ABSTRACT 
 
 
 
 
Title of Document: CULTURAL DIFFERENCES IN PREJUDICE 
BETWEEN INDIVIDUAL- AND GROUP-
ORIENTED CULTURES   
  
 Hyeyoung Shin, Doctor of Philosophy, 2014 
  
Directed By: Prof. Charles Stangor  
Department of Psychology 
 
 
The present dissertation investigated cultural differences in the degree and dynamics of 
prejudice between individual- and group-oriented cultures.  In Study 1, in the US where 
personal responsibility and individual?s capitalistic/meritocratic achievements are 
emphasized, participants reported greater distance to groups based on personal qualities 
(e.g., heavy drinkers) than in South Korea, and competition for employment was 
positively associated with prejudice toward various groups (but not in South Korea).  In 
South Korea where the holistic/essential quality, the self-ingroup overlap, and 
relationships within ingroups are emphasized, participants reported greater distance to 
groups perceived as essentially different from the majority (e.g., different race) than in 
the US.  In Study 2, the emphasis on individual achievements consistently predicted 
social hierarchy beliefs in the US (but not in South Korea), whereas the emphasis on 
roles/positions within ingroups consistently predicted both social and biological hierarchy 
beliefs in South Korea (but not in the US).  In Study 3, the emphasis on individual 
  
uniqueness was negatively associated with social distance to non-normative groups (e.g., 
homosexuals) only in the US, whereas the value of conformity with norms/conventions 
predicted social distance to low SES (e.g., poor/uneducated/homeless), non-normative, 
and value-based (e.g., people whose opinions are different from mine in religious issues) 
target groups both in the US and South Korea.  Conformity with norms/conventions also 
predicted social distance to racial/ethnic outgroups (e.g., non-Koreans to South Korean 
participants) only in South Korea.  In addition, essentialism was associated with social 
distance to low SES groups in the US, whereas essentialism was associated with social 
distance to low SES, non-normative, and racial/ethnic groups in South Korea.  Overall, 
the present research provided empirical evidence that cultural norms/values are 
associated with differences in the degree and dynamics of prejudice between individual- 
and group-oriented cultures.    
  
 
 
 
 
 
 
 
 
 
 
CULTURAL DIFFERENCES IN PREJUDICE BETWEEN INDIVIDUAL- AND 
GROUP-ORIENTED CULTURES    
 
 
 
By 
 
 
Hyeyoung Shin 
 
 
 
 
 
Dissertation submitted to the Faculty of the Graduate School of the  
University of Maryland, College Park, in partial fulfillment 
of the requirements for the degree of 
Doctor of Philosophy 
2014 
 
 
 
 
 
 
 
 
 
 
Advisory Committee: 
Professor Charles G. Stangor, Chair 
Professor John F. Dovidio 
Professor Seung-kyung Kim 
Assistant Professor Jaime L. Napier 
Professor David D. Yager 
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
? Copyright by 
Hyeyoung Shin 
2014 
 
 
 
 
 
 
 
 
 
 
 
 
 
  ii 
 
Table of Contents 
 
 
Table of Contents .......................................................................................................... ii 
List of Tables ............................................................................................................... iii 
List of Figures .............................................................................................................. iv 
Chapter 1: General Introduction ................................................................................... 1 
Chapter 2: Study 1 ...................................................................................................... 18 
Chapter 3: Study 2 ...................................................................................................... 37 
Chapter 4: Study 3 ...................................................................................................... 63 
Chapter 5:  General Discussion................................................................................... 82 
Appendices ................................................................................................................ 109 
Bibliography ............................................................................................................. 111 
 
 
 
 
 
 
 
 
 
 
  iii 
 
List of Tables 
 
 
 
Table 1 Cultural differences in proportions of negative responses (social distance) 
             to the target Groups  ....................................................................................... 88 
Table 2 Associations between competition for employment and negative responses 
            (social distance) to target groups in the US and South Korea   ...................... 89 
Table 3 Descriptive statistics (means & SDs) and cultural differences in the degree 
             of hierarchy beliefs and norms/values between the US and South Korea ..... 90 
Table 4 The baseline measurement models in the US and South Korea .................... 91 
Table 5 Configural and metric invariance in measurement models ........................... 92 
Table 6 The structural models in the US and South Korea......................................... 93 
Table 7 Cultural differences in structural models between the US and South Korea . 94 
Table 8 Means (SDs) and cultural differences in the degree of perceived social  
             distance to four target groups and individual- and group-oriented   
             norms/values between the US and South Korea ............................................ 95 
Table 9 The baseline measurement models in the US and South Korea .................... 96 
Table 10 Configural and metric invariance in measurement models ......................... 98 
Table 11 Structural models in the US and South Korea ............................................. 99 
Table 12 Cultural differences in the structural models between the US and South 
               Korea .......................................................................................................... 101 
 
 
 
 
  iv 
 
List of Figures 
Figure 1 Proportions of mean negative responses (social distance) to groups  
              perceived as personally responsible and essentially different in the US   
              (Whites) and South Korea ........................................................................... 102 
Figure 2 Social and biological hierarchy beliefs in the US and South Korea  .......... 103 
Figure 3 Standardized parameter estimates of the two-factor structural models in 
              the US and South Korea .............................................................................. 104 
Figure 4 Standardized parameter estimates of the three-factor structural models in the   
              US and South Korea.................................................................................... 105 
Figure 5 Mean social distance to four types of target groups in the US and South Korea 
               .................................................................................................................... 106 
Figure 6 Standardized parameter estimates of the structural models in the US and 
              South Korea ................................................................................................ 107 
Figure 7 Standardized parameter estimates of the structural models in the US and 
              South Korea ................................................................................................ 108 
 
 
 
 
 
 
 
 
 
 
 
              
       
 
 
 
 
 
 
  1 
 
Chapter 1: General Introduction 
Prejudice is a negative attitude toward an outgroup and its members that creates 
or maintains hierarchical relations between the groups (Dovidio, Hewstone, Glick, & 
Esses, 2010).  It is widely considered a universal phenomenon, observable across all 
cultures.  Intergroup bias is hypothesized to be rooted in evolutionary-based process 
(Sidanius & Pratto, 1999) and embedded in the ?normal? ways people think about and 
process information about groups (Dovidio & Gaertner, 2010; Fiske, 1998; Tajfel & 
Turner, 1979).  Nevertheless, relatively little research has examined the dynamics of 
prejudice cross-culturally (cf. Shin, Dovidio, & Napier, 2013).  Like other attitudes, 
beliefs, and values, prejudice and group-based hierarchies can also be shaped, expressed, 
applied, and interpreted differently across cultures.  The present dissertation thus 
integrated work in social and cross-cultural psychology to explore potential similarities 
and differences in the degree and dynamics of prejudice across cultures, specifically 
between the United States (US) and South Korea, the most prototypical individual-
oriented and group-oriented countries in national comparisons, respectively (see Kim, 
Triandis, Kagitcibasi, Choi, & Yoon, 1994; Nisbett, 2003).   
The goal of the dissertation is to illuminate general and culture-specific processes 
in prejudice and discrimination in individual- and group-oriented cultures.  The research 
presented in the dissertation investigates not only how prejudice toward different groups 
is expressed in various degrees across cultures, but also how the dynamics of bias may 
vary as a function of dimensions of cultural norms and values.  This chapter considers 
basic concepts, including culture and prejudice, and reviews the general literature 
relevant to understanding the relationship between culture and prejudice.  I then examine 
  2 
 
the specific ways that culture can shape the expression of prejudice, developing a general 
framework that guides the empirical work and theoretical refinement in later chapters of 
the dissertation.   
The remainder of this chapter (a) defines culture and reviews the history and 
methodology of the research in this area; (b) examines cultural differences, drawing on 
research and theory relevant to the dimensions of cultural differences investigated in the 
dissertation; and (c) considers culture and prejudice based on the nature of prejudice and 
its measurement in ways that could be applied across cultures. 
Culture 
In psychology, culture is conceptualized as a dynamic system of practices and 
meanings that is inseparable from the content and process of the mind (Cohen & 
Kitayama, 2007).  This definition implies that culture is continuously recreated through 
interactions between an individual person and social situations, including the influence of 
other individuals, groups, and situations.  Culture is also communicated and sustained by 
norms (ways of thinking, feeling, and behaving that are perceived as appropriate or 
normal within a group or society; see Stangor, 2004) and values (concepts that are 
emphasized and perceived as important and desirable; see Rokeach, 1979; Schwartz, 
1992, 1994).  In a more macro or anthropological aspect, a culture, thus, involves not 
only ecological conditions and historical backgrounds but also religious/philosophical 
traditions and values.  That is, different historical influences, philosophical traditions, and 
conventions through socialization shape different cultural orientations, norms, and values 
(Kluckhohn, 1951; Rokeach, 1973; Simmel, 1917; Whiting & Whiting, 1975; Williams, 
1968), which become standards for judgments and behaviors of people living in the 
  3 
 
culture.  For example, early (age 9-12) and late (age 15-18) adolescents from six different 
ethnic groups in two different countries (native Germans, Turkish, and former Soviet 
Union immigrants in Germany and native Israelis, former Soviet Union immigrants, Arab 
Israelis in Israel) showed that the association between the value (hierarchy beliefs vs. 
egalitarianism) shared by one?s own cultural group and negative attitudes toward 
outgroups was stronger at the cultural group level than at individual level, and the effect 
was greater for the older group than the younger group (Schiefer et al., 2010).  Culture, 
thus, can be more broadly defined as a dynamic system in which all of these 
psychological, situational, and anthropological factors interact with each other.   
In psychology, research on culture in the 1960s and 1970s focused on describing 
national characteristics or values.  More recently, while interest in this issue has 
continued, the field has also been emphasizing the influence of culture on psychological 
processes, such as cultural differences in perception, attribution, motivation, or emotion 
(see Kitayama & Cohen, 2007).  Currently, research is also investigating how these 
differences are associated with differences in brain activity (see Chiao & Ambady, 2007).   
Methodologically, the term, cultural psychology is used for the research on 
culture in psychology in general, but it is also used for the research that focuses on one 
culture using ethnographic methods.  Indigenous psychology focuses on core concepts or 
phenomena that exist only in a specific geographical area only during a specific time 
period.  Cross-cultural psychology uses samples from different cultures and examines 
similarities and differences between the cultures.  Each of these three approaches has 
both advantages and disadvantages.  Cultural psychology provides in-depth knowledge 
about a particular culture but, because the interpretation of findings relies solely on data 
  4 
 
from one culture, the conclusions may be limited in generalizability.  Indigenous 
psychology is particularly useful to understand a specific problem or phenomenon that 
has not been observed in other areas or in other time periods.  However, because the 
research focus is limited in a problem or phenomenon that temporarily exists only one 
specific area, the findings may not be applicable to other areas or in other time periods.  
Cross-cultural psychology provides more comprehensive knowledge, compared to 
cultural psychology, based on both similarities and differences between cultures and 
illuminates processes that operate pan-culturally or are culture-specific. However, 
conducting cross-cultural research requires cross-cultural equivalence in measurement 
procedure (i.e., measurement invariance, see Byrne, 2008; Chen, 2008) in order to avoid 
methodological biases that can invalidate results.  For example, a concept can have 
different meanings across cultures (e.g., ?being individualistic? mean a normal and 
positive quality in an individual-oriented North American country but can mean ?being 
self-centered? with a negative connotation in a group-oriented East Asian country), and a 
certain concept (e.g., ?filial piety?) is concerned more in certain cultures than other 
cultures.  Thus, cross-cultural psychology requires demonstration of the equivalence of a 
measure in its meaning across cultures to interpret findings on the measure correctly.  
The perspectives of cultural, indigenous, and cross-cultural psychology can offer 
complementary insights.  Triandis (2000) suggested that cross-cultural comparison based 
on the knowledge established by work in cultural psychology represents an ideal 
approach.   
 
 
  5 
 
Cultural Differences 
Much of the research in cultural and cross-cultural psychology has identified 
fundamental dimensions on which cultures systematically differ in ways that influence 
relations within and between groups.  In this section, I review two related dimensions of 
cultures: (a) individualism versus collectivism and (b) individual- versus group-
orientations.  I also consider the origins of these cultural dimensions, on which Western 
or Northern European and East Asian cultures substantially differ.  
Individualism versus collectivism.  In the previous research in both cultural and 
cross-cultural psychology, one of the most widely studied topics is individualism versus 
collectivism (first defined by Hofstede, 1980, 1991; see also Kitayama & Cohen, 2007; 
Nisbett, 2003).  Individualism and collectivism refer to the ways in which people relate 
with others and social environments.  Triandis (1995) initially defined individualism as 
?a social pattern that consists of loosely linked individuals who view themselves as 
independent of collectives; are primarily motivated by their own preference, needs, 
rights, and the contracts they have established with others; give priority to their personal 
goals over the goals of others; and emphasize rational analyses of the advantages and 
disadvantages to associating with others? (p. 2).  He also defined collectivism as ?a social 
pattern consisting of closely linked individuals who see themselves as parts of one or 
more collectives (family, co-workers, tribe, nation); are primarily motivated by the norms 
of and duties imposed by those collectives; are willing to give priority to the goals of 
these collectives over their own personal goals; and emphasize their connectedness to 
members of these collectives? (p. 2).  Triandis?s initial definitions helped have 
preliminary understanding that becomes the base of many following research on 
  6 
 
individualism and collectivism.  Triandis (2007) later defined individualism as a cultural 
syndrome in which the individual is the basic unit of societal structure and the value of 
individual happiness is supported, and collectivism as a cultural syndrome in which the 
group is the basic unit and the value of preservation and enhancement of group resources 
is supported.      
Systematic differences between individualistic and collectivistic cultures have 
been found in various areas in social psychology including the self-concept (e.g., 
independence vs. interdependence, Markus & Kitayama, 1991), cognitive processes (e.g., 
analytic vs. holistic cognition, Nisbett, Peng, Choi, & Norenzayan, 2001), perception 
(e.g., the difference in the filed-dependency, Masuda & Nisbett, 2001), attribution (e.g., 
attribution to person vs. situation, Choi, Nisbett, & Norenzayan, 1999; Morris & Peng, 
1994), motivation (e.g., the difference in the self-enhancement motivation, Heine, 
Lehman, Markus, & Kitayama, 1999), and emotion (e.g., guilty vs. shame, Markus & 
Kitayama, 1994). Western European (particularly Northern European/North American) 
and East Asian countries have been identified as most representative individualistic and 
collectivistic cultures, respectively.  Because of the fundamental differences in 
psychological orientations between Western European (and North American) and East 
Asian cultures, work in cross-cultural psychology notes that theories and principles 
developed in the Western European cultures may not uniformly reflect those of non-
Western cultures (see Heine, 2010; Henrich, Heine, & Norenzayan, 2010; Markus, 
Kitayama, & Heiman, 1998).  
The origins and backgrounds.  The distinct differences between Western or 
Northern European and East Asian cultures are hypothesized to originate from ecological 
  7 
 
conditions of ancient Greek and Chinese cultures.  According to Berry (1994) and Nisbett 
(2003), because of ecological conditions that facilitated maritime trade, ancient Greek 
culture emphasized individuals? personal choices and efforts to succeed in trading and 
developed monetary values that are useful in exchanging goods.  By contrast, because the 
ecological conditions were more suitable for agriculture, ancient Chinese culture stressed 
cooperation with others as a group to harvest more and encouraged attention to 
relationships within the group.  Berry and Nisbett each explained that when these lifestyle 
patterns are repeated for hundreds and thousands of years, these lifestyle patterns become 
the base of the core values of corresponding culture.  These core values also provide the 
foundation of religious or philosophical traditions and the religious or philosophical 
traditions enhance the core values again establishing more stable norms and values of 
each culture.   
In countries with the Western or Northern European heritage (such as Canada, 
Finland, Germany, Netherlands, Norway, Sweden, Switzerland, the United Kingdom, and 
the United States), the Protestant Ethic has been a dominant philosophical tradition.  
Protestantism, established by Martin Luther (1483-1546), is a branch of Judeo-
Christianity beliefs that began in Germany in the 16th century in protest to aspects of the 
medieval Catholic traditions.  Protestantism was developed into many denominations but 
commonly emphasizes that the salvation should be based solely on the beliefs in the 
Bible, Christ, grace, faith, and God.  Max Weber (1905/2008) argued that the Calvinist 
(John Calvin, 1509-1564, a major denomination of Protestantism) value of calling (a 
belief in which one?s vacation is from God; the religious pursuit of wealth) provided the 
rationalization of the pursuit of economic gain and consequently led to the development 
  8 
 
of capitalism in the Protestant regions, such as countries with Western or Northern 
European heritage.  Psychological evidence has revealed that the Protestant Ethic is 
associated with personal diligence and responsibilities and encourages personal 
achievements including individual wealth (Heaven, 1990; Katz & Hass, 1988; Quinn & 
Croker, 1999; Rokeach, 1973).   
Meritocracy beliefs of the countries with the Western or Northern European 
heritage also emphasize values similar to those highlighted in the Protestant Ethic 
(Sidanius & Pratto, 1999).  Initially, meritocracy (Young, 1958) referred to an ideology 
that people should gain their social status based on their intellectual merits; it contrasted 
with aristocracy beliefs, in which people?s status was given by birth.  In modern societies, 
meritocracy is regarded as a belief in which individuals who have more talents and/or 
give more efforts (consequently achieve more) deserve more rewards (e.g., higher socio-
economic status) (Major, Kaiser, O'Brien, & McCoy, 2007).     
Although the Protestant Ethic and meritocracy beliefs seem to originate from 
different sources, they are the core norms and values of the particular geographical 
regions and have the common emphasis on personal achievements.  The difference 
between the two is that the Protestant Ethic is based on the religious pursuit of wealth and 
is more closely related to capitalistic achievements, whereas meritocracy beliefs are more 
general and nonreligious norms and values relating to individual competence and 
achieved social status.  The emphasis on achieved socio-economic status in the countries 
with the Western or Northern European heritage may thus be rooted in a combination of 
the Protestant Ethic and meritocracy beliefs.   
  9 
 
