari, A. A. (2007). Role of
personality traits and learning approaches on academic achievement of
university students. Psychological Research, 9(3-4), 65-84.
Siemens, G. (2004, April 5, 2005). Connectivism. Retrieved from
http://devrijeruimte.org/content/artikelen/Connectivism.pdf
Skinner, B. F. (1968). The technology of teaching. East Norwalk, CT, US:
Appleton-Century-Crofts.
Skinner, N. F., Hutchinson, L., Lukenda, A., Drake, G., & Boucher, J.
(2003). National personality characteristics: II. Adaption-innovation in
210 References
Canadian, American, and British samples. Psychological Reports, 92(1),

21-22.
Skogsberg, K., & Clump, M. (2003). Do psychology and biology majors
differ in their study processes and learning styles? College Student
Journal, 37(1), 27-33.
Slavin, R. E. (2003). Educational psychology: Theory and practice (7th
ed.). Boston: Allyn & Bacon.
Smith, J. P., & Cage, B. N. (2000). The effects of chess instruction on the
mathematics achievement of southern, rural, Black secondary students.
Research in the Schools, 7(1), 19-26.
Smith, S. N., & Miller, R. J. (2005). Learning Approaches: Examination
type, discipline of study, and gender. Educational Psychology, 25(1),
43-53.
Spera, C. (2005). A Review of the Relationship Among Parenting Practices,
Parenting Styles, and Adolescent School Achievement. Educational
Psychology Review, 17(2), 125-146. Retrieved from
Sridevi, K., & Rao, P. V. K. (2003). Temporal effects of meditation on
cognitive style. Journal of Indian Psychology, 21(1), 38-51.
Sridevi, K., Sitamma, M., & Rao, P. V. K. (1995). Perceptual organisation
and yoga training. Journal of Indian Psychology, 13(2), 21-27.
Stahl, S. A., Erickson, L. G., & Rayman, M. C. (1986). Detection of
inconsistencies by reflective and impulsive seventh-grade readers.
National Reading Conference Yearbook, 35, 233-238.
Stansbury, V. K., & Coll, K. M. (1998). Myers-Briggs Attitude Typology-
The Influence of Birth Order with Other Family Variables. The Family
Journal, 6(2), 116-122.
Steel, P., Schmidt, J., & Shultz, J. (2008). Refining the relationship between
personality and subjective well-being. Psychological Bulletin, 134(1),
138-161.
Steffe, L. P., & Gale, J. E. (1995). Constructivism in education. Hillsdale,
N. J.: Lawrence Erlbaum.
Steinberg, L. (2001). We know some things: Parent-adolescent relationships
in retrospect and prospect. Journal of Research on Adolescence, 11(1),
1-19. Retrieved from
Steinberg, L., Lamborn, S. D., Darling, N., Mounts, N. S., & Dornbusch, S.
M. (1994). Over-time changes in adjustment and competence among
adolescents from authoritative, authoritarian, indulgent, and neglectful

families. Child Development, 65(3), 754-770.
Steinberg, L., & Morris, A. S. (2001). Adolescent development. Annual
Review of Psychology, 52(1), 83-110. Retrieved from
http://search.ebscohost.com/login.aspx?direct=true&db=bth&AN=444
Stericker, A., & LeVesconte, S. (1982). Effect of brief training on sex-
related differences in visual-spatial skill. Journal of Personality and
Social Psychology, 43(5), 1018-1029.
Sternberg, R. J. (1988). Mental self-government: A theory of intellectual
styles and their development. Human Development, 31, 197-224.
Sternberg, R. J. (1994). Thinking styles: Theory and assessment at the
interface between intelligence and personality. In R. J. Sternberg & P.
Ruzgis (Eds.), Intelligence and personality (pp. 169-187). New York:
Sternberg, R. J. (1997). Thinking styles. New York: Cambridge University
Press.
Sternberg, R. J., & Grigorenko, E. L. (1993). Thinking styles and the gifted.
Roeper Review, 16(2), 122-130.
Sternberg, R. J., & Grigorenko, E. L. (1995). Styles of thinking in the
school. European Journal of High Ability, 6, 201-219.
Sternberg, R. J., & Wagner, R. K. (1992). Thinking Styles Inventory.
Unpublished test. Yale University.
Sternberg, R. J., Wagner, R. K., & Zhang, L. F. (2003). Thinking Styles
Inventory-Revised. Unpublished test. Yale University.
Sternberg, R. J., Wagner, R. K., & Zhang, L. F. (2007). Thinking Styles
Inventory-Revised II. Unpublished test. Yale University. New Haven,
CT.
Suldo, S. M., & Huebner, E. S. (2004). The role of life satisfaction in the
relationship between authoritative parenting dimensions and adolescent
problem behavior. Social Indicators Research, 66(1), 165-195.
Retrieved from
http://dx.doi.org/10.1023/B:SOCI.0000007498.62080.1e
Tai, W.-y. (2012). The impact of teaching styles on students’ learning styles
and career interests. Unpublished doctoral dissertation. The University
of Hong Kong.
Tang, J. M. (2004). Are asian thinking styles different? acculturation and
thinking styles in a Chinese Canadian population. Dissertation Abstracts
International: Section B: The Sciences and Engineering, 65(3-B), 1573.
Taris, T. W., & Bok, I. A. (1996). Effects of parenting style upon
psychological well-being of young adults: Exploring the relations
212 References
among parental care, locus of control and depression. Early Child

