The Trajectory of Cultural Taste: Influence of Intergenerational Educational and Class Mobility on Cultural Taste

(1)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₁₉₃

Journal of Research in Education Sciences 2017, 62(3), 193-223

doi:10.6209/JORIES.2017.62(3).07

The Trajectory of Cultural Taste: Influence of

Intergenerational Educational and Class

Mobility on Cultural Taste

Chih-Chia Chuang

_{Da-Sen Lin}

Department of Sociology,

National Dong Hwa University Department of Sociology, Fo Guang University

Abstract

Over the past two decades, Taiwan has undergone considerable industrial and social transformation. However, few studies have examined the relationship between cultural taste and social mobility from the perspective of social change. This study adopted Bourdieu’s perspectives on habitus, cultural taste, and social trajectory as the theoretical framework and used the 1997 and 2007 data from the Taiwan Social Change Survey to investigate the effects of changes in cultural taste influenced by intergenerational social mobility. The results show that an extensive variety of cultural taste permeated cultural consumption behaviors in both survey years, a finding that contradicts the logic of class distinction proposed by Bourdieu; cultural taste can be used to effectively distinguish groups according to educational level and social class. In addition, the results of this study support the argument that social mobility trajectories affect cultural taste variety. The trajectories of intergenerational educational and class mobility were the key factors shaping variety in cultural taste. People whose class had been elevated intergenerationally exhibited more diverse cultural tastes than did those without upward mobility. Moreover, the number of strata crossed through upward mobility is positively and significantly related to cultural taste variety; people who experienced intergenerational educational upward mobility and those who crossed more strata through upward mobility exhibited greater variety in cultural taste than did those whose educational level was identical to that of their fathers. However, notably, people who experienced intergenerational

Corresponding Author: Chih-Chia Chuang, E-mail: chihchia@gms.ndhu.edu.tw

(2)

194 The Trajectory of Cultural Taste Chih-Chia Chuang & Da-Sen Lin

downward educational mobility also exhibited greater variety in cultural taste than did those with no mobility. This observation deviates from observations among people who experienced downward mobility in class. Therefore, the trajectories of intergenerational educational and class mobility differ in their influences on cultural taste variety.

Keywords: cultural omnivores, cultural taste, habitus, intergenerational class mobility, intergenerational educational mobility

(3)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₁₉₅

Introduction

Since the late 1980s, numerous European and American studies have proposed revisions for Bourdieu’s single correspondence1 between class and cultural taste (Chan & Goldthorpe, 2007, 2010; DiMaggio, 1987; Erickson, 1996; Katz-Gerro, 2002; Peterson, 2002; Peterson & Kern, 1996; Peterson & Simkus, 1992; Rimmer, 2012; Sullivan & Katz-Gerro, 2007; van Eijck, 2001; van Eijck & Lievens, 2008). However, these studies have not provided further empirical evidence concerning the reason for the cultural taste of elites becoming omnivorous. No related empirical study in Taiwan has investigated the possible turn in cultural taste in Taiwanese society. Therefore, the present study investigated the change of cultural taste distinction in Taiwanese society. Bourdieu’s perspectives on habitus, cultural taste, and social trajectory were adopted as a theoretical framework for examining the influence of intergenerational social mobility on the cultural taste of the child generation.

In the past two decades, Taiwan has undergone considerable industrial and social transformation (Lin, T.-H., 2009, 2013; Lin & Hu, 2011). However, few studies have examined the relationship between cultural taste and social mobility from the context of social change. Previous studies have mostly investigated the formation of cultural taste and its validity from the perspective of family background (De Graaf, De Graaf, & Kraaykamp, 2000; DiMaggio, 1982; Lareau, 1989, 2011), overlooking the vital role of social mobility in analyzing cultural taste to distinguish class.

When social transformation brings about a change in social structure, the structural and individual opportunities for social mobility are enhanced. Therefore, this study addressed the following questions: What changes in cultural taste are caused by these trajectories of social mobility? Does high social mobility lead to individuals or groups having omnivorous cultural tastes, or is cultural taste more closely connected to a person’s present class position? Through exploring these questions, this article aims to investigate changes in cultural taste and the influence of social mobility on cultural taste in Taiwan.

1_{Bourdieu (1984) emphasized that a single connection exists between cultural taste and social class. In other words,}

people of higher social class prefer highbrow or elite culture, whereas those of lower social classes tend to enjoy popular or mass culture. Through daily practices such as leisure activities and art appreciation, class distinction manifests in taste difference between classes.

(4)

Literature Review

Homology and Omnivore Argument

Early studies in this field have discussed how cultural consumption became a critical social practice (e.g., Simmel, 1971; Veblen, 1931). However, a relatively complete theoretical and empirical foundation did not emerge until the seminal work Distinction: A Social Critique of the Judgement of Taste (Bourdieu, 1984) (hereafter referred to as “Distinction”). The homological argument in Distinction was derived from Bourdieu’s rethinking of Weber’s concepts of class and status groups (Bourdieu, 1984, p. xii). To escape the binary opposition between class subjectivism and objectivism in Marxist theory, Bourdieu adopted and reinterpreted two basic concepts from Weber’s theory of stratification (Wright, 1985): class and status group (Bourdieu, 1987, p. 7). Bourdieu considered that in daily life, class always manages to appear legitimately in a status group (Swartz, 1997, pp. 150-153; Weininger, 2005, p. 95). Consequently, under Bourdieu’s theoretical context, homologous associations are generated between class and status groups as well as between class and the cultural tastes of a status group.

Related empirical studies in Taiwan have mostly adopted the single correspondence between class and cultural taste as the theoretical framework for analysis. Several studies have investigated the difference in the cultural preferences of different social groups (Chiu, 1997; Huang, Chang, & Wang, 2010; Hwang, 2000); other studies have examined the validity of cultural capital in various fields (Chang, 2006, 2010; Hsieh & Chuang, 2016; Huang, Wu, & Yu, 2015; Lin & Wu, 2007). However, most studies have focused on the differences in cultural capital among various social groups and the effect of this difference on student academic achievements (Chang, 2011; Hwang & Wu, 2011; Lee & Yu, 2005; Lee & Hwang, 2004; Lin, P.-F., 2009; Su & Hwang, 2009; Sun & Hwang, 1996). However, no study conducted in Taiwan has investigated the influence of social mobility on cultural taste.

Ever since their detection in the U.S. in the 1980s, omnivores have received much attention in cultural sociology. DiMaggio (1987) and Wilensky (1964) have described the phenomenon of omnivorous cultural taste in Western society. Furthermore, Peterson (1992), Peterson and Kern (1996), and Peterson and Simkus (1992) have explicitly identified that the cultural taste of elites in Western society has become increasingly omnivorous. Peterson and Simkus showed that in American society, people with a higher vocational status enjoy not only fine musical genres, but also popular music. Peterson and Kern compared the changes in cultural tastes in American society from

(5)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₁₉₇

1982 to 1992 and observed a decrease in the number of cultural highbrows who merely preferred fine culture and despised popular culture, showing a trend of American high society becoming cultural omnivores.

To date, only a few studies in Taiwan have incorporated the omnivore concept into their analysis (Lee & Hwang, 2004; Su & Hwang, 2009). These studies have consistently shown that higher cultural variety is inversely related to academic performance. Furthermore, the negative effects of cultural variety were considered to be caused by an exclusion effect of time. None of these studies have explored the class-distinguishing implications of cultural variety. Therefore, when investigating the relationship between social mobility and cultural taste, we first examined whether cultural tastes in Taiwanese society developed towards an omnivorous pattern as they have done in Western societies.

