CULTURAL CAPITAL, ECONOMIC CAPITAL, AND ACADEMIC ACHIEVEMENT:
SOME EVIDENCE FROM TAIWAN
By
SHIUH-TARNG CHENG

A DISSERTATION
Submitted to Michigan State University
in partial fulfillment of the requirements
for the degree of
DOCTOR OF PHILOSOPHY
SOCIOLOGY
2012

ABSTRACT
CULTURAL CAPITAL, ECONOMIC CAPITAL, AND ACADEMIC ACHIEVEMENT:
SOME EVIDENCE FROM TAIWAN
By
SHIUH-TARNG CHENG
Pierre Bourdieu’s (1984) concept of cultural capital is popular among educational researchers in
Europe and the U.S., but it has rarely been applied in Confucian societies such as Taiwan.
Bourdieu frames education as an arena of cultural conflict in which social inequalities are
reproduced through the unequal distribution of “capitals” both within and without the
educational system. In contrast, educational researchers who study Asian societies often assume
cultural homogeneity, social harmony, and limited inequality. Inspired by Bourdieu’s theory,
this study analyzed secondary data from first and second panel of junior high school students
(n=12,527) in the Taiwan Educational Panel Survey (TEPS). The results show that parental
cultural capital is strongly associated with parental economic capital (income) and has a
significant effect on students’ cultural capital, which in turn influences their academic
achievement. While Confucian influences facilitate certain educational aspirations and practices
across social classes, different levels of economic capital and cultural capital possessed by the
parents and their children differentiate educational outcomes. By applying a Western-developed
concept in a non-Western context, this study contributes to the theoretical and methodological
development of research on cultural capital and demonstrates how social, cultural, and
institutional contexts outside the West condition the process of educational stratification.

Copyright by
SHIUH-TARNG CHENG
2012

ACKNOWLEDGMENTS
I would like to express my deepest gratitude to my co-chairs Professor Stan Kaplowitz and
Professor Desiree Baolian Qin for their guidance, encouragement, and useful critiques of this
research work. I would also like to thank two other members of my committee, Professor
Barbara Schneider and Professor Steven J. Gold, for their advice and support. My grateful
thanks are also extended to Professor Alesia Montgomery for guiding me in those formative
years, as well as Prof. Harry Schwarzweller for his relentless push around my project, and to Dr.
Chyi-In Wu and Dr. Stanly Lee for the insights they shared. I would also like to thank Dr. Mark
Cummings and Dr. Marion Cummings whose warmth and friendship make the path I took much
smoother. Further, this dissertation would not be possible without the faith and support of my
mother, my sisters, my brother-in-law Tony, my colleagues, and my friends back home in
Taiwan. Finally, I would to thank my wife Karen for standing by my side during the darkest
moments and my two boys Jie-Ming and Jun-Wei who constantly remind me what a beautiful
world we live in.

iv

TABLE OF CONTENTS
LIST OF TABLES

vii

LIST OF FIGURES

viii

CHAPTER ONE - INTRODUCTION
Statement of the Problem
Purpose of the Study
Research Questions and Hypotheses

1
1
4
4

CHAPTER TWO – LITERATURE REVIEW
Family SES and Educational Outcomes in the West and in Taiwan
Bourdieu‘s Theory of Capital
Cultural Capital in Educational Research

6
6
11
17

CHAPTER THREE - METHODS
Sample
Measurement of Variables
Parental Economic Capital
Parental Cultural Capital
Students‘ Cultural Capital
Students‘ Academic Achievement
Data Analysis
Developing the Models

21
21
21
21
21
22
22
23
23

CHAPTER FOUR – RESEARCH FINDINGS
Finding the Best Measurement Model
Assessment of the Best Measurement Model
Assessment of the Structural Model

26
26
28
29

CHAPTER FIVE – DISCUSSION AND CONCLUSION
LIMITATIONS
CONCLUSION

31
36
36

APPENDICES
Appendix A
Appendix B

37
38
40
41
42
43
44
45
46
47

Appendix C
Appendix D
Appendix E

Table 1 List of Factors, Indictors, and Item Content
Figure 1 Results from Structural Equation Modeling for Model A
Figure 2 Results from Structural Equation Modeling for Model E
Figure 3 Results from Structural Equation Modeling for Model K
Figure 4 Results from Structural Equation Modeling for Model L
Table 2 Summary of Differences between Models
Table 3 Model Fit Indices for Model A, E, K, and L
Table 4 Factor Loadings for Model L
Table 5 Structural Coefficients for Model L
v

REFERENCES

48

vi

LIST OF TABLES
Table 1 List of Factors, Indictors, and Item Content

37

Table 2 Summary of Differences between Models

43

Table 3 Model Fit Indices for Model A, E, K, and L

44

Table 4 Factor Loadings for Model L

45

Table 5 Structural Coefficients for Model L

46

vii

LIST OF FIGURES
Figure 1 Results from Structural Equation Modeling for Model A

39

Figure 2 Results from Structural Equation Modeling for Model E

40

Figure 3 Results from Structural Equation Modeling for Model K

41

Figure 4 Results from Structural Equation Modeling for Model L

42

viii

CHAPTER ONE - INTRODUCTION
Statement of the Problem
Pierre Bourdieu‘s (1984) concept of cultural capital is popular among educational
researchers in Europe and the U.S. (De Graaf et al., 2000; DiMaggio, 1982; Dumais, 2002;
Lareau, 2003; Sullivan, 2001). Bourdieu (1973) defines cultural capital as ―instruments for the
appropriation of symbolic wealth socially designated as worthy of being sought and possessed‖;
this wealth appears to be the ―undivided property of the whole society‖—accessible to all on the
basis of individual ability and effort—but social origins shape its appropriation (p. 73). In his
theory, Bourdieu (1984) frames education as an arena of cultural conflict in which social
inequalities are reproduced through the various forms of ―capital‖ both within and without the
system. Nevertheless, due to researchers‘ diverse ways of interpretation, past research
examining the link between cultural capital and school outcomes has yielded inconclusive results
(DiMaggio, 1982; Kalmijn & Kraaykamp, 1996; Katsillis & Robinson, 1990; Robinson &
Garnier, 1985; Werfhorst et al., 2003).
In contrast to Bourdieu‘s critical view of the educational system, researchers who study
Asian societies often assume cultural homogeneity, social harmony, and limited inequality.
These researchers argue that common Confucian values and beliefs about education—within the
context of nationally standardized school curriculums and funding—result in comparable
academic achievement among middle class and working class children (Bond, 1996; Broaded,
1997; Chao, 1996; Smith, 1981; Stevenson & Lee, 1996; Wu, 1996; Yu, 1996).
Some scholars believe that in today‘s globalizing world, the social and cultural differences
between traditional Confucian Asian society and Western capitalist society are becoming less
clear. The ongoing processes of globalization may facilitate convergence between these two

1

societies. But many unique historical and institutional factors continue to shape the ways each
society changes and evolves in the path of modernization. Under the backdrop of widening
income gap between the rich and the poor on a global scale, this research aims to examine how
Western-developed concepts such as cultural capital can be applied to explain different
educational outcomes in the non-Western context.
While Taiwan does have less educational and income inequality than the U.S. (Buchmann
1

& Hannum, 2001; Chu, 1989, Thornton & Lin, 1994), family background still plays a crucial
role in shaping educational trajectories (Chang, 2006; Chen, 2005; Kelly, 2004; Tsai & Shavit,
2007; Wu, 2009). Students from low-income families (monthly income less than US$1,515) are
more likely to attend less academically vigorous high schools. For instance, students from lowincome families are less likely than others to go to academic-oriented (public) senior high school
(24.78% vs. the national average of 32.57%) and more likely to study in less prestigious (public)
vocational high school (26.48% vs. the average 20.93%) (Ministry of Education, R.O.C., 2009a)..
In addition, low-income students in Taiwan are less likely to attend college than are highincome students. In 2002, 15.4% of the students from mid-to-low income families (total annual
household incomes less than US$32, 941) attended public colleges/universities (7.15% below the
national average of 22.5%) while 84.6% of them attended private colleges/universities (7.15%
above the national average of 77.45%). A recent national survey indicates that in 2008, the
college attendance rate for students from low-income families was 8%, a starkly lower figure
than the national average of 22.5% (Ministry of Education, R.O.C., 2009a).

1 According to government statistics, in 2000, the Gini coefficient of income inequality was .343

for Taiwan (Directorate-General of Budget, Accounting, and Statistics, Executive Yuan, R.O.C.,
2012), compared with .447 for the U.S (U.S. Census Bureau, 2001). Note that 0 indicates
absolute equality and 1 indicates maximum inequality in which all income is concentrated in one
person.
2

Furthermore, among those who attend university, students from high-income families are
more likely than those from low-income families to attend public universities, which in Taiwan
are of higher prestige than private universities. At the college and university level, in 2002,
15.4% of the students from mid-to-low income families (total annual household incomes less
than US$32,941) attended public colleges/universities (7.15% below the national average of
22.5%) while 84.6% of them attended private colleges/universities (7.15% above the national
average of 77.45%). Despite having experienced rapid economic growth and expansion of
education during the past few decades, evidence of social inequalities in education persists.
The paradoxical mix of century-old Confucian influences and persisting educational
inequality makes Taiwan an interesting case for studying. Confucianism is a system of thought
that originated in ancient China during the Han dynasty, 206BCE-220CE. According to scholars
such as Oldstone-Moore (2002) and Shun and Wong (2004), Confucian thought centers on the
notions of social harmony, social hierarchy, family ethics, and individual responsibility. Based
on the Confucian tradition, an ideal person should be devoted to li (ritual and protocol),
motivated by virtue, and dedicated to serving the government and education (Oldstone-Moore,
2002, p. 7). In this research, I posit that Confucianism (as a widely held, conscious system of
thought) is analytically distinct from cultural capital (which, in its embodied form, is largely
unconscious and unequally distributed in society). While Confucianism produces similarities in
certain educational aspirations and practices across social classes, the different levels of
economic capital and cultural capital possessed by the students and their parents differentiate
2

their educational strategies and outcomes. In other words, I accept that the process of social

2 Not much research in the past has touched upon the issue of how equally Confucian beliefs are

shared across Taiwan. Ma and Smith‘s (1992) study shows that the extent of support for
Confucian ethical beliefs varied across occupation, residence, and place of origin but does not
3

reproduction of inequality is mitigated by cultural resources associated with Confucianism that
are widely held among both middle class and working class parents. Nevertheless,
Confucianism reduces, but does not eliminate, the inequalities in the distribution of cultural
capital and thus it does not eradicate class-based differences in school performance.
Purpose of the Study
The purpose of this dissertation research is to test, in the Taiwanese context, the
applicability of Bourdieu‘s view that cultural capital is unequally distributed and leads to
unequal educational outcomes. The study analyzes secondary data of junior high school students
(ninth graders, n=12,527) from the Taiwan Educational Panel Survey (TEPS). Different
structural equation models are tested to understand the causal relationships among parental
economic capital, parental cultural capital, students‘ cultural capital, and students‘ academic
achievement. Through testing different models, the study also aims to identity key components
of cultural capital that are unique to the Taiwanese context.
Research Questions and Hypotheses
To empirically assess the applicability of Bourdieu‘s concept of cultural capital to the
Taiwanese context, this study investigates (1) the extent to which parental economic capital
(income) is associated with parental cultural capital; (2) the extent to which parental economic
capital and cultural capital is associated with students‘ cultural capital; and (3) the extent to
which the student‘s cultural capital is converted to (positive) school outcomes. Therefore, the
research questions for this proposed study are:
1. To what extent is parental economic capital associated with parental cultural capital?