The Protestant Ethic and meritocracy beliefs, however, can be used to legitimate 
the position of dominant group or to justify the status quo by assuming that dominant 
groups deserve the high status because they are more talented and/or diligent than the 
lower status groups without considering other systematic problems.  For example, Social 
Dominance Theory (Sidanius & Pratto, 1999) explains that dominant groups use beliefs 
(e.g., the Protestant Ethic and meritocracy beliefs in the US) that can justify their 
dominant positions within the society.  System Justification Theory (Jost & Banaji, 1994; 
Jost, Banji, & Nosek, 2004) also discusses that both advantaged and disadvantaged 
groups use stereotypes on different status groups (e.g., the low socioeconomic status 
groups do not deserve to be rewarded because they are not competent enough to earn it) 
in order to justify the status quo and support the legitimacy of the existing social order.  
Also, the low-status group members who endorse a meritocracy worldview (e.g., ?anyone 
who works hard can get ahead and succeed?) report lower self-esteem than those who do 
not endorse meritocracy worldview when they perceived discrimination against their own 
group (Major et al., 2007).  This tendency of dominant groups to legitimate their 
dominant position using norms and values of the culture is universal across cultures 
(Sidanius & Pratto, 1999), but dominant groups in different cultures use different norms 
and values (i.e., norms/values that are most acceptable in their culture to justify their 
dominant position more easily).  Dominant groups in the countries with the Western or 
Northern European heritage, for example, generally use the Protestant Ethic and 
meritocracy beliefs to justify their dominant status.       
In East Asian countries, such as China, Japan, Korea, Taiwan, and Vietnam, 
Confucianism and Buddhism have played important roles in shaping cultural norms and 
  10 
 
values. Confucianism (Confucius, 551-478 BC, see Tu, 1998a) has been a major 
philosophical tradition since about the 4th century, and Confucian values still permeate 
various facets of modern East Asian societies (Cha, 1994; Kim, 1994).  For example, 
although 33.42% of South Koreans identified themselves as a ?Christian? as the largest 
religious group in South Korea in 2010 
(http://www.thearda.com/internationalData/countries/Country_124_1.asp) and younger 
generations are increasingly westernized, Confucian traditions are reflected in many 
current public norms and conventions (see Cho, K., 2007) including norms and 
convention within Christian communities (e.g., pastors have the highest position within 
the church community and hierarchies within the community are emphasized).   
Confucianism focuses on how people should be related with each other.  
Confucianism emphasizes ?humanity in relations? (rather than individual morality) and 
teaches that individuals? preferences are less important than the needs of groups to which 
individuals belong (Gardner & Seeley, 2001).  Confucianism assumes hierarchical social 
stratification given by birth (more systematically in the past) and emphasizes hierarchical 
relations within a group or community (e.g., Cho, H., 1998; Cho, K., 2007; Jordan, 1998; 
Lebra, 1998; Tu, 1998b; Yim, 1998).  Confucianism teaches lifetime roles and duties of 
each class or social role within an extended family, communities, and a country, 
emphasizing roles/duties of the lower status (e.g., emphasize roles and duties of children 
rather than of parents, see Jordan, 1998).  During the time in which the traditional feudal 
age was moved toward modern times (e.g., between the 15th and 19th centuries in Korea), 
Confucianism also emphasized individual?s intellectual competence, allowing individuals 
to gain a higher status within the class given by birth (see Kazin, Edwards, & Rothman, 
  11 
 
2010).  However, the emphasis on individual intellectual competence in Confucianism 
was based on a more stable and stronger value of social stratification given by birth and 
the emphasis on hierarchical relations (i.e., roles and duties) within the society.       
Buddhism (the teachings of Buddha or ?the awakened one,? around 500 BC) has 
also been another major religious and philosophical tradition in East Asian countries.  In 
Buddhism, the life is deemed as full of suffering and it continues through the 
reincarnation of various living things.  In Buddhism, thus, there is no unique and 
independent self-concept as in the Western culture (see Spector et al., 2002; Weisz, 
Rothbaum, & Blackburn, 1984); the self is a part of nature (see Mahler, 1974; Parsons & 
Scheider, 1974) and related with all other existences around it (see Masuda & Nisbett, 
2001; Nisbett et al., 2001).  The cultural difference in the perception of the self and the 
relation between the self and the world was also found in an empirical study.  Kashima et 
al. (2005) found that the individual person was perceived as more agentic (having internal 
states of thinking, feeling, wanting, and intending) than groups (as an example of the 
world around oneself) in Australia, the United Kingdom, and the United States, whereas 
both individuals and groups were perceived as having equal level of agency in Hong 
Kong, Japan, and South Korea.  This holistic/circular worldview and the perception of the 
self is as a part of (not separated from) a well-organized world in East Asian cultures (see 
Nisbett, 2003) leads to the emphasis of predetermined or essential qualities of people?s 
nature and destiny (i.e., essentialism).  An emphasis on essential qualities promotes 
stereotyping and contributes to the legitimation of existing social inequalities as 
inevitable (e.g., Haslam, Bastian, Bain & Kashima, 2006; Yzerbyt, Rocher, & Schadron, 
1997).   
  12 
 
In summary, Western or Northern European and East Asian cultures have 
developed distinctive core cultural norms and values.  Based historically on ecological 
conditions, Western and Northern European countries came to value an individual?s 
personal choice and efforts, as well as wealth and capitalistic principles.  The Protestant 
Ethic and meritocracy beliefs also helped shape these norms and values.  By contrast, 
East Asian countries came to primarily value cooperation as a group and relations within 
the group.  Confucianism reinforced the value of group, roles and hierarchies within the 
group, and the holistic/circular worldview that emphasize predetermined or essential 
qualities of people?s nature and destiny (i.e., essentialism)  helped organize relations 
within the groups more stably.  These two distinctive sets of core cultural norms and 
values form the foundation for individual-oriented and group-oriented cultures.         
Individual versus group orientations.  Individual- and group-orientations are 
based on individualism and collectivism but are more specific.  That is, the individualism 
versus collectivism distinction reflects multiple facets, especially for collectivism (see 
Ashmore, Deaux, & McLaughlin-Volpe, 2004; Brewer & Chen, 2007).  For example, the 
interdependent self-concept can include the association of the self with other individuals 
(e.g., Kim, 1994; Markus & Kitayama, 1991), groups (e.g., Menon, Morris, Chiu, & 
Hong, 1999; Smith & Henry, 1996; Vijver & Watkins, 2006), or situations and contexts 
(Choi & Nisbett, 1998; Masuda & Nisbett, 2001).  Among these various aspects of 
individualism and collectivism that have been identified, one key dimension that is most 
likely relevant to prejudice, an intergroup phenomenon, is individual- versus group-
orientation.  The research that describe characteristics of various countries has also 
shown that Northern European/North American and East Asian cultures are the most 
  13 
 
representative individual- and group- oriented cultures, respectively (see Kim et al.,  
1994; Menon et al.,  1999; Nisbett, 2003; Smith & Henry, 1996; Triandis, 1994; Vijver & 
Watkins, 2006).   
In individual-oriented cultures, as discussed earlier, the core cultural norms and 
values are based on individual?s personal qualities and capitalistic/meritocratic 
achievements.  Consequently, each individual?s uniqueness, independence, right, pursuit 
of self-interest, self-determination, self-control, and personal responsibility are 
emphasized (Markus & Kitayama, 1991; Markus, Mullally, & Kitayama, 1997).  In 
addition, the value of diversity is promoted based on the emphasis of individual 
uniqueness and pursuit of self-interest.  The self and ingroups are perceived as separate 
entity based on the emphasis of independence of the self and individuals have various 
types of ingroups based on their personal choice and interests.  Also, the emphasis on 
individual?s capitalistic/meritocratic achievements may facilitate social hierarchies 
providing justification for negative attitudes toward the low-status or less achieving 
groups, and dominant groups would use these cultural norms/values to justify their 
dominant status.   
In group-oriented cultures, the core cultural norms and values are based on group, 
relations within the group, and essential qualities.  Consequently, the interdependence 
between the self and ingroups (Smith & Henry, 1996; Vijver & Watkins, 2006), roles and 
hierarchies within ingroups (Slote & De Vos, 1998), conformity with ingroup norms 
(rather than personal attitudes, Kashima, Siegel, Tanaka, & Kashima, 1992; Kim & 
Markus, 1999; Smith & Bond, 1993; Williams & Sogon, 1984), and 
essentially/biologically-based group qualities are emphasized (see also Fiske, Kitayama, 
  14 
 
Markus, & Nisbett, 1998; Triandis, 1994).  These group-oriented norms and values help 
organize relations within a group more stably and facilitate group functioning.  Also, the 
emphasis of the interdependence between the self and ingroups, relations within 
ingroups, and conformity with ingroup norms leads groups to be perceived as more 
cohesive and entitative, especially for outgroups (see Hamilton & Sherman, 1996).  Also, 
the value of diversity tends to be considered less positively, and deviants from the 
majority or ingroup norms are evaluated more negatively in group- than individual-orient 
cultures.  In addition, in group-oriented cultures, ingroups are commonly defined based 
primarily on kinship or essential qualities (e.g., the same ethnic group), rather than 
personal choice and interests.  The interdependence between the self and ingroups also 
leads individuals to perceive ingroup/outgroup boundaries to be less permeable (Tajfel & 
Turner, 1979) and view the distinction between ingroups and outgroups as more 
permanent and stronger (Triandis, McCusker, & Hui, 1990).  The emphasis on essential 
qualities promotes a focus on biologically-based group characteristics, and dominant 
groups in these cultures are likely to use the norms and values of essential qualities to 
justify their dominant status.     
Culture and Prejudice    
Prejudice is a negative attitude toward an outgroup and its members that creates 
or maintains hierarchical relations between the groups (Dovidio et al., 2010).  However, 
despite social psychology?s long-term emphasis on social influence and the fact that each 
culture and society has its own norms and values, it is surprising that cultural differences 
in prejudice have received relatively little attention in the field.  Moreover, many of the 
studies that have attempted to generalize existing theories of prejudice have focused on 
  15 
 
regional differences within a particular country (e.g., the North and South in the US; 
Pettigrew, 1959) or across Western European countries (e.g., Zick, Pettigrew, & Wagner, 
2008).  Although some research has included a broader range of countries (for example, 
work on the Stereotype Content Model; Cuddy et al., 2009; also Jost, Kivetz, Rubini, 
Guermandi, & Mosso, 2005; Pettigrew, 2001, 2003; Pratto et al., 2000, 2013), relatively 
little comparative research on prejudice and associated underlying processes between 
Western or Northern European and East Asian cultures has been conducted.   
Perhaps one reason for the dearth of research on cross-cultural differences in 
prejudice is because the targets of prejudice substantially differ across cultures for 
historical, economic, and political reasons (Allport, 1954/1979; see also Dovidio, Major, 
& Crocker, 2000; Shin et al., 2013).  Because of these distinctive influences across 
cultures, research on prejudice frequently reflect culture-specifically (i.e., indigenous or 
cultural psychologically) more urgent or immediate issues (e.g., bias expressed in terms 
of opposition to school busing in the US, see McConahay, 1986 or political stances 
related to symbolic racism in the US, see Henry & Sears, 2002).  In addition, research on 
prejudice within a culture is typically dominated by work on groups that received the 
public attention historically or politically (e.g., anti-Semitism or Islamophobia in Europe; 
Maori in New Zealand).  Thus, although there is substantial empirical interest in 
prejudice within a culture (Dovidio et al., 2010), research on similarities and differences 
in prejudice between cultures is rare. 
Methodologically, social distance (psychological distance to a social group or its 
members) represents a fundamental manifestation of prejudice that would be well-suited 
to cross-cultural research.  In general, individuals intentionally distance themselves from 
  16 
 
devalued groups and their members (e.g., Kaiser & Miller, 2001).  Allport (1954/1979) 
suggested that distancing oneself from a devalued group provides psychological benefits 
(e.g., self-enhancement) but it represents a basic form of prejudice against the devalued 
group.  Also, what groups are perceived as devalued likely depend on the norms and 
values of the given social context that provide the standards of what is ?normal,? ?better,? 
or ?more desirable.?   
Bogardus (1933), who made classic contributions in this area, created a social 
distance scale, which forms the basis for current measures.  The original social distance 
scale asked respondents how willing they would be to have a member of a specified 
group as (a) a visitor in my country, (b) a citizen in my country,  (c) a co-worker, (d) a 
neighbor, (e) a close personal friend, and (f) a close relative by marriage (in order of 
greater intimacy).  Social distance measures correlate with a wide range of other 
measures of prejudice, including measures tailored to specific target groups.  For 
instance, social distance strongly relates to attitudes toward Blacks (e.g., Implicit-
Association Test (IAT), Greenwald, McGhee, & Schwartz, 1998), homosexuals (e.g., 
Modern Homonegativity Scale (MHS), Morrison & Morrison, 2003) and the low-SES 
(e.g., Social Dominance Orientation (SDO) scale, Pratto, Sidanius, Stallworth, & Malle, 
1994 ).  Social distance has been used as a reliable measure of prejudice toward different 
race, gender, social class, religion, and nationality (e.g., Jackman, 1994; Triandis & 
Triandis, 1960) and stigmatized ingroup members (e.g., Steele, Spencer, & Aronson, 
2002).  Because of its fundamental relationship to prejudice and its adaptability for 
measuring responses to a wide range of groups, social distance is employed as one of the 
primary measures of prejudice in my dissertation.   
  17 
 
In summary, my dissertation investigates how cultural differences in norms and 
values between individual- and group-oriented cultures have important implications for 
prejudice.  Different norms or values can provide different bases for prejudice leading to 
different degrees of bias toward the same groups or affect the types of groups that are 
targeted for prejudice.  In the subsequent chapters in this dissertation, I develop a 
theoretical framework for understanding prejudice cross-culturally and testing the 
implications empirically. Chapter 2 (Study 1) examines cultural similarities and 
differences in the degree of prejudice toward groups that are based on either personal 
qualities or essential qualities and how realistic competition is associated with prejudice 
toward various groups across cultures.  Chapter 3 (Study 2) investigates cultural 
similarities and differences in the degree of supporting hierarchies that are based on 
either social or biological factors and how the cultural norms/values of individual 
achievements, the self-ingroup overlap, and roles/positions within ingroups associated 
with the social and biological hierarchy beliefs.  Chapter 4 (Study 3) considers cultural 
similarities and differences in the degree of prejudice toward low SES, non-normative, 
racial/ethnic, and value-based groups and how the cultural norms/values of essential 
qualities, individual uniqueness, and conformity with norms/conventions associated with 
prejudice toward these four types of outgroups.  Chapter 5 discusses the overall findings, 
consistent themes, and conclusions across the three studies and their implications.  
  18 
 
 
Chapter 2: Study 1 
In Study 1, I investigated, using nationally representative data from the US and 
South Korea available in the World Values Survey (WVS; n.d.), whether there are 
cultural differences in the degree of prejudice toward the same target groups between the 
two countries.  In particular, in the WVS, respondents are asked whether they would be 
willing to have as neighbors a range of groups.  In previous research (Shin et al., 2013), I 
investigated cultural differences in prejudice between Northern European/North 
American (Canada, Finland, Germany, the Netherlands, Norway, Sweden, Switzerland, 
the United Kingdom, and the United States) and East Asian (China, South Korea, 
Taiwan, and Vietnam) countries in terms of attitudes toward stigmatized groups 
representing Goffman?s (1963) distinction between stigmas based ?blemishes of 
character? (e.g., homosexuals) and ?tribal stigmas? (e.g., people of a different race).  Shin 
et al. (2013) found that bias was greater in East Asian countries than in Northern 
European/North American countries for both types of stigmas, and cultural values of 
individual uniqueness and behavioral conformity mediated the cultural differences in 
prejudice.  In Study 1, I examined the degree of prejudice focusing on the US (Whites) 
and South Korea, highly representative of individual- and group-oriented cultures, 
respectively, across a broader range of stigmatized groups.  I also examined cultural 
differences in associations between competition for employment and prejudice.   
In the present research, I examined social distance toward eight groups that more 
broadly represent social categories characterized based on a personal quality (e.g., heavy 
drinkers) or groups that have a relatively more entitative (i.e., more similar and cohesive) 
  19 
 
and essential quality (e.g., people of a different race).  While these groups generally 
correspond to Goffman?s (1963) distinction between blemishes of character and tribal 
stigmas, I refer to these groups as personal quality groups and essentially different 
groups, respectively, to acknowledge historical shifts in the way some groups may be 
perceived and to consider cultural differences in perceptions of these groups due to 
different norms/values in each culture.  Among various aspects of the personal and 
essential qualities, I focused on groups perceived as having common personal qualities 
that determine their group membership and groups perceived as being essentially 
different from the majority or typical ingroups of the society.  Personal quality groups 
included drug addicts, heavy drinkers, people with a criminal record, and emotionally 
unstable people.  Essentially different groups were Muslims, people of a different race, 
immigrants/foreign workers, and homosexuals. I note that Goffman (1963) classified 
homosexuals as individuals with blemishes of character, a personal quality.  However, 
World Health Organization (WHO) officially removed homosexuality from the list of 
mental disorders in 1990 and widely publicized research identified potential biological 
predispositions in homosexuality (e.g., LeVay, 1991).  Also, gay and lesbian groups have 
coordinated collective actions and achieved greater legal acceptance in many countries, 
including the US (see Marcus, 2002). Thus, I included homosexuals to essentially 
different groups.   
I hypothesized that prejudice toward groups based on personal qualities would be 
greater in the US than in South Korea, because people are typically seen as responsible 
for their personal qualities and perceived controllability of a stigmatizing ?mark? is one 
of the most potent predictors of bias (Dovidio et al., 2000) in individual-oriented cultures.  
  20 
 
Because in individual-oriented cultures, personal choices, efforts, and responsibilities are 
emphasized (Katz & Hass, 1988; Quinn & Croker, 1999; Rokeach, 1973), stigmatized 
groups whose memberships are perceived as being controllable by personal choice and 
efforts (e.g., heavy drinkers) would become targets of greater prejudice (see Crandall & 
Martinez, 1996) than in group-oriented cultures.   
I also hypothesized that prejudice toward groups perceived as being essentially 
different would be greater in South Korea than in the US, because of the greater emphasis 
on the holistic/essential quality (see Nisbett, 2003) and relations within ingroups (Fiske et 
al., 1998; Gardner & Seeley, 2001, Triandis, 1994) in group-oriented cultures.  Because 
in group-oriented cultures, holistic/essential qualities of individuals and groups are 
emphasized and the relation among ingroup members is a primary concern (e.g., Brett & 
Gelfand, 2006), groups that are perceived as being essentially different from the majority 
or the typical ingroups (e.g., different ethnic groups) would likely become targets of 
prejudice than in individual-oriented cultures.   
I also examined whether perceived realistic competition for employment, one of 
the well-established predictors of prejudice (see Dovidio et al., 2010), is associated with 
prejudice toward the eight target groups universally across cultures.  I examined support 
for a policy favoring ingroup members over immigrants when jobs are scarce as a 
measure of competitive intergroup orientation.  In the prejudice literature, competition 
typically refers to an individual?s perception that another person?s or another group?s gain 
will result in a loss of resources for one?s self or for one?s group (Campbell, 1965; Esses, 
Jackson, & Armstrong, 1998; Sherif, Harvey, White, Hood, & Sherif, 1961).   
  21 
 