Social Mobility and Cultural Taste Profiles

Regarding the class-habitus relationship, Bourdieu analyzed all possible changes and tendencies that individuals or groups are likely to experience in the habitus structure under the effects of various forces in all social spaces (Bourdieu, 1990, p. 45). This discourse served as a theoretical framework for this article in investigating the relationship between social mobility and cultural taste. Bourdieu asserted that the habitus structure a person derives from their primal life experiences constitutes the basis and core of the entire habitus. In other words, family background exerts a crucial influence on habitus. Although individuals may be influenced by power struggles and resource distribution in various spaces later in life, the habitus structure established in their childhood is persistent and tenaciously resistant.

Nevertheless, the habitus of individuals differs according to the specific social life they experience. Bourdieu (1990, pp. 101-102) indicated that “the distinguishing principle of individual habitus lies in the specificity of social trajectory. These social trajectories correspond to the sequence of chronicle determinants. Furthermore, these social trajectories are not mutually reducible.” However, individual life experiences differ by era and environment among people from different classes or even the same class. Individual habitus displays a unique mark and style for individuals, even for those coming from the same class. Bourdieu’s concept of social trajectory depicts the class trajectory developed through social mobility. Therefore, the outcome of personal struggles in a field is also reflected in the intergenerational changes of socioeconomic status and cultural practices.

To date, a vast number of studies demonstrate that cultural practices of socially mobile individuals are predominantly shaped by childhood socialization or they are more in line with the

(6)

socialization context the individuals eventually end up in. Psychological studies find musical preferences to be very stable across time (Delsing, Ter Bogt, Engels, & Meeus, 2008; Rentfrow & Gosling, 2003).2 These studies consider musical preferences as a stable personality characteristic and, hence, socially mobile individuals stick to the cultural profile characteristic of their context of origin (Ter Bogt, Delsing, van Zalk, Christenson, & Meeus, 2011). However, some authors question the lasting influence of socialization during early childhood (Daenekindt & Roose, 2013; Erickson, 1996; Lahire, 2008). Lahire argues that individuals are multi-socialized and new situations may force individuals to relinquish schemes adopted during early childhood and adapt to new social situations. Empirical research has found evidence for the arguments, as esthetic dispositions (Daenekindt & Roose, 2013) and cultural practices (Daenekindt & Roose, 2013) of socially mobile individuals are influenced by the social position of destination. In this article, then, we focus on the association of social mobility and cultural taste profiles. If we find an association between the heterogeneity of socializing experience and practices related to cultural activities, it is very likely that this heterogeneity will also cause dissonance between dispositions, between social contexts, and all possible combinations between them.

Research Hypotheses

The theoretical evolution from homological to omnivore arguments shows that the relationship between cultural taste and social stratification in Western societies changes as society changes. As Taiwan transformed from an industrial society to a service-oriented society, the producer society gradually transformed into a consumer society. Shifts in social structure generated considerable social mobility, thereby altering the links between cultural consumption patterns, education, and class. Similar to Western societies, high social mobility was generated in Taiwanese society as a result of social structural transformation. In line with this reasoning, we expect the change in cultural

2_{In the context of globalization, the freedom and diversity of media communication have shaped personal cultural}

tastes. Many scholars have begun to question the validity of using social class to explain cultural tastes in contemporary society. Against this background, the individualization argument was formed. Beck (1992, pp. 139, 158) posited that in a highly industrialized society, individuals and society are no longer mediated by class or status, but rather via new and direct links. Giddens (1991) also postulated that lifestyle and social status are not necessarily correlated. Lifestyle can be a product of objective environmental conditions or individuals’ subjective choices. Warde (1997, p. 8) argued that this challenge is a drastic shift from habitus to freedom. Therefore, in modern society, cultural tastes may be gradually deviating from the foundation of their original connection to social class, transforming into more of an individual self-realization. To avoid losing focus on the analysis, coupled with the fact that no variables from the research data were found fit enough to analyze the influence of the freedom of media communication on cultural tastes, no research hypotheses were tested.

(7)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₁₉₉

taste observed in Western societies to have also occurred in Taiwanese society. Regarding the relationship between class and cultural taste, the following hypotheses are proposed:

H1: Class properties of the child generation are the main influencing factors of cultural taste.

H2: High-status groups exhibit a notable characteristic of cultural omnivorism in their cultural tastes.

Social transformation brings opportunities for social mobility. Various experiences of social mobilization further develop the specific social trajectory of individuals and social groups, thereby generating distinctive habitus and influencing daily life practices (Bourdieu, 1984, p. 123). Consequently, even for members of the same class, their individual habitus exhibits specific styles because of differences in their experiences according to the era and environment they grew up in. Such styles are externalized as distinguishing cultural consumption genres. Therefore, the experience trajectory, both education and class-generated by social mobility, is reflected in their preferences for certain cultural tastes.

Because the structural upward mobility in education generated by education expansion does not necessarily correspond to class mobility, the association between educational mobility and class mobility cannot be reduced to a simple causal relation. According to Bourdieu’s discussion on the education field, habitus, and cultural capital, the present study considers that education exerts a crucial effect on cultural taste development. Therefore, the cultural tastes of the child generation are prone to be influenced by their educational attainment, whereas the effect of the paternal generation’s educational attainment is limited. This article adopts intergenerational education and intergenerational class mobility as indicators of social mobility to examine the influence of social mobility on cultural taste. Related research hypotheses are proposed as follows:

H3: Education exerts a significant influence on individual cultural taste, and the trajectory of intergenerational educational mobility lead to differences in the variety of cultural taste.

H4: Class exerts a significant influence on individual cultural taste, and the trajectory of intergenerational class mobility leads to differences in the variety of cultural taste. H5: The more strata the child generation crosses through intergenerational class mobility,

(8)

Methods

Data Source

We used date from two waves of the Taiwan Social Change Surveys “Social Stratification” module conducted in 1997 (Chiu, 1998) and 2007 (Chang & Liao, 2008) respectively.3 Both data sources were national sampling surveys that covered items on lifestyle such as music appreciation and leisure activity participation. The data sources also include demographic statistical data on career, education, income, subjectively perceived status, and paternal educational attainment and class. This study analyzes the items shared by these two surveys.

Measures

The dependent variables in this study comprise the constructs that are built according to the number of respondents and the frequency with which they participate in 13 cultural activities. The independent variables comprised demographic statistical variables (i.e., sex and age), family background (province of origin, father’s class, and father’s educational attainment), stratification properties (educational attainment, income, and subjectively perceived status), and social mobility. The father’s class and educational attainment were matched with those of the child generation, thus constructing a variable of intergenerational social mobility. Monthly income was adopted as an indicator of income. Subjectively perceived status refers to the status at which respondents place themselves in society. The class classification framework proposed by Erikson and Goldthorpe (1992) was employed to identify the objective classes of the paternal and child generations. The dependent and independent variables are detailed as follows.

Dependent Variables: Constructs of Cultural Taste

The dependent variables were derived from the number of respondents and the frequency with which they participate in 13 cultural activities. Participation in cultural activities comprises four

3_{It is difficult to find data that are fit for use in examining the correlation among changes in cultural tastes and social}

mobility in Taiwanese society. To discuss whether there have been changes in cultural tastes, it was necessary for the present study to consider at least two sets of survey data (the time interval needed to not be too short, while the questions related to cultural consumption in the questionnaire needed to be considerably consistent). Data provided by the Taiwan Social Change Survey: Social Stratification are more appropriate and fit to conduct the survey in the current study. The Social Stratification Questionnaire was conducted in 1992, 1997, 2002, 2007, and 2012, for a total of five sets of data. Among the questionnaires, data collected in 1997, 2002, and 2007 are comprised of more details about cultural consumption. Hence, the current study selected the two data sets of 1997 and 2007 (with an interval of ten years in between) for data analysis.