substantially differ between socioeconomic groups. Therefore, I posit that Confucian beliefs,
especially their emphasis on education, are still widely held by both middle and working class
parents and students in Taiwan. However, such a proposition is not tested in this research.
4

2. To what extent is parental cultural capital associated with students‘ cultural capital?
3. To what extent is parental economic capital associated with students‘ cultural capital?
4. To what extent is parental economic capital associated with students‘ school outcomes?
5. To what extent is parental cultural capital associated with students‘ school outcomes?
6. To what extent is the student‘s cultural capital associated with his or her school
outcomes?
If Bourdieu‘s argument is valid, we should find evidence of the unequal distribution of parental
cultural capital based on income levels. We should also find that parents transmit economic
capital and cultural capital to their children, and that this is associated with positive school
outcomes of those students who receive these advantages. Specific research hypotheses to be
tested are:
1. Parents‘ cultural capital is positively associated with parents‘ economic capital.
2. Parents‘ cultural capital is positively associated with students‘ cultural capital.
3. Parents‘ economic capital is positively associated with students‘ cultural capital.
4. Parents‘ economic capital is positively associated with students‘ academic achievement.
5. Parents‘ cultural capital is positively associated with students‘ academic achievement.
6. Students‘ cultural capital is positively associated with students‘ academic achievement.

5

CHAPTER TWO – LITERATURE REVIEW
Family plays a crucial role in shaping the educational trajectories of children through the
transmission of economic, social, and cultural resources. Previous research has established the
independent effect of each of these different types of resources on children‘s educational
outcomes in different social, cultural, and national contexts. To understand the mechanism
behind persistent educational stratification, Bourdieu‘s social reproduction theory provides us
with a framework for explaining how economic, cultural, and social resources combine to
influence educational outcomes. In this review of literature, I will first discuss research
regarding the influence of family socioeconomic status (SES) on educational outcomes in the
West and in Taiwan. I then discuss Bourdieu‘s theory of capital and studies that apply his
concept. Finally, I will discuss key issues concerning the application of cultural capital in
educational research.
Family SES and Educational Outcomes in the West and in Taiwan
A copious body of educational research in the West has tracked the impact of family
income and wealth on educational outcomes of children. Students from higher socioeconomic
backgrounds generally achieve better school outcomes than their less well-off counterparts.
Studies conducted in the U.S, Canada, and U.K. repeatedly showed that socioeconomic factors
have a substantial and persistent influence on school attainment and achievement (Acemoglu &
Pischke, 2001, Blanden & Gregg, 2004; Coleman et al., 1966; Duncan & Murnane, 2011;
Ferguson et al., 2007). For example, in the U.S., Acemoglu & Pischke (2001) found that a 10%
increase in family income is associated with a 1.4% increase in the probability of attending a
four-year college. Under the backdrop of widening income gap, a body of recently published
studies indicates a growing achievement gap as well as disparities in educational experiences

6

between the poor students and their more affluent peers in the U.S. (Philips, 2011; Reardon,
2011).
In the U.K., studies show that children from low-income households leave school for
work earlier and generally have fewer formal qualifications than their more affluent peers.
Blanden and Gregg‘s (2004) study shows that ―of [all] children born in 1970, some 26% failed to
3

achieve any O levels or equivalent by the age of 30, whilst 23% went on to get a degree.
Among children from the poorest 20% of households at age 16, only 11% went on to get a
degree and 41% failed to achieve any O levels‖. In a similar vein, Ermisch and Francesconi‘s
(1997) study demonstrates that financial constraints affect parents‘ investment in their children‘s
human capital. In addition, scarcity of resources (both of money and of time) in larger families
has adverse effects on the educational attainment of children.
The substantial effect of family income in shaping educational trajectories of children has
inspired research applying western-developed theories in non-western settings. The extent to
which family income matters—and the ways in which it matters—are not uniform crossnationally (Buchmann & Hannum, 2001). Despite the common view that Taiwan has less
income inequality than some countries in the West such the U.S. and U.K., (Buchmann &
Hannum, 2001; Chu, 1989, Thornton & Lin, 1994), evidence suggests that here too, students
from higher income families have an advantage over those from lower income backgrounds. For
instance, Hung and Marjoribanks (2005) concluded from their study of eleven-year-old
Taiwanese children (n=261), that family social status has an unmediated effect on children‘s
academic achievement, independent of parents‘ aspirations and parental involvement. Studying
3

O-levels (Ordinary Level) are subject and exam-based qualifications in the U.K. They were
the main examinations for 16 years olds in England, Wales and Northern Ireland from 1951 until
1988.
7

the impact of family and school factors on students‘ academic achievement at the junior high
school level, Lin (2007) showed that family SES exerts profound effects on school achievement
when compared with various resources provided by the school (which explained only 4% of the
variance). Using a national survey data set, Han and colleagues (2003) found a positive
relationship between family income and college attendance rates. Furthermore, Wu‘s (2009)
study of 1,510 undergraduates from five national universities in Taiwan show that compared to
non-elite universities, a larger proportion of students at the elite universities come from middle
and high SES backgrounds. In addition, Wu also found that the college experience for students
from different socioeconomic backgrounds are quite different and that students from less welloff families often engaged in paid employment during their university studies. Wu thus
concluded that ―the inequality in higher education participation among students from different
social classes is no longer an inequity in participation rates but an inequity in educational
quality‖ (p. 404).
Wu‘s findings are supported by Tsai and Shavit‘s (2007) study of access to higher
education in Taiwan. Tsai and Shavit argue that the educational expansion in Taiwan was
accompanied by stratification in the type (quality) of education the students receive. While
students from less-advantaged family backgrounds now have greater assess to higher education,
differences exist in the types of universities and colleges they go to (public vs. private) and types
of educational experiences they have (high vs. low teaching quality). Luca‘s (2001) Effectively
Maintained Inequality (EMI) theory states that for levels of education that are universal,
competition takes place in the type (quality) of education attained. This theory has found support
in the evidence from Taiwan.

8

To understand the educational field in Taiwan, we need to consider both the cultural and
institutional factors that shape ―the rule of the game‖ and at the same time, structure individual
practices (Bourdieu & Wacquant, 1992, p. 18). Like most nations, both the content and structure
of Taiwan‘s educational system is intricately linked with its economic, societal, and political
development (Yang, 1994). In 1968, the compulsory phase of schooling was extended to nine
years (six years of elementary and three years of junior high school). After completing the
elementary school, usually at age 12, students are assigned to local junior high schools based on
their registered places of residence. In 2008, 99.8% of elementary school graduates continued to
junior high school (Ministry of Education, ROC, 2009).
The transition from junior high school to senior high school can be regarded as the first
critical point of academic transition because students will take competitive examinations and be
assigned to tracked schools according to their scores. At the post-secondary level, schools are
hierarchically ranked and differentiated by their functions (academic oriented vs. non-academic
oriented). Three main types of schooling are available to junior high school graduates: academic
senior high schools, vocational high schools, and five-year junior colleges (consist of three years
of high school curricula and two years of specialized technical training).
Students who are admitted to the academic senior high school will undergo rigorous
preparation for the joint college entrance examination in order to gain admission to colleges and
universities. On the other hand, students who are admitted to either vocational high schools or
five-year junior college will engage in specialized technical and vocational training and later join
the job market. In 2008, 96.06% of the girls and 94.6% of the boys advanced to the postsecondary level (Ministry of Education, ROC, 2009). The same year, the ratio of senior high

9

school students to vocational high school students was 46:54 (Ministry of Education, ROC,
2009).
Due to the restriction imposed by the government on the proportion of academic-tracked
students at the secondary level, the competition for these spots is fierce among junior high school
students. However, the association between family background and children‘s school
achievement is likely to be weaker than that in the United States due to Taiwan‘s more equal
provision of basic education and the ―incentive structure‖ within which teachers strive to help
those who can perform well at the standardized tests regardless of their family backgrounds
(Broaded, 1997, p. 39, also see Stevenson & Lee, 1996, p. 141).
Despite this, some formal and informal features of the educational systems can contribute
to unequal educational outcomes. First, the prevalence of after-school tutoring services, ranging
from pre-kindergarten English conversation lessons to subject-specific exam preparation, have
long been a predominant part of the student experience. Because these tutoring services often
require substantial financial investment, students from families with higher socioeconomic
backgrounds can expect to enjoy more of such resources and support.
Second, due to the fact that students often attend the elementary and junior high school
within the school districts closest their residence, patterns of residential segregation based on
income, social class, or ethnicity can be seen in the composition of the student body of the school
(Broaded, 1997). Furthermore, junior high schools are informally ranked based on the
percentage of their graduates admitted to top academic senior high schools. Therefore, the
practices of ‗skipping‘ from one‘s neighborhood school to better school or enrolling students in
the private junior high school to be better prepared for the joint college entrance exam are

10

common. This introduces a ―potentially inegalitarian thrust‖ in the system of equal educational
provision (Broaded, 1997, p. 38). As Jao & McKeever state:
Academic secondary schools now serve as the most crucial gate through which
Taiwanese students must pass if they hope to gain access to higher education and greater
opportunities in the labor market, and our analysis shows how crucial it is to distinguish
between these two tracks within secondary schools….In Taiwan, it seems that social
elites take clear advantage of not merely levels of education, but distinctions within
specific levels of the educational system, to ensure that their children obtain specific
educational credentials even as the educational systems expands and average
achievement rises. (Jao & McKeever, 2006, p.149)
Bourdieu’s Theory of Capital
While it is an important factor, family income captures only one aspect of how parental
status influences school performance. To explain educational outcomes and social reproduction,
Bourdieu (1986) specifies four generic types of capital: economic capital, cultural capital, social
4

capital, and symbolic capital. According to Bourdieu, people maximize their standings by
accumulating and using these four different types of capital. The value of capital is specific to
the field within which individuals (or more precisely, individuals of the same class conditions)
compete and maneuver.
Field is the termed used by Bourdieu to capture ―the rule of the game‖ and to symbolize
struggle and competition within different social spheres (Bourdieu & Wacquant, 1992, p. 18). It
can be conceived as ―structured spaces that are organized around specific types of capital or
combinations of capital‖ (Swartz, 1997, p.117). The patterns and dynamics of competitions and
maneuvering in a given field result in social structures. These structures do not determine action