Based on a range of theories, such as the Realistic Group Conflict Theory 
(Campbell, 1965) and the Unified Instrumental Model of Group Conflict (Esses, Jackson, 
Dovidio & Hodson, 2005), the perception of competition with immigrants for 
employment is closely associated with prejudice.  However, cross-cultural comparisons 
on competition has found that the zero-sum competition (see Duckitt, 2005; Esses, 
Dovidio, Jackson, & Armstrong, 2001; Sidanius & Pratto, 1999), in which one?s gain or 
loss equals others? complete losses or gains, is observed more in individual- than group-
oriented cultures (Leung, Au, Fernandez-Dols, & Iwawaki, 1992; Leung, Bond, Carment, 
Krishman, & Liebrand, 1990).  Individual-oriented cultures also have a greater emphasis 
on equity of outcomes, whereas group-oriented cultures emphasize equality (Kashima, 
Siegal, Tanaka, & Isaka, 1988).  In addition, the Protestant Ethic encourages the 
achievement of individual wealth and consequently capitalistic concerns become the core 
value of the US culture (Sidanius & Pratto, 1999), whereas social networks rather than 
maximizing economic capital are emphasized in South Korean culture (Brett & Gelfand, 
2006).  Thus, I hypothesized that perceived competition for employment would be more 
strongly and positively associated with prejudice toward more various groups in 
individual-oriented cultures than in group-oriented cultures.        
Method 
Participants.  I examined data from the World Values Survey (WVS; 
www.wordlvaluessurvey.org), an international study of socio-cultural and political 
changes by social scientists from approximately 100 different countries.  Participants 
were recruited by a representative sampling method from the US (n = 3008 Whites) and 
South Korea (n = 3649). Across the two samples, there were slightly more women 
  22 
 
(51.5%) than men (48.5%), X2 (N = 6657, df = 1) = 6.07, p = .014, and more women in 
the US (53.0%) than in South Korea (50.3%), X2 (N = 6657, df = 1) = 4.83, p = .028, ? = 
-.027, p = .028.  Ages ranged between 17 and 94 years (M = 43.44, SD = 15.81).  US 
participants (M = 48.16, SD = 17.42) were older than South Koreans (M = 39.57, SD = 
13.14), t(5465.80) = 22.27, p < .001, ?2p = .07.  Education levels were coded on an 8-
point WVS standardized scale from 1, ?did not complete elementary education? to 8, 
?university with degree/higher education.?  Education ranged from 1 to 8 (M = 5.81, SD 
= 1.94).  South Koreans (M = 6.18, SD = 1.80) were more educated than US participants 
(M = 5.36, SD = 2.00), t(6108.30) = -17.22, p < .001, ?2p = .04.  Income levels were 
coded on a 10-point WVS standardized scale.  Income ranged between 1 and 10 (M = 
5.47, SD = 2.21), with US participants (M = 5.92, SD = 2.43) having slightly higher 
income than South Korean participants (M = 5.12, SD = 1.95), t(5203.77) = 14.09, p = 
.056, ?2p = .03. 
Procedure.  The American (Gallup, 1995; Inglehart, 1999; Inglehart, Larsen, & 
Miller, 2006) and South Korean (Auh, 2001; Auh & Han, 1996; Auh & Han, 2005) 
investigators collected data through face-to-face interviews at three time points between 
1994 and 2008.  Responses were made available to other investigators at the World 
Values Survey website.  I selected prejudice and competition items in addition to 
demographic variables for analyses.     
The main dependent variable, prejudice, was assessed by perceived social 
distance toward eight target groups.  The survey asked respondents whether there were 
any groups (from a standard list) that they ?would not like to have as neighbors.?  The 
responses were coded as 0 (?not mentioned?) or 1 (?mentioned?) for each group.  
  23 
 
Negative responses (i.e., mentioned as a group participants would not like to have as their 
neighbor) reflected social distance to each target group.  There was no measure of the 
degree of bias against the target groups.  I included the eight groups that were asked at 
least once across the two samples and that are generally targets of prejudice across the 
two countries.  The eight groups were people with a criminal record; people of a different 
race; heavy drinkers; emotionally unstable people; Muslims; immigrants/foreign workers; 
drug addicts; and homosexuals (presented in this order).  I conducted analyses separately 
for each target group and for two combined group types (personal quality and essentially 
different groups) across and between the two cultures.  The independent variable, culture 
(individual- versus group-oriented) was examined with the US and South Korean samples 
to test cultural differences in the degree of prejudice.  In addition, perceived competition 
for employment was examined as a predictor/mediator of prejudice between the US and 
South Korea.  One item related to feelings of competitiveness with other groups: ?When 
jobs are scarce, employers should give priority to [US/ South Korean] people over 
immigrants.?  The responses were coded as 1 (?disagree?), 2 (?neither?), or 3 (?agree?).   
Preliminary data on the target groups.  I collected preliminary data in the US 
(n = 27 Whites) and South Korea (n = 34) to assess whether the perception of 
membership changeability of each target group is similar across the two cultures and to 
explore how the target groups clustered.  I first asked participants to rate each of the eight 
groups on ?How easy or difficult is it to change the characteristics or identities of 
people?? from 1, ?very difficult to change? to 9, ?very easy to change.?  Ratings on the 
changeability of membership would indicate the extent to which participants perceive 
controllability versus the essential quality of each target group.  Mean ratings of the eight 
  24 
 
target groups between the two samples were highly correlated, r(6) = .92, p < .001, and 
across the eight target groups, greater perceived essential quality predicted more 
prejudice (the mean proportions of negative responses to each target group presented in 
Table 1) in South Korea relative to the US, r(6) = -.64, p = .043.   
I, then, conducted an exploratory factor analysis (principle component analysis) 
on the perceptions of membership changeability of eight target groups with varimax 
rotation.  Across the two cultures, the exploratory factor analysis yielded two factors.  
One factor (eigenvalue = 1.58, 19.68% of the variance) reflected groups characterized 
based on a personal quality:  drug addicts (loading = .84), heavy drinkers (.79), people 
with a criminal record (.56), and emotionally unstable people (.68).  The other factor 
(eigenvalue = 2.60, 32.53% of the variance) reflected groups perceived as having an 
essential quality that is different from the characteristics of the majority or the typical 
ingroups (i.e., essentially different groups):  immigrants/ foreign workers (.60), Muslims 
(.61), people of a different race (.64), and homosexuals (.64).  In addition, we conducted 
a 2 (Culture: US vs. South Korea) x 2 (Group Type: personal quality vs. essentially 
different groups) mixed-model analysis of variance (ANOVA) on the mean ratings of the 
two group types by each participant.  There was a significant main effect of Group Type, 
F(1, 59) = 58.60, p < .001, ?2p = .50.  Essentially different groups were rated as having 
qualities that were less amenable to change (a lower score) than personal quality groups 
(M = 3.30, SD = 1.12 vs. 4.83, SD = 1.31).  There was no main effect of Culture (p = 
.419, ?2p = .01) and no Culture x Group Type interaction (p = .627, ?2p = .004).   
The main analyses were thus based on the two clusters of groups:  (a) personal 
quality groups (drug addicts, heavy drinkers, people with a criminal record, and 
  25 
 
emotionally unstable people) and (b) essentially different groups (Muslims, people of a 
different race, immigrants/foreign workers, and homosexuals).   
There were some missing data.  For the US sample, ?people with a criminal 
record,? ?emotionally unstable people? and ?Muslims? were not asked in 2006.  For 
South Koreans, ?people of a different race? was not asked in 1996; ?Muslims? was not 
asked in 1996 and 2005; and ?people with a criminal record? and ?emotionally unstable 
people? were not asked in 2005.  The three waves were combined to include more target 
groups.  There was no systematic missing data for competition across the two samples. 
Results 
For Study 1, I first tested cultural differences between the US and South Korea in 
the degree of prejudice (assessed by social distance) against the eight target groups 
individually and by two types.  I then examined cultural differences in the hypothesized 
mediator of bias, perceived competition for employment.     
Cultural differences in the degree of prejudice.  Table 1 presents cultural 
differences in proportions of negative responses (mentioned as a group participants 
would not like to have as their neighbor) to each of the eight target groups between the 
US and South Korea.  The negative response to each target group was interpreted as the 
degree of perceived social distance to each target group.  As predicted, US participants 
perceived greater distance to personal quality groups (drug addicts, heavy drinkers, 
people with a criminal record, and emotionally unstable people)  than South Korean 
participants, whereas South Korean participants perceived greater distance to essentially 
different groups (Muslims, people of a different race, immigrants/foreign workers, and 
homosexuals) than US participants.  All differences were statistically significant (p < 
  26 
 
.001) and the differences remained significant when they were controlled for gender, age, 
education, and income.   
I, then, conducted a 2 (Culture: US vs. South Korea) x 2 (Group Type: personal 
quality vs. essentially different groups) mixed-model analysis of variance (ANOVA) on 
the mean proportions of negative responses (perceived social distance) to the two types of 
groups by each participant.  There was a significant main effect of Culture, F(1, 6649) = 
323.71, p < .001, ?2p = .05.  Across the two types of target groups, South Korean 
participants (M = .54 SD = .34) perceived greater distance than US participants (M = .42, 
SD = .21).  There was a significant main effect of Group Type, F(1, 6649) = 3900.60, p < 
.001, ?2p = .37.  Across the two cultures, the distance to personal quality groups (M = .64, 
SD = .38) was greater than the distance to essentially different groups (M = .33, SD = 
.38).  Also, there was a significant Culture x Group Type interaction, F(1, 6649) = 
1653.20, p < .001, ?2p = .20.  As hypothesized, US participants (M = .69, SD = .32) 
perceived greater distance to personal quality groups than South Korean participants (M = 
.60, SD = .42), F(1, 6651) = 89.70, p < .001, ?2p = .01.  By contrast, South Korean 
participants (M = .49, SD = .41) perceived greater distance to essentially different groups 
than US participants (M = .14, SD = .24), F(1, 6649) = 1659.95, p < .001, ?2p = .20.  The 
cultural difference was greater for essentially different groups (M = .14 vs. M = .49; ?2p = 
.20) than for personal quality groups (M = .69 vs. M = .60; ?2p = .01).  Within the US, the 
distance was greater for personal quality groups than for essentially different groups (M = 
.69 vs. M = 14), F(1, 3004) = 6269.09, p < .001, ?2p = .68.  Within South Korea, the 
distance was also greater for personal quality groups than for essentially different groups 
(M = .60 vs. M = .49), F(1, 3645) = 221.50, p < .001, ?2p = .06.  The group type 
  27 
 
difference was greater in the US (M = .69 vs. M = .14; ?2p = .68) than in South Korea (M 
= .60 vs. M = .49; ?2p = .06).  Figure 1 shows mean proportions of negative responses 
(social distance) to personal quality and essentially different groups in the US and South 
Korea.   
I also examined cultural differences in the pervasiveness of prejudice (the number 
of the groups a respondent did not want as a neighbor).  I conducted a 2 (Culture: US vs. 
South Korea) x 2 (Group Type: personally quality vs. essentially different groups) mixed-
model analysis of variance (ANOVA) on the number of groups mentioned by each 
participant.  There was a significant main effect of Culture, F(1, 6649) = 13304.45, p < 
.001, ?2p = .67.  South Korean participants (M = 3.21, SD = 2.48) mentioned more groups 
than US participants (M = 2.78, SD = 1.53).  There was a significant main effect of Group 
Type, F(1, 6649) = 5028.07, p < .001, ?2p = .43.  Participants mentioned personal quality 
groups (M = 2.02, SD = 1.33) more than essentially different groups (M = .99, SD = 
1.20).  Also, there was a significant Culture x Group Type interaction, F(1, 6649) = 
1677.37, p < .001, ?2p = .20.  As hypothesized, US participants (M = 2.25, SD = 1.15) 
mentioned personally quality groups more than South Korean participants (M = 1.84, SD 
= 1.44), F(1, 6651) = 166.40, p < .001, ?2p = .02.  By contrast, South Korean participants 
(M = 1.37, SD = 1.29) mentioned essentially different groups more than US participants 
(M = .52, SD = .90), F(1, 6649) = 934.51, p < .001, ?2p = .12.  The cultural difference was 
greater for essentially different groups (M = .52 vs. M = 1.37; ?2p = .12) than for 
personally quality groups (M = 2.25 vs. M = 1.84; ?2p = .02).  Within the US, participants 
mentioned personal quality groups more than essentially different groups (M = 2.25 vs. M 
= .52), F(1, 3004) = 4715.54, p < .001, ?2p = .61.  Within South Korea, participants also 
  28 
 
mentioned personal quality groups more than essentially different groups (M = 1.84 vs. M 
= 1.37), F(1, 3645) = 600.48, p < .001, ?2p = .14.  The group type difference was greater 
in the US (M = 2.25 vs. M = .52; ?2p = .61) than in South Korea (M = 1.84 vs. M = 1.37; 
?2p = .14).     
Cultural differences in the degree of norms/values.  Perceived competition for 
employment was higher in South Korea (M = 2.80, SD = .47) than in the US (M = 2.31, 
SD = .88), t(4293.28) = -27.27, p < .001, ?2p = .11.  The difference remained significant 
when controlled for gender, age, education, and income.   
Cultural differences in dynamics of prejudice.  To examine the relationship 
between perceived competition for employment and prejudice, including the potential 
moderating effect of culture, I conducted logistic regressions with Culture (US vs. South 
Korea), Competition, and their interaction (see Aiken & West, 1991) as predictors of 
perceived social distance to each of eight target groups (not mentioned vs. mentioned as a 
group they would not like to have as their neighbor).  The main effect of Culture for each 
of the eight target groups was examined in cultural differences in the degree of prejudice 
(see Table 1).   
Table 2 presents the main effects of Competition in the total sample and in each 
culture and the interaction effects of Culture x Competition for each of the eight target 
groups based on odds ratios and Wald statistics.  The main effect of Competition was 
significant and positive only for essentially different groups.  Across cultures, perceived 
competition for employment was significantly positively associated with perceived social 
distance to the four essentially different groups (i.e., Muslims, people of a different race, 
immigrants/ foreign workers, and homosexuals).  Competition was significantly but 
  29 
 
negatively associated with social distance to drug addicts.  These associations remained 
significant when they were controlled for gender, age, education and income.  There was 
no association between perceived competition for employment and perceived social 
distance to the other three personal quality groups (i.e., heavy drinkers, people of a 
criminal record, and emotionally unstable people).   
However, I found significant Culture x Competition interactions in the opposite 
direction (i.e., positive associations in the US and negative/marginally significant 
associations in South Korea) for all groups except for Muslims and homosexuals, and 
these interactions remained significant when analyses controlled for gender, age, 
education and income.  The interaction for Muslims was significant and in the opposite 
direction based on odds ratio but not significant based on Wald statistics.  The interaction 
for homosexuals was significant and in the same direction (i.e., positive associations both 
in the US and in South Korea), but attenuated to non-significance when controlled for 
gender, age, education, and income.   
I, then, examined associations between competition and social distance to each of 
eight target groups separately in the US than in South Korea.  As hypothesized, in the 
US, associations between competition and social distance were significant and positive 
for all eight groups, and these associations remained significant when they were 
controlled for gender, age, education, and income.  The association was particularly large 
for immigrants/foreign workers and homosexuals.  In South Korea, however, associations 
between competition and social distance were significant and positive only in the four 
essentially different groups (i.e., Muslims, people with a different race, 
immigrants/foreign workers, and homosexuals), and these associations were much 
  30 
 
smaller than those in the US and attenuated to marginally significance (Muslims and 
immigrants/foreign workers) or non-significance (people of a different race) when they 
were controlled for gender, age, education and income, except for homosexuals.  Only 
the association between competition and social distance to homosexuals was significant 
and positive in South Korea.  Moreover, competition was significantly negatively 
associated with social distance to the four personal quality groups (drug addicts, heavy 
drinkers, people with a criminal record, and emotionally unstable people) based on odds 
ratios, although, based on Wald statistics, the association for people with a criminal 
record was non-significant and the association for emotionally unstable people was 
marginally significant when controlling for gender, age, education, and income.   
I also conducted regression analyses with Culture, Competition, and their 
interaction as predictors of mean negative responses (perceived social distance) to each of 
two types of target groups.  The results were consistent with the results based on the 
individual target groups.  Across cultures, competition did not predict social distance to 
personal quality groups, b = .01, SE = .01, p = .29, B = .01, but predicted essentially 
different groups, b = .12, SE = .01, p < .001, B = .23.  The Culture x Competition 
interaction was significant for personal quality groups.  The simple slope tests showed 
that competition predicted social distance to personal quality groups positively in the US, 
b = .12, SE = .02, p < .001, B = .19, but negatively in South Korea, b = -.12, SE = .02, p < 
.001, B = -.12.  All significant effects remained significant when analyses controlled for 
gender, age, education and income.  The Culture x Competition interaction was also 
significant for essentially different groups.  The simple slope tests showed that 
competition marginally significantly predicted social distance to essentially different 
  31 
 