(9)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₀₁

levels: often, sometimes, seldom, and never. By adopting categorical principal component analysis (CatPCA),4 this study extracted latent structural aspects underlying the number of respondents and frequency with which they participated in cultural activities. In total, three factors were extracted: Variety, Popularity, and Traditionalism (Table 2). However, the total variance explained in Factor 1 (Variety) was 62.9% (38.923/61.872) for the 1997 data and 60.2% (36.438/60.497) for the 2007 data, showing that Variety is the main characteristic explaining cultural taste. Therefore, in a multivariate analysis, this study adopted Variety as a dependent variable in examining the association between social mobility and cultural taste to achieve conciseness in interpreting the analysis. The scores attained by the respondents on Variety were obtained from the scores they attained through the regular expression of the construct.

Demographic Statistical Variables

1. Sex: In the multivariate analysis, male and female sexes were employed as dummy variables, with female as the control group.

2. Age: Respondent age was categorized into five groups: ≤ 29, 30-39, 40-49, 50-59, and ≥ 60 years. Respondents younger than 29 years were the control group.

Family Background Variables

1. Province of origin: The respondents were grouped into Hokkienese Taiwanese, Hakka, mainland Chinese, and indigenous people according to their father’s province of origin, with Hokkienese Taiwanese serving as the control group.

2. Father’s educational attainment: Father’s educational attainment is classified into five categories. Below primary school refers to uneducated, self-study, illiterate, and primary school level education. Junior high school comprises compulsory and noncompulsory junior high schools and junior vocational schools. Senior high school consists of academic and vocational senior high schools, senior vocational schools, and noncommissioned officer schools. Junior college or above comprises 2-, 3-, and 5-year colleges, military and police special programs, colleges and academies,

4_{CatPCA differs from PCA in that before data reduction, CatPCA first converts ordinal or nominal data to interval}

data through optional scaling. Subsequently, PCA is conducted. However, PCA involves a basic assumption that the data are measured on an interval scale. Surveys on social science often cannot satisfy this assumption. The mathematical logic of optional scaling involves designating a value to each item of the categorical variables, and further analyzing the variations in the influence that the quantified items have on the dependent variables. In addition, under the premise that the relationships among the categorical variables are linear after the conversion, an alternating least squares iterative estimation is adopted to conduct nonlinear repeated iteration operations on the data to determine the optimal quantification of each item in the original categorical variables. In a corresponding model, the optimal quantifications replace the original categorical variables in subsequent statistical operations. Thus, ordinal and nominal data can be used in CatPCA (Meulman, van der Kooij, & Heiser, 2004).

(10)

open colleges and open universities, technical colleges and universities, and bachelor, Master’s, and Ph.D. programs. These categories were operationalized as dummy variables in the multivariate analysis, with below primary school educational attainment serving as the control group.

3. Father’s objective class: This study simplified the class classification framework proposed by Erikson and Goldthorpe (1992) into six types: service class (general professionals, managers, high-level managers and technicians, and office supervisory staff); routine non-manual workers (administrative and business staff, and salespersons in routine administrative and business roles); petty bourgeoisie (self-employed workers in small industries or general store owners); farmers and agricultural laborers (agriculture, forestry, fishery, and animal husbandry practitioners); skilled workers (technicians, site supervisors, and other skilled workers); and non-skilled workers (physical laborers, machine operators, and plant operators). These classes were handled as dummy variables in the multivariate analysis, with the farmers and agricultural labors serving as the control group.

Class Properties Variables of the Child Generation

1. Educational attainment: The categories for this variable are identical to those used in the father’s educational attainment variable.

2. Income: The respondents’ monthly income was divided into four groups: ≤ NT$20,000 (control group), NT$20,000-40,000, NT$40,000-NT$60,000, and ≥ NT$60,000.

3. Subjectively perceived status: This variable adopts the score from an item in the questionnaire, which asks, “On a scale of 1 to 10, how would you rate your social status?” (higher scores indicate higher perceived status). Social status was grouped into three types according to the respondents’ scores: 1-3, 4-6, and 7-10 points.

4. Objective class: The types for this variable are identical to those used in the father’s objective class variable.

Social Mobility Variables

This study adopted intergenerational mobility in education and class as indicators of social mobility. Three classification methods were used in this study to measure social mobility. In the first method, the “non-high-status groups” in the child generation were regarded as one group, whereas the high-status groups in the child generation were further categorized into “father high-child high,” “father middle-child high,” and “father low-child high” according to the class or educational attainment of the paternal generation. Among the intergenerational educational mobility variables, “non-high-status groups” in the child generation refers to respondents whose educational attainment is below junior high school or high school. “High-high” indicates that the educational attainment of

(11)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₀₃

both the respondents and their fathers is junior college or above. “Middle-high” means that the educational attainment of the fathers is senior high school and that of the respondents is junior college or above. “Low-high” means that the educational attainment of the father is below junior high school and that of the respondents is junior college or above. Regarding the intergenerational class mobility, “non-high-status groups in the child generation” refers to respondents who are petty bourgeoisie, skilled workers, non-skilled workers, or farmers and agricultural labors. “High-high” shows that the class of both the respondents and their fathers are service class or routine non-manual workers. “High-middle” indicates that the fathers are petty bourgeoisie or skilled workers, whereas the respondents belong to the service class or are routine non-manual workers. Finally, “low-middle” means that the fathers are non-skilled workers or farmers and agricultural labors, whereas the respondents belong to the service class or are routine non-manual workers. This method of measurement facilitates clarifying whether within-groups cultural taste variations were generated by intergenerational social mobility among the high-status child generation. If the variety of cultural taste is positively related to the number of strata that the child generation has crossed through upward social mobility and the variety is a valid indicator for the high-status groups to distinguish themselves from other groups, then the variety of cultural taste observed in the “father low and child high” group is expected to be significantly higher than that of other groups.

If the variety in cultural taste is assumed to be mainly influenced by the number of strata crossed through upward or downward social mobility, then the variety of cultural taste is enhanced as the number of strata crossed increases. Therefore, the second method involved dividing the social mobility variable into “no mobility,” “one stratum across,” and “two strata across.” The third measuring method was adopted to distinguish the social mobility variable into “downward mobility,” “no mobility,” “one stratum upward,” and “two strata upward.” This method enabled identifying the influence of the number of strata crossed and the orientation of social mobility (i.e., upward or downward) on cultural taste. These three methods were applied to the intergenerational educational and class mobility variables in this study.