4

Bourdieu (1984, 1990) also use ―academic capital‖ to indicate the value that the society
accords to academic qualifications . In this paper I will consider academic capital as a subset of
cultural capital that refers to one‘s educational level.
11

but exert influences through shaping the disposition of the individual, while at the same time the
structure can be shaped by individuals who have gained and accumulated a lot of capital.
In Bourdieu‘s theory, social actors pursue strategies, but not as conscious maximizers of
available means to achieve desirable goals (Swartz, 1997). To Bourdieu (1992), human minds
are ―socially bounded‖ and ―socially structured‖ in a way that their courses of action are more
tacit, practical, and dispositional than simply rule-following or rational calculation (p. 126).
In the education field, for instance, parents of different social-economic statuses all strive
to do what they think is best for their children despite their differences in income, education
levels, interpersonal networks, and individual disposition. These differences in specific types of
capitals and combinations of capitals results in different educational practices and strategies,
which can either reinforce or challenge the existing order.
Bourdieu (1986) sees economic capital as the ―root‖ of all other forms of capital and
treats all others types of capital as ―transformed, disguised forms of economic capital‖ (p. 91).
Other forms of capitals can be converted into economic capital (such as converting educational
credentials into a high paying professional position) and economic capital can be used in pursuit
of other forms of capital (e.g. family income can be used to pay for school tuition and tutoring
services). Bourdieu (1986) sees social capital as actual or potential resources associated with ―a
durable network of more or less institutionalized relationship of mutual acquaintance and
recognition‖ (p. 88). To Bourdieu, social capital can be accumulated and deployed, collectively
or individually, for instrumental (tangible) or symbolic gains. Finally, Bourdieu (1990) uses
symbolic capital to denote the power of the dominant class to impose meanings through
legitimation. To Bourdieu, symbolic capital is a disguised form of power that demands
recognition, deference, and obedience with legitimacy (Swartz, 1997). It involves culturally

12

valued attributes (such as one‘s accent) that can be material but are not recognized as such
(McDonough & Nunez, 2007, p. 148).
There is nothing novel about suggesting that money and networks matter. Bourdieu‘s
(1986) concept of cultural capital is arguably the most innovative component of his theorization
of the forms of capital. The concept of cultural capital is central to Bourdieu‘s analyses of how
the educational system contributes to the process of social stratification. Examining the popular
post-World War II public policies of expanding educational opportunities in order to alleviate
societal inequalities, Bourdieu found that despite the tremendous improvement made in all
Western democracies, glaring inequalities in wealth, income, and status persisted (Bourdieu &
Passeron, 1977). Bourdieu (1977a; Bourdieu & Passeron, 1977) argues that rather than
functioning as an equalizer, the educational system actually reproduces the unequal distribution
of inherited cultural differences and therefore, is the institution most culpable for the
transmission of social inequalities in modern societies (Swartz, 1997).
Bourdieu (1973) defines cultural capital as ―instruments for the appropriation of symbolic
wealth socially designated as worthy of being sought and possessed‖; this wealth appears to be
the ―undivided property of the whole society‖—accessible to all on the basis of individual ability
and effort—but social origins shape its appropriation (p. 73). According to Bourdieu, cultural
capital exists in embodied, objectified (physical), and institutionalized states. The embodied
state has roots in the family environment of early childhood, in which values, skills, and manners
are cultivated that contribute to the forming of ―long lasting dispositions of the mind and body‖
(Bourdieu, 1986, p. 84). The embodied form of cultural capital is related to Bourdieu‘s concept

13

5

of habitus.

Bourdieu uses the term habitus to denote a ―socialized subjectivity‖ that

internalizes the externality (incorporation) and externalizes the internality (objectification), and
in the process, contributes to the reproduction of individuals and classes (Bourdieu, 1977b, p. 72;
Bourdieu & Wacquant, 1992, p. 126). The objectified state of cultural capital refers to physical
possessions of cultural goods such as pictures, books, dictionaries, instruments, machines, etc.
Finally, the institutionalized state refers to academic credentials and professional certifications.
Bourdieu‘s framing of cultural capital varies at different stages of his intellectual life and
6

performs different roles in his various writings. Lamont and Lareau (1988) note that ―in
Bourdieu‘s global theoretical framework, cultural capital is alternatively an informal academic
standard, a class attribute, a basis for social selection, and a resource for power which is salient
as an indicator/basis of class position‖ (p. 156). Overall, his theory of social reproduction is a
powerful tool for unmasking hidden relationships of culture, power, and stratification
(McDonough & Nunez, 2007). As Lareau and Weininger (2004) note: ―The concept of ‗capital‘
has enabled researchers to view culture as a resource – one that provides access to scarce
rewards, is subject to monopolization, and under certain conditions, may be transmitted from one
generation to the next‖ (p. 105). More importantly, focusing on the accumulation and
conversion of capital, Bourdieu‘s theory can deepen our understanding of the processes and
mechanisms associated with class reproduction both within and outside the educational system.
5

Wacquant (1989) notes that ―the roots of habitus are found in Aristotle‘s notion of hexis,
elaborated in his doctrine of virtue, meaning an acquired yet entrenched state of moral character
that orients our feelings and desires in a situation, and thence our actions‖ (p. 315). Bourdieu
himself also used ‗cultural unconscious,‘ ‗habit-forming force,‘ ‗set of basic, deeply interiorized
master-patterns,‘ ‗mental habit,‘ ‗mental and corporeal schemata of perceptions, appreciations,
and action,‘ and ‗generative principle of regulated improvisations‘ to designate the concept of
habitus (Swartz, 1997, p. 101).
6
See Lamont & Lareau, 1988, p. 155, for different definitions of cultural capital by Bourdieu
and Passeron.
14

Bourdieu‘s cultural explanation of unequal educational attainment and achievement differs
from the blaming-the-victim version of culture-of-poverty arguments. Rather than emphasizing
the cultural origins of persistent deviant behavior, Bourdieu‘s theory focus on the mechanism
and processes within which individuals adapt to limited (or structured) opportunities. It shows
how structural disadvantages can be incorporated into relatively stable dispositions through
intergenerational socialization and in the process, reproduce social structure.
Despite these merits, several issues in Bourdieu‘s theory need to be further explored.
While Bourdieu‘s theory highlights the contrasting and competitive practices of individuals in
different class positions, it may be less useful in explaining social actions that are cooperative in
nature. Culture does not simply structure social conflict and the reproduction of inequality; it
can also shape production and cooperation (Lauglo, 2000; Swartz, 1997). Thus, applying
Bourdieu‘s concept of cultural capital to societies influenced by Confucian thought—such as
Taiwan—provides an opportunity for understanding how cultural and social contexts condition
processes of stratification.
7

Over the years, characteristics of Confucian values that stress the acquisition of
academic skills, human malleability, persistence, restraint of emotion, deference to the group,
8

parental authority, filial piety, environmentalism, moral development, and parental training of

7 A value is ―a conception, explicit or implicit, distinctive of an individual or a group, of the

desirable which influences the selection from available modes, means, and end of actions‖
(Kluckhohn, 1951, cited in Bond, 1996, p. 209). Yu (1996) notes that the three highest concerns
of Confucianism are lide (attaining virtue), ligong (rendering meritorious service), and liyan
(establishing words) (p. 232). Also see Oldstone-Moore, 2002, for a discussion of key concepts
of Confucianism.
8
Although filial piety (meaning respects for parents and ancestors) is often regarded as an
important value associated with Chinese parenting, Ho (1996) notes: ―the research results
reviewed point to two broad generalizations, both of which reinforce the view that filial piety no
longer commands absolute observance as it did in the past. First, the extent to which traditional
15

children have been identified as having direct or indirect impacts on the educational beliefs and
practices of parents and students (Bond, 1996; Broaded, 1997; Chao, 1996; Ho, 1996; Li, 2003;
Schneider & Lee, 1990; Smith, 1981; Stevenson & Lee, 1996; Tweed & Lehman, 2002; Wu,
1996; Yu, 1996). As a result of the high premium the society places on academic achievement,
over the years, students from Taiwan have achieved high scores in the international standardized
testing. For instance, in the latest results of the Program for International Student Assessment
(also known as PISA), students from Taiwan were ranked 5th in math, 12th in science, and 25th in
reading among all participating countries and economies (Taiwan PISA National Center, 2012).
At the same time, the Confucian concept of ‗open education‘ (or ‗education without
9

discrimination‘) and the belief in mobility through educational attainment and achievement
have survived two thousand years of Chinese history and are still alive and well in Taiwan today
(Smith, 1981). With regard to the concept of open education, Stevenson and Lee (1996) note
that much of the research literature in Chinese on academic achievement advocates the provision
of equal educational opportunities to socioeconomically less advantaged students through the
help of the government and the school. To suggest that parents of different SES engage in
similar educational beliefs and practices due to the Confucian influence is taking the proposition
further than evidence allows. Nevertheless, the evidence does suggest that educational
achievement values are entrenched into the lower reaches as well as the upper reaches of the
class structure in Taiwan (Broaded, 1997). Next I will discuss issues concerning the use of
cultural capital in educational research.

filial attitudes are reflected in actual behavior seems rather limited. Second, present-day Chinese
are becoming selective in their filial beliefs and actions‖ (p. 155).
9
Confucius once said, ―Education knows no class‖ (cited in Smith, 1981, xxxi).
16

Cultural Capital in Educational Research
Despite the common consensus of treating culture as a powerful force shaping ones‘
educational trajectory, the theoretical usefulness inherent in Bourdieu‘s concept of cultural
capital in empirical research have not been fully realized. For instance, past research using the
concept of cultural capital in explaining different educational outcomes was mainly conducted in
countries in the West and has yielded inconclusive results. Part of this is due to researchers‘
diverse ways of defining and operationalizing the concept of cultural capital. For instance, some
researchers associate the term with high culture while others treat it as dominant institutional
standards used for social and cultural exclusion. Another reason could be that Bourdieu himself
has not offered a clear definition despite the richness of his writing on cultural reproduction. As
Lamont and Lareau (1988) note, ―This proliferation of definitions [of cultural capital],
undoubtedly a sign of intellectual vitality—and possibly, of the fruitfulness of the concept—has
created sheer confusion‖ (p. 153).
Perhaps the most common way of operationalizing cultural capital is to treat it as related
to knowledge of or competence with ―highbrow‖ culture (such as fine art, classical music, and
literature). Researchers taking this approach assume that cultural competence serves as a
―signal‖ of high social status that is implicitly rewarded in the educational system. In this line of
research, variables such as one‘s familiarity with high culture (including attitude, activities, and
information on art, music, literature, theater, etc.) as well as linguistic ability are often used as
indicators of the ―amount‖ of cultural capital the student or the parent possesses (e.g. De Graaf et
al., 2000; DiMaggio, 1982; Dumais, 2002; Sullivan, 2001).
Interpreting cultural capital as familiarity with high-culture, some researchers have found
positive effects of cultural capital on grades or educational attainment (DiMaggio, 1982; Kalmijn

17

& Kraaykamp, 1996; Werfhorst et al., 2003). Others have found parental reading behavior to be
positively associated with children‘s educational attainment whereas Beaux-arts participation is
not (e.g. De Graaf et al., 2000). Still others have found no effect of cultural capital on
educational outcomes (Katsillis & Robinson; 1990; Robinson & Garnier, 1985).
Several methodological issues can be identified to explain the inconclusive results. First,
when the concept of cultural capital was first developed, it was used by Bourdieu to describe
how the French educational system imposes certain evaluative standards upon students and thus
contributes to the reproduction of the social distribution of cultural capital within the society.