groups positively in the US, b = .03, SE = .01, p = .055, B = .04, but negatively in South 
Korea, b = -.03, SE = .01, p = .055, B = -.03, and these results became significant when 
controlling for gender, age, education, and income, b = .03, SE = .01, p = .024, B = .05, 
and b = -.03, SE = .01, p = .024, B = -.03, respectively.   
I, then, examined associations between competition and social distance to the two 
types of target groups separately in the US and in South Korea.  As expected, in the US, 
competition predicted social distance to both personal quality and essentially different 
groups significantly and positively, b = .06, SE = .01, p < .001, B = .17, and b = .06, SE = 
.005, p < .001, B = .20, respectively, and these associations remained significant when 
they were controlled for gender, age, education, and income.  In South Korea, 
competition predicted social distance to personal quality groups significantly negatively, 
b = -.06, SE = .02, p < .001, B = -.06.  Competition predicted social distance to essentially 
different groups significantly positively, b = .03, SE = .01, p = .038, B = .04, but the 
association attenuated to non-significance when controlled for gender, age, education, 
and income, b = .02, SE = .02, p = .169, B = .02.   
Discussion 
Study 1 demonstrated both systematic similarities and differences in the degree of 
prejudice between the United States (Whites) and South Korea, countries representing 
individual- and group-oriented cultures.  I examined two types of target groups that are 
relevant to prejudice, groups whose membership was based on personal qualities (drug 
addicts, heavy drinkers, people with a criminal record, and emotionally unstable people) 
and groups perceived as essentially different from the majority or typical ingroups of the 
  32 
 
society (whose members are seen as less personally responsible for their membership; 
Muslims, people with a different race, immigrants/foreign workers, and homosexuals).   
Based on the perceived social distance to these eight target groups, Study 1 
revealed systematic differences between the US and South Korea in the degree of 
prejudice.  White US (vs. South Korean) participants reported greater social distance to 
personal quality groups, whereas South Korean (vs. US) participants reported greater 
social distance to groups perceived as essentially different from the majority of typical 
ingroups.  These cultural differences in the degree of prejudice were greater for 
essentially different groups than for personal quality groups.  Across the two cultures, 
prejudice was greater for personal quality groups than for essentially different groups.   
These findings imply that norms/values of each culture or society are closely 
associated with prejudice and targets of prejudice can be decided when a social group or 
category is perceived as being wrong by the standard of norms/values of the culture or 
society they belong to or being different from the majority or typical ingroups of the 
culture or society they belong to.  These two criteria can be used in both individual- and 
group-oriented cultures to decide whether a group/category can be a ?legitimate? target of 
prejudice or not.  Particularly, whether a group/category is right or wrong by the standard 
of norms/values of the society seems to be a more universal criterion to decide who the 
target of prejudice can be.  Across the two cultures, respondents expressed greater social 
distance to groups perceived as being personally responsible for their group membership 
than to groups perceived as essentially different from the majority or typical ingroups of 
the society.  This is also consistent with Jones et al.?s (1984) conclusion that perceptions 
of controllability of a stigma is one of the most important determinants of negative 
  33 
 
behavior toward the groups.  Our results showed the cross-cultural generalizability of that 
conclusion.  By contrast, whether a group/category is similar to or different from the 
majority or typical ingroups of the society can be a more important criterion to decide the 
target of prejudice in group-oriented cultures than in individual-oriented cultures.   
Study 1 also revealed that there were cultural differences in dynamics of 
prejudice.   As hypothesized, associations between the perception of competition for 
employment with immigrants and prejudice were significantly positive for all eight target 
groups in the US, in which capitalistic/meritocratic achievements are emphasized based 
on the Protestant Ethic, but not in South Korea.  The perception of competition predicted 
prejudice toward essentially different groups in both cultures, but more strongly in the 
US, and competition predicted prejudice toward personal quality groups in the US, but 
not in South Korea.  Greater associations between competition and prejudice in the US 
can be based on the core cultural values of the US.   
The associations were particularly large for immigrants/foreign workers and 
homosexuals.  Among the eight target groups, immigrants/foreign workers were the 
direct competitor for employment in the competition measure of this study.  In the US, 
the perception of competition predicted prejudice significantly toward each of the groups 
but primarily toward immigrant/foreign workers (see Table 2).  By contrast, in South 
Korea the perception of competition predicted prejudice marginally significantly toward 
essentially different groups ? and not distinctively toward immigrant/foreign workers in 
this set ? and not toward personal quality groups.  Thus, in both the US and South Korea, 
these perceptions of realistic competition were associated with prejudice toward other 
targets (personal quality and essentially different groups in the US and essentially 
  34 
 
different groups in South Korea).  This pattern suggests that perceptions of competition 
may be rooted in a more general norm/value that promotes intergroup biases generally, 
such as social dominance (Sidanius & Pratto, 1999) or concerns for status and safety in 
general.  Consequently, competition predicted prejudice toward groups that are not 
directly related to economic threat, such as homosexuals.  Bias toward homosexuals 
typically represents symbolic threat (Stephan & Stephan, 2000).  The lack of a 
distinctively strong relationship between perceived competition with immigrant/foreign 
workers compared to other essentially different groups in South Korea may reflect not 
only the emphasis on group-based perceptions of differences among groups in South 
Korea but also the fact that associations between competition and prejudice are based on 
a more general intergroup orientation or ideology (like social dominance orientation or 
concerns for status and safety), which is further accentuated by the emphasis on 
capitalistic/meritocratic values in the US. 
Unexpectedly, stronger perceptions of competition with immigrants for 
employment were associated with less prejudice toward personal quality groups in South 
Korea. In South Korea, associations between competition for employment and prejudice 
were negative (i.e., greater competition was associated with less negative attitudes) 
toward personal quality groups, especially for drug addicts.  This unanticipated finding 
may be because competition between groups deflects negative attention away from 
individual-level biases in group oriented cultures (i.e., biases against personal quality 
groups were relieved focusing on biases that are more culturally important at the group 
level).  The higher level of competition for employment in South Korea may be because 
of the national financial crisis during the survey period (the International Monetary Fund 
  35 
 
Era [1997-2001]) in South Korea or reflect high employment competition rates in many 
Asian countries in general.   
Although the overall pattern of findings was supportive of the predictions that 
guided Study 1, I note some methodological and conceptual limitations of the study.  One 
methodological limitation of Study 1 is that the World Values Survey assessed perceived 
competition for employment in a specific way, asking participants? agreement with the 
item, ?When jobs are scarce, employers should give priority to [US/South Korean] people 
over immigrants.?  In general, the consequences of competition may vary depending on 
who competes with whom for what.  However, the item used in the study not only 
focused on employment opportunities that are closely associated with the Protestant Ethic 
and capitalistic/meritocratic achievement (see Campbell, 1965; Esses, Jackson, & 
Armstrong, 1998) as we intended, but also had immigrants as competing targets.  Having 
immigrants as competing targets can lead participants to focus either on the competition 
for employment or on the competition between our nation people and immigrants 
measuring threats from xenophobia or ingroup favoritism.  It is also possible that other 
measures of competition focused more on group-level symbolic threats to the safety and 
wellness of ingroups could predict prejudice more strongly in group-oriented cultures. 
The limitation was considered in Study 2 using more specific and concrete items.  In 
addition, the World Value Survey asked participants whether or not they were prejudiced 
against certain groups, but Study 2 examined the degree of intergroup bias.   
Conceptually, a limitation of Study 1 involves that we measure prejudice with 
social distance (the unwillingness to have members of certain groups as neighbors), thus 
we may need to measure prejudice with a different way (e.g., hierarchy beliefs).  Also, I 
  36 
 
examined two types of groups, personal quality groups, whose membership is perceived 
as primarily determined by social factors, and essential quality groups, whose 
membership is perceived as primarily determined by biological factors and found 
systematic differences in the degree of prejudice toward these two types of target groups 
between the two culture.  Thus, the distinction between groups that are based on social 
and biological criteria may be an important aspect of targets of prejudice with 
implications for different types of hierarchically-based intergroup biases.  The former 
(being wrong by the standard of norms/values of the society) can be more related to the 
hierarchy beliefs that are based on the achieved social status or reputations, whereas the 
later (being different from the majority or typical ingroups of the society) can be more 
related to the hierarchy beliefs that are based on the given genetic quality or physical 
conditions.  I thus further examined these two types of hierarchy beliefs in Study 2. 
  37 
 
Chapter 3: Study 2 
Study 2 continued to investigate cultural differences in the degree and dynamics 
of prejudice (assessed by hierarchically-based intergroup biases) between individual- and 
group-oriented cultures, specifically the US and South Korea.  However, Study 2 
extended Study 1 by measuring prejudice with hierarchically-based intergroup biases and 
by examining other norms/values that were not examined in Study 1.  I examined one 
variable that represents individual-oriented norms/values (the emphasis on individual 
achievements) and two variables that represent group-oriented norms/values (the 
emphasis on the self-ingroup overlap and roles/positions within ingroups) as potential 
predictors of hierarchically-based intergroup biases.   
Also, whereas Study 1 investigated social distance to two types of groups, 
classified based on personal or essential qualities, Study 2 examined more general 
concepts of prejudice that are related to the two types of groups examined in Study 1:  the 
support for intergroup hierarchies that are based on social or biological factors.  The 
personal quality groups in Study 1 (e.g., heavy drinkers) are the groups defined by their 
social behavior (socially-based), whereas the essentially different groups (e.g., people of 
a different race) are the groups whose membership is perceived as less changeable or 
controllable than personal quality groups (biologically-based).   
Social and biological hierarchy beliefs have been considered as the key 
components and major facets of prejudice (see Dovidio et al., 2010; Sidanius & Pratto, 
1999).  Social hierarchy beliefs reflects the endorsement of social order based on 
achieved status or social reputations (e.g., beliefs in school rankings), whereas biological 
hierarchy beliefs represent the support of social order based on given genetic quality or 
  38 
 
physical conditions (e.g., beliefs in the genetic superiority/inferiority of a race or nation).  
These two types of hierarchy beliefs also represent two types of targets of prejudice.  
Social hierarchy beliefs lead to prejudice toward groups or categories that are perceived 
as achieving relatively less or having inferior social reputations than other groups in the 
society (e.g., low socio-economic status groups).  Biological hierarchy beliefs lead to 
prejudice toward groups or categories that are perceived as having relatively inferior 
given genetic or essential qualities or physical conditions.   
I examined social and biological hierarchy beliefs separately, not only because the 
distinction between the two would provide more concrete and unique associations with 
predictors of prejudice investigated in Study 2, but also because I can investigate how 
various cultural norms/values are differently associated with the perception of the 
achieved (socially-based) and ascribed (biologically-based) group status within societal 
hierarchies.  I hypothesized that both social and biological hierarchy beliefs would be 
stronger in South Korea than in the US because of the greater emphasis on hierarchical 
relations in group-oriented cultures.  I also hypothesized that social hierarchy beliefs 
would be stronger than biological hierarchy beliefs across the two cultures because of the 
greater emphasis on achieved status than ascribed status in general.   
In addition to competition for employment (the realistic/capitalistic concern) in 
Study 1, I examined the emphasis on individual achievements (the meritocratic concern) 
in Study 2, another variable that represents individual-oriented norms/values.  
Meritocracy, an ideology based on the Protestant Ethic, emphasizes individual 
achievements.  Meritocracy beliefs often lead individuals to overlook the influence of 
situational (see Gilbert & Malone, 1995; Ross, 1977) or structural factors relevant to the 
  39 
 
current achievement status, and evaluate the socio-economically lower status negatively 
(making them a target of prejudice) to legitimize the position of dominant groups 
(Sidanius & Pratto, 1999) or justify the status quo (Jost et al., 2004).  Thus, the emphasis 
on individual achievements would be more closely associated with socially-based 
hierarchy beliefs than biologically-based.  However, the emphasis on individual 
achievements would directly predict prejudice particularly toward the socio-economically 
lower status or social hierarchy beliefs only in individual-oriented cultures, because, 
corresponding to their own norms/values, group-oriented cultures may not consider 
individual achievement much when they decide targets of prejudice.  Cross-cultural 
research also showed that motivation to achieve personally is associated with 
accumulation of economic capital in most Western countries (Schwartz, 1992, 1994) and, 
in negotiation, the primary concern of Western cultures is maximizing economic capital; 
by contrast, in non-Western cultures, the dominant motive resides in social networking or 
relations (social or relational capital) (Brett & Gelfand, 2006).  I, thus, hypothesized that 
the emphasis on individual achievements would directly predict social hierarchy beliefs 
(i.e., prejudice toward groups characterized as relatively less achieving within a 
hierarchy) in individual-oriented cultures but not in group-oriented cultures.   
Because Study 1 suggested that, in group-oriented cultures, there would be other 
predictors of prejudice that are more strongly associated with prejudice than competition 
for employment, in Study 2, I examined two group-oriented norms/values, the emphasis 
on the self-ingroup overlap and roles/positions within ingroups, that would be associated 
with prejudice.  These two group-oriented norms/values are also related to the well-
known cultural differences in the self-concept, independent versus interdependent 
  40 
 
(Markus & Kitayama, 1991), because the self-concept would reflect the most stable 
norms/values of the society.  In this study, I elaborated on being interdependent (the self-
concept of group-oriented cultures) by examining being interdependent with ingroups 
(the emphasis on the self-ingroup overlap) and ingroup members (the emphasis on 
roles/positions within ingroups) separately, because we have different predictions for 
each of them and for the latent factor that represents the two.       
I expected that the two group-oriented norms/values would have a common latent 
factor, group-oriented norms/values.  I hypothesized that this latent factor would directly 
predict both social and biological hierarchy beliefs in both individual- and group-oriented 
cultures because prejudice is an intergroup phenomenon rooted in perceptions of the self 
in relation to group memberships (ingroup and outgroup classifications; Tajfel & Turner, 
1979).  However, I also hypothesized that the associations between the latent factor that 
represents group-oriented norms/values and hierarchically-based intergroup biases would 
be greater in group-oriented than individual-oriented cultures because the greater 
centrality of group-oriented norms/values in group-oriented than in individual-oriented 
cultures.   
I also expected that the emphasis on the self-ingroup overlap would be associated 
with both social and biological hierarchy beliefs in both individual- and group-oriented 
cultures, but indirectly, for example, as an indicator of the latent factor that represents 
group-oriented norms/values because of the lack of a direct relation between the two.  In 
group-oriented cultures, the self-concept is developed based more on memberships of 
various ingroups or social categories (e.g., family, hometown, nation, and organization) 
than on individual?s unique characteristics, so ingroups are perceived as the extended self 
  41 
 
and the degree of self-ingroup overlap is greater than in individual-oriented cultures.  As 
consequences, people in group-oriented cultures perceive the common fate with ingroups 
(Triandis, 1994), support ingroups? values and decisions (Cha, 1994), and frequently 
place primary weight on the existence and importance of groups over individuals 
(Gardner & Seeley, 2001) than those in individual-oriented cultures.  Schaberg (2002) 
also found that explicitly expressing or emphasizing an individual?s preference or choice 
in a group setting is considered ?immature? or ?self-centered? in group-oriented cultures.  
Thus, we hypothesized that the emphasis on the self-ingroup overlap would be greater in 
group-oriented cultures than in individual-oriented cultures.  I also hypothesized that the 
emphasis on the self-ingroup overlap would indirectly predict both social and biological 
hierarchy beliefs both in individual- and group-oriented cultures (i.e., as an indicator of 
the latent factor that represents group-oriented norms/values in the two-factor model and 
in the relations with other predictors in three-factor models).   
I also expected that the emphasis on roles/positions within ingroups would 
indirectly predict both social and biological hierarchy beliefs in both individual- and 
group-oriented cultures indirectly as an indicator of the latent factor that represents 
group-oriented norms/values, as discussed earlier.  The emphasis on roles/positions 
within ingroups, however, would also directly predict both social and biological 
hierarchy beliefs particularly in group-oriented cultures.  In general, hierarchical relations 
are necessary to make a group of individuals to function as a group, which is particularly 
important in group-oriented cultures where groups are vital units of the society.  In 
group-oriented cultures, relationships among and with ingroup members are emphasized, 
so individuals tend to define self and others largely based on relations with others (e.g., 
  42 
 
I?m a student of Dr. Kim) emphasizing the roles or positions within a group (Naoi & 
Schooler, 1985), which enhance the existing hierarchies within the group.  Also, because 
most roles/positions within a relation are not identical or equal providing some power 
difference between them, the emphasis on roles/positions within a relation would also 
induce hierarchies within the relation.  These hierarchically networked relations among 
ingroup members may foster the endorsement of hierarchically-based intergroup biases.  I 
hypothesized that the emphasis on roles/positions within ingroups would be greater in 
group-oriented than in individual-oriented cultures.  I also hypothesized that the emphasis 
on roles/positions within groups would directly predict both social and biological 
hierarchy beliefs only in group-oriented cultures, because, corresponding to their own 
norms/values, individual-oriented cultures may not consider roles/positions within a 
group when they decide targets of prejudice.  The emphasis on roles/positions within 
ingroups would also indirectly predict both social and biological hierarchy beliefs in both 
individual- and group-oriented cultures as an indicator of the latent factor that represents 
group-oriented norms/values.   
Study 2, therefore, employed questionnaires that measure the tendency to 
emphasize (a) individual achievements, (b) the self-ingroup overlap, and (c) 
roles/positions within ingroups, as well as the tendency to support (d) social hierarchy 
beliefs and (e) biological hierarchy beliefs.  I used college student samples in the US (the 
total sample and Whites) and South Korea.  I first examined cultural differences in the 
degree of hierarchically-based intergroup biases.  I hypothesized that both social and 
biological hierarchy beliefs would be greater in South Korea than in the US.  I also 
  43 
 
hypothesized that social hierarchy beliefs would be stronger than biological hierarchy 
beliefs across the two cultures.    
I then examined cultural differences in dynamics of hierarchically-based 
intergroup biases proposing two structural equation models.  In the two-factor model, 
where I examined the two latent factors that represent individual- and group-oriented 
norms/values as predicts of prejudice, I hypothesized that individual-oriented 
norms/values would directly predict social hierarchy beliefs in the US (but not in South 
Korea), whereas group-oriented norms/values would directly predict both social and 
biological hierarchy beliefs both in the US and South Korea.  I also hypothesized that the 
associations between the group-oriented norms/values and hierarchically-based 
intergroup biases would be greater in South Korea than in the US.  In the three-factor 
model, where I examined the three norms/values as predictors of prejudice, I 
hypothesized that the emphasis on individual-achievements would directly predict social 
hierarchy beliefs in the US (but not in South Korea).  I also hypothesized that the 
emphasis on the self-ingroup overlap would only indirectly predict hierarchy beliefs both 
in the US and South Korea.  I also hypothesized that the emphasis on roles/positions 
within ingroups would directly predict both social and biological hierarchy beliefs in 
South Korea (but not in the US).      
Method 
Participants.  Altogether 1010 (508 US and 502 South Korean) college students 
in undergraduate psychology courses participated for credits or as a class activity.  The 
US sample was consisted of 57.5% (n = 292) non-Hispanic European, 13.6% (n = 69) 
African, 8.1% (n = 41) East Asian, 7.3% (n = 37) Hispanic, 5.7% (n = 29) other Asian 
  44 
 