Results

Properties of 1997 and 2007 Samples

Considerable variation in intergenerational class structure occurred over the study period, including that a greater proportion of the child generation belonged to the service class compared to the paternal generation (Table 1). Additionally, a significantly smaller proportion of the child

(12)

Table 1

Descriptive Statistics of Selected Variables

Survey Year 1997 2007 N Percentage (%) N Percentage (%) Pearson X2 Male 1,290 49.7 1,005 49.3 Gender Female 1,306 50.3 1,035 50.7 0.083*** ≤ 29 465 17.9 434 21.3 30-39 786 30.3 373 18.3 40-49 748 28.8 433 21.2 50-59 377 14.5 354 17.4 Population Statristic Age ≥ 60 220 8.5 446 21.9 246.535*** Hokkien 2,023 78.0 1,510 74.3 Hakka 282 10.9 272 13.4 Mainlander 241 9.3 217 10.7 Ethnicity Aborigines 46 1.8 33 1.6 10.399*** Service Class 211 8.5 255 13.5 Routine Non-manual Worker 93 3.8 73 3.9 Petty Bourgeoisie 555 22.5 434 23.0 Farmer and Agricultural 980 39.7 640 33.9 Skilled Worker 355 14.4 282 14.9 Father’s Class

Non-skilled Worker 274 11.1 205 10.9

34.701***

Elementary School Below 1,928 77.3 1,278 65.7 Junior High School 240 9.6 236 12.1

High School 197 7.9 258 13.3

Family Background

Father’s Education

Junior College or Above 130 5.2 173 8.9

79.183*** Low 413 16.4 532 26.8 Middle 1,556 61.7 1,253 63.1 Subjective Status High 551 21.9 202 10.2 148.467*** Service Class 402 20.7 374 28.5 Routine Non-manual Worker 245 12.6 173 13.2 Petty Bourgeoisie 408 21.0 219 16.7 Farmer and Agricultural 184 9.5 82 6.2 Skilled Worker 365 18.8 243 18.5 Progeny Stratification Class Non-skilled Worker 339 17.4 222 16.9 37.898*** (continued)

(13)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₀₅

Table 1

Descriptive Statistics of Selected Variables (continued)

Survey Year 1997 2007 N Percentage (%) N Percentage (%) Pearson X2

Elementary School Below 818 31.5 507 24.9 Junior High School 407 15.7 284 13.9

High School 770 29.7 569 27.9

Education

Junior College or Above 600 23.1 678 33.3

63.781*** 2 Million or Below 435 23.8 604 38.3 2-4 Million 771 42.1 593 37.6 4-6 Million 402 22.0 225 14.3 Wage 6 Million or Above 222 12.1 156 9.9 94.089***

Father High-child High 147 8.0 144 11.7 Father Middle-child High 273 14.8 231 18.8 Father Low-child High 184 10.0 130 10.6 Class

Mobility (1)

Child Non-high 1,240 67.2 722 58.8

26.693***

Father High-child High 92 3.7 130 6.7 Father Middle-child High 111 4.5 181 9.3 Father Low-child High 388 15.6 344 17.7 Education

Mobility (1)

Child Non-high 1,903 76.3 1,288 66.3

77.225***

Non Mobility 807 43.8 526 42.9 One Stratum Across 828 44.9 551 44.9 Class

Mobility (2)

Two Strata Across 209 11.3 150 12.2

0.636***

Non Mobility 1,285 51.5 906 46.6 One Stratum Across 813 32.6 684 35.2 Education

Mobility (2)

Two Strata Across 396 15.9 353 18.2

10.887***

Non Mobility 807 43.8 526 42.9 One Stratum Upward 629 34.1 415 33.8 Two Strata Upward 184 10.0 130 10.6 Class

Mobility (3)

Downward Mobility 224 12.1 156 12.7

0.619***

Non Mobility 1,285 51.5 906 46.6 One Stratum Upward 764 30.6 635 32.7 Two Strata Upward 388 15.6 344 17.7 Social Mobility Education Mobility (3) Downward Mobility 57 2.3 58 3.0 11.866*** **p < .01. ***p < .001.

(14)

generation worked as farmers and agricultural laborers. The results show that Taiwanese society experienced a transformation in its industrial structure. The results for Variable 1 of intergenerational class mobility indicate that the upward mobility of the child generation increased over the study period, and that the intergenerational replication in the upper class was also enhanced. The results for Variable 3 show that intergenerational class mobility was frequent. Variation in the intergenerational class proportion reflects the social structural change that Taiwan underwent after transforming into a mature industrialized society in the 1990s, and shows the influence of structural change on intergenerational social mobility. This study considered that intergenerational social mobility not only influenced the class identification of the child generation, but also generated considerable variety in the cultural tastes of the child generation.

In the study period, variations in the educational attainment of the paternal and child generations were considerable. Compared with the results derived from the 1997 data, the educational attainment of the paternal generation in 2007 was generally higher, with the proportion of junior and senior high schools and junior college or above increasing significantly. Furthermore, the educational attainment of the child generation was notably higher than that of the paternal generation for the study period. The results for Variable 1 of educational mobility show an increase in the proportion of the child generation who moved upward to the top level in the study period. Moreover, the results for Variable 3 of educational mobility indicate considerable intergenerational educational mobility. Therefore, one of the main research objectives of this study was to examine the influence of intergenerational educational mobility on the cultural tastes of the child generation.

The relationships among the subjectively perceived status, objective class, and cultural taste of the child generation have vital theoretical implications (Chiang, 2001; Giddens, 1973; Lin & Hwang, 2008; Marsh & Hsu, 1994). In the 1997 data, more members of the child generation considered themselves as having a high status rather than a low one. However, the results became inverse over the study period. The results illustrate that the variation in upward mobility in the objective class structure of the child generation did not correspond to the enhancement in the subjectively perceived status. The incoherence between the objective class and subjective status in the child generation was more marked in the 2007 data than in the 1997 data. Therefore, if a strong connection exists between perceived status and cultural taste, then the associations between cultural taste and social stratification properties found in the 1997 and 2007 data should generate different results.

Properties of and Distinction Among Cultural Tastes

(15)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₀₇

factors with eigenvalues higher than 1. Table 2 indicates that the cultural activities observed in the 1997 and 2007 data can be illustrated by three main factors. In the 1997 and 2007 data, the factor loadings of Dimension 1 in the 13 cultural activities exhibit relatively high positive values, showing that Dimension 1 can be viewed as a comprehensive behavioral indicator of the agents’ consumption of various cultural activities. Therefore, this study derived Dimension 1 from the 1997 and 2007 data as the cultural variety construct. This construct shows that the respondents’ consumption behaviors in cultural activities were characteristically omnivorous. However, this construct also reveals no significant change in the variety of cultural consumption behaviors in the study period.

Table 2

Pattern Matrix of CatPCA for the 1997 and 2007 Cultural Activities

1997 2007

DIM 1 DIM 2 DIM 3 DIM 1 DIM 2 DIM 3

(Diversity) (Popular) (Traditional) (Diversity) (Traditional) (Popular) Mandarin pop 00.703 00.388 ---.249 00.687 ---.262 0.414 Taiwanese pop 00.400 00.766 ---.064 00.410 00.143 0.723 Western pop 00.780 -0-.033 ---.133 00.758 ---.293 -.089 Japanese pop 00.556 00.299 00.140 00.652 ---.034 0.069 Taiwanese Opera, Puppetry 00.256 00.275 00.761 －－－ Taiwanese Opera －－－ 00.349 00.762 0.138 Puppetry －－－ 00.420 00.654 0.221 Chinese Music, Peking Opera 00.544 －.177 00.586 －－－ Chinese Music －－－ 00.618 00.387 -.357 Peking Opera －－－ 00.510 00.514 -.310 Classical Music, Opera 00.705 ---.307 00.214 00.730 ---.031 -.301 Karaoke and KTV 00.648 00.170 ---.230 00.559 ---.281 0.288 Novel and Book 00.753 ---.252 ---.195 00.703 ---.325 -.169

Gallery 00.709 ---.418 ---.046 00.702 ---.224 -.230 Outdoor Activity 00.599 ---.169 ---.181 00.568 ---.195 0.034 Eigenvalues 04.282 01.326 01.199 04.737 01.873 1.255 Variance Explained 38.923 12.052 10.896 36.438 14.407 9.652 Total Variance 61.872 60.497 Explained (%)

(16)

The factor loadings on Dimension 2 ranged from positive to negative values in the descending order of Taiwanese pop songs, Mandarin pop songs, Japanese pop songs, Taiwanese opera, puppetry, and other cultural activities. Dimension 2 presents a spectrum of cultural properties from popular culture to fine cultures. Therefore, Dimension 2 in 1997 was constructed as the popularity construct. However, in 2007, Dimension 2, which exhibits the second highest explained variance, presents a spectrum of cultural properties from traditionalism to modernity. This dimension developed into the Traditionalism construct. From 1997 to 2007, the factor that exhibited the second-highest explained variance changed from Popularity to Traditionalism.