Development and Care, 132, 93-104.
Taylor, K. M., & Betz, N. E. (1983). Applications of self-efficacy theory to
the understanding and treatment of career indecision. Journal of
Vocational Behavior, 22(1), 63-81. doi: http://dx.doi.org/10.1016/0001-
8791(83)90006-4
Tennant, M. (1997). Psychological and adult learning (2nd ed.). London:
Routledge.
Thompson, M. N., & Subich, L. M. (2006). The relation of social status to
the career decision-making process. Journal of Vocational Behavior,
69(2), 289-301.
Tian, X. (2007). Do assessment methods matter? A sensitivity test.
Assessment and Evaluation in Higher Education, 32(4), 387-401.
Tierney, P. (1997). The influence of cognitive climate on job satisfaction
and creative efficacy. Journal of Social Behavior & Personality, 12(4),
831-848.
Tobacyk, J. J., Livingston, M. M., & Robbins, J. E. (2008). Relationships
between Myers-Briggs type indicator measure of psychological type and
neo measure of big five personality factors in polish university students:
A preliminary cross-cultural comparison. Psychological Reports,
103(2), 588-590. doi: http://dx.doi.org/10.2466/pr0.103.2.588-590
Tobias, S., & Duffy, T. M. (2009). The success or failure of constructivist
instruction: An introduction. In S. Tobias & T. M. Duffy (Eds.),
Constructivist Instruction: Success or Failure? (pp. 3-10). New York,
London: Routledge.
Tobin, K., & Tippins, D. (1993). Constructivism as a referent for teaching
and learning. In K. Tobin (Ed.), The practice of constructivism in science
education (pp. 3-22). Hillsdale, New Jersey: Lawrence-Erlbaum
Associates.
Torbit, G. (1981). Counselor learning styles: A variable in career chioce.
Canadian Counsellor, 15(4), 57-74.
Torrance, E. P. (1988). Style of learning and thinking: Administrator's
manual. Bensenville, IL: Scholastic Testing Service.
Torrance, E. P., & Reynolds, R. C. (1978). Images of the future of gifted
adolescents: Effects of alienation and specialized cerebral functioning.
Gifted Child Quarterly, 22, 40-54.
Tse, P. (2003). The relationship among thinking styles, learning
approaches, value attachment, and academic performance of junior
secondary school students in Hong Kong. Unpublished manuscript. The
City University of Hong Kong.
Tucker, R. W. (1999). An examination of accounting students' thinking