Americans, 4.7% (n = 24) were interracial/multiracial, 2.0% (n = 10) Middle Easterner, 
and 0.4% (n = 2) American Indian or Alaskan Native Americans; 0.8% (n = 4) did not 
report their ethnic backgrounds.  The South Korean sample was consisted of 96% (n = 
482) Koreans and 1.8% (n = 9) non-Korean or interracial/multiracial; 2.2% (n = 11) did 
not report their ethnic backgrounds.  Across the two samples, 53.1% (n = 536) were 
women and 35.5% (n = 359) were men; 11.4% (n = 115) did not report their gender.  The 
percentage of female participants was greater in the US (64.4% female, 30.3% male, and 
5.3% missing) than in South Korea (41.6% female, 40.8% male, and 17.5% missing), X2 
(N = 895, df = 1) = 28.37, p < .001, ? = .18.  Ages ranged from 17 to 50 years (M = 
20.70, SD = 2.88).  South Korean participants (M = 21.35, SD = 3.42) were older than US 
participants (M = 20.08, SD = 2.06), t(990) = -7.10, p < .001, ?2p = .05, but they were all 
college students.   
Procedure.  Participants indicated the extent to which they agree with a list of 
statements on a 9-point scale from 1, ?totally disagree? to 9, ?totally agree,? and provided 
demographic information.  To prevent participants from recognizing the purpose of the 
different measures and creating an acquiescence bias for specific scales, items from 
different scales were listed randomly with various other statements (see also Locke & 
Baik, 2009).  The questionnaire was translated into Korean and translated back into 
English to avoid possible item biases in cross-cultural research.   
Predictor variables.  I proposed one individual- and two group-oriented 
norms/values as potential predictors of hierarchically-based intergroup biases.  An 
exploratory factor analysis (principle component analysis) on ratings of eight items with 
varimax rotation yielded three factors consistently for both cultures.  The first factor (US:  
  45 
 
eigenvalue = 2.82, 35.20% of the variance / Korea: eigenvalue = 3.10, 38.81% of the 
variance) reflected the emphasis on individual achievements, the second factor (US:  
eigenvalue = 1.86, 23.25% of the variance / Korea: eigenvalue = 1.45, 18.16% of the 
variance) reflected the emphasis on the self-ingroup overlap, and the third factor (US:  
eigenvalue = .94, 11.81% of the variance / Korea: eigenvalue = .94, 11.74% of the 
variance) reflected the emphasis on roles/positions within ingroups.  All the loadings 
were greater than .50.  Based on the exploratory factor analysis, for later analyses, we 
computed three latent factor scores using principle axis factoring (PAF) in order to 
include measurement errors in the analyses (see Winter & Dodou, 2012).  The individual-
oriented norms/values, the emphasis on individual achievements (Cronbach?s ? = .85; US 
= .85, Korea = .84) was consisted of three items.  Two group-oriented norms/values were 
computed based on five items (Cronbach?s ? = .69; US = .66, Korea = .67); the emphasis 
on the self-ingroup overlap (Cronbach?s ? = .68; US = .67, Korea = .67) was consisted of 
three items, and the emphasis on roles/positions within ingroups (r = .43; US = .45, 
Korea = .38) was consisted of two items (see Appendix for items used).     
Outcome variables.  In order to measure hierarchically-based intergroup biases 
separately for the socially- and biologically-based, I created ten items guided by work on 
the Social Dominance Orientation scale (SDO; Pratto et al, 1994) representing concrete 
examples of hierarchical relations between groups based specifically on social (e.g., 
educational attainment, income) or biological (e.g., genetic superiority, physical 
inferiority) factors.  An exploratory factor analysis (principle component analysis) on 
ratings of the ten items with varimax rotation yielded two factors in the total sample, in 
the US, and mostly in South Korea (one social item loaded better with the biological 
  46 
 
factor and one biological item loaded better with the social factor in Korea).  One factor 
(US:  eigenvalue = 5.53, 55.28% of the variance / Korea: eigenvalue = 4.96, 49.64% of 
the variance) reflected social hierarchy beliefs and the other factor (US:  eigenvalue = 
1.10, 11.02% of the variance / Korea: eigenvalue = .93, 9.34% of the variance) reflected 
biological hierarchy beliefs.  All loadings for principle components were greater than .50.  
Based on the exploratory factor analysis, we computed two latent factor scores using 
principle axis factoring (PAF) for later analyses.  Social hierarchy beliefs (Cronbach?s ? 
= .89; US = .91, Korea = .85) was consisted of five items.  Biological hierarchy beliefs 
(Cronbach?s ? = .81; US = .82, Korea = .75) was consisted of five items (see Appendix 
for items used).     
Preliminary data on the social-biological basis.  When I collected preliminary 
data in the US (n = 27 Whites) and South Korea (n = 34) to assess the membership 
changeability in Study 1, I also collected data to assess whether the perceptions of the 
social-biological basis of each target groups are similar across the two cultures.  I asked 
participants to rate each of the eight groups on ?How much are the characteristics or 
identities of people socially or biologically-based?? from 1, ?biologically-based? to 9, 
?socially-based.?  Mean ratings of the eight target groups between the two samples were 
highly correlated, r(6) = .92, p < .001, and across the eight target groups, ratings of 
membership changeability and the social-biological bases were also positively correlated 
both in the US and South Korea, r(6) = .76, p = .015 and r(6) = .68, p = .031, 
respectively.   
Results  
  47 
 
For Study 2, I first tested cultural differences between the US and South Korea in 
the degree of prejudice assessed by hierarchically-based intergroup biases (social and 
biological hierarchy beliefs).  I then tested two structural equation models to examine 
cultural differences in dynamics between individual/group-oriented norms/values and 
hierarchically-based intergroup biases.     
Preliminary analyses.  To prepare structural equation modeling analyses with 
latent variables, we conducted missing values analyses and examined data distributions.  
Missing values (US = .07%, Korea = .13%) were analyzed separately for predictor and 
outcome variables and separately for the US and Korean samples.  The missing values 
were missing completely at random (MCAR tests, Little, 1988) in each of these four data 
sets, so they were imputed using the expectation maximization (EM) algorithm (see Little 
& Rubin, 2002).  Data distributions were examined at both univariate and multivariate 
levels for each factor.  Most of individual indicators were significantly skewed and/or 
kurtotic and a few of them were considered severe at univariate level (Curren, West, 
Finch, 1996), but were not problematic in terms of the multivariate kurtosis (Bentler & 
Wu, 2002; Mardia, 1970).     
I also examined gender differences for each factor.  Across the cultures, the 
tendency to emphasize individual achievements (women:  M = 4.89, SD = 1.84 / men:  M 
= 5.20, SD = 1.66) and to support biological hierarchy beliefs (women:  M = 2.61, SD = 
1.49 / men:  M = 3.53, SD = 1.68) were higher for men than women, t(818.23) = -2.61, p 
= .009, ?2p = .01 and t(893) = -8.59, p < .001, ?2p = .08, respectively.  However, 
individual achievements was higher for South Korean men (M = 5.52, SD = 1.51) than 
US men (M = 4.77, SD = 1.75), t(301.03) = -4.27, p < .001, ?2p = .05, and higher for 
  48 
 
South Korean women (M = 5.27, SD = 1.80) than US women (M = 4.65, SD = 1.82), 
t(534) = -3.82, p < .001, ?2p = .03.  Also, biological hierarchy beliefs was higher for 
South Korean men (M = 3.94, SD = 1.46) than US men (M = 2.98, SD = 1.80), t(287.63) 
= -5.41, p < .001, ?2p = .08 and higher for South Korean women (M = 3.31, SD = 1.38) 
than US women (M = 2.17, SD = 1.39), t(534) = -9.33, p < .001, ?2p = .14.  The results 
indicated that gender differences in the emphasis on individual achievements and 
biological hierarchy beliefs may not confound cultural differences in these variables.        
Cultural differences in the degree of prejudice.  Table 3 presents descriptive 
statistics and cultural differences in the degree of social and biological hierarchy beliefs 
between the US and South Korea, as well as between Whites in the US and Koreans in 
South Korea.  As hypothesized, both social and biological hierarchy beliefs were stronger 
in South Korea than in the US.  I repeated the cultural difference tests using latent factor 
(PAF) scores.  The results were consistent with the results of cultural differences in the 
mean scores in terms of significance and effect size.      
I also conducted a 2 (Culture: US vs. South Korea) x 2 (Belief Type: social vs. 
biological hierarchy beliefs) mixed-model analysis of variance (ANOVA) on the mean 
tendency to support two types of hierarchy beliefs by each participant.  There was a 
significant main effect of Culture, F(1, 1008) = 141.13, p < .001, ?2p = .12.  Across the 
two types of hierarchy beliefs, South Korean participants (M = 3.89 SD = 1.45) supported 
hierarchy beliefs more than US participants (M = 2.76, SD = 1.56).  There was a 
significant main effect of Belief Type, F(1, 1008) = 194.14, p < .001, ?2p = .16.  Across 
the two cultures, the support of social hierarchy beliefs (M = 3.60, SD = 1.82) was greater 
than the support of biological hierarchy beliefs (M = 3.04, SD = 1.64).  There was also a 
  49 
 
significant Culture x Belief Type interaction, F(1, 1008) = 7.50, p = .006, ?2p = .007.  The 
Culture difference was greater for biological (M = 2.42 vs. M = 3.66; ?2p = .14) than for 
social (M = 3.09 vs. M = 4.11; ?2p = .08) hierarchy beliefs.  As hypothesized, South 
Korean participants supported social (M = 4.11 vs. M = 3.09) and biological (M = 3.66 
vs. M = 2.42) hierarchy beliefs more than US participants.  Within the US, the support of 
social hierarchy beliefs (M = 3.09) were greater than biological hierarchy beliefs (M = 
2.42), F(1, 507) = 113.64, p < .001, ?2p = .18.  Within South Korea, the support of social 
hierarchy beliefs (M = 4.11) was also greater than biological hierarchy beliefs (M = 3.66), 
F(1, 501) = 81.22, p < .001, ?2p = .14.   
When I repeated the 2 x 2 mixed-model analysis with Whites in the US and 
Koreans in South Korea, the main effects of Culture (M = 2.82 vs. M = 3.88) and Belief 
Type (M = 3.60 vs. M = 3.09) were significant, F(1, 772) = 94.07, p < .001, ?2p = .11 and 
F(1, 772) = 120.14, p < .001, ?2p = .14, respectively, and in the same direction.  However, 
differently from the total sample, the Culture x Belief Type interaction was not 
significant, F(1, 772) = 1.30, p = .26, ?2p = .002.  The Culture differences for social (M = 
3.10 vs. M = 4.10; ?2p = .08) and biological (M = 2.54 vs. M = 3.65; ?2p = .11) hierarchy 
beliefs were not significantly different from each other, because the support of biological 
hierarchy beliefs was greater among Whites than in the total sample of the US.  However, 
as hypothesized, Koreans in South Korea supported both social (M = 4.10 vs. M = 3.10) 
and biological (M = 3.65 vs. M = 2.54) hierarchy beliefs more than Whites in the US.  
Among Whites in the US, the support of social hierarchy beliefs (M = 3.10) was greater 
than biological hierarchy beliefs (M = 2.42), F(1, 291) = 44.42, p < .001, ?2p = .13.  
Among Koreans in South Korea, the support of social hierarchy beliefs (M = 4.10) was 
  50 
 
also greater than biological hierarchy beliefs (M = 3.65), F(1, 481) = 79.47, p < .001, ?2p 
= .14.  Figure 2 shows mean social and biological hierarchy beliefs in the US and South 
Korea for the two sets of samples.      
Cultural differences in the degree of norms/values.  Table 3 also presents 
descriptive statistics and cultural differences in the degree of norms/values between the 
US and South Korea, as well as between Whites in the US and Koreans in South Korea.  
In the total samples, the emphasis on individual achievements (M = 4.67 vs. M = 5.40), 
the self-ingroup overlap (M = 4.63 vs. M = 5.23), roles/positions within ingroups (M = 
4.67 vs. M = 5.34), and group-oriented norm/values (the mean of five group-oriented 
items; M = 4.65 vs. M = 5.28) were all greater in South Korea than in the US.  The 
emphasis on individual achievements (M = 4.66 vs. M = 5.39), the self-ingroup overlap 
(M = 4.52 vs. M = 5.24), roles/positions within ingroups (M = 4.45 vs. M = 5.33), and 
group-oriented norms/values (M = 4.49 vs. M = 5.28) were also greater among Koreans 
in South Korea than among Whites in the US.  In the US, Whites tended to show less 
emphasis on group-oriented norms/values than the total sample.  I repeated the cultural 
difference tests using latent factor (PAF) scores.  The results were consistent with the 
results of cultural differences in the mean scores in terms of significance and effect size.   
Cultural differences in dynamics of prejudice.  I examined associations 
between the individual- and group-oriented norms/values and hierarchically-based 
intergroup biases in each culture and cultural differences in these associations.  To test 
associations among all predictor and outcome variables in a model not disregarding 
measurement errors in the two cultural samples, I conducted structural equation modeling 
(SEM) analyses with latent variables for each sample and multigroup SEM analyses 
  51 
 
across the two samples.  All SEM models were tested based on the variance-covariance 
matrix using LISREL 8.8 (J?reskog & S?rbom, 2006).  The Satorra-Bentler (SB) scaling 
method (Satorra & Bentler, 1988, 1994) was applied to estimate X2 and standard errors 
based on the diagnostic examination of data distributions (Finney & DiStefano, 2006) 
and X2 differences were tested using the SB scaled X2 difference computation procedure 
for LISREL 8 users (Bryant & Satorra, 2012).  Model fits were evaluated based on 
RMSEA (Root Mean Squared Error or Approximation, Steiger, 1990; see also Breivik & 
Olsson, 2001; Browne & Cudeck, 1993; for SB scaled RMSEA, Nevitt & Hancock, 
2000) with its 90% C.I. (Olsson, Foss & Breivik, 2004) and the close fit test (Hayduk et 
al., 2005; Kline, 2011) for the parsimony-adjusted fit, CFI (Comparative Fit Index, 
Bentler, 1990; see also Hu & Bentler, 1999) for incremental fit, and SRMR (standardized 
root mean squared residual; Hu & Bentler, 1999) for the absolute fit, in addition to X2 
statistics.  We also referred to cutoff points suggested by Chen (2007) to evaluate 
differences in model fit statistics in measurement invariance tests.       
 Analytic strategy.  I first tested measurement invariance (whether measurement 
models are equivalent across samples, Byrne, 2008; Horn & McArdle, 1992; J?reskog, 
1971; van de Vijver & Leung, 1997).  Because I hypothesized that the degree and 
associations of indicators for each latent factor (cultural norms/values and hierarchy 
beliefs) would differ between the two cultures, among various levels and types of 
measurement invariance, I focused on configural invariance (whether a common baseline 
model has equivalent structure and a good model fit across two samples) and metric 
invariance (whether a common baseline model has equivalent factor loadings across two 
samples).  Because I expected that some factor loadings would differ across two cultures, 
  52 
 
I did not remove the non-invariant items and conducted multigroup SEM analyses 
assuming both metric invariance (all factor loadings are constrained to be equal across 
two samples) and partial metric invariance (invariant factor loadings are constrained to 
be equal and non-invariant factor loadings are freed across two samples, see Byrne, 
Shavelson, & Muth?n, 1989).          
To test configural invariance, I first conducted a confirmatory factor analysis for 
each latent factor separately for each culture to test whether the baseline model for each 
latent factor is valid (show a good model fit) having common structure between the two 
cultures.  I then conducted a multigroup SEM analysis for each latent factor 
simultaneously for the two cultural samples to test whether the global model has a good 
model fit when factor loadings and error covariance are constrained to be equal across the 
two samples.  To test metric invariance, I conducted a series of multigroup SEM analyses 
for each latent factor simultaneously for the two cultural samples.  I compared a global 
model in which all factor loadings (and error covariance if applicable) are constrained to 
be equal across two cultures (H0) to a global model in which target factor loadings are 
freed and other factor loadings are constrained to be equal across two cultures (H1) letting 
error variances and the factor variance be free across two cultures in both models.  I, 
then, computed the chi-square difference and compared other model fit statistics of the 
two models to see whether the freed target factor loadings lead to a difference in model 
fits.  I inferred that the target factor loadings significantly differ between the two cultures 
if the differences in chi-square and other model fit statistics are significant or statistically 
meaningful considering sample sizes and invariance patterns (see Chen, 2007; Meade & 
Lautenschlager, 2004).   
  53 
 