The third factor derived from the 2007 data also differs from that derived from the 1997 data. The factor loadings from cultural activities in the 2007 data were in the descending order of Taiwanese opera, Taiwanese puppetry, traditional Chinese music, Peking opera, classic music, and opera, whereas the cultural activities in the 2007 data were in the order of Taiwanese pop songs, Mandarin pop songs, Karaoke and KTV, and other cultural activities. Consequently, Dimension 3 in the 1997 data present a spectrum of cultural properties ranging from traditionalism to modernism, whereas that in the 2007 data ranges from the majority of popular culture to the minority of fine culture. Therefore, the traditionalism construct was developed from Dimension 3 in 1997, whereas the popularity construct was built from Dimension 3 in 2007.

These three factors and their explained variance illustrate the behavioral properties of cultural activity participation and the distinguishing effect of cultural taste on social groups. In Taiwan, the consumption of cultural activity is mainly manifested in the variety of cultural taste. In the 1997 and 2007 data, the cultural variety factor exhibits the highest explained variance, which is far greater than that of the other two factors, clearly indicating the omnivorous property of cultural taste and that no obvious change occurred over the study period. Furthermore, adopting Traditionalism and Popularity as factors facilitates delineating the properties of cultural activities, such as the spectrum of cultural consumption behaviors covering mass popular activities and minority fine cultural activities, and these cultural activities differed in their properties ranging from traditionalism to modernism. Subsequently, a comparison between these two time points regarding the factors with the second highest and lowest explained variance shows that for the 1997 data, using the popularity-fine culture construct to distinguish social groups generated stronger validity than did adopting the traditionalism-modernism construct. However, for the 2007 data, the traditionalism-modernism construct exerted a more significant distinguishing effect. This result illustrates that in the study period, the validity of adopting the popularity-fine culture construct of cultural taste as a distinguishing indicator decreased and was replaced by taste differences that manifested in the traditionalism-modernism construct.

(17)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₀₉

Associations Among Cultural Taste, Education, and Class

Participation in cultural activities was characterized by Variety, Traditionalism, and Popularity. According to the total variance explained (Table 2), Variety was identified as the main factor. For conciseness in explaining the patterns, Variety was adopted as a dependent variable in the multivariate model for investigating the relationship between cultural taste and social mobility. The multivariate analysis results in Table 3 show the influential factors in cultural taste variety and the effect of intergenerational mobility on cultural taste. Furthermore, changes in cultural taste were also examined by comparing the results derived from the 1997 and 2007 data.

Table 3

Multiple Regression Analysis Table of the Variety Constructs in the 1997 and 2007 Data

Survey year 1997 2007 Gender (Femalea) (.001)*

*

(-.001)*

*

(.003)*

*

(.001)*

*

(.001)*

*

(.000)*

*

Male -.005*** .009*** -.033*** .010*** .015*** -.006*** Age (≤ 29a) (.648)*

*

(.559)*

*

(.150)*

*

(.524)*

*

(.459)*

*

(.080)*

*

30-39 -.070*** -.085*** -.040*** -.024*** -.035*** -.040*** 40-49 -.229*** -.245*** -.099*** .000*** -.007*** .028*** 50-59 -.326*** -.323*** -.129*** -.121*** -.119*** -.001*** Population Statistic ≥ 60 -.351*** -.344*** -.136*** -.351*** -.340*** -.089*** Ethnicity (Hokkiena) (.025)*

*

(.023)*

*

(-.001)*

*

(.008)*

*

(.007)*

*

(-.009)*

*

Hakka .057*** .051*** .020*** .013*** .005*** .000*** Mainlander .035*** .039** -.007*** .008*** .010*** -.033*** Aborigines .019*** .024*** .042*** -.025*** -.019*** .011***

Father’s Objective Class (Farmer and Agricultural Labora)

(.209)*

*

(.174)*

*

(.032)*

*

(.217)*

*

(.187)*

*

(-.042)*

*

Service Class .136*** .134*** .034*** .140*** .126*** .020***

Routine Non-manual Worker .092*** .077*** .023*** .053*** .053*** -.003***

Petty Bourgeoisie .151*** .141*** .036** .155*** .149*** .055**

Skilled Worker .072*** .065*** .023*** .122*** .118*** .065***

Non-skilled Worker .028*** .026*** .001*** .081*** .084*** .047**

Father’s Education

(Elementary School or Belowa)

(.118)*

*

(.087)*

*

(.016)*

*

(.250)*

*

(.210)*

*

(.101)*

*

Junior High School .084*** .072*** .029*** .079*** .071*** .047**

High School .071*** .065*** .019*** .157*** .144*** .095*** Family

Background

Junior College or Above .059*** .040** -.003*** .135*** .122*** .083*** (continued)

(18)

Table 3

Multiple Regression Analysis Table of the Variety Constructs in the 1997 and 2007 Data (continued)

Survey year 1997 2007

Education (Elementary School or Belowa)

(.500)*

*

(.461)*

*

Junior High School .175*** .162***

High School .359*** .302***

Junior College or Above .367*** .409***

Wage (2 Million or Belowa) (.038)*

*

(.063)*

*

2-4 Million .032*** .068***

4-6 Million .069*** .070***

6 Million or Above .045** .058***

Subjective Status (Lowa) (.158)*

*

(.047)*

*

(.135)*

*

(.054)*

*

Middle .166*** .079*** .141*** .069***

High .259*** .112*** .168*** .101***

Objective Status (Farmer and Agricultural Labora)

(.215)*

*

(.209)*

*

Service Class .243*** .279***

Routine Non-manual Worker .191*** .210***

Petty Bourgeoisie .131*** .136*** Skilled Worker .122*** .153*** Progeny Stratification Non-skilled Worker .066*** .060*** R2_.274_.302_***_.450_***_.289 _.315_***_.437_*** N 1,650.274 1,650.302*** 1,650.450*** 1,127.2891,127315*** 1,127.437***

Note. “a” as reference. Importance coefficients are in parentheses.

_{p < .1. *p < .05. **p < .01. ***p < .001.}

The 1997 model shows that the importance of family background factors decreased when the social stratification properties of the child generation were controlled. The educational attainment of the child generation was the most crucial factor influencing cultural taste variety (importance coefficient = .5), followed by objective class.5 The results show that cultural taste variety is mainly related to the education and objective class of an individual. Therefore, groups of educational attainment and objective class position can be distinguished according to their cultural taste variety. This result explains that the major property of cultural activity participation in the 1997 data was the

5_{Likewise, Chan and Goldthorpe (2007) found that the relationship between the genres of musical consumption and}

(19)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₁₁

significant difference in between-class cultural taste variety; specifically, the elite class displayed significantly higher variety than did the non-elite classes.

In the 2007 data, the educational attainment and objective class of the child generation remained the two key factors influencing cultural taste variety. However, the influence of the father’s educational attainment was significantly enhanced in the 2007 data (importance coefficient = .101) relative to the 1997 data (importance coefficient = .016). Respondents whose father’s educational attainment was senior high school or junior college or above exhibited greater variety than did those whose father’s educational attainment was below primary school. Moreover, the importance of the father’s objective class increased slightly over the study period. These results indicate that the family background factors had a stronger influence on the cultural variety of the child generation over time. In addition, in the study period, the elite class displayed their distinct lifestyles by participating in greater diversity of cultural activities. According to the aforementioned research findings, H1 and H2 are supported.