styles. Dissertation Abstracts International (Section B): The Sciences
and Engineering, 60(6B), 2977.
Updegraff, J. A., & Suh, E. M. (2007). Happiness is a warm abstract
thought: Self-construal abstractness and subjective well-being. The
Journal of Positive Psychology: Dedicated to furthering research and
promoting good practice, 2(1), 18 - 28. Retrieved from
http://www.informaworld.com/10.1080/17439760601069150
van Meel, J. (1991). Proximate social influences on field-dependence-
independence. In S. Wapner & J. Demick (Eds.), Field dependence-
independence: Cognitive style across the life span (pp. 325-351).
Hillsdale, NJ: Erlbaum.
Van Merriënboer, J. J. G., & Paas, F. (2003). Powerful learning and the
many faces of instructional design: Toward a framework for the design
of powerful learning environments. In E. D. Corte, L. Verschaffel, N.
Entwistle, & J. J. G. v. Merriënboer (Eds.), Powerful learning
environments: Unravelling basic components and dimensions (pp. 3-
20). Oxford, UK: Elsevier Science.
Varma, O. P., & Thakur, M. (1992). Cognitive style and scholastic
achievements. Psycho-Lingua, 22(2), 81-92.
Veenhoven, R., Ehrhardt, J., Ho, M. S. D., & de Vries, A. (1993). Happiness
in nations: Subjective appreciation of life in 56 nations 1946-1992.
Rotterdam, Netherlands: Erasmus University Rotterdam.
Verma, S. (2001). A study of thinking styles of tertiary students. Psycho-
Lingua, 31(1), 15-19.
Vermunt, J. D. (2003). The power of learning environments and the quality
of student learning. In E. D. Corte, L. Verschaffel, N. Entwistle, & J. J.
G. v. Merriënboer (Eds.), Powerful learning environments: Unravelling
basic components and dimensions (pp. 109-124). Oxford, UK: Elsevier
Science.
Volet, S. E., Renshaw, P. D., & Tietzel, K. (1994). A short-term longitudinal
investigation of cross-cultural differences in study approaches using
Biggs' SPQ questionnaire. British Journal of Educational Psychology,
64, 301-318.
von Wittich, D., & Antonakis, J. (2011). The KAI cognitive style inventory:
Was it personality all along? Personality and Individual Differences,
50(7), 1044-1049. doi: http://dx.doi.org/10.1016/j.paid.2011.01.022
Vygotsky, L. S. (1978). Mind in society: the development of higher
psychological process. Cambridge, Mass: Havard University Press.
Watkins, D. A., & Dahlin, B. (1997). Assessing study approaches in
Sweden. Psychological Reports, 81, 131-136.
214 References
Watson, D., Clark, L. A., & Tellegen, A. (1988). Development and

validation of brief measures of positive and negative affect: The PANAS
scales. Journal of Personality & Social Psychology, 54(6), 1063-1070.
Wegner, R. C. (1980). The relationship of field independence-dependence
cognitive styles with managerial disposition. Dissertation Abstracts
International, 40(8B), 4012.
Wentzel, K. R., & Battle, A. A. (2001). Social relationships and school
adjustment. In T. Urdan & F. Pajares (Eds.), Adolescence and education:
general issues in the education of adolescents (pp. 93-118). Greenwich,
Conn: Information Age Publishing.
Wierstra, R. F. A., Kanselaar, G., Linden, J. L. V. D., & Lodewijks, H. G.
L. C. (1999). Learning environment perceptions of european university
students. Learning Environments Research, 2, 79-98.
Wilding, J., & Andrews, B. (2006). Life goals, approaches to study and
performance in an undergraduate cohort. British Journal of Educational
Psychology, 76(1), 171-182.
Williams, M. E. (2001). The effects of conceptual model provision and
cognitive style on problem-solving performance of learners engaged in
an exploratory learning environment. Dissertation Abstracts
International (Section A): Humanities and Social Sciences, 62(3A), 983.
Wilson, D. J., Mundy-Castle, A., & Sibanda, P. (1990). Field differentiation
and LOGO performance among Zimbabwean school girls. Journal of
Social Psychology, 130(2), 277-279.
Wilson, K., & Fowler, J. (2005). Assessing the impact of learning
environments on students' approaches to learning: Comparing
conventional and action learning designs. Assessment and Evaluation in
Higher Education, 30(1), 87-101.
Wise, T. N., Hall, W. A., & Wong, O. (1978). The relationship of cognitive
styles and affective status to post-operative analgesic utilization.
Journal of Psychosomatic Research, 22, 513-518.
Witkin, H. A. (1954). Personality through perception: An experimental and
clinical study. New York: Harper.
Witkin, H. A. (1962). Psychological differentiation: Studies of
development. New York: Wiley.
Witkin, H. A., & Goodenough, D. R. (1981). Cognitve styles: essence and
origins. New York: International Unversities Press.
Wong, Y. C. (2003). Constructivist online learning environment for social
work education: An evaluation of students' learning process and
outcome. Unpublished Dissertation. The University of Hong Kong.
Hong Kong.
Wu, X., & Zhang, H. C. (1999). The preliminary application of the Thinking
Style Inventory in college students. Psychological Science China, 22(4),
293-297.
Wunderley, L. J., Reddy, W. B., & Dember, W. N. (1998). Optimism and
pessimism in business leaders. Journal of Applied Social Psychology,
28(9), 751-760.
Xie, Q. (2015). Intellectual styles: Their associations and their relationships
to ability and personality. Journal of Cognitive Education and
Psychology, 14(1), 63-76. doi:
http://dx.doi.org/10.1891/1945-8959.14.1.63
Yang, S. C., & Lin, W. C. (2004). The relationship among creative, critical
thinking and thinking styles in Taiwan high school students. Journal of
Instructional Psychology, 31(1), 33-45.
Ye, S. (2008). A Longitudinal Study of Subjective Well-being among
Chinese University Students: the Roles of Personality, Attribution, and
Coping. Unpublished Doctorial Dissertation. The University of Hong
Kong. Hong Kong.
You, Z., & Jia, F. (2008). Do they learn differently? An investigation of the
pre-service teachers from US and China. Teaching and Teacher
Education, 24(4), 836-845.
Yu, D. M. (2008). Teachers' teaching styles and interpersonal behaviors.
Unpublished manuscript.
Yu, T. M. (2012). Do teachers’ interaction styles affect secondary school
students’ thinking styles and academic performance? . Unpublished
doctoral dissertation. The University of Hong Kong.
Zhang, L. F. (1999). Further cross-cultural validation of the theory of mental
self-government. The Journal of Psychology, 133(2), 165-181. doi:
https://doi.org/10.1080/00223989909599731
Zhang, L. F. (2000). University students' learning approaches in three
cultures: An investigation of Biggs's 3P model. The Journal of
Psychology, 134(1), 37-55.
Zhang, L. F. (2001a). Approaches and thinking styles in teaching. Journal
of Psychology: Interdisciplinary and Applied, 135(5), 547-562.
Zhang, L. F. (2001b). Do styles of thinking matter among Hong Kong
secondary school students? Personality and Individual Differences,
31(3), 289-301.
Zhang, L. F. (2001c). Do thinking styles contribute to academic
achievement beyond abilities? The Journal of Psychology, 135(6), 621-
637.
Zhang, L. F. (2001d). Thinking styles, self-esteem, and extracurricular
experiences. International Journal of Psychology, 36(2), 100-107.
216 References
Zhang, L. F. (2002a). Measuring thinking styles in addition to measuring