10

While it may be true that there exists ―congruity between educational norms and status practices‖
in the French society, it may not be necessary for it to take a highbrow aesthetic form, especially
when the concept is applied in a non-French context such as the United States or other societies
(Lareau & Weininger, 2004, p. 117).
Another problem associated with equating cultural capital with high-culture activities and
consumption is that it creates the distinction between ―social‖ (non-cognitive skills, habits, and
styles) and ―technical‖ (cognitive skills and grades) competence, which does not fully capture the
complexity of Bourdieu‘s original theorization (Lareau & Weininger, 2004). To Bourdieu
(1986), both forms of competence are interwoven in the embodied state (habitus) and are in part
socially constructed, and therefore, should not be treated separately (Bourdieu, 1986).

10

Bourdieu (1977a) states: ―It is in terms of this logic that must be understood the prominent
value accorded by the French educational system to such subtle modalities in the relationship to
culture and language as affluence, elegance, naturalness, or distinction….‖ (p. 495).
18

Interestingly, much of mainstream research does not consider the embodied form of cultural
capital (or habitus) in its analysis, making it difficult to see how Bourdieu‘s model functions.

11

A different understanding of the concept of cultural capital animates the work of
Blackledge (2001), Carter (2003), Lareau (2000, 2003), Lareau & Horvat (1999), McDonough et
al. (2000) and Reay (1998). Considering cultural capital as a ―resource‖ that facilitates the
compliance of institutional standards, this group of scholars examines how dimensions such as
gender, ethnicity, and social class function to generate distinctions that also serve as a basis for
social exclusion. Employing mostly micro-interpersonal methods (ethnography and interviews),
their findings not only highlight the arbitrary nature of institutional imposition within different
social contexts, but also underscore the importance of considering how the individual strategic
use of knowledge, abilities, and skills produce benefits within different fields (Lareau &
Weininger, 2004; Lamont & Lareau, 1988). For example, Lareau‘s (2003) study shows that
when interacting with educational authorities (teachers and school administrators), middle-class
parents exhibit a sense of entitlement and pursue strategies and deploy cultural resources that are
absent among their working-class and poor counterparts (e.g. making requests based on their
children‘s special needs). In doing so, middle-class parents not only better able to have their
demands met but also appear to be more capable of complying with the dominant institutional
standards that call for active and engaged parents.
To summarize, despite the broad acceptance of treating culture as a key element of
stratification within the disciplines of sociology and education, the theoretical potential inherent
in Bourdieu‘s concept of cultural capital remains underdeveloped. In this regard, Bourdieu‘s
11

Dumais‘ (2002) study is an exception. However, she operationalized the concept of habitus as
students‘ occupational aspirations, which may not fully reflect the dynamic nature of the concept
itself.
19

own conceptual ambiguity and researchers‘ insufficient consideration for both the context (what
constitutes cultural capital in a particular social and cultural setting) and content (different forms
of cultural capital that reflect Bourdieu‘s original theorization) in applying the concept have
hindered the advance in this line of inquiry.

20

CHAPTER THREE - METHODS
Sample
The data of this study is from the first and second panels of the Taiwan Education Panel
Survey (TEPS), a survey based on a nationally representative sample of 20,055 seventh-grade
students, their parents, and their teachers in 2001 and 2003. Sponsored by Academia Sinica, the
Ministry of Education, and the National Science Council, TEPS is a series of island-wide surveys
of students, teachers, and parents at junior high schools, senior high schools, and five-year
colleges. The data was collected with the intention of measuring cultural capital. The
questionnaires used in the survey contain items on cultural variables (such as reading and
cultural activities), parental educational involvement variables (such as parents‘ attitude toward
education and learning and educational practices), and teacher-reported academic skills and
performance measures (such as grades, level of efforts, and problem-solving skills of the
student).
Measurement of Variables
Parental economic capital.
The parental economic capital variable in this study will be based on two questions
related to the financial situation of the family in the TEPS parents‘ questionnaire. One question
asks ―What is the total monthly family income?‖ The other question asks ―How has the family‘s
financial situation been during the past ten years?‖
Parental cultural capital.
The parental cultural capital variable is constructed based on questions in the TEPS
parents‘ questionnaire, which includes items on (1) parents‘ educational level, (2) parents‘
reading, (3) parents‘ cultural activities, and (4) parents‘ self-reported ability, skills, and level of

21

confidence in various daily situations (see Appendix A Table 1 for a list of questions from the
TEPS parent questionnaire). Combining these four dimensions will better reflect Bourdieu‘s
original conceptualization of cultural capital.
Students’ cultural capital.
The students‘ cultural capital variable is constructed based on in questions from the TEPS
students‘ questionnaire on (1) student‘s reading, (2) student‘s cultural activities, (3) teachers‘
evaluations of students‘ aptitudes and skills, (4) students‘ self-reported ability, skills, and level of
confidence in various daily situations, and (5) students‘ academic aspirations (see Appendix A
Table 1 for a list of questions from the TEPS student questionnaire).
Integrating all these dimensions not only takes into account the ‗embodied‘ form of
cultural capital but also the ability of the students to comply with the institutional standards from
the perspective of the teacher. Such conceptualization of cultural capital will better reflect the
versatile nature of cultural capital as ―an informal academic standard,‖ ―a class attribute,‖ and ―a
basis for social selection‖ in Bourdieu‘s original theorization (Lamont & Lareau, 1988, p. 156).
Students’ academic achievement.
Students‘ academic achievement will be measured by (1) students‘ IRT (Item Response
Theory)

12

scores in 2001 and 2003 and (2) teacher-reported grades in the first semester of the

student‘s third year. In the TEPS questionnaire, teachers were asked to rank students
academically on a scale of 1 (among top five students in the class) to 5 (among bottom five
students in the class).

12 Item Response Theory (IRT), also known as latent trait theory, is based on the idea that the

probability of a correct response to an item is a mathematical function of a person and item
parameters. It is generally regarded as superior to the classical test theory and has been used to
measure abilities, attitudes, and intelligences.
22

Data Analysis
Structural equation modeling (SEM) with AMOS 17 was used to analyze the association
between four latent variables: parental economic capital, parental cultural capital, students‘
cultural capital, and students‘ academic achievement. The use of SEM is well-suited for
analyzing the data for the present study because of its ability to analyze a multi-equation model
within which some of the latent variables have multiple indicators (Kline, 2004). Table 1 (see
Appendix A) shows the names, context of the latent variables, and observed indicators.
Several steps were taken in preparing the data for analysis. Variables were re-coded
based on the content to ensure the directionality of respondents‘ answers are uniform (e.g.
1=never, 5=always; 1=no, 2=yes). Those whose answers are ―I don‘t know‖ were recoded into
―system missing.‖ The percentage of missing responses for most variables in the final model (L)
ranges 0.1 % to 10.2 %. Exceptions are SAS1 (Student‘s Academic Aspiration 1), 19.9%, SAS2
(Student‘s Academic Aspiration 2), 26.9%, and SAA3 (Student‘s Academic Achievement 3),
18.6%. The full information maximum likelihood (FIML) estimation in AMOS, which avoids
the massive loss of cases resulting from listwise deletion, also avoids the biases associated with
pair-wise deletion. This method was used to deal with the missing data.

13

To approximate

normality, several variables that have a skew greater than +2 were logarithmically transformed.
Developing the Models
Bourdieu‘s theory states that parental economic and cultural capital can be transmitted
into students‘ cultural capital and students with a greater amount of cultural capital will do better
at school. To test his theorization, four different models were developed. All these models
consist of four latent constructs (or factors)— parental economic capital, parental cultural capital,
13

Also see Allison (2002) and Arbuckle (1996)
23

students‘ cultural capital, and students‘ academic achievement. Each of these latent constructs
has its own second-order factors and indicators (the content and change of factors and indicators
will be explained later). Reliabilities were estimated for observed indicators of the same
construct (latent variable) and Cronbach‘s alpha ranged from 0.7 to 0.9.

14

Despite the

differences between models, they all reflect the same theoretical constructions outlined by
Bourdieu. That is, parental economic capital and parental cultural capital positively affect
student‘s cultural capital, which in term influences his or her academic achievement.
Before testing a substantive or theoretical model, one should first find a good
measurement model (Kline, 2005). Hence four different models were run (see Appendix B for
the diagram of each model and Table 2, Summary of differences between models). The model
fit indices are presented in Table 3 in Appendix C. All of these measurement models reflect the
same theoretical constructions outlined by Bourdieu.
The original model, model A (see Figure 1 in Appendix B), consists of four latent
constructs: parental economic capital, parental cultural capital, students‘ cultural capital, and
students‘‘ academic achievement. Parental economic capital is measured by two questions in the
TEPS survey. One of the questions asks about total family monthly income and the other
question asks about family financial situation in the past ten years.
Parental cultural capital is a second-order factor that is measured by parents‘ education
and three (first-order) factors: parents‘ reading (PR), parents‘ cultural activities (PC), parents‘
self-reported ability and confidence in various daily situations (PA), and their indicators.
Students‘ cultural capital is a second-order factor that is measured by five first-order factors:
student‘s reading (SR), student‘s cultural activities (SC), students‘ academic aspiration (SAS),
14 With exceptions SC (students‘ cultural activities) and SR (students‘ reading) for which

Cronbach‘s alpha was 0.6 and 0.4 respectively.
24

student‘s self-reported ability and confidence (SA), teachers‘ evaluation of students (TE), and
their indictors.
Incorporating reading and cultural activities variables as part of students‘ cultural capital
and parents‘ cultural capital constructs allows the examination of the relationship between the
two. Treating parents‘ and students‘ self-evaluated ability and confidence as part of their
cultural capital in the original model (A) is intended to capture the ―embodiment‖ of cultural
advantage in Bourdieu‘s theorization. In addition, the inclusion of teachers‘ evaluation of
student‘s skills and abilities (e.g. psychological maturity, abstract thinking, and problem solving)
as part of students‘ cultural capital is an attempt to capture the ―imposition of evaluative criteria‖
by the educational authority (Lareau & Weininger, 2004, p. 126). Lareau and Weininger (2004)
contend that the ways that students are evaluated at school (both formally and informally) often
reflect the dominant institutional standards and these standards often favor students from certain
family backgrounds. Finally, the outcome factor, students‘ academic achievement, is measured
by the students‘ IRT (Item Response Theory) scores in both 2001 and 2003 and teacher-reported
student grades in 2003.