Influences of Intergenerational Social Mobility on Cultural Taste

Model 1 in Table 4 shows that when the demographic properties and family background variables were controlled, the cultural variety of the child generation in the high-status groups was significantly higher than that of the non-high-status groups. This phenomenon can be observed in both 1997 and 2007. However, does the variety in the cultural tastes of the high-status child generation differ according to the educational attainment or class of their fathers? In this study, no significant difference was observed in the 1997 data. However, in the 2007 data, members of the high-status child generation, whose father was from a lower class, exhibited significantly greater variety in their cultural tastes when compared with the members of the child generation who inherited their fathers’ higher class (see Appendix). In addition, among the groups in the high-status child generation, differences in their trajectories of intergenerational educational mobility did not reflect a significant difference in the variety of their cultural tastes (see Appendix).

Does the number of the strata crossed through intergenerational mobility influence the variety in cultural tastes in the child generation? Model 2 in Table 4 reveals that in the 1997 and 2007 data, people who crossed strata through intergenerational class mobility exhibited more diversified cultural tastes than those whose class was a replica of their fathers’. Moreover, the variety in cultural taste for people who crossed two strata through intergenerational class mobility was significantly higher than that of people who crossed only one stratum. Therefore, H5 is supported. However, people who crossed one or no stratum exhibited a less significant difference in the variety

(20)

Table 4

Multi-regression Model

Model 1 Model 2 Model 3

1997 2007 1997 2007 1997 2007 Gender (Femalea) Male -.005*** .010*** -.063*** -.026*** -.063*** -.019*** Age (≤ 29a) 30-39 -.100*** -.089*** -.097*** -.154*** -.096*** -.149*** 40-49 -.257*** -.055** -.227*** -.155*** -.225*** -.157*** 50-59 -.311*** -.148*** -.269*** -.226*** -.268*** -.227*** Population Statistic ≥ 60 -.313*** -.349*** -.271*** -.411*** -.270*** -.415*** Ethnicities (Hokkiena) Hakka -.041*** -.008*** .026*** -.007*** .025*** -.008*** Mainlander -.035** .024*** .070*** .062*** .074*** .062*** Family Background Aborigines -.023*** -.019*** .030*** -.024*** .031*** -.028***

Subjective Status (Lowa)

Middle .111*** .109*** .106*** .126*** .104*** .126*** High .163*** .111*** .175*** .141*** .172*** .143***

Wage (2 Million or Belowa)

2-4 Million .085*** .096*** .087*** .088*** .086*** .085*** 4-6 Million .135*** .109*** .161*** .135*** .159*** .126*** Progeny Stratification 6 Million or Above .124*** .112*** .146*** .151*** .144*** .146*** (1) Class Mobility (Child Non-higha)

Father High-child High .106*** .098*** Father Middle-child High .158*** .129*** Father Low-child High .151*** .172*** (1) Education Mobility

(Child Non-higha)

Father High-child High .041** .116*** Father Middle-child High .076*** .121*** Social

Mobility (1)

Father Low-child High .105*** .133***

(21)

Table 4

Multi-regression Model (continued)

Model 1 Model 2 Model 3

1997 2007 1997 2007 1997 2007 (2) Class Mobility

(No Mobilitya)

One Stratum Across .041*** .047**

Two Strata Across .084*** .118*** (2) Education Mobility

(No Mobilitya)

One Stratum Across .229*** .134*** Social

Mobility (2)

Two Strata Across .241*** .154*** (3) Class Mobility

(No Mobilitya)

One Stratum Upward .037** -.075***

Two Strata Upward .089*** .131***

Downward Mobility .030*** -.020***

(3) Education Mobility (Child Non-higha)

One Stratum Upward .230*** .121***

Two Strata Upward .239*** .143***

Social Mobility (3)

Downward Mobility .059*** .084***

R2_.371_*** _.377_***_.377_***_.339_***_.379_***_.346_***

Note. “a” as reference.

_{p < .1. *p < .05. **p < .01. ***p < .001.}

of their cultural tastes in the 2007 data compared with the 1997 data. Moreover, the cultural taste variety of respondents who crossed strata through intergenerational educational mobility maintained significantly higher variety than did those whose class was a replica of their fathers’. Furthermore, the more strata the child generation crossed, the more diversified their cultural tastes became. Further integration of the effects of the social mobility variables in Models 1 and 2 (Table 4) shows that the trajectories of intergenerational class and educational mobility are reflected in cultural taste variety. However, no significant difference in within-groups cultural taste variety was observed in the child generation groups at higher levels of class or education.6 Therefore, the research results

6_{Except in the 2007 data, the “father low-child high” intergenerational class mobility group exhibited significantly}

higher variety in cultural tastes compared with the “father high-child high” group; no significant difference in Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₁₃

(22)

show that H3 and H4 are only partially supported.

To further clarify the influence of social mobility trajectories on cultural taste variety, social mobility was classified into upward and downward mobility. Model 3 (Table 4) shows that no significant difference was observed between the cultural taste variety of people who experienced downward class mobility and those with no class mobility. By contrast, people who crossed two strata through upward mobility show the greatest variety in cultural tastes, followed by those who crossed one stratum upward. The variety difference between people who crossed one or no stratum through class mobility is more significant in the 2007 data than in the 1997 data. However, the trajectories of educational mobility generated different effects than those caused by the trajectories of class mobility. Specifically, the cultural taste variety observed in the members of the child generation that experienced upward or downward mobility was significantly higher than that of the members with no mobility. The effect of intergenerational educational mobility trajectories in the 1997 data is similar to that of the 2007 data.

In summary, whether the child generation belongs to the higher-status groups of class or education, and how many strata were crossed through intergenerational class mobility were key factors in shaping the more diversified cultural tastes of the child generation. Therefore, greater cultural taste variety was observed in the members of the higher-status child generation whose status was a replica of their fathers’ status, as well as in those who acquired the status through intergenerational class and educational mobility. The trajectories of intergenerational class upward mobility and those of the intergenerational educational upward and downward mobility also facilitate enhancing the variety of cultural tastes. However, the effects of educational mobility trajectories on cultural taste variety differed from the effects of class mobility trajectories. A salient difference is that the people who experienced downward mobility in education exhibited higher variety in cultural tastes than did those with no mobility; however, no significant difference in variety was observed between those who experienced downward or no mobility in class. A comparison of the results drawn from the 1997 and 2007 data revealed no significant change in the effects of intergenerational mobility.

Discussion and Conclusions

In the past two decades, Taiwan has undergone considerable industrial and social transformation. within-groups cultural taste variety caused by the trajectories of the social mobility was observed in the child generation groups at the higher levels of class or education.

(23)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₁₅

When social transformation brings about a change in social structure, the structural and individual opportunities for social mobility are enhanced. This study addressed the following question: What changes do these trajectories of social mobility cause to class cultural tastes? Our finding shows that both cultural activity participation behaviors in 1997 and 2007 exhibited an omnivorous characteristic in cultural taste. The results indicate that the agents engaged in a wider range of cultural genres, indicating their diverse cultural tastes. In addition, the results show that the omnivore argument has greater explanatory power for Taiwanese society between 1997 and 2007 than Bourdieu’s logic of distinction. Therefore, social groups with varying attributes can be effectively distinguished according to the level of variety in their cultural tastes.