personality traits? Personality and Individual Differences, 33, 445-458.
doi: http://dx.doi.org/10.1016/S0191-8869(01)00166-0
Zhang, L. F. (2002b). Thinking styles and cognitive development. The
Journal of Genetic Psychology, 163(2), 179-195.
Zhang, L. F. (2002c). Thinking styles and modes of thinking: Implications
for education and research. The Journal of Psychology, 136(3), 245-261.
Zhang, L. F. (2002d). Thinking styles and the Big Five Personality Traits.
Educational Psychology, 22(1), 17-31. doi:
http://dx.doi.org/10.1080/01443410120101224
Zhang, L. F. (2002e). Thinking styles: Their relationships with modes of
thinking and academic performance. Educational Psychology, 22(3),
331-348.
Zhang, L. F. (2003a). Are parents’ and children’s thinking styles related?
Psychological Reports, 93, 617-630. doi:10.2466/pr0.2003.93.2.617
Zhang, L. F. (2003b). Contributions of thinking styles to critical thinking
dispositions. Journal of Psychology: Interdisciplinary and Applied,
137(6), 517-544.
Zhang, L. F. (2003c). Does the big five predict learning approaches?
Personality and Individual Differences, 34, 1431-1446.
Zhang, L. F. (2004a). Contributions of thinking styles to vocational purpose
beyond self-rated abilities. Psychological Reports, 94, 697-714.
Zhang, L. F. (2004b). Field-dependence/independence: Cognitive style or
perceptual ability?--Validating against thinking styles and academic
achievement. Personality and Individual Differences, 37(6), 1295-1311.
doi: http://dx.doi.org/10.1016/j.paid.2003.12.015
Zhang, L. F. (2004c). Learning approaches and career personality types:
Biggs and Holland united. Personality and Individual Differences, 37,
65-81.
Zhang, L. F. (2004d). Revisiting the predictive power of thinking styles for
academic performance. The Journal of Psychology, 138(4), 351-370.
doi: https://doi.org/10.3200/JRLP.138.4.351-370
Zhang, L. F. (2004e). Thinking styles: University students' preferred
teaching styles and their conceptions of effective teachers. The Journal
of Psychology, 138(3), 233-252.
Zhang, L. F. (2005a). Predicting cognitive development, intellectual styles,
and personality traits from self-rated abilities. Learning and Individual
Differences, 15(1), 67-88.
Zhang, L. F. (2005b). Validating the theory of mental self-government in a
non-academic setting. Personality and Individual Differences, 38, 1915-
1925.
Zhang, L. F. (2006a). Preferred teaching styles and modes of thinking