25

CHAPTER FOUR – RESEARCH FINDINGS
Finding the Best Measurement Model
Among all models tested, Model A (see Figure 1 in Appendix B) is the most
comprehensive in terms of factors and indicators included in the model. The Chi-square for
Model A is 19,923, with 582 degrees of freedom and a p-value less than .001 (See Table 3 in
15

Appendix C). RMSEA

equals .052 and CFI, NFI, and TLI are .873, .869, and .854,

respectively. These results are not satisfactory because CFI and NFI are below .90.
In the second model, Model E (see Figure 2 in Appendix B), two paths were added for
theoretical reasons: the path between Parental Economic Capital and Student‘s Academic
Achievement and the path between Parental Cultural Capital and Student‘s Academic
Achievement. In addition, three second-order factors (SC, SA, PA) and an indicator (SR3) were
eliminated due to their low standardized regression weights. The results for Model B showed
that Chi-square equals 11,084 with 198 degrees of freedom. RMSEA equals .066 and CFI, NFI,
and TLI are .901, .900, and .874, respectively. While there were some improvements on CFI and
NFI, and TLI, RMSEA had become worse (from .052 to .066) because the model was less
parsimonious.
In the third model, Model K (see Figure 3 in Appendix B), a second-order factor TE
(teachers‘ evaluation of students‘ confidence and various abilities) was removed due to its high
correlation with Student‘s Cultural Capital (.793). Such a high correlation contributed to the
exceptionally high correlation between Student‘s Cultural Capital and Student‘s Academic
Achievement (.947). It seemed that in the TEPS survey, students who did well academically

15 RMSEA (Root Mean Square Error of Approximation) is a ―badness of fit‖ index in that a

value of zero indicates perfect fit and the higher the values, the worse the fit (Kline, 2005, p.
138).
26

were also ranked highly by their teachers in their confidence levels and abilities, making TE a
―contaminated‖ construct in the model. As a result, TE became a predictor of Student‘s
Academic Achievement. In other words, items in TE were quite from others in Student‘s
Cultural Capital in that they indicated student‘s effort and work habits (and teacher‘s perception
of them) rather than the academic skills the students might have. Once TE was removed, the
results in Model K showed some further improvements in the model fit, with Chi-square equaled
4,582 (df=125, p<.001), RMSEA equaled .053, and CFI, NFI, and TLI were .938, .937, and .915.
Among all the models tested, Model L has the best fit indices. In Model L (see Figure 4
in Appendix B), two more indicators (PR3 and PC3) in Model K were eliminated due to their
relatively low regression weights (factor loadings). The Chi-square is 2,611, with 94 degrees of
freedom and a p-value less than .001. However, Chi-square is proportional to sample size and
with a sample of over 12,000 rejecting the null hypothesis of perfect fit is almost certain. Hence,
as with the other models, fit indices that do not depend on sample size were used to evaluate the
model. Results showed that for Model L, RMSEA equals .046 and CFI, NFI, and TLI
are .961, .960, and .944, respectively, all of which are considered good by many experts in CFA
such as Bentler (1990).
To measure convergent and discriminant validity, a pure CFA measurement model was
run. This model was created by eliminating all causal relations between second-order factors
while estimating the correlations between them and the results showed that there was no
excessively (greater than .85) high inter-correlation among factors and therefore that the factors
can be considered distinct.

27

Assessment of the Best Measurement Model
In evaluating the measurement model, another focus is on the relationships between
factors and their indicators. As shown Table 4 (in Appendix D), in Model L, standardized
loadings between factors and their indicators (e.g between SAS—Student‘s Academic Aspiration
and SAS1, SAS2) range from moderate to high (from .46 to .916) and all are significant at p<.001.
The standardized loadings between Student‘s Reading (SR) and its indicators SR1 (.538)
and SR2 (.462) are relatively low when compared with other loadings. Nevertheless, SR is kept in
the model for two reasons. First, statistically, removing SR in the model worsens model fit
(RMSEA increases from .046 to .052 and squared multiple correlation for the outcome factor
student‘s academic achievement decreases from .600 to .451). Second, we keep it for theoretical
reasons. Reading as a habit (or linguistic competence) has been shown by past research on
cultural capital as being beneficial for school outcomes (e.g. De Graaf et al., 2000; Dumais, 2002;
Sullivan, 2001). Moreover, it is an essential element in Bourdieu ‗s theorization.
The low loadings in SR (SR1=.583, SR2=.462) may be a result of how the concept is
measured in the TEPS survey (based on (1) frequency of visiting bookstores and book fairs and (2)
time spent reading literature). Similarly, the low loading for EC2 (.534) could be due to the lessthan-ideal measures of parental economic status. These are based on (1) total monthly family
income and (2) financial situation in the past ten years, without information on family wealth such
as property ownership. Nevertheless, these are the only two questions related to parental financial
situation being asked in the TEPS survey. Overall, most of the latent constructs in the model can
be regarded as valid and reliable.

28

Assessment of the Structural Model
In evaluating the structural part of the model, the focus is on the relationships between
factors. The goal here is to see whether the directions (positive or negative) and magnitudes of
effects between factors specified in the model are supported by the data (Diamantopoulos &
Siguaw, 2000). As shown in Figure 4 (in Appendix B) Table 5 (in Appendix E), in Model L (as
well as in all other models), the signs of almost all parameters are consistent with the
hypothesized relationships among factors and all causal paths are significant at p<.001. That is,
parental economic capital is positively associated with parental cultural capital, which has a
positive effect on students‘ cultural capital and ultimately, their academic achievement.
The two non-significant paths are the one between parental economic capital and students‘
academic achievement and the path between parental cultural capital and students‘ academic
achievement. Nevertheless, the non-significance of these two paths does not undermine
Bourdieu‘s theory or contradict research hypotheses. Rather, it strengthens his argument that
parental cultural capital acts as a ―mechanism‖ through which family advantages (economical
and cultural) are transmitted. The correlation between parental cultural capital and parental
economic capital was .721, indicating that parents‘ income is highly associated with their
cultural capital (measured by parents‘ education, reading, and cultural activities). In other words,
parents of higher income tend to be better educated and are more engaged in reading and cultural
activities.
The standardized effect of parental economic capital on students‘ cultural capital is much smaller
(r=.108) than the effect from parental cultural capital (r=.532). Furthermore, the strong and
significant relationship between students‘ cultural capital and students‘ academic achievement is
evidenced by the structural coefficient of .727. While money matters, it is the transmission of

29

cultural capital from parents to students that has the largest effect on students‘ performance. The
squared multiple correlations (explained variance) for the two endogenous variables in the model
are .600 for student‘s academic achievement and .377 for students‘ cultural capital. Taken
together, the structural part of the model provides strong support for Bourdieu‘s theory that
parental economic capital and cultural capital have positive impacts on students‘ cultural capital,
which in turn, affects their academic achievement.

30

CHAPTER FIVE – DISCUSSION AND CONCLUSION
The findings from this study support Bourdieu‘s theory that parental cultural capital,
unequally distributed based on income levels, has positive effects on students‘ cultural capital
and subsequently, their school outcomes in Taiwan. As shown in the final structural equation
model (L), parental economic capital (income), the ‗root‘ of all forms of capital, is closely
associated with parental cultural capital (.721). While both forms of parental capital exert
positive direct effects on students‘ cultural capital, the effect from parental cultural capital is
much greater (.532 vs. .108.).
The study further indicates that in Taiwan, cultural capital takes the form of education level,
reading (PR), and cultural activities (PC) for the parents while for the students, the amount of
reading (SR) and high academic aspirations (SAS) are the most essential. Finally, the findings
also indicate that students‘ cultural capital has a strong effect on their academic achievement
(.727) and that its effect are much greater than the effects from parental economic capital (.039,
non-significant) and parental cultural capital (.042, non-significant). Overall, the findings
validate Bourdieu‘s argument that cultural capital act as a disguised mechanism that transmits
family advantages. The unequal distribution of cultural capital based on different income levels
leads to unequal distribution of cultural capital among the students, differentiating their
educational outcomes.
By testing different models based on Bourdieu‘s framework, we learned several lessons that
have implications for future research. First, the more ―conventional‖ (or in Lareau and
Weininger‘s term ‗the dominant‘) interpretation of cultural capital in the English-language
literature often treats cultural capital as a construct closely associated with high-brow cultural
activities. Such treatment unnecessarily creates a partition between one‘s sense of artistic

31

appreciation and linguistic abilities on the one hand, and cognitive skills and abilities on the
other.
On the surface, such an approach may appear to fall in line with Bourdieu‘s (1984)
contention that cultural consumption serves as a basis for social distinction. However, it does
not truly reflect his original theoretical intent, particularly his signature concept of habitus.
Indeed, to Bourdieu (1986), both cultural competence (such as art appreciation) and technical
competence (such as scientific knowledge and math skills) are analytically inseparable.

16

These

two forms of competence merge in the ―embodied‖ state of an individual and are activated in
different social spheres. Thus, as this study has shown, incorporating academic skills as one
dimension of cultural capital better captures the essence of Bourdieu‘s theory and further realizes
the concept‘s empirical potentials.