For structural models, because the measurement models showed configural 
invariance and partial metric invariance, I first tested the proposed structural models (the 
two-factor model and the three-factor model) separately for each culture and compared 
the patterns of associations between the two cultures without testing whether the 
differences are statistically significant.  In addition to this, because different degrees of 
measurement invariance requirement have been discussed for cross-cultural models (see 
Chen, 2008; Milfont & Fischer, 2010, Millsap & Kwok, 2004), I also tested the proposed 
structural models simultaneously across the two cultures conducting multigroup SEM 
analyses, with which I can infer whether the differences between the two cultures are 
statistically significant or not.  I tested the moderating effect of culture separately for 
assuming metric invariance and assuming partial metric invariance.  In both analyses, I 
compared a global model in which all parameter estimates are constrained to be equal 
across two cultures (H0) to a global model in which all parameter estimates (for cultural 
differences in overall models) or target paths are freed and other paths are constrained to 
be equal across two cultures (H1).  In all H1 models, we freed all error variances, error 
covariances, latent factor variance, and latent factor covariances across two cultures.  I 
then compared the two global models (H0 vs. H1) based on the chi-square difference.  I 
inferred that the overall models or target paths significantly differ between the two 
cultures if the chi-square difference is significant.     
Theoretically, I proposed two structural models as an alternative to each other.  
The two-factor model tested associations between two predictor factors (individual and 
group-oriented norms/values) and two outcome factors (social and biological hierarchy 
beliefs), whereas the three-factor model tested associations between three predictor 
  54 
 
factors (the emphasis on individual achievements, the self-ingroup overlap, and 
roles/positions within ingroups) and the two same outcome factors.  In both models, I 
tested all possible paths between the predictor factors and outcome factors, all possible 
covariance among the predictor factors, and the covariance between the two outcome 
factors, because we assumed that all predictors would be associated with both hierarchy 
beliefs at least indirectly and/or at least in one of the two cultures.  I also assumed that all 
predictors would be associated with each other as a predictor of hierarchy beliefs at least 
to some degree and/or at least one of the two cultures, and that the two hierarchy beliefs 
would be associated with each other in both cultures.  I also tested the final models only 
with Whites in the US and Koreans in South Korea to see whether Whites (the focal 
group in the US) are different from the total sample in the US in terms of the association 
between the norms/values and hierarchy beliefs.           
Measurement models.  Table 4a presents factor loadings and model fit statistics 
in the common baseline measurement models (that have equivalent structure across the 
two cultures) in the US and South Korea.  All common baseline models showed a good 
model fit based on all significant factor loadings in both cultures.  Table 5 presents model 
fit statistics of the global models in which all factor loadings and error covariances (the 
model configuration) are constrained to be equal across two cultures (H0).  The model fit 
statistics indicated that the common baseline measurement model for each of four latent 
factors (individual- and group-oriented norms/values and social and biological hierarchy 
beliefs) is valid (configural invariance).  Table 5 also presents differences in chi-squares 
(Bryant & Satorra, 2012) and other model fit statistics.  When I consider the significance 
of differences, sample sizes, and invariance patterns, one factor loading of individual-
  55 
 
oriented norms/values, all of group-oriented norms/values, one of social hierarchy 
beliefs, and one of biological hierarchy beliefs were regarded as different between the 
two cultures.  The omnibus tests (the comparison between the model in which all factor 
loadings are constrained to be equal and the model in which all factor loadings are freed) 
also indicated that the overall factor loadings for group-oriented norms/values, social 
hierarchy beliefs and biological hierarchy beliefs differ between the two cultures.   
Structural models.  Table 8 presents parameter estimates and model fit statistics 
for the two-factor and three-factor structural model in each sample in the US and South 
Korea.  Both two-factor and three-factor structural models in each sample in the two 
cultures showed a good model fit.  In the two-factor model, the total variance accounted 
by the two predictor factors were greater in South Korea than in the US for both social 
(13% vs. 49%) and biological (12% vs. 31%) hierarchy beliefs.  In the three-factor 
model, the total variance accounted by the three predictor factors were also greater in 
South Korea than in the US for both social (15% vs. 42%) and biological (13% vs. 26%) 
hierarchy beliefs.  Similar patterns were found in the comparison between Whites in the 
US and Koreans in South Korea.   
Figure 3 presents standardized parameter estimates of the two-factor structural 
model in the US and South Korea.  As hypothesized, in the US, the path from individual-
oriented norms/values (based on the three individual-oriented items) to social hierarchy 
beliefs was significant and positive both in the total sample (B = .20) and among Whites 
(B = .18).  The path from individual-oriented norms/values to biological hierarchy beliefs 
was also significant and positive, but only in the total sample (B = .14 vs. B = .05).  In 
South Korea, the paths from individual-oriented norms/values to both social (B = -.14 in 
  56 
 
the total sample) and biological (B = -.11 in the total sample) hierarchy beliefs were not 
significant in both samples.   
The paths from group-oriented norms/values (based on the five group-oriented 
items) to both social and biological hierarchy beliefs were significant and positive in all 
four samples.  As hypothesized, the paths from group-oriented norms/values to social (B 
= .23 vs. B =.77) and biological (B = .26 vs. B =.62) hierarchy beliefs were greater in 
South Korea than in the US.  In addition, the covariance between the two predictor 
factors was greater in South Korea (B = .61) than in the US (B = .47), and Whites in the 
US showed the least covariance (B = .37) among the four samples.  The covariance 
between the two hierarchy beliefs was similar (the range from B = .59 to B = .64) across 
the four samples.   
Figure 4 presents standardized parameter estimates of the three-factor structural 
model in the US and South Korea.  As hypothesized, in the US, the path from individual 
achievements to social hierarchy beliefs was significant and positive both in the total 
sample (B = .21) and among Whites (B = .19).  The path from individual achievements to 
biological hierarchy beliefs was also significant and positive, but only in the total sample 
(B = .14 vs. B = .04).  In South Korea, the paths from individual achievements to both 
social (B = -.05 in the total sample) and biological (B = -.04 in the total sample) hierarchy 
beliefs were not significant in both samples.   
As hypothesized, the self-ingroup overlap predicted hierarchy beliefs only 
indirectly; the paths from the self-ingroup overlap to both social and biological hierarchy 
beliefs were not significant in all four samples.  The magnitude of parameter estimates 
  57 
 
between the self-ingroup overlap and the two hierarchy beliefs were similar across the 
samples (the range from B = .13 to B = .17), except for Whites in the US (B = .04).   
The path from roles/positions within ingroups to social hierarchy beliefs, in the 
US, was not significant both in the total sample (B = .12) and among Whites (B = .13).  
The path from roles/positions within ingroups to biological hierarchy beliefs was also not 
significant in both samples in the US, but was greater among Whites (B = .28) than in the 
total sample (B = .17).  In South Korea, as hypothesized, the paths from roles/positions 
within ingroups to social (B = .58 in the total sample) and biological (B = .41 in the total 
sample) hierarchy beliefs were significant and positive in both samples.     
In addition, the covariance between individual achievements and the self-ingroup 
overlap was smaller in the US (B = .19 in the total sample) than in South Korea (B = .40 
in the total sample).  The covariance between individual achievements and roles/positions 
within ingroups was significant and positive in all four samples, but the magnitude was 
smaller among Whites (B = .40) than in the total sample in the US (B = .50) and in South 
Korea (B = .58 in the total sample).  The covariance between the two group-oriented 
norms/values was significant and positive in all four samples, but the magnitude was 
greater among Whites (B = .74) than in the total sample in the US (B = .61) and in South 
Korea (B = .63 in the total sample).  The covariance between the two hierarchy beliefs 
was also significant, positive, and similar across the four samples (the range from B = .60 
to B = .66).      
Table 7 presents model fit statistics of global models and the moderating effect of 
culture on each target parameter estimates between the US and South Korea, as well as 
Whites in the US and Koreans in South Korea.  I tested the moderating effect of culture 
  58 
 
separately for assuming metric invariance and assuming partial metric invariance (see 
analytic strategy).  The chi-square differences (?X2, Bryant & Satorra, 2012) indicated 
that the overall models, all paths, and each target path significantly differ between the 
two cultures.  This confirms that the overall structural models, overall associations 
between norms/values and hierarchy beliefs, and all of each single path are different 
between the two cultures, not only in terms of the pattern of magnitude and significance, 
but also statistically significantly.     
I also tested whether the path between individual-oriented norms/values and 
biological hierarchy beliefs of the two-factor models and the path between individual 
achievements and biological hierarchy beliefs of the three-factor models differ between 
the total sample and Whites in the US.  Although the magnitude and significance differed 
between the two samples when they were tested separately in single-group analyses, the 
chi-square differences in multigroup analyses indicated that the differences may not be 
statistically significant in the two factor model (metric invariance is assumed:  ?X2 (?df = 
37) = 13.64, p > .05 / partial metric invariance is assumed: ?X2 (?df = 44) = 14.64, p > 
.05) and in the three factor model (metric invariance is assumed:  ?X2 (?df = 38) = 13.52, 
p > .05 / partial metric invariance is assumed: ?X2 (?df = 44) = 14.52, p > .05).          
Discussion  
Study 2 provided empirical evidence for cultural differences in the degree of 
hierarchically-based intergroup biases.  As hypothesized based on the cultural 
norms/values of group-oriented cultures that emphasize hierarchical relations within a 
group (see Cha, 1994; Triandis, 1994), both social and biological hierarchy beliefs were 
stronger in South Korea than in the US.  The emphasis on the holistic/essential nature in 
  59 
 
individuals and groups in group-oriented cultures may also lead to endorsing biological 
hierarchies more and perceiving existing hierarchies as pre-determined and/or permanent.  
Also, as hypothesized, the support of social hierarchy beliefs was greater than the support 
of biological hierarchy beliefs across the two cultures.  The cultural difference in 
biological hierarchy beliefs was greater than the cultural difference in social hierarchy 
beliefs in the total samples, but the cultural differences in social and biological hierarchy 
beliefs became similar when we compared Whites in the US and Koreans in South Korea, 
because Whites (as a social majority in the US) supported biological hierarchy beliefs 
more than the total sample in the US.   
In Study 1, I found that the social distance to personal quality groups was greater 
in the US than in South Korea, whereas the distance to groups perceived as being 
essentially different was greater in South Korea than in the US.  However, in Study 2, the 
endorsement of both social and biological hierarchy beliefs was greater in South Korea 
than in the US.  This may be because, although the emphasis on personal responsibility 
and social hierarchy beliefs can be closely related to each other, the target groups of 
Study 1 and Study 2 can be perceived differently.  The target groups of Study 1 (drug 
addicts, heavy drinkers, people with a criminal record, and emotionally unstable people) 
can be perceived more likely as being wrong or problematic, whereas the potential target 
groups of social hierarchy beliefs in Study 2 (low-ranked schools, low-status jobs, low 
income people, no professional degree, and no college degree) can be perceived more 
likely as being incompetent.  In Study 2, hierarchy itself can also be emphasized rather 
than the potential target groups of the hierarchy beliefs.  These differences imply that the 
  60 
 
degree of prejudice can vary to similar target groups depending on how the target groups 
are perceived and why.   
Study 2 also demonstrated cultural differences in the dynamics of hierarchically-
based intergroup biases. In particular, we investigated potential differences between the 
US and South Korea in the degree to which individual-oriented norms/values (the 
emphasis on individual achievements) and group-oriented norms/values (the emphasis on 
the self-ingroup overlap and roles/positions within ingroups) mediated social and 
biological hierarchy beliefs in the two alternative structural models (the two-factor model 
and the three-factor model).   
Emphasis on the importance of individual achievements was higher, on average, 
in South Korea than in the US.  This unanticipated finding may be related to the priority 
given to education and achievements in school in South Korea (see Im, 2011).  However, 
as hypothesized, the emphasis on individual achievements predicted social hierarchy 
beliefs only in the US, where the emphasis on individual achievements is a central 
norm/value.  The result suggests that people in the US justify their prejudice toward 
socially lower status groups based on central norms/values of their culture (the emphasis 
on personal choice, efforts, and responsibility; see Markus et al., 1997).  I tested only the 
emphasis on individual achievements in the present study, but future research might 
consider both meritocratic and Protestant Ethic beliefs, norms/values that could justify 
the endorsement of social hierarchy beliefs in the US, particularly among Whites, the 
socially dominant group in the US (Major et al., 2007).  Moreover, future research might 
consider differences in these processes among different racial and ethnic groups in the 
US.  I found that, among Whites who comprised the substantial proportion of participants 
  61 
 
in our sample, individual achievements predicted social hierarchy but not biological 
hierarchy beliefs, whereas in the total sample that includes members of other racial/ethnic 
groups, the path from individual achievements to biological hierarchy beliefs became 
significant.  Although multigroup analyses indicated the differences between the two 
samples in the US may not be statistically significant and the greater power associated 
with a larger sample may explain this effect, the results also suggest that there may be 
variations in values and their influences among different ethnic groups in the US.  
Group-oriented norms/values directly predicted both social and biological 
hierarchy beliefs both in the US and South Korea.  This finding implies that group-
oriented norms/values are general and common predictors of hierarchically-based 
intergroup biases across the two cultures.  That is, even in individual-oriented cultures, 
any social context that emphasizes group-oriented norms/values can encourage people to 
endorse hierarchically-based intergroup biases.  The three-factor model further revealed 
that the component of group-oriented norms/values representing the emphasis on 
roles/positions within ingroups primarily accounted for the association with social and 
biological hierarchy beliefs; the component reflecting the emphasis on the self-ingroup 
overlap did not directly predict the endorsement of social or biological hierarchy beliefs 
either in the US or South Korea.  In addition, the effects of group-oriented norms/values 
on hierarchy beliefs were greater in South Korea than in the US.   
The individual and group-oriented norms/values examined in Study 2 were related 
to each other both in the US and South Korea probably as predictors of prejudice, but, 
overall, more closely in South Korea than in the US.  This may be because, in individual-
oriented cultures, the self, and thus individuals, are clearly distinguished from groups 
  62 
 
(Brewer & Chen, 2007; Markus & Kitayama, 1991), whereas in group-oriented cultures, 
the self is not emphasized and the distinction between the self and ingroups are relatively 
less distinctive (Cha, 1994; Menon et al., 1999; Smith & Henry, 1996; Triandis, 1994; 
Vijver & Watkins, 2006).  Also, the total variances accounted by these norms/values 
were also consistently greater in South Korea than in the US.  This is probably because I 
have more group-oriented norms/values than individual-oriented norms/values in the 
models, and because group-oriented norms/values, which are stronger in South Korea 
than in the US, are more influential predictors of hierarchically-based intergroup biases 
than individual-oriented norms/values.         
One limitation of Study 2 involves sampling.  Especially studies that involve 
cultural norms/values, representative sampling is valuable and college students may not 
represent the population of a specific culture.  However, because two cultural samples 
have similar socio-demographic backgrounds, the findings of this study may still be 
informative and meaningful.  I also note that the measures I introduced in Study 2 may 
require further validation.  I used only two or three items to compute a latent factor for 
each norm/value and some statements can measure more than one concept we proposed.  
For example, the measure of individual achievements can measure emphasizing 
?individual? achievement or emphasizing individual ?achievements.?  Nevertheless, we 
believe that the different aspects we studied appeared to coherently reflect individual- and 
group-oriented norms/values.   
  63 
 
Chapter 4: Study 3 
Study 3 investigated cultural differences in the degree and dynamics of prejudice 
(assessed by social distance) between individual- and group-oriented cultures, the US and 
South Korea, respectively, with a more comprehensive set of target groups (low-SES 
groups, non-normative groups, racial/ethnic, and value-based) and individual/group-
oriented norms/values (the emphasis on essentialism, individual uniqueness, and 
conformity with norms/conventions) that are not examined in the previous studies.      
In Study 1 and Study 2, I focused on two general types of prejudice.  One is based 
on personal qualities (Study 1) or the endorsement of social order based on achieved 
status or social reputations (Study 2).  The other is based on essential qualities (Study 1) 
or the support of social order based on given genetic qualities or physical conditions 
(Study 2).  I assumed that the focus on the personal quality or (personally) achieved 
status or reputation is a more individually oriented perspective that is more closely 
related with individual-oriented norms/values such as the emphasis on individual?s 
capitalistic/meritocratic achievements (based on the Protestant Ethic) of individual-
oriented cultures.  I also assumed that the focus on essential quality or given 
genetic/physical conditions is relatively more group-oriented perspective that helps a 
group function as a group.  That is, in order for a group to function as a group, the group 
needs stably organized systems or relations among group members (based on the 
emphasis on relations in Confucianism, see Gardner & Seeley, 2001).  This leads to 
focusing more on essentially or naturally given qualities and focusing more on essentially 
or naturally given qualities also help the group has a more stably organized system or 
relations among the group members (based on the holistic/circular worldview in East 
  64 
 
Asian cultures, see Nisbett, 2003).  Study 3, thus, investigated the influence of 
essentialism on the four types of target groups, in addition to the emphasis on individual 
uniqueness and conformity with norms/conventions, other representative individual- and 
group-oriented norms/values that are not directly examined in the previous studies.                
The results of Study 1 and Study 2 indicated that prejudice that is based on 
personal qualities or achieved status/reputations is consistently greater than the one based 
on essential qualities or given genetic/physical conditions in both individual- and group-
oriented cultures.  Based on this finding, we predicted that low-SES groups (groups 
characterized as poor and uneducated) would be one of the most vulnerable target groups 
of prejudice in both cultures and particularly for individual-oriented cultures in which the 
core cultural value focuses on individual?s personal choice and the consequent individual 
achievements.  I hypothesized that prejudice toward low-SES groups would be greater in 
the US than in South Korea.  Another highly vulnerable target group is non-normative 
groups (groups characterized as different from the majority or typical ingroups of the 
society).  Whether a group is normative or non-normative is determined based on the 
current norms/values of the society, thus concrete examples of non-normative groups 
would vary across different eras and different cultures.  However, once a group is 
perceived as non-normative, the group would more likely be a target of prejudice in 
group- (vs. individual-) oriented cultures in which conformity with norms/conventions 
are emphasized.  I hypothesized that prejudice toward non-normative groups would be 
greater in South Korea than in the US.  Racial/ethnic outgroups are relatively more 
obvious outgroups because of the visible differences, but, usually, prejudice toward 
racial/ethnic outgroups are not directly or explicitly displayed because of the lack of 
  65 
 
justifications (i.e., being different is not enough to express negative attitudes in a social 
setting).  However, prejudice toward racial/ethnic outgroups can more frequently be 
expressed in group-oriented cultures where essential qualities are emphasized.  I 
hypothesized that prejudice toward racial/ethnic outgroups would be greater in South 
Korea than in the US.  In addition, I included value-based outgroups, the group of people 
who has different views of value, for example, in political or religious issues.  However, I 
predicted that the distinction between individual- and group-orientations would not 
directly relate to prejudice toward value-based outgroups.     
I also predicted cultural differences in associations among the four types of target 
groups.  In individual-oriented cultures where the core cultural values emphasize personal 
choice and responsibility, rather than given essential qualities, low-SES and non-
normative groups would more likely be perceived as having different values (i.e., the 
values that are different form the Protestant Ethic or traditional family values), which 
may less likely occur in group-oriented cultures.  I hypothesized that associations of 
value-based outgroups with low-SES and non-normative groups would be greater in the 
US than in South Korea.  In group-oriented cultures, values-based outgroups would be 
more associated with racial/ethnic outgroups, the groups that are essentially different in 
many ways including values.  I hypothesized that the association between value-based 
outgroups and racial/ethnic outgroups would be greater in South Korea than in the US.  
Also, in both cultures, non-normative groups are usually low-SES groups of the society 
because it may not be easy to achieve status with their non-normative backgrounds.  
However, it would be more difficult for non-normative groups to achieve status in group- 
(vs. individual-) oriented cultures in which the core cultural values emphasize conformity 
  66 
 