among university students in mainland China. Thinking Skills and
Creativity, 1(2), 95-107.
Zhang, L. F. (2006b). Thinking styles and the Big Five Personality Traits
revisited. Personality and Individual Differences, 40, 1177-1187. doi:
Zhang, L. F. (2007a). From career personality types to preferences for
teachers' teaching styles: A new perspective on style match. Personality
and Individual Differences, 43(7), 1863-1874.
Zhang, L. F. (2007b). Intellectual styles and academic achievement among
senior secondary school students in rural China. Educational
Psychology, 27(5), 675-692.
Zhang, L. F. (2007c). Revisiting thinking styles' contributions to the
knowledge and use of and attitudes towards computing and information
technology. Learning and Individual Differences, 17, 17-24.
Zhang, L. F. (2008a). Preferences for teaching styles matter in academic
achievement: Scientific and practical implications. Educational
Psychology, 28(6), 615-625.
Zhang, L. F. (2008b). Thinking styles and emotions. The Journal of
Psychology, 142(5), 497-515.
Zhang, L. F. (2008c). Thinking styles and identity development among
Chinese university students. American Journal of Psychology, 121(2),
255-271.
Zhang, L. F. (2009). Anxiety and thinking styles. Personality and Individual
Differences, 47, 347-351. doi:
Zhang, L. F. (2010). Further investigating thinking styles and psychosocial
development in the Chinese higher education context. Learning and
Individual Differences, 20(6), 593-603.
doi:10.1016/j.lindif.2010.04.011
Zhang, L. F. (2012). Why schools should care about intellectual styles.
Journal of Cognitive Education and Psychology, 11(3), 256-270. doi:
http://dx.doi.org/10.1891/1945-8959.11.3.256
Zhang, L. F. (2013). The Malleability of Intellectual Styles. New York:
Zhang, L. F. (2014). The malleability of intellectual styles. New York, NY:
Zhang, L. F. (2017). The value of intellectual styles. New York, NY:
Zhang, L. F., & Fan, W. (2011). The Theory of Mental Self-Government
Grows Up: Where Has It Led the Field after 21 Years? . In S. Rayner &
218 References
E. Cools (Eds.), Style differences in cognition, learning, and

management (pp. 46-59). New York: Routledge.
Zhang, L. F., Fu, H., & Jiao, B. (2008). Accounting for Tibetan university
students' and teachers' intellectual styles. Educational Review, 60(1), 21-
37.
Zhang, L. F., & He, Y. F. (2003). Do thinking styles matter in the use of and
attitudes toward computing and information technology among Hong
Kong university students? . Journal of Educational Computing
Research, 29(4), 471-293. doi: https://doi.org/10.2190/3M8D-1D56-
V60C-BHR5
Zhang, L. F., & He, Y. F. (2011). Thinking styles and the Eriksonian stages.
Journal of Adult Development, 18(1), 8-17. doi:
http://dx.doi.org/10.1007/s10804-010-9101-z
Zhang, L. F., & Higgins, P. (2008). The predictive power of socialization
variables for thinking styles among adults in the workplace. Learning
and Individual Differences, 18, 11-18.
Zhang, L. F., & Huang, J. F. (2001). Thinking styles and the five-factor
model of personality. European Journal of Personality, 15, 465-476.
doi: http://dx.doi.org/10.1002/per.429
Zhang, L. F., Huang, J. F., & Zhang, L. L. (2005). Preferences in teaching
styles among Hong Kong and U.S. university students. Personality and
Individual Differences, 39, 1319-1331.
Zhang, L. F., & Postiglione, G. A. (2001). Thinking styles, self-esteem, and
socio-economic status. Personality and Individual Differences, 31,
1333-1346. doi: https://doi.org/10.1016/S0191-8869(00)00227-0
Zhang, L. F., & Postiglione, G. A. (2005). Thinking styles, culture, and
economy: Comparing Tibetan minority students with Han Chinese
majority students. Manuscript submitted for publication.
Zhang, L. F., & Sachs, J. (1997). Assessing thinking styles in the theory of
mental self-government: A Hong Kong validity study. Psychological
Reports, 81, 915-928.
Zhang, L. F., & Sternberg, R. J. (1998). Thinking styles, abilities, and
academic achievement among Hong Kong university students.
Educational Research Journal, 13(1), 41-62.
Zhang, L. F., & Sternberg, R. J. (2002). Thinking styles and teacher
characteristics. International Journal of Psychology, 37(1), 3-12.
doi:10.1080/00207590143000171
Zhang, L. F., & Sternberg, R. J. (2005). A threefold model of intellectual
styles. Educational Psychology Review, 17(1), 1-53.
Zhang, L. F., & Sternberg, R. J. (2006). The nature of intellectual styles.
Mahwah, NJ: Lawrence Erlbaum.
Zhang, L. F., & Sternberg, R. J. (2009). Revisiting the value issue in