The educational attainment and objective class of the child generation are the two main factors influencing cultural taste variety, in which educational attainment is more crucial than class. This research finding show that H1 is supported. People with higher positions in educational attainment and objective class tend to become cultural omnivores who appreciate diverse cultural activities, whereas those who occupy lower positions tend to become cultural snobs who enjoy homogenous cultures. Therefore, H2 is supported. The results show that between 1997 and 2007, Taiwanese society experienced a development similar to that of Western society, as observed in previous studies (e.g., Peterson, 2002; Peterson & Kern, 1996; Peterson & Simkus, 1992; Sullivan & Katz-Gerro, 2007; Rimmer, 2012); in other words, the cultural tastes of elites diversified.

From a theoretical perspective of social mobility trajectories, this study explains the omnivore phenomenon in cultural activity participation behaviors. The emergence of cultural omnivores among highly educated individuals is likely related to the increase in intergenerational educational mobility caused by education expansion. Changes in intergenerational educational structures enabled children whose fathers had lower educational attainment to acquire a higher educational attainment. Consequently, the child generation gained access to various cultures during their growth process. Moreover, the high variety in cultural taste among the child generation who belonged to the service class may be an outcome of intergenerational class upward mobility. Frequent experiences in intergenerational class mobility enabled the child generation to be more tolerant of diverse cultures instead of developing exclusive tastes. The results of the present study support the argument that social mobility trajectories affect cultural taste variety. People whose class moved upward intergenerationally exhibited more diverse cultural tastes than those without upward mobility. These research findings show that H5 is supported. Moreover, the number of strata crossed through upward mobility is positively and significantly related to cultural taste variety. People who experienced intergenerational educational upward mobility, and those who crossed more strata through upward

(24)

mobility, displayed greater variety in their cultural tastes than did those whose educational rank was a replica of their fathers’. However, a notable phenomenon was observed: people who experienced intergenerational downward educational mobility also exhibited greater variety in their cultural tastes than those with no mobility. This situation differs from that observed among people who underwent downward mobility in class. Therefore, the trajectories of intergenerational educational and class mobility differ in their influences on cultural taste variety. According to the aforementioned research findings, H3 and H4 are only partially supported. Future studies should investigate this difference.

Doubts and unanswered questions remain in terms of understanding how cultural tastes are being shaped and developed in Taiwanese society. This study found that the impact of family background will be replaced by attributes of social class in future generations. Hence, it would be worthy for future studies to analyze how the role of the father’s education might impact children’s cultural tastes, and to understand the related family operating mechanism via interviews. Furthermore, the present study did not find any significant diversity in cultural tastes among the group of children from the elite class or in those with a high educational level due to the differences in their trajectories of social mobility. If an individual’s social trajectory is the mechanism that creates upward mobility, leading to diversity in cultural tastes among children, then why do the children who have inherited their fathers’ elite status due to class reproduction also show differences in cultural tastes? These issues need to be further explored. As Taiwanese society is facing a wave of freedom and openness in the media due to globalization, longer observation time is needed to collect more data; this is required to answer the question of whether discrepancies in people’s cultural tastes might shift to individual self-realization, as predicted by Bauman (1998, 2001), Beck (1992), and Giddens (1991). If more suitable research data can be collected for a comparative analysis at multiple points in time, the complex relationship among the changes in cultural tastes and social mobility can be further investigated. These are all issues worthy of future in-depth research.

Acknowledgements

The data used in this study are based on Taiwan Social Change Survey, which was conducted by the National Science Council & Institute of Sociology, Academia Sinica. We are grateful to Ruey-Ming Tsay, Jeng Liu, Teng-Lin Yu for their helpful comments and suggestions. Two anonymous reviewers also provided useful comments. Nevertheless we assume responsibility for any errors in the analysis and findings. The writing of this paper was supported in part by a grant from the Ministry of Science and Technology.

(25)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₁₇

References

Bauman, Z. (1998). Work, consumerism and the new poor. Buckingham, UK: Open University Press.

Bauman, Z. (2001). The individualized society. Cambridge, UK: Polity Press. Beck, U. (1992). Risk society: Towards a new modernity. London, UK: Sage.

Bourdieu, P. (1984). Distinction: A social critique of the judgement of taste. Cambridge, MA: Harvard University Press.

Bourdieu, P. (1987). What makes a social class? On the theoretical and practical existence of groups. Berkeley Journal of Sociology, 32, 1-17.

Bourdieu, P. (1990). The logic of practice. Stanford, CA: Stanford University Press.

Chan, T. W., & Goldthorpe, J. H. (2007). Social stratification and cultural consumption: Music in England. European Sociological Review, 23(1), 1-19. doi:10.1093/esr/jcl016

Chan, T. W., & Goldthorpe, J. H. (2010). Introduction: Social status and cultural consumption. In T. W. Chan (Ed.), Social status and cultural consumption (pp. 1-27). New York, NY: Cambridge University Press.

Chang, F.-C. (2006). Explore the relations among the socioeconomic status, cultural capital, education aspiration, and academic achievement by structural equation Modeling. Psychological Testing, 53(2), 261-296. (in Chinese)

Chang, F.-C. (2010). By SEM test the student’s mathematics achievement impacted factors: Evidence from four Asian countries’ grade 8 in TIMSS 2003. Forum of Educational Administration, 2(2), 1-33. (in Chinese)

Chang, F.-C. (2011). The relation among parents’ education, cultural capital, self-aspiration, students’ interesting and mathematics achievement. Journal of National Taichung University: Education, 25(1), 29-56. (in Chinese)

Chang, L.-Y., & Liao, P.-S. (2008). Taiwan social change survey report 2007, phase 5, wave 3. Taipei, Taiwan: Institute of Sociology, Academia Sinica. (in Chinese)

Chiang, T.-H. (2001). Critique on Marxist proletarianization and its meanings on teaching profession and class consciousness. Taiwan Journal of Sociology of Education, 1(2), 33-57. (in Chinese) Chiu, H.-Y. (1997). Social stratification, cultural identification, and music preferences in Taiwan. In

L.-Y. Chang, Y.-H. Lu, & F.-C. Wang (Eds.), Taiwanese society in 1990s: Taiwan social change survey symposium series II (pp. 189-228). Taipei, Taiwan: Institute of Sociology, Academia

(26)

Sinica. (in Chinese)

Chiu, H.-Y. (1998). Taiwan social change survey report 1998, phase 3, wave 3. Taipei, Taiwan: Institute of Sociology, Academia Sinica. (in Chinese)

Daenekindt, S., & Roose, H. (2013). Cultural chameleons: Social mobility and cultural practices in the private and the public sphere. Acta Sociologica, 56(4), 309-324. doi:10.1177/0001699313 496589

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. doi:10.2307/2673239

Delsing, M. J. M. H., Ter Bogt, T. F. M., Engels, R. C. M. E., & Meeus,W. H. J. (2008). Adolescents’ music preferences and personality characteristics. European Journal of Personality, 22(2), 109- 130. doi:10.1002/per.665

DiMaggio, P. (1982). Cultural capital and school success: The impact of status culture participation on the grades of U.S. high school students. American Sociological Review, 47(2), 189-201. doi: 10.2307/2094962

DiMaggio, P. (1987). Classification in art. American Sociological Review, 52(4), 440-455. doi:10. 2307/2095290

Erickson, B. H. (1996). Culture, class, and connections. American Journal of Sociology, 102(1), 217- 251. doi:10.1086/230912

Erikson, R., & Goldthorpe, J. H. (1992). The constant flux: A study of class mobility in industrial societies. Oxford, UK: Clarendon.

Giddens, A. (1973). The class structure of the advanced societies. London, UK: Hutchinson. Giddens, A. (1991). Modernity and self-identify: Self and society in the late modern age. Cambridge,

UK: Polity.