with norms and conventions of the society.  I hypothesized that the association between 
non-normative and low SES groups would be greater in South Korea than in the US.         
Study 3 also investigated cultural differences in dynamics of prejudice.  In Study 
1, I inferred the effects of individual- and group-oriented norms/values from the results 
on cultural differences in the degree of prejudice toward personal and essential quality 
groups.  I assumed that the cultural differences in the degree of prejudice toward the two 
types of target groups (i.e., groups based on personal vs. essential qualities) would be due 
to the cultural difference in individual- or group-oriented norms/values that emphasize 
either personal or essential qualities.  The emphasis on essential qualities (i.e., 
essentialism) has been known to promote stereotyping and legitimate existing social 
inequalities as inevitable (e.g., Haslam et al., 2006; Yzerbyt et al., 1997).  Essentialism 
also leads to perceiving ingroups as having more human essence than outgroups (Leyens 
et al., 2001).  In addition, when biological (vs. social) qualities of racial groups are 
emphasized, racial inequalities are more accepted and racial outgroups are perceived as 
more unrelated to the self (Williams & Eberhardt, 2008).   
Some early work on essentialism (Yzerbyt, Corneille, & Estrada, 2001) discussed 
cultural differences in essentialism in terms of dispositional bias (fundamental attribution 
error, Ross, 1977; correspondence bias, Choi et al., 1999), which is in line with findings 
on cultural differences in the perception of the locus of control.  That is, in individual-
oriented cultures, people perceive the self as the locus of control and tend to explain their 
and other?s behavior focusing on the internal self (see Spector et al., 2002; Weisz, 
Rothbaum, & Blackburn, 1984), whereas, in group-oriented cultures, people perceive 
both the self and the world as the locus of control and tend to explain their and other?s 
  67 
 
behaviors focusing relatively more on external or situational factors (Mahler, 1974; 
Parsons & Scheider, 1974).  These cultural differences implies that, in individual-oriented 
cultures, people more likely recognize individual achievements and blame individual 
failure attributing both success and failure to the self, whereas, in group-oriented cultures, 
people less likely recognize individual achievements and blame individual failure 
attributing both success and failure to the given social/physical situations.  In other 
words, individual-oriented cultures emphasize personal responsibility assuming the 
possibility to change one?s status; both the self and the world are relatively more 
malleable and mutable.  However, group-oriented cultures suggest that the locus of 
control is perceived as beyond the self and the self is perceived as a part of essentially 
predetermined and well-organized world, which leads people to focus more on essentially 
given qualities and to perceive a limit to change one?s status.  Keller (2005) also found 
that when essentialist information is salient, the level of prejudice and ingroup bias 
increased particularly for those who hold chronic essentialist beliefs (such as individuals 
in group-oriented cultures).  In Study 3, I examined cultural differences in the effect of 
essentialism on prejudice toward the four types of target groups.  I hypothesized that the 
degree of emphasis on essential qualities would be greater in South Korea than in the US, 
and essentialism would directly predict prejudice toward more various target groups (low 
SES, non-normative, and racial/ethnic groups) in South Korea than in the US.   
I also examined the emphasis on individual uniqueness (Markus et al., 1997) and 
conformity with norms/conventions (Kashima et al., 1992; Kim & Markus, 1999; Smith 
& Bond, 1993; Williams & Sogon, 1984), the most representative individual- and group-
oriented norms/values (in addition to the emphasis on individual achievements, the self-
  68 
 
ingroup overlap, and roles/positions within ingroups in Study 2).  In the previous research 
on cultural differences in stigmatization (Shin et al., 2013), I examined the value of 
individual uniqueness and behavioral conformity as mediators of cultural differences in 
prejudice toward outgroups with blemishes of character (e.g., homosexuals) and tribal 
outgroups (e.g., people with a different race).  However, because I used preexisting data, 
the value of individual uniqueness and behavioral conformity was measured based on a 
single item for each variable (?to think up new ideas and be creative; to do things one?s 
own way? and ?always behave properly; to avoid doing anything people would say is 
wrong,? respectively) and we examined two types of target groups based on Goffman?s 
(1963) distinction.  In Study 3, I computed a latent factor for the emphasis on individual 
uniqueness and conformity with norms/conventions using more items and examined the 
associations with the four types of target groups.   
I predicted that the emphasis on individual-uniqueness would directly predict less 
prejudice toward non-normative groups but only in individual-oriented cultures, because 
individual uniqueness is not a core cultural value of group-oriented cultures.  I found in 
Study 1 and Study 2 that if a norm or value (e.g., realistic competition or the emphasis on 
individual achievements) is not a core cultural value of a specific culture, it does not 
directly predict prejudice, even when the degree to agree with the norm/value is greater.  
I hypothesized that the emphasis on individual uniqueness would directly predict less 
prejudice only in the US.  I also predicted that the emphasis on conformity with 
norms/conventions would directly predict more prejudice toward various target groups in 
both cultures.  The emphasis on conformity with norms/values is one of the most 
representative group-oriented norms/values and we found in Study 2 that group-oriented 
  69 
 
norms/values are general and common predictors of prejudice across cultures.  I predicted 
that, however, the effect of the emphasis on conformity with norms/values would be 
greater in group-oriented cultures than in individual-oriented cultures.  I thus 
hypothesized that the emphasis on conformity with norms/conventions would be 
positively associated with prejudice toward various target groups both in the US and 
South Korea, but more consistently in South Korea.      
In sum, in Study 3, I hypothesized that a) prejudice toward low-SES groups 
would be greater in the US than in South Korea, b) prejudice toward non-normative 
groups would be greater in South Korea than in the US, c) prejudice toward racial/ethnic 
outgroups would be greater in South Korea than in the US, d) associations of value-based 
outgroups with low-SES and non-normative groups would be greater in the US than in 
South Korea, e) the association between value-based outgroups and racial/ethnic 
outgroups would be greater in South Korea than in the US, f) the association between low 
SES and non-normative groups would be greater in South Korea than in the US, g) 
essentialism would predict prejudice toward more various targets (low SES, non-
normative, and racial/ethnic groups) in South Korea than in the US, h) the emphasis on 
individual uniqueness would be negatively associated with prejudice toward non-
normative groups only in the US, and i) the emphasis on conformity with 
norms/conventions would be positively associated with prejudice toward various target 
groups both in the US and South Korea, but more consistently in South Korea.                   
Method   
Participants.  Altogether 612 (306 US and 302 South Korean) college students in 
undergraduate psychology courses participated for credits or as a class activity.  The US 
  70 
 
sample was consisted of 56.2% (n = 172) non-Hispanic European, 13.1% (n = 40) 
African, 8.2% (n = 25) Hispanic, 8.2% (n = 25) East Asian, 5.5% (n = 17) other Asian 
Americans, 5.2% (n = 16) were interracial/multiracial, 2.3% (n = 7) Middle Easterner, 
and 0.3% (n = 1) American Indian or Alaskan Native Americans; 0.7% (n = 2) did not 
report their ethnic backgrounds.  The South Korean sample was consisted of 94.0% (n = 
284) Koreans and 2.0% (n = 6) non-Korean or interracial/multiracial; 4.0% (n = 12) did 
not report their ethnic backgrounds.  Across the two samples, 57.6% (n = 350) were 
women and 31.3% (n = 190) were men; 11.2% (n = 68) did not report their gender.  The 
percentage of female participants was greater in the US (68.3% female, 26.5% male, and 
5.2% missing) than in South Korea (46.7% female, 36.1% male, and 17.2% missing), X2 
(N = 540, df = 1) = 14.43, p < .001, ? = .16.  Ages ranged from 17 to 50 years (M = 
21.06, SD = 3.16).  South Korean participants (M = 21.95, SD = 3.90) were older than US 
participants (M = 20.21, SD = 1.91), t(409.76) = -6.85, p < .001, ?2p = .08, but they were 
all college students.   
Procedure.  Participants indicated the extent to which they agree with a list of 
statements (norms/values) on a 9-point scale from 1, ?totally disagree? to 9, ?totally 
agree.?  They, then, indicated the extent to which they ?feel distance from the target 
group? from 1, ?very close? to 9, ?very distant.?  They also provided demographic 
information.  To prevent participants from recognizing the purpose of the different 
measures and creating an acquiescence bias for specific scales, items of different scales 
were listed randomly with various other statements (see also Locke & Baik, 2009).  The 
questionnaire was translated into Korean and translated back into English to avoid 
possible item biases in cross-cultural research.   
  71 
 
Predictor variables.  I examined one individual-oriented (the emphasis on 
individual uniqueness) and two group-oriented (the emphasis on conformity with 
norms/conventions and essentialism) norms/values as potential predictors of prejudice.  
The emphasis on individual uniqueness (Cronbach?s ? = .70; US = .70, Korea = .62) was 
consisted of four items; the emphasis on conformity with norms/conventions (Cronbach?s 
? = .81; US = .80, Korea = .78) was consisted of four items; and the emphasis on 
essential qualities or essentialism (r = .32; US = .31, Korea = .31) was consisted of two 
items; (see Appendix B for items used).  I also conducted a principle axis factoring (PAF) 
analysis for each of the three norms/values in order to include measurement errors in the 
analyses (see Winter & Dodou, 2012).  All loadings were greater than .50 in the total 
sample and in each culture, except for one loading to individual uniqueness item in the 
US (.473) and one loading to individual uniqueness item in South Korea (.451).  The PAF 
latent factor scores were used to test cultural differences in the degree of emphasizing 
each norms/values.        
Outcome variables.  I examined ten individual groups 
(poor/unemployed/homeless, uneducated, born in a low-status country, homosexuals, 
born with physical/mental disability, non-Americans/Korean, immigrants to the 
US/Korea, people whose personal interests or tastes are different from mine, people 
whose opinions are different from mine in social/political/economic issues, and people 
whose opinions are different from mine in religious issues) as potential targets of 
prejudice across cultures.  An exploratory factor analysis (principle component analysis) 
on social distance ratings of these ten target groups with varimax rotation yielded four 
factors in the total sample and in each culture.  The first factor (Total:  eigenvalue = 3.55, 
  72 
 
35.46% of variance / US:  eigenvalue = 1.66, 16.64% of the variance / Korea: eigenvalue 
= 3.23, 32.34% of the variance) reflected prejudice toward low SES groups 
(poor/unemployed/homeless, uneducated, and born in a low-status country / Cronbach?s 
? = .77; US = .79, Korea = .72).  The second factor (Total:  eigenvalue = 1.62, 16.24% of 
variance / US:  eigenvalue = 3.83, 38.30% of the variance / Korea: eigenvalue = 1.82, 
18.22% of the variance) reflected prejudice toward value-based outgroups (people whose 
personal interests or tastes are different from mine, people whose opinions are different 
from mine in social/political/economic issues, and people whose opinions are different 
from mine in religious issues / Cronbach?s ? = .76; US = .77, Korea = .75).  The third 
factor (Total:  eigenvalue = 1.17, 11.73% of variance / US:  eigenvalue = 1.12, 11.16% of 
the variance / Korea: eigenvalue = 1.15, 11.49% of the variance) reflected prejudice 
toward racial/ethnic outgroups (non-Americans/Korean and immigrants to the US/Korea 
/ r = .71; US = .79, Korea = .55).  The forth factor (Total:  eigenvalue = 1.03, 10.30% of 
variance / US:  eigenvalue = .93, 9.30% of the variance / Korea: eigenvalue = .96, 9.63% 
of the variance) reflected prejudice toward non-normative groups (homosexuals and born 
with physical/mental disability / r = .36; US = .38, Korea = .30).  I also conducted a 
principle axis factoring (PAF) for each of these four types of target groups (see Winter & 
Dodou, 2012).  All loadings were greater than .50 in the total sample and in each culture.  
The PAF latent factor scores were used to test cultural differences in the degree of 
prejudice toward each type of target groups.   
Results 
For Study 3, I first tested cultural differences between the US and South Korea in 
the degree of prejudice (assessed by social distance) toward four types of target groups, 
  73 
 
as well as three individual/group-oriented norms/values.  I then tested cultural differences 
in dynamics of prejudice toward the four types of target groups proposing two structural 
equation models, one for essentialism and the other for the emphasis on individual 
uniqueness and conformity with norms/conventions.  For Study 3, I did not repeat the 
analyses using only Whites in the US (vs. only Koreans in South Korea) because Study 2 
showed the differences between the US total sample and Whites were not significantly 
different.             
Preliminary analyses.  To prepare structural equation modeling analyses with 
latent variables, I conducted missing values analyses and examined data distributions.  
Missing values (US = .28%, Korea = .84%) were analyzed separately for predictor and 
outcome variables and separately for the US and Korean samples.  The missing values 
were missing completely at random (MCAR tests, Little, 1988) in each of these four data 
sets, so they were imputed using the expectation maximization (EM) algorithm (see Little 
& Rubin, 2002).  Data distributions were examined at both univariate and multivariate 
levels for each factor.  Most of individual indicators were skewed and/or kurtotic, but not 
severe at univariate level (Curren et al., 1996) and were not problematic in terms of the 
multivariate kurtosis (Bentler & Wu, 2002; Mardia, 1970).         
I also examined gender differences for each factor.  Across the cultures, social 
distance to non-normative groups (women:  M = 5.69, SD = 1.68 / men:  M = 6.31, SD = 
1.54) was higher for men than for women, t(538) = -2.44, p = .015, ?2p = .01, t(538) = -
2.45, p = .014, ?2p = .01, t(538) = -3.93, p < .001, ?2p = .03, and t(538) = -4.20, p < .001, 
?2p = .03, respectively.  However, the social distance to non-normative groups (M = 6.04 
vs. M = 6.52) was higher for South Korean men than for US men, t(133.83) = -3.21, p = 
  74 
 
.002, ?2p = .06, t(188) = -6.39, p < .001, ?2p = .18, t(139.28) = -2.97, p = .003, ?2p = .05, 
and t(188) = -2.14, p = .033, ?2p = .02, respectively.  The social distance to non-normative 
groups (M = 5.50 vs. M = 5.98) was also higher for South Korean women than for US 
women, t(341.07) = -5.14, p < .001, ?2p = .06, t(348) = -8.76, p < .001, ?2p = .18, 
t(331.39) = -3.64, p < .001, ?2p = .03, and t(338.95) = -2.72, p = .007, ?2p = .02, 
respectively.  The results indicated that the gender difference in social distance to non-
normative groups may not confound the cultural difference in social distance to non-
normative groups.          
Cultural differences in the degree of prejudice.  Table 8 presents descriptive 
statistics and cultural differences in the degree of prejudice (social distance) toward four 
types of target groups between the US and South Korea.  As hypothesized, perceived 
social distance to low-SES groups (M = 6.17 vs. M = 5.73) was greater in the US than in 
South Korea, whereas the distance to non-normative groups (M = 5.60 vs. M = 6.23) and 
racial/ethnic outgroups (M = 4.87 vs. M = 5.15) was greater in South Korea than in the 
US.  The distance to value-based outgroups (M = 5.65 vs. M = 5.69) did not differ 
between the two cultures.  I repeated the cultural difference tests using latent factor 
(PAF) scores.  The results of mean scores and latent factor scores were similar in terms of 
significance and effect size.  Figure 5 shows the mean social distance to the four types of 
target groups in the US and South Korea.           
Cultural differences in the degree of norms/values.  Table 8 also presents 
descriptive statistics and cultural differences in the degree of three individual/group-
oriented norms/values between the US and South Korea.  As predicted, the emphasis on 
individual uniqueness (M = 6.85 vs. M = 6.01) was greater in the US than in South Korea, 
  75 
 
whereas the emphasis on conformity with norms/conventions (M = 4.72 vs. M = 5.70) 
and essentialism (M = 5.60 vs. M = 6.13) were greater in South Korea than in the US.  I 
repeated the cultural difference tests using latent factor (PAF) scores.  The results of 
mean scores and latent factor scores were similar in terms of significance and effect size.   
Cultural differences in associations between the norms/values and prejudice.  
I examined associations between three individual/group-oriented norms/values and 
perceived social distance to four types of target groups in each culture and cultural 
differences in these associations.  To test associations among the predictor and outcome 
variables in a model not disregarding measurement errors in the two cultural samples, I 
conducted structural equation modeling (SEM) analyses with latent variables for each 
sample and multigroup SEM analyses across the two samples.  All SEM models were 
tested based on the variance-covariance matrix using LISREL 8.8 (J?reskog & S?rbom, 
2006).  The Satorra-Bentler (SB) scaling method (Satorra & Bentler, 1988, 1994) was 
applied to estimate X2 and standard errors in structural models and the measurement 
models for individual uniqueness based on the diagnostic examination of data 
distributions (Finney & DiStefano, 2006) and X2 differences were tested using the SB 
scaled X2 difference computation procedure for LISREL 8 users (Bryant & Satorra, 
2012).  I, then, applied the analytic strategy we used in Study 2.   
I proposed two structural models, one tested essentialism as a predictor factor and 
the other tested the emphasis on the individual uniqueness and conformity with 
norms/values as predictor factors.  I examined the emphasis on individual uniqueness and 
conformity with norms/values together because the two variables are theoretically 
relevant.  In both models, perceived social distance to each of the four types of target 
  76 
 
groups were outcome factors.  I tested all possible covariance among the four types of 
target groups, because I hypothesized cultural differences in associations among the 
target groups.  I also tested all possible paths between the predictor and outcome factors, 
because we assumed that all predictors would be associated with prejudice at least 
indirectly and/or at least in one of the two cultures. 
Measurement models.  Table 9 presents factor loadings and model fit statistics in 
the common baseline measurement models (that have equivalent structure across the two 
cultures) in the US and South Korea.  The common baseline models showed a good 
model fit based on all significant factor loadings in each culture, which indicates that the 
common baseline models are valid in each culture.  Table 10 presents model fit statistics 
of the global models in which all factor loadings and error covariances (the model 
configuration) are constrained to be equal across two cultures (H0).  The model fit 
statistics of the common baseline measurement models across two cultures were not 
good, which indicates that the common baseline models may not have configural 
invariance across two cultures.  Table 10 also presents differences in chi-squares (Bryant 
& Satorra, 2012) and other model fit statistics.  When we consider the significance of 
differences, sample sizes and invariance patterns (Chen, 2007), almost all factor loadings 
were non-invariant between two cultures.  Also, model fit statistics of the models in 
which all factor loadings and error covariances are freed across two cultures were good 
and significantly different from the model fit statistics of the model in which all factor 
loadings and error covariances are constrained to be equal across two cultures, which 
indicates the common baseline model for each latent factor may not have metric 
  77 
 
invariance across two cultures.  The measurement model for essentialism was not tested 
because the model was under-identified.       
Structural models.  Table 11 presents parameter estimates and model fit statistics 
for the structural model in the US and South Korea.  Both structural models (one for 
essentialism and the other for uniqueness-conformity) showed a good model fit in each 
culture.  Overall, the total variances of social distance to each of the four types of target 
groups accounted by essentialism were consistently greater in South Korea than in the 
US.  This indicates that essentialism, one of the core cultural values of group-oriented 
cultures, is a more influential predictor of prejudice in group-oriented cultures than in 
individual-oriented cultures.  The total variances of social distance to each of the four 
types of target groups accounted by the emphasis on individual uniqueness and 
conformity with norms/conventions were slightly greater in South Korea than in the US 
for all target groups except one case; the total variance of social distance to non-
normative groups was greater in the US than in South Korea (18% vs. 6%).  This may be 
because both individual uniqueness and conformity with norms/conventions predict 
social distance to non-normative groups in the US, but only conformity with 
norms/conventions predicted social distance to non-normative groups in South Korea. 
Figure 6 presents standardized parameter estimates of the structural model on 
essentialism in the US and South Korea.  As hypothesized, essentialism predict social 
distance to low SES (B = .39), non-normative (B = .42), and racial/ethnic (B = .38) 
groups in South Korea, but predicted only social distance to low SES (B = .26) groups in 
the US.  Figure 7 presents standardized parameter estimates of the structural model on 
individual uniqueness and conformity with norms/conventions in the US and South 
  78 
 