intellectual styles. In L. F. Zhang & R. J. Sternberg (Eds.), Perspectives
on the Nature of Intellectual Styles (pp. 63-85). New York: Springer
Publishing Company.
Zheng, L., & Tao, G. (2001). The influential factors on quality of life in
children. Chinese Journal of Clinical Psychology, 9(2), 105-107.

Thinking Styles - Jieqiong Fan

Uploaded by

Copyright:

Available Formats

Thinking Styles - Jieqiong Fan

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Thinking Styles - Jieqiong Fan

Uploaded by

Copyright:

Available Formats

Identity, Value, and Malleability

This book first published 2020

Cambridge Scholars Publishing

Lady Stephenson Library, Newcastle upon Tyne, NE6 2PA, UK

British Library Cataloguing in Publication Data

Copyright © 2020 by Jieqiong Fan

ISBN (10): 1-5275-5817-7

List of Tables ............................................................................................. ix

Preface ...................................................................................................... xii

List of Abbreviations ................................................................................ xv

Chapter One ................................................................................................ 1

Chapter Two ............................................................................................. 26

Chapter Three ........................................................................................... 40

Chapter Four ........................................................................................... 100

Chapter Five ........................................................................................... 139

Chapter Six ............................................................................................. 154

Appendices ............................................................................................. 162

References .............................................................................................. 186

Table 1.1: Descriptions of the 13 Thinking Styles in the Theory

Table 2.1: Partial Correlations Between Thinking Styles and Personality

Table 2.2: Regressions of Thinking Styles on Personality Traits (EIV) ... 30

Table 2.3: Regressions of CDSE on Personality Traits and Thinking

Table 2.4: Regressions of SWB on Personality Traits and Thinking

Table 2.5: Repeated Measures on Thinking Styles and Personality

Table 3.1: Mean-Level Change and Rank-Order Consistency

Table 3.2: The Distribution of the Reliable Change Index ....................... 49

Table 3.3: The Distribution of the Reliable Change Index

Table 3.4: Zero-Order Correlations Between Socioeconomic Status

Table 3.5: Predicting Thinking Styles from the Three Dimensions

Table 3.6: Predicting Thinking Styles from the Three Dimensions

Table 3.7: Style Differences Based on the Types of Parenting

Table 3.8: Style Differences Based on the Types of Parenting Styles

Table 3.9: Multiple Comparisons for Thinking Styles Based on Types

Table 3.10: Multiple Comparisons for Thinking Styles Based on Types

Table 3.11: Predicting Thinking Styles from Learning Environments

Table 3.12: Predicting Thinking Styles from Learning Environments

Table 3.13: Predicting Thinking Styles from Parenting Styles, Learning

Table 3.14: Predicting Thinking Styles from Parenting Styles, Learning

Table 3.15: Conditional Change Model of Thinking Styles ..................... 98

Table 4.1: Predicting Career Decision Self-Efficacy from Thinking

Table 4.2: Predicting Career Decision Self-Efficacy from Thinking

Table 4.3: Predicting the Change of Career Decision Self-Efficacy from

Table 4.4: Predicting Subjective Well-being from Thinking Styles

Table 4.5: Predicting Subjective Well-being from Thinking Styles

Table 4.6: Predicting the Change of Subjective Well-being from Thinking

Table 4.7: Testing Longitudinal Relationships that are Potentially

Fortunately, scholars have made tremendous efforts to integrate the field of

Finally, I would like to express heartfelt appreciation to my family,

CDSE: Career decision self-efficacy

ISPLE: Inventory of students’ perceived learning environment

KAI: Kirton’s adaption-innovation styles

LE: Learning environments

PS: Parenting styles

SCCT: Social cognitive career theory

SES: Socioeconomic status

SWB: Subjective well-being

TS: Thinking styles

TSI: Thinking style inventory

ȕopenness .58* .49* .21* .52* .29*** -.12* .41***

ȕagreeableness -.26* -.19* -.35* -.21* -.10* .21* -.43* .26***