Hsieh, C.-L., & Chuang, C.-C. (2016). Effects of intergenerational transfer and conversion of cultural capital on the educational achievements of junior high school students. Journal of Research in Education Sciences, 61(3), 163-195. doi:10.6209/JORIES.2016.61(3).06

Huang, C.-Y., Wu, I-H., & Yu, C.-Y. (2015). Relationships among family socio-economic status, social capital, cultural capital, financial capital, and the learning outcomes of junior high school students with disabilities. Journal of Research in Education Sciences, 60(4), 129-160. doi:10. 6209/JORIES.2015.60(4).05

Huang, L.-H., Chang, T.-S., & Wang, T.-W. (2010). The comparison between the gifted and regular elementary students regarding family socioeconomic backgrounds and cultural capital. Journal

(27)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₁₉

of Educational and Multicultural Research, 2, 59-94. (in Chinese)

Hwang, Y.-J. (2000). Cultural capital, Social network and strata identification, class boundaries. The NCCU Journal of Sociology, 30, 1-42. (in Chinese)

Hwang, Y.-J., & Wu, Y.-I (2011). Is the prestigious junior high school a bridge or a rainbow? Testing Coleman’s theory among eighth graders Taitung. Taiwan Journal of Sociology of Education, 11(1), 41-75. (in Chinese)

Katz-Gerro, T. (2002). Highbrow cultural consumption and class distinction in Italy, Israel, West Germany, Sweden, and the United States. Social Forces, 81(1), 207-229. doi:10.1353/sof.2002. 0050

Lahire, B. (2008). The individual and the mixing of genres: Cultural dissonance and self-distinction. Poetics, 36(2), 166-188. doi:10.1016/j.poetic.2008.02.001

Lareau, A. (1989). Home advantage: Social class and parental intervention in elementary education. Philadelphia, PA: Falmer Press.

Lareau, A. (2011). Unequal childhoods: Race, class, and family life (2nd ed.). Oakland, CA: University of California Press.

Lee, D.-R., & Yu, M.-N. (2005). The verification of a structural equation model on SES, siblings, household education resources and educational achievement: Using the empirical data of the 2001 TEPS. Taiwan Journal of Sociology of Education, 5(2), 1-48. (in Chinese)

Lee, W.-Y., & Hwang, Y.-J. (2004). The study on relationship among cultural capital, social capital and students’ achievement－An example of National Taitung Teachers College. NTTU Educational Research Journal, 15(2), 23-58. (in Chinese)

Lin, C.-Y., & Wu, Y.-Y. (2007). Impact of family and school factors on students’ academic achievement: An analysis of hierarchical linear modeling. Bulletin of Educational Research, 53(4), 107-144. (in Chinese)

Lin, J.-X., & Hwang, Y.-J. (2008). Objective class position, cultural capital, and subjective class identification for elementary and junior high school teachers in Taiwan: Compared with other occupations. Bulletin of Educational Research, 54(3), 99-136. (in Chinese)

Lin, P.-F. (2009). An empirical study of the relationships between talent classes and learning achievement: A cultural capital perspective. Formosan Education and Society, 17, 111-134. (in Chinese)

Lin, T.-H. (2009). Post-industrializing Taiwan: Changing class structure and social inequality, 1992-2007. Taiwanese Journal of Sociology, 43, 93-158. (in Chinese)

(28)

for Humanities and Social Sciences, 25(4), 689-734. (in Chinese)

Lin, T.-H., & Hu, A. K.-W. (2011). Cross-strait trade and class politics in Taiwan. Thought and Words, 49(3), 95-134. (in Chinese)

Marsh, R. M., & Hsu, C. K. (1994). White-collar proletarianization? The case of Taiwan. Research in Social Stratification and Mobility, 13, 43-69.

Meulman, J. J., van der Kooij, A. J., & Heiser, W. J. (2004). Principal components analysis with nonlinear optimal scaling transformations for ordinal and nominal data. In D. Kaplan (Ed.), The sage handbook of quantitative methodology for the social sciences (pp. 49-70). London, UK: Sage. doi:10.4135/9781412986311.n3

Peterson, R. A. (1992). Understanding audience segmentation: From elite and mass to omnivore and univore. Poetics, 21(4), 243-258. doi:10.1016/0304-422X(92)90008-Q

Peterson, R. A. (2002). Roll over Beethoven, there’s a new way to be cool. Contexts, 1(2), 34-39. doi:10.1525/ctx.2002.1.2.34

Peterson, R. A., & Kern, R. M. (1996). Changing highbrow taste: From snob to omnivore. American Sociological Review, 61(5), 900-907. doi:10.2307/2096460

Peterson, R. A., & Simkus, A. (1992). How musical tastes mark occupational status groups. In M. Lamont and M. Fournier (Eds.), Cultivating differences: Symbolic boundaries and the making of inequality (pp. 152-186). Chicago, IL: University of Chicago Press.

Rentfrow, P. J., & Gosling, S. D. (2003). The do re mi’s of everyday life: The structure and personality correlates of music preferences. Journal of Personality and Social Psychology, 84(6), 1236-1256. doi:10.1037/0022-3514.84.6.1236

Rimmer, M. (2012). Beyond omnivores and univores: The promise of a concept of musical habitus. Cultural Sociology, 6(3), 299-318. doi:10.1177/1749975511401278

Simmel, G. (1971). The metropolis and mental life. In D. Levine (Ed.), On individuality and social forms (pp. 324-339). Chicago, IL: University of Chicago Press. doi:10.7208/chicago/97802269 24694.001.0001

Su, C.-L., & Hwang, Y.-J. (2009). Influence of cultural capital on academic performance through school social capital: A study of eighth graders in Taitung. Bulletin of Educational Research, 55(3), 99-129. (in Chinese)

Sullivan, O., & Katz-Gerro, T. (2007). The omnivore thesis revisited: Voracious cultural consumers. European Sociological Review, 23(2), 123-137. doi:10.1093/esr/jcl024

Sun, C.-S., & Hwang, Y.-J. (1996). Shadow education, cultural capital and educational attainment. Taiwanese Journal of Sociology, 19, 95-139. (in Chinese)

(29)

Chih-Chia Chuang & Da-Sen Lin The Trajectory of Cultural Taste ₂₂₁

Swartz, D. (1997). Culture and power: The sociology of Pierre Bourdieu. Chicago, IL: University of Chicago Press. doi:10.1093/sf/77.3.1232

Ter Bogt, T. F. M., Delsing, M. J. M. H., van Zalk, M., Christenson, P. G., & Meeus, W. H. J. (2011). Intergenerational continuity of taste: Parental and adolescent music preferences. Social Forces, 90(1), 297-319. doi:10.1093/sf/90.1.297

van Eijck, K. (2001). Social differentiation in musical taste patterns. Social Forces, 79(3), 1163-1185. doi:10.1353/sof.2001.0017

van Eijck, K., & Lievens, J. (2008). Cultural omnivorousness as a combination of highbrow, pop, and folk elements: The relation between taste patterns and attitudes concerning social integration. Poetics, 36(2), 217-242. doi:10.1016/j.poetic.2008.02.002

Veblen, T. (1931). The theory of the leisure class. New York, NY: Viking Press. Warde, A. (1997). Consumption, food and taste. Thousand Oaks, CA: Sage.

Weininger, E. B. (2005). Foundations of Pierre Bourdieu’s class analysis. Approaches to Class Analysis, 4, 82-118.

Wilensky, H. L. (1964). Mass society and mass culture: Interdependence or independence? American Sociological Review, 29(2), 173-197. doi:10.2307/2092122