17

Furthermore, the findings suggest a lack of correspondence between what constitutes
cultural capital for the parents and for their children. As shown in the final model, Model L (see
Figure 4 in Appendix B), parental cultural capital in Taiwan is indicated via parents‘ education
level, parents‘ reading, and parents‘ cultural activities. But for the students, academic aspiration
and reading are the most salient attributes of cultural capital.
This lack of correspondence has important methodological implications for future
investigation. First, in Bourdieu‘s theory of social and cultural reproduction, the underlining
assumption is that children acquire, intentionally and unintentionally, certain societally-valued
16 As Bourdieu (1986) states in his criticism of Becker and other human capital theorists, ―their

studies of the relationship between academic ability and academic investment show that they are
unaware that ability or talent is itself the product of an investment of time and cultural capital‖ (p.
85).
17
As Lareau and Weininger (2004) note, ―academic skills should not be excluded from the
purview of cultural capital research. Academic skills are, instead, part of what we should be
conceiving of as cultural capital‖ (p. 136).
32

qualities (tangible or intangible) through their family upbringing. It is further assumed that these
qualities later give them advantages in the process of education and in the job market. While
there may be ―leakage‖ in the process, the transmission is pervasive and systematic. Thus,
children whose parents listen to classical music and read often are more likely to have greater
affinity toward classical music and reading themselves.
Following this logic, researchers often take corresponding sets of questions asked to
parents and students as indicators of their cultural capital in assessing its effect on academic
attainment and achievement. For instance, Ganzeboom, De Graaf, and Robert (1990),
Aschaffenburg and Maas (1996), and Sullivan‘s (2001) work on cultural capital all invoke highculture practices or cues in both parents and their children. Although an association between
parents‘ cultural capital and students‘ cultural capital helps establish evidence on the intra-family
transmission of cultural capital, their correspondence (in terms of the constituents of cultural
capital) cannot be taken for granted and deserves scrutiny.
Such caution is especially warranted when analyzing data cross-culturally. As this study
suggests, unlike their parents, junior high school students in Taiwan are pre-occupied with heavy
school work and the preparation for national exams, leaving them no time for other activities.
This is evidenced by the elimination of SC (students‘ cultural activities) as one component of
students‘ cultural capital from the original model (A) due to low standardized regression weights
(.107). If this study had unquestionably taken students‘ cultural activities as the sole indicators
of their cultural capital measure, we might not be able to find an effect of students‘ cultural
capital on their academic achievement. Similarly, if we had relied only on parents‘ high-status
cultural signals (behaviors, tastes, and attitudes) and assumed that such signals would ―reproduce‖
themselves in their children, we may have failed to capture the real mechanism of the

33

transmission (and transmutation) of cultural capital, a process that may not be universal and is
shaped by local institutional contexts. This point is best explicated by Lamont and Lareau (1988)
who argue ―before the effects of cultural capital could be analyzed in a given context, its content
had to be empirically specified (cited in Lareau & Weininger, 2004, p. 117, emphasis in original).
In addition, the findings (as shown in Model L) also suggest that the amount of students‘
reading (SR) and how high their academic aspirations (SAS) were are the two key components
of student‘s cultural capital. This is a rather interesting (yet expected) outcome given that for
centuries Taiwan has been influenced by Confucian thought that places a premium on academic
excellence. Under such cultural ethos, students who internalized this motivational drive to
succeed academically are considered to have embodied ―symbolic wealth socially designated as
worthy of being sought and possessed‖ (Bourdieu, 1973, p. 73). While Confucian influences
facilitate certain educational aspirations and practices across social classes, the unequal
distribution of this motivational drive within the students and different levels of cultural capital
and economic capital possessed by the parents differentiate their educational strategies and
outcomes.
If this is true, where does this motivational drive come from? To Bourdieu, such
motivational disposition is inculcated (mostly unconsciously) mainly through one‘s family
upbringing. Once acquired, this inculcation becomes part of one‘s forgotten history (the
unconscious part of habitus) and later takes on a symbolic form to signal ―legitimate competence‖
in a given field. He writes:
Cultural capital can be acquired, to a varying extent, depending on the period, the society,
and the social class, in the absence of any deliberate inculcation, and therefore quite
unconsciously….It thus manages to combine the prestige of innate property with the
merits of acquisition….Because the social conditions of its transmission and acquisition
are more disguised than those of economic capital, it is predisposed to function as

34

symbolic capital, i.e., to be unrecognized as capital and recognized as legitimate
competence, as authority exerting an effect of (mis)recognition. (Bourdieu, 1986, p. 86)
Thus, according to Bourdieu, having high academic aspiration is not merely a random personal
trait. Rather, it is a product of social investment by the parents. It serves as a ―classifier‖ that
affects a student‘s ability to comply with the dominant educational norms. Nevertheless, more
future research is needed in this area to further establish this proposition.
This study has attempted to advance the theoretical and methodological development of
research on cultural capital in several important ways. First, this study attempts to improve on
the ways previous research operationalized cultural capital by taking into consideration the
―embodied‖ form of cultural capital (such as students‘ academic aspiration). In addition, unlike
research that mainly focuses on the net effect of cultural capital on student
attainment/achievement, this study also considers all the theoretical constructs outlined in
Bourdieu‘s theory (parental economic capital, parental cultural capital, students‘ cultural capital,
and students‘ academic achievement), allowing better evaluation of his model.
Finally, by applying Bourdieu‘s concept of cultural capital to the Taiwanese society, this
study demonstrates how social, cultural, and institutional contexts beyond the West condition the
process of educational stratification. Although Taiwan has lesser educational and social
inequality than some countries in the West (Buchmann & Hannum, 2001; Chu, 1989, Thornton
& Lin, 1994), parental cultural capital is closely associated with family income and has a strong
effect on students‘ cultural capital, which in turn influences their academic achievement. In this
regard, Confucianism reduces, but does not eliminate the effects of economic capital on
academic achievement and thus it does not eradicate differences between students from different
socio-economic statuses.

35

Limitations
Several limitations due to the design of the questionnaire have to be acknowledged here.
First, while it would be desirable to also have information on family wealth (e.g. property and
stocks ownership), the TEPS questionnaire does not contain such items. In addition, the
questionnaire does not ask questions about family size and without such information, monthly
family income can only serve as an imperfect indicator of parental economic capital. In addition,
the data analyzed is cross-sectional and does not allow us to examine the long term effect of
cultural capital. It is possible that students with greater amount of cultural capital at this point in
time may engage in more cultural activities later in their life than those with lesser amount of
cultural capital.
Conclusion
Applying a Western-developed concept in a non-Western context, this dissertation
research demonstrates that parental cultural capital in Taiwan is strongly associated with parental
economic capital (income) and has a significant effect on both students‘ cultural capital and their
academic achievement. Despite the fact that over the years Taiwan has achieved high
educational proficiency amid the legacy of Confucian influences, cultural and financial resources
within the family matter, both in the process of students‘ learning and their educational outcomes.
Bourdieu‘s theory of social and cultural reproduction is a powerful tool for unmasking
the often hidden mechanism that perpetuates structural inequality, yet the complexity of social
life makes producing a parsimonious definition of cultural capital a difficult task. Capital, be it
social, cultural, or economic, tends to accumulate. However, the ways through which different
types of capital converge and operate within different contexts will always remain a fruitful area
for sociological studies.

36

APPENDICES

37

APPENDIX A
Table 1 List of Factors, Indicators and Item Content
Parental Economic Capital
―What is the total monthly family income?‖
―How has the family‘s financial situation been during the past ten
years?‖
Parental Cultural Capital
―What is your educational level?‖

EC1
EC2

PE1
Latent
Variable

Observed
Indicators
PR1
PR2
PR3

PR
PR4

PR5
PC1
PC

PC2
PC3

PA

SR

PA1
PA2
PA3
PA4
PA5
SR1
SR2
SR3

―Do you often read?‖/ ―Does your spouse often read?‖ (taking the
mean of both parents)
―Did you make trips to bookstores, book fairs or borrow books
from the library during the past year?‖ (taking the mean of both
parents)
―Did you take your child to bookstores, book fairs, or any kind of
exhibits when he or she was in elementary school?‖
Taking sum of the following questions:
―Do you subscribe to Chinese language journals or magazines?‖
―Do you subscribe to English language journals or magazines?‖
―Do you have encyclopedia at home?‖
―Do you have Internet connection at home?‖
―During the past year, how much time did you spend on reading
books or magazines?‖ (taking the mean of both parents)
―During the past year, how many times did you go to classical
music concerts or dance/opera performances?‖ (taking the mean
of both parents)
―During the past year, how many times did you go to exhibitions
in art galleries or museums?‖ (taking the mean of both parents)
―Did you take your child to classical music concerts or
dance/opera performances when he or she was in elementary
school?‖
―You can easily communicate a relatively complex matter.‖
―You are well-organized no matter how big or small the thing is.‖
―You can work well with any person.‖
―You learn how to use products of high technology fairly fast.‖
―There isn‘t a lot of things in life that you can not handle.‖
Students’ Cultural Capital
―Do you frequent book stores, libraries, or book fairs?‖
―How much time do you spend reading literature on the subject of
history, philosophy, biography, politics, economics, and
technology?‖
―How many books did you borrow from the library in 8th grade?‖
38

SC

TE

SA

SAS

SAA

SC1
SC2
SC3
TE1
TE2
TE3
TE4
SA2
SA4
SA5
SA6
SA7
SAS1
SAS2

SAA1
SAA2
SAA3

―Do you listen to classical music or go to dance performances?‖
―Do you go to art or literary exhibitions at the gallery or
museum?‖
―Do you know how to play a musical instrument?‖
―What is the student‘s level of academic progress?‖
―What is the student‘s level of effort?‖
―How well does the student turn in assignments on time?‖
―How well does the student ask and answer questions in class?‖
―I am good at learning new technology.‖
―I am good at expressing my opinions, making reports, and
explaining things.‖
―I am very organized.‖
―I am good at working with others‖
―There isn‘t a lot of things in life that I cannot handle.‖
―What level of education do you expect yourself to obtain?‖
―Considering your ability, what level of education do you think
you can obtain?‖
Students’ Academic Achievement
Students‘ IRT Scores in 2001 (wave 1)
Students‘ IRT Scores in 2003 (wave 2)
―What is the grade of the student this semester?‖ (the question
was asked in the first semester of students‘ third year at the junior
high school in 2003, answered by their homeroom teachers)

39

APPENDIX B
Figure 1. Results from Structural Equation Modeling for Model A
(errors terms are not included)
Note: PEC=Parental Economic Capital, PCC=Parental Cultural Capital, SCC=Student‘s Cultural
Capital, SAA=Student‘s Academic Achievement
EC1
1

EC2

.779

SR2

PE1
PR1

SR

PEC

SR3

.678

.391
.132

.711

PR2

SC1
SC

@

SC2

PR3

.107

PR

PCC

PR4
PR5

SR1

.539

.963

SCC

.646

.416

SC3
SAS1

SAS

.717

SAS2

PC1
PC

PC2

.377

.076

PC3

.980

SA2
SA

PA1

SA3
SA4

PA2
PA3

SA1

.781

PA

PA4

SA5
TE1

PA5

TE2

SAA

TE
TE3
TE4

SAA1

SAA2

40

SAA3

Figure 2. Results from Structural Equation Modeling for Model E
(errors terms are not included)
Note: PEC=Parental Economic Capital, PCC=Parental Cultural Capital, SCC=Student‘s Cultural
Capital, SAA=Student‘s Academic Achievement

EC1

EC2

1

.785

SR2

PE1
PR1

SR1

.535
SR

PEC
.680

.382
.704

PR2

SAS1

.069

SAS

SAS2

@
PR3
PR
PR4
PR5
PC1
PC2
PC3

.429

PCC

.639

TE1

SCC

.959

TE2
.793

.717
.069
.947

SAA

SAA1

SAA2

SAA3

41

TE3
TE4

PC
.009

TE

Figure 3. Results from Structural Equation Modeling for Model K
(errors terms are not included)
Note: PEC=Parental Economic Capital, PCC=Parental Cultural Capital, SCC=Student‘s Cultural
Capital, SAA=Student‘s Academic Achievement
EC1
1

EC2

.783
PE1
PR1

.536

PEC
.682
.706

PR2

.113

@
PR3

SR1
.515

.545

SR
SR2

PR

PCC

PR4

SCC

SAS1

.958

SAS

PR5
PC1
PC2

.749
.717

.051

PC
.773

PC3
.006

SAA

SAA1

SAA2

SAA3

42

SAS2

Figure 4. Results from Structural Equation Modeling for Model L
(errors terms are not included)
Note: PEC=Parental Economic Capital, PCC=Parental Cultural Capital, SCC=Student‘s Cultural
Capital, SAA=Student‘s Academic Achievement
EC2

EC1

.786
PE1

.534

PEC
.721
.694

.108

PR1

SR

.039

PR2

SR1
SR2

PR
.509

PR4
PR5

SCC

PCC
.945

.782
.532

.640

SAS1
SAS

PC1
PC
PC2

.727

.042

SAA

SAA1

SAA2

SAA3

43

SAS2

Table 2 Summary of Differences between Models
Model A
Key components:
Parental Economic Capital
– EC1, EC2
Parental Cultural Capital
PE1
PR – PR1, PR2, PR3, PR4,
PR5
PC – PC1, PC2, PC3
PA – PA1, PA2, PA3, PA4,
PA5
Student’s Cultural Capital
SR – SR1, SR2, SR3
SC – SC1, SC2, SC3
SAS – SAS1, SAS2
SA – SA2, SA4, SA5, SA6,
SA7
TE – TE5, TE12, TE13,
TE14
Student’s Academic
Achievement
– SAA1, SAA2, SAA3

Differences between Models
Model E
Model K
Differences from
Differences from
Model A:
Model E:
Added (based on
theoretical
reasons):
(1) the path
between
Parental
Economic
Capital and
Student’s
Academic
Achievement
(2) the path
between
Parental
Cultural
Capital and
Student’s
Academic
Achievement
Eliminated (due to
low standardized
reg. weights):
(1) SR3
(2) SC
(3) SA
(4) PA

44

Eliminated TE
due to high intercorrelation
between TE,
Student’s
Cultural Capital,
and Student’s
Academic
Achievement
Items in TE were
quite from others
in Student’s
Cultural Capital
in that they
indicated
student‘s effort
and work habits
(and teacher‘s
perception of
them) rather than
student‘s
academic skills

Model L
Differences from
Model K:
Eliminated PR3
and PC3 due to
low standardized
reg. weights.

APPENDIX C
Table 3 Model Fit Indices for Model A, E, K, and L
Fit Measure
Chi-Square
(CMIN)
df
RMSEA

NFI
RFI
IFI
TLI (NNFI)
CFI
PNFI
PCFI
CN
AIC

ECVI

Desirable
Range
p >.05
<.05 good fit
.05 --- .079
reasonable fit
.08 ---.10
mediocre fit
>.10 poor fit
>.90
>.90
>.90
>.90
>.90
>.50
>.50
>200
Less than
those of the
Saturated and
Independence
models
Less than
those of the
Saturated and
Independence
models

Model
A
19923
p<.001
df=582

E
11084
p<.001
df=198

K
4582
p<.001
df=125

L
2611
p<.001
df=94

.052

.066

.053

.046

.869
.851
.873
.854
.873
.760
.763
418
20163<
152660
20163>
1404

.900
.872
.902
.874
.901
.704
.706
280
11238<
110785
(Ind.)
11238>
550 (Sat.)
.897<
8.844 (Ind.)
.897>
.044 (Sat.)

.937
.913
.938
.915
.938
.685
.686
416
4713<
72317 (Ind.)
4713>
378 (Sat.)

.960
.942
.961
.944
.961
.663
.664
565
2727<
64980 (Ind.)
2727>
304 (Sat.)

.376<
5.773 (Ind.)
.376>
.030 (Sat.)

.218<
5.188 (Ind.)
.218>
.024 (Sat.)

1.6<
12.187
1.6>
.112

45

APPENDIX D
Table 4 Factor Loadings for Model L










EC2
EC1
PE1
PR1
PR2
PR4
PR5

Standardized
Coefficients
.534
.786***
.694
.696
.734***
.633***
.742***

PC1

.621






PC2
SR1
SR2

.893***
.538
.462***

SAS1

.814




SAS2

.800***

SAA3

.666




SAA2
SAA1

.916***
.875***

Parameters
Parental_Economic_Capital
Parental_Economic_Capital
Parental_Cultural_Capital
PR (Parent‘s Reading)

PC (Parent‘s Cultural
Activities)
SR (Student‘s Reading)
SAS (Student‘s Acad.
Aspirations)
Student's_Academic_Achieve
ment

*** = p<.001
Note: The unstandardized coefficients for variables EC2, PE1, PR1, PC1, SR1, SAS1, and SAA3
were set to be 1 and no significance test is reported in AMOS output.

46

APPENDIX E
Table 5 Structural Coefficients for Model L
Parameters
Parental_Economic_Capital
Parental_Economic_Capital
Parental_Cultural_Capital
Parental_Cultural_Capital
Parental_Cultural_Capital
Student's_Cultural_Capital
Student's_Cultural_Capital









Student's_Cultural_Capital



Parental_Economic_Capital



Parental_Cultural_Capital



Parental_Cultural_Capital
Student's_Cultural_Capital
Student's_Cultural_Capital
PR
PC
SAS
SR
Student's_Academic_
Achievement
Student's_Academic_
Achievement
Student's_Academic_
Achievement

Standardized
Coefficients
.721***
.108***
.532***
.945***
.640***
.782***
.509
.727***
.039 (p=.073)
.042 (p=.181)

*** = p<.001
Note: The unstandardized coefficients for variable SR was set to be 1 and no significance test is
reported in AMOS output.

47

REFERENCES

48

REFERENCES
Acemoglu, D., & Pischke, J. S. (2001). Changes in the wage structure, family income,
and children's education. European Economic Review, 45(4-6), 890-904.
Allison, P. D. (2002). Missing Data. Thousand Oaks, CA: Sage.
Arbuckle, J. (1996). Full information estimation in the presence of incomplete data. In G. A.
Marculides & R. Schumacker (Eds.), Advanced Structural Equation Modeling: Issues and
Techniques (pp. 243-276). NJ: Erlbaum.
Aschaffenburg, K., & Maas, I. (1997). Cultural and educational careers: The dynamics of social
reproduction American Sociological Review, 62(4), 573-587.
Berg, B. L. (2007). Qualitative Research Methods for Social Sciences. Boston: Pearson/Allyn &
Bacon.
Blackledge, A. (2001). The wrong sort of capital: Bangladeshi women and their children's
schooling in Birmingham, U.K. International Journal of Bilingualism, 5(3), 345-369.
Blanden, J., & Gregg, P. (2004). Family income and educational attainment: A review of
approaches and evidence for Britain. Oxford Review of Economic Policy, 20(2), 245-263.
Bond, M. H. (1996). Chinese values. In M. H. Bond (Ed.), Chinese Psychology (pp. 208-226).
Hong Kong: Oxford University Press.
Bourdieu, P. (1973). Cultural Reproduction and Social Reproduction. In R. Brown (Ed.),
Knowledge, Education, and Cultural Change (pp. 71-112). London: Tavistock Publications
Limited.
Bourdieu, P. (1977a). Cultural reproduction and social reproduction. In A. H. Halsey (Ed.),
Power and Ideology in Education. New York: Oxford University Press.
Bourdieu, P. (1977b). Outline of A Theory of Practice. Cambridge ; New York: Cambridge
University Press.
Bourdieu, P. (1984). Distinction: A social critique of the judgment of taste. Cambridge, Mass.:
Harvard University Press.
Bourdieu, P. (1986). Forms of capital. In Sadovnik (Ed.), Sociology of Education: A Critical
Reader (pp. 83-95). New York: Routledge.
Bourdieu, P. (1990). The Logic of Practice. Cambridge: Polity.

49

Bourdieu, P., & Passeron, J. C. (1977). Reproduction in Education, Society, and Culture (1990
ed.). London ; Newbury Park, Calif.: Sage in association with Theory, Culture & Society, Dept.
of Administrative and Social Studies, Teesside Polytechnic.
Bourdieu, P., & Wacquant, L. J. D. (1992). An Invitation to Reflexive Sociology. Chicago:
University of Chicago Press.
Breen, R., & Johsson, J. O. (2005). Inequality of opportunity in comparative perspective: Recent
research on educational attainment and social mobility. Annual Review of Sociology, 31, 223243.
Broaded, C. M. (1997). The limits and possibilities of tracking: Some evidence from Taiwan.
Sociology of Education, 70(1), 36-53.
Brown, M. W., & Cudeck, R. (1993). Alternative ways of assessing model fit. In K. A. Bollen &
J. S. Long (Eds.), Testing Structural Equation Models (pp. 136-162). Newbury Park, CA: Sage.
Brubaker, R. (2004). Rethinking classical theory: The sociological vision of Pierre Bourdieu. In
D. Swartz & V. L. Zolberg (Eds.), After Bourdieu: Influence, Critique, Elaboration (pp. 25-64).
Dordrecht ; Boston: Kluwer Academic Publishers.
Buchmann, C. (2002a). Getting ahead in Kenya: Social capital, shadow education, and
achievement. In B. Fuller & E. Hannum (Eds.), Schooling and Social Capital in Diverse
Cultures. Boston: JAI.
Buchmann, C. (2002b). Measuring family background in international studies of educationi:
Conceptual issues and methodological challenges. In A. C. Porter & A.
Gamoran (Eds.), Methodological Advances in Cross-National Surveys of Educational
Achievement (pp. 150-197). Washington, DC: National Academy Press.
Buchmann, C., & Hannum, E. (2001). Education and stratification in developing countries: A
review of theories and research. Annual Review of Sociology(27), 77-102.
Carter, P. L. (2003). "Black" cultural capital, status positioning, and schooling conflicts for lowincome African American Youth. Social Problems, 50(1), 136-155.
Chang, H.-C. (2006). The dynamics of social changes and family attainment norms: The effect of
parental education, sibship structure, and gender differences on school transitions in Taiwan.
Unpublished Doctoral Dissertation, University of Wisconsin-Madison.
Chang, L.-Y. (2003). The Taiwan Education Panel Survey (TEPS): First Wave (2001) Student
Survey (Public Version), First Wave (2001) Parent Survey (Public Version), Center for Survey
Research, The Academia Sinica, Taipei, Taiwan

50

Chang, L.-Y. (2005). The Taiwan Education Panel Survey (TEPS): Second Wave (2003) Student
Survey (Public Version), Second Wave (2003) Parent Survey (Public Version), Center for
Survey Research, The Academia Sinica, Taipei, Taiwan
Chang, L.-Y. (2005). The Taiwan Education Panel Survey (TEPS): First Wave (2001) and
Second Wave (2003) Teacher Survey (Restricted Version), Center for Survey Research, The
Academia Sinica, Taipei, Taiwan. Authorization No.: Teps1a015098
Chao, R. K. (1996). Chinese and European American mothers' beliefs about the role of parenting
in children's school success. Journal of Cross-cultural Psychology, 27(4), 403-423.
Chen, W.-C. (2005). Does social origin weight more when a society gets richer? Over-time
analysis of social origin's influence on educational attainment in postwar Taiwan. Unpublished
Doctoral Dissertation, The University of Chicago.
Chu, Y.-P. (1989). 1980 zhi 1986 nian jian Taiwan Suode Fenpei Biandong de Fenxi [Changes
in Taiwan's income distribution bewteen 1980 and 1986]. In C.-C. Yi & C. R. Chu (Eds.),
Taiwan Shehui Xianxiang de Fenxi [Social Phenomena in Taiwan - An analysis] (pp. 437-455).
Taipei: Academic Sinica Press.
Coleman, J. S., Campbell, E. Q., Hobson, C. J., McPartland, J., & Mood, A. M. (1966). Equality
of educational opportunity. Washington, DC: Department of Health, Education, and Welfare.
De Graaf, N. D., De Graaf, P. M., & Kraaykamp, G. (2000). Parental cultural capital and
educational attainment in the Netherlands: A refinement of the cultural capital perspective.
Sociology of Education, 73(2), 92-111.
Diamantopoulos, A., & Siguaw, J. A. (2000). Introducing LISREL. London: SAGE Publications
Ltd.
DiMaggio, P. (1982). Cultural capital and school success. American Sociological Review, 47(2),
189-201.
Directorate-General of Budget, Accounting, and Statistics, Executive Yuan, R.O.C. (2012).
Report on percentage share of disposable income by percentile group of households and income
inequality indexes, 1964－2003. Retrived May 31, 2012, from
http://www.dgbas.gov.tw/public/Data/2261644171.dot
Dumais, S. (2002). Cultural capital, gender, and school success: The role of habitus. Sociology of
Education, 75(1), 44-68.
Duncan, G., & Murnane, R. (Eds.). (2011). Whither Opportunity? New York: Russell Sage
Foundation.
Ermisch, J., & Francesconi, M. (1997). Family Matters, CEPR/Royal Economic Society Meeting.

51

Ferguson, H., Bovaird, S., & Mueller, M. (2007). The impact of poverty on educational
outcomes for children. Paediatr Child Health, 12(8), 701-706.
Field, A. P. (2005). Discovering Statistics Using SPSS (2nd ed.). London: Sage.
Ganzeboom, H. B. G., De Graaf, P. M., & Robert, P. (1990). Cultural reproduction theory on
Socialist ground. Research in Social Stratification and Mobility, 9, 79-104.
Han, W.-J., Huang, C.-C., & Garfinkel, I. (2003). The importance of family structure and family
income on family's educational expenditure and children's college attendance: Empirical
evidence from Taiwan. Journal of Family Issues, 24(6), 753-786.
Ho, D. Y. F. (1996). Filial piety and its psychological consequences. In M. H. Bond (Ed.),
Chinese Psychology (pp. 155-165). Hong Kong: Oxford University Press.
Horvat, E. M. (2003). The interactive effects of race and class in educational research:
Theoretical insights from the work of Pierre Bourdieu. Penn GSE Perspectives on Urban
Education, 2(1), 1-25.
Hung, C.-L., & Marjoribanks, K. (2005). Parents, teachers, and children's school outcomes: a
Taiwanese study. Educational Studies, 31(1), 3-13.
Jao, J.-C., & McKeever, M. (2006). Ethnic inequalities and educational attainment in Taiwan.
Sociology of Education, 79(2), 131-152.
Kalmijn, M., & Kraaykamp, G. (1996). Race, cultural captial, and schooling: An analysis of
Trends in the United States. Sociology of Education, 69, 22-34.
Katsillis, J., & Robinson, R. (1990). Cultural capital, student achievement, and educational
reproduction: The case of Greece. American Sociological Review, 55, 270-279.
Kelly, S. (2004). Do increased levels of parental involvement account for social class differences
in track placement? Social Science Research, 33, 626-659.
Kline, R. B. (2005). Principles and Practices of Structural Equation Modeling (2nd ed.). New
York: The Guilford Press.
Lamont, M., & Lareau, A.(1988). Cultural capital: Allusions, gaps, and glissandos in recent
theoretical developments. Sociological Theory, 6(2), 153-168.
Lareau, A. (2000). Home Advantage : Social Class and Parental Intervention in Elementary
Education. London New York: Falmer Press.
Lareau, A. (2001). Linking Bourdieu's concept of capital to the broader field. In B. J. Biddle
(Ed.), Social Class, Poverty, and Education: Policy and Practice (pp. 77-100). New York:
RoutledgeFalmer.
52

Lareau, A. (2003). Unequal Childhoods: Class, Race, and Family Life. Berkeley: University of
California Press.
Lareau, A., & Horvat, E. M. (1999). Moments of social inclusion and exclusion: Race, class, and
cultural capital in family-school relationships. Sociology of Education, 72(1), 37-53.
Lareau, A., & Weininger, E. B. (2004). Cultural capital in educational research. In D. L. Swartz
& V. L. Zolberg (Eds.), After Bourdieu: Influence, Critique, Elaboration (pp. 105-144).
Dordrecht, The Netherlands: Kluwer Academic Publishers.
Lauglo, J. (2000). Social capital trumping class and cultural capital? In S. Baron, J. Field & T.
Schuller (Eds.), Social Capital: Critical Perspectives (pp. 142-167). Oxford; New York: Oxford
University Press.
Leung, K. (1996). The role of beliefs in Chinese culture. In M. H. Bond (Ed.), Chinese
Psychology (pp. 247-262). Hong Kong: Oxford University Press.
Li, J. (2003). U.S. and Chinese cultural beliefs about learning. Journal of Educational
Psychology, 95(2), 258-267.
Lin, C.-Y. (2007). The impact of individual-family and school factors on students' academic
achievement: To analyze the educational equality and the relevant issues of junior high school
levels in terms of SEM: National Kaohsiuing Normal University.
Louie, V. (2001). Parent' aspirations and investment: The role of social class in the educational
experiences of 1.5- and second-generation Chinese Americans. Harvard Education Review,
71(3), 438-474.
Lucas, S. R. (2001). Effectively maintained inequality: Education transition, track mobility, and
social background effects. The American Journal of Sociology, 106(6), 1642-1690.
Ma, L.-C., & Smith, K. B. (1992). Social correlates of Confucian ethics in Taiwan. Journal of
Social Psychology, 132(5), 655-659.
McDonough, P. M., & Nunez, A.-M. (2007). Bourdieu's sociology of education: Identifying
persistent inequality, unmasking domination, and fighting social reproduction. In C. A. Torres &
A. Teodore (Eds.), Critique and Utopia (pp. 139-154). Lanham, Md.: Rowman & Littlefield.
McDonough, P. M., Ventresca, M., & Outcalt, M. (2000). Field of dreams: Organizational field
approaches to understanding the transformation of college access. Higher Education: Handbook
of Theory and Research, 15, 371-405.
Ministry of Education, R. O. C. (2005). Summary of Statistics.
Ministry of Education, R. O. C. (2009a). A Report on Low-income Families.
53

Ministry of Education, R. O. C. (2009b). Summary of Statistics.
Oldstone-Moore, J. (2002). Confucianism. New York: Oxford University Press.
Pearce, R. R. (2006). Effects of cultural and social structural factors on the achievement of White
and Chinese American students at school transition points American Educational Research
Journal 43(1), 75-101.
Phillips, M. (2011). Parenting, time use, and disparities in academic outcomes. In G. Duncan &
R. Murnane (Eds.), Whither Opportunity? (pp. 207-228). New York: Russell Sage Foundation.
Reardon, S. F. (2011). The widening academic achievement gap between the rich and the poor:
New evidence and possible explanations. In G. Duncan & R. Murnane (Eds.), Whither
Opportunity? (pp. 339-358). New York: Russell Sage Foundation.
Reay, D. (1998). Cultural reproduction: Mothers Involvement in their children's primary
schooling. In Bourdieu and Education (pp. 55-71). London: Falmer Press.
Robbins, D. (2005). The origins, early development and status of Bourdieu's concept of 'cultural
capital'. The British Journal of Sociology, 56(1), 13-30.
Robinson, R., & Garnier, M. (1985). Class reproduction among men and women in Frence:
Reproduction theory on its home ground. American Journal of Sociology, 91, 250-280.
Schneider, B., & Lee, Y. (1990). A model for academic success: The school and home
environment of East Asian students. Anthropology and Education Quarterly, 21(4), 358-377.
Shun, K.-L., & Wong, D. B. (2004). Confucian Ethics. Cambridge: Cambridge University Press.
Smith, D. C. (1981). An Island of Learning. Taipei, Taiwan: Pacific Cultural Foundation.
Stevenson, H. W., & Lee, S. Y. (1996). The academic achievement of Chinese students. In M. H.
Bond (Ed.), Chinese Psychology (pp. 124-142). Hong Kong: Oxford University Press.
Sullivan, A. (2001). Cultural capital and educational attainment. Sociology, 35(4), 893-912.
Swartz, D. (1997). Culture & Power: The Sociology of Pierre Bourdieu. Chicago: University of
Chicago Press.
Thornton, A., & Lin, H.-S. (1994). Social Change and the Family in Taiwan. Chicago:
University of Chicago Press.
Tsai, S.-L., & Shavit, Y. (2007). Taiwan: Higher Education - Expansion and Equality of
Educational Opportunity. In D. B. Grusky & P. England (Eds.), Stratification in Higher
Education. Stanford, CA: Stanford University Press.
54

Tweed, R. G., & Lehman, D. R. (2002). Learning considered within a cultural context. American
Psychology, 57(2), 89-99.
Taiwan PISA National Center. (2012). Rankings in PISA. Retrived May 31, 2012, from:
http://pisa.nutn.edu.tw/link_rank_en.htm
U.S. Census Bureau. (2001). Current Population Survey.
Wacquant, L. J. D. (1989). Towards a reflexive sociology: A workshop with Pierre Bourdieu.
Sociological Theory, 7(1), 26-63.
Werfhorst, H. G. V. D., Sullivan, A., & Cheung, S.-Y. (2003). Social class, ability and choice of
subject in secondary and tertiary education in Britain. British Educational Research Journal,
29(1), 41-62.
Wu, C.-C. (2009). Higher education expansion and low-income students in Taiwan.
International Journal of Educational Development, 29, 399-405.
Wu, D. Y. H. (1996). Chinese childhood socialization. In M. H. Bond (Ed.), Chinese Psychology
(pp. 143-154). Hong Kong: Oxford University Press.
Yang, Y.-R. (1994). Education and national development: The case of Taiwan. Chinese
Education and Society(November-December), 7-22.
Yates, J. F., & Lee, J. W. (1996). Chinese decision-making. In M. H. Bond (Ed.), Chinese
Psychology (pp. 338-351). Hong Kong: Oxford University Press.
Yu, A. B. (1996). Ultimate life concerns, self, and Chinese achievement motivation. In M. H.
Bond (Ed.), Chinese Psychology (pp. 227-246). Hong Kong: Oxford University Press.

55