Korea.  As hypothesized, the emphasis on individual uniqueness predicted less social 
distance to non-normative groups (B = -.22) only in the US.  The emphasis on conformity 
with norms/conventions predicted more social distance to low SES (B = .26), non-
normative (B = .33), and value-based (B = .21) groups in the US and predicted low SES 
(B = .25), non-normative (B = .23), racial/ethnic (B = .18), and value-based (B =.24) 
groups in South Korea.                     
In both models, as hypothesized, associations of value-based outgroups with low-
SES (B = .49 vs. B = .17 in the essentialism model and B = .44 vs. B = .17 in the 
uniqueness-conformity model) and non-normative (B = .45 vs. B = .19 in the essentialism 
model and B = .39 vs. B = .21 in the uniqueness-conformity model) groups were greater 
in the US than in South Korea.  The association between value-based outgroups and 
racial/ethnic outgroups was greater in South Korea than in the US (B = .18 vs. B = .43 in 
the essentialism model and B = .17 vs. B = .47 in the uniqueness-conformity model).  The 
association between low SES and non-normative groups was also greater in South Korea 
than in the US (B = .60 vs. B = .71 in the essentialism model and B = .58 vs. B = .81 in 
the uniqueness-conformity model).   
Table 12 presents model fit statistics of global models and the moderating effect 
of culture on each target parameter estimates between the US and South Korea.  All 
variant factor loadings are freed across the samples based on the measurement invariance 
tests.  The chi-square differences (Bryant & Satorra, 2012) indicated that the overall 
models and each target parameter estimates (including all hypothesized cultural 
differences) significantly differ between the two cultures.   
 
  79 
 
Discussion 
Study 3 showed empirical evidence for cultural differences in the degree of 
prejudice toward a more comprehensive set of target groups.  As hypothesized based on 
the known norms/values of each culture and findings of Study 1 and Study 2 (e.g., the 
emphasis on personal responsibility, capitalistic/meritocratic achievements, and 
individual uniqueness in individual-oriented cultures vs. the emphasis on essential 
qualities and conformity with norms/conventions in group-oriented cultures), prejudice 
toward low-SES groups was greater in the US than in South Korea, whereas prejudice 
toward non-normative and racial/ethnic groups was greater in South Korea than in the 
US.    
As expected, low-SES groups (groups characterized as poor and uneducated) 
were one of the most vulnerable targets of prejudice particularly for individual-oriented 
cultures in which the core cultural norm/value focuses on individual?s personal choice 
and capitalistic/meritocratic achievements.  Prejudice toward low SES group was also the 
greatest among the four types of target groups within the US.  It suggests that if a target 
of prejudice has a low status (e.g., Blacks with a low SES), prejudice toward the target 
group would be harsher than when the target has a high status (e.g., Blacks with a high 
SES).  I also found that the association between prejudice toward low-SES groups and 
prejudice toward value-based outgroups were greater in the US than in South Korea 
(prejudice toward value-based outgroups was more associated with prejudice toward 
racial/ethnic outgroups in South Korea).  This implies that low SES groups are more 
likely perceived as having different values (e.g., values that are different form the 
Protestant Ethic).  In addition, overall associations among the prejudice toward each of 
  80 
 
the four types of target groups were consistently greater in the US than in South Korea in 
both structural models.  This suggests that there are more chances for a target of prejudice 
to have all or some of the four types simultaneously (e.g., low-status ethnic/racial 
outgroups who have different norms/values) in the US than in South Korea.   
In group-oriented cultures, non-normative groups (groups characterized as 
different from the majority or typical ingroups of the society) can be the most vulnerable 
targets of prejudice.  Prejudice toward non-normative groups was the greatest among the 
four types of target groups within South Korea.  I also found that the association between 
low SES and non-normative groups was greater in South Korea than in the US.  This 
probably because being different from the majority makes achieving a high status more 
difficult in group-oriented cultures in which core cultural norms/values emphasize 
conformity with norms/values.   
Study 3 also examined cultural differences in the dynamics of prejudice.  As 
hypothesized, essentialism, a core cultural norm/value of group-oriented cultures, 
predicted more various types of target groups, including racial/ethnic outgroups, in South 
Korea than in the US.  The emphasis on individual uniqueness, a core cultural norm/value 
of individual-oriented cultures, predicted prejudice toward non-normative groups only in 
the US and did not predicted prejudice in South Korea, even when the degree to agree 
with individual uniqueness was slightly higher than the degree to agree with conformity 
with norms/conventions within South Korea.  This implies that, as we found in Study 1 
(perceived competition for employment) and Study 2 (the emphasis on individual 
achievements), the degree to agree with certain norms/values does not always mean the 
norms/values are the core norms/values of the culture and only the core norms/values are 
  81 
 
associated with prejudice toward predicted target groups.  The emphasis on conformity 
with norms/conventions, one of the most representative group-oriented norms/values, 
predicted prejudice toward various types of target groups both in the US and South 
Korea, as a general and common predictor of prejudice across cultures.   
  82 
 
Chapter 5:  General Discussion 
The present research demonstrated cultural similarities and differences in 
prejudice between individual- and group-oriented cultures.  As expected, prejudice was 
prevalent in both cultures.  Social distance to groups perceived as being personally 
responsible and essentially different (Study 1), social and biological hierarchy beliefs 
(Study 2), and social distance to low SES, non-normative, racial/ethnic, and value-based 
groups (Study 3) were observed in both cultures.  I also found that group-oriented 
norms/values (Study 2) and specifically the emphasis on roles/positions within ingroups 
(Study 2) and conformity with norms/conventions (Study 3) are common predictors of 
prejudice across the two cultures.  In addition, individuals in both cultures expressed their 
prejudice more understandably and explicitly when the norms/values of their culture or 
society provided a basis to justify their prejudice (Study 1, 2, & 3).   
However, based on the different norms/values of individual- and group-oriented 
cultures, the degree of prejudice toward different types of target groups varied between 
the two cultures.  Individual-oriented cultures emphasize individual?s personal choice, 
efforts, responsibility, thus groups perceived as personally responsible for the negative 
attitudes toward them (groups with a controllable membership, Study 1) and low SES 
groups (Study 3) were targets of prejudice more in the US than in South Korea.  By 
contrast, because group-oriented cultures emphasize the holistic/essential nature of 
individuals and groups and the relationship among ingroup members, the groups 
perceived as essentially different from the majority or typical ingroups of the society (i.e., 
groups with a relatively more essential, cohesive, and permanent quality, Study 1) and 
other various groups (low SES, non-normative, and racial/ethnic, Study 3) were targets of 
  83 
 
prejudice more in South Korea than in the US.  Also, because group-oriented cultures 
emphasize hierarchical relations such as roles/positions within ingroups, both social and 
biological hierarchy beliefs were greater in South Korea than in the US (Study 2).   
In addition, because specific norms/values are associated with specific target 
groups, dynamics of prejudice differed between the two cultures.  Perceived competition 
for employment (Study 1) and the emphasis on individual achievement (Study 2) were 
associated with prejudice only in the US, whereas the emphasis on roles/positions within 
ingroups (Study 2) was associated with prejudice only in South Korea.  In addition, 
essentialism (Study 3) was associated with prejudice more in South Korea, whereas the 
emphasis on individual uniqueness (Study 3) predicted only in the US.  Group-oriented 
norms/values (Study 2) and the emphasis on conformity with norms/conventions (Study 
3) were associated with prejudice in the two cultures, but more strongly in South Korea 
than in the US.  Taken together, these findings highlighted both cultural differences and 
similarities in the dynamics of prejudice. 
The findings across the three studies identified the most vulnerable target groups 
of prejudice across the two cultures and in each culture.  Across the two cultures, groups 
perceived as being personally responsible for the prejudice and groups with lower status 
in a socially-based hierarchy were primary targets of prejudice.  In individual-oriented 
cultures, groups perceived as being personally responsible or relatively less achieving 
(e.g., the poor or less educated) were more likely to be targets of prejudice.  In group-
oriented cultures, groups perceived as being essentially different or inferior/lower status 
in the given hierarchical relation (e.g., foreign workers from the lower status countries) 
were common targets of prejudice. 
  84 
 
Our findings about the most vulnerable target groups of prejudice also suggest 
different ways to reduce biases toward these groups in different cultures.  For example, 
interventions that emphasize the uncontrollability or lack of personal responsibility for a 
stigmatizing condition or system-blame rather than victim-blame explanations (i.e., 
focusing more on biological or external causes of the stigmatizing condition, Bobocel & 
Hafer, 2007; Jones, 1986; Weiner, 1995) would likely be effective for reducing prejudice 
against them in both individual- and group-oriented cultures, but particularly in 
individual-oriented cultures.  By contrast, interventions that emphasize shared identity 
(Gaertner & Dovidio, 2012) through legislative measures (see Esses et al., 2001) would 
be a way to reduce prejudice toward different or low-status ethnic groups in in group-
oriented cultures.  These interventions would also be more effective if they are led by 
high-status leaders (because relationships are more hierarchically organized in group-
oriented cultures, Cha, 1994; Triandis, 1994) or pronounced through public routes 
(because the tendency to conform with norms and the majority are higher in group-
oriented cultures, Kashima, et al., 1992; Kim & Markus, 1999; Smith & Bond, 1993; 
Williams & Sogon, 1984) emphasizing the common fates and benefits they would share 
as members of a society.   
Despite the convergent findings across the two studies employing different 
measures and methodologies, we acknowledge some general limitations of the current 
work and suggest promising directions for future research.  The current work focused on 
two countries that were prototypic of individual- and group-oriented cultures, but future 
research would benefit from investigating a wider range of national samples to address 
differences between individual- and group oriented cultures.  Also, although Study 2 and 
  85 
 
3 employed causal modeling analytic techniques, both of the current studies used cross-
sectional survey designs, which limit the ability to draw firm causal inferences.  Although 
culturally internalized norms/values may not be easily manipulated entirely in their 
naturalistic form, it is possible to prime individual or collective self-concepts in 
experimental contexts (Brewer & Gardner, 1996).  Gardner, Gabriel, and Lee (1999), for 
example, demonstrated that priming students in the US and Hong Kong with independent 
or interdependent self-construal produced within-culture shifts in judgments that 
corresponded to between-culture differences.  Thus, future research might investigate 
how priming of cultural norms/values predicts prejudice in an experimental setting.  In 
addition, the current work examined either norm/value that was known as a predictor of 
prejudice in previous research (realistic competition) or the most representative 
individual- and group-oriented norms/values (the emphasis on individual achievements, 
individual uniqueness, the self-ingroup overlap, roles/positions within ingroups, 
conformity with norms/conventions, and essentialism), but future research can include 
other cultural norms/values that are not directly tested in the current work.      
In addition, more attention could be devoted to considering other factors that vary 
between countries and that are known as important variables for prejudice and intergroup 
relations.  For instance, the current work did not include measures of diversity experience 
or diversity support in the analysis.  However, a meta-analytic review of the contact 
hypothesis (Pettigrew & Tropp, 2006) revealed that greater intergroup contact generally 
reduces prejudice.  With respect to our research, the net migration rates per a thousand 
population of the US (3.64 in 2013) and South Korea (0 in 2013) indicate different levels 
of diversity experience between the countries.  Thus, the greater tendency to express 
  86 
 
prejudice toward essentially different groups (Muslims, people of a different race, 
immigrants/foreign workers, and homosexuals) in South Korea than in the US (Study 1) 
may be interpreted as the result of less contact (i.e., less familiarity) with these groups in 
South Korea compared to the US.  Also, the greater association between competition and 
prejudice in the US than in South Korea (Study 1) may be related to higher rates of 
immigration in the US than in South Korea.  That is, more contact with new immigrants 
in the form of competition over resources can exacerbate, rather than reduce, prejudice by 
emphasizing zero-sum outcomes likely more pronounced in the US than in South Korea 
(Esses et al., 2001).  Thus, future research should consider the role of intergroup contact 
(in terms of both quantity and quality) as an additional factor in cultural differences in 
prejudice as well as cultural norms/values in the dynamics of prejudice.     
In conclusion, the present dissertation suggests the value of cross-cultural 
research to understand prejudice more comprehensively.  Different cultural backgrounds 
may lead individuals to endorse different types of prejudice and intergroup biases to a 
different extent.  Studies on cultural differences in prejudice can also help identify when 
and where certain relationships, often assumed to be universal (e.g., competition and 
prejudice), hold more or less strongly.  The present research demonstrated the importance 
of recognizing the different cultural forces that not only systematically shape the level of 
prejudice toward different groups but also influence the different mechanisms that may 
underlie intergroup bias across different cultures.  Understanding prejudice in a broader 
cultural context illuminates both commonalities and differences in the processes that lead 
to intergroup bias, which produces a more comprehensive understanding of the 
  87 
 
psychology of bias and can form the basis of culturally sensitive interventions to reduce 
prejudice and discrimination. 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  88 
 
 
 
 
 
Table 1 
Cultural Differences in Proportions of Negative Responses (Social Distance) to the Target Groups  
   
United 
States 
 
South  
Korea 
 
 
X2  
 
?    
 
Wald  
 
Wald 
controlling for 
COVs 1 
Personal  
quality  
groups 
 
Drug addicts .84 .64 357.77 -.23 340.61 351.24 
Heavy drinkers .64 .55 51.96 -.09 51.80 44.04 
People with a criminal record .59 .49 41.34 -.10 41.21 25.33 
Emotionally unstable people .53 .48 10.57 -.05 10.56 13.44 
       
Essentially 
different  
groups 
 
Muslims .11 .57 833.90 .50 697.62 574.46 
People of a different race .08 .34 677.47 .35 555.57 588.06 
Immigrants/foreign workers .12 .41 711.16 .33 634.78 588.96 
Homosexuals .27 .84 1729.93 .56 1464.81 1281.07 
Note:  All statistics were statistically significant, p < .001. 
1 Controlling for gender, age, education, and income 
 
 
 
 
 
 
 
 
 
  89 
 
 
 
 
Table 2 
Associations between Competition for Employment and Negative Responses (Social Distance) to Target Groups in the US and South Korea   
 
  
Competition and Negative Responses 
 
 
 
Culture x Competition 
Interaction  
Total 
 
 
US 
 
South Korea 
Personal 
quality 
groups 
 
 
Drug addicts .79 (.72, .85) ? 
34.53? 
1.31 (1.30, 1.31)* 
22.91*  
.65 (.65, .66)? 
27.24? 
.50 (.50, .50) ? 
49.16? 
Heavy drinkers 1.02 (.95, 1.09) 
.20 
1.24 (1.24, 1, 24)* 
24.59* 
.85 (.85, .85)? 
4.93? 
.69 (.69, .69) ? 
19.79? 
People with a criminal record 1.07 (.99, 1.16) 
3.01 
1.33 (1.33, 1.34)* 
34.69*  
.88 (.88, .89)? 
1.79 
.66 (.66, .66) ? 
15.73? 
Emotionally unstable people 
 
 
1.08 (1.00, 1.16) 
3.52 
1.28 (1.28, 1.28)* 
26.08* 
.81 (.81, .82)? 
5.06? 1 (p=.059)  
.64 (.64, .64) ? 
19.19? 
Essentially  
different  
groups 
Muslims 2.21 (1.96, 2.50)* 
161.59* 
1.55 (1.55, 1.55)* 
24.47* 
1.47 (1.47, 1.47)* 1 (p=.055)  
9.73* 
.95 (.94, .95) ? 
.13 
People of a different race 1.93 (1.72)* 
120.63* 
1.65 (1.64, 1.65)* 
23.44* 
1.23 (1.23, 1.23)* 
5.14*1 
.75 (.75, .75) ? 
4.48? 
Immigrants/foreign workers 2.10 (1.91, 2.32)* 
220.72* 
2.30 (2.30, 2.30)* 
81.57* 
1.17 (1.17, 1.17)* 
4.64*1 (p=.096) 
.51 (.51, .51) ? 
32.66? 
Homosexuals 2.28 (2.11, 2.47)* 
425.55* 
1.61 (1.61, 1.61)* 
82.60* 
2.00 (2.00, 2.01)* 
51.82* 
1.24 (1.24, 1.25)* 
3.98*1 
 
Note.  Odds ratios (95% C.I.) are stated on the top line and Wald statistics are stated on the bottom line in each cell.   
* Significant and positive associations or interactions in the same direction 
? Significant and negative associations or interaction in the opposite direction 
 
1 Attenuated to non-significance controlling for gender, age, education, and income     
 
 
 
 
  90 
 
 
 
 
Table 3 
Descriptive Statistics (Means & SDs) and Cultural Differences in the Degree of Hierarchy Beliefs and Norms/Values between the US and South Korea 
  US Korea Cultural differences 1  US-W KR-K Cultural differences 1 
 
Social  
hierarchy beliefs 
 
 3.09 
(1.85) 
4.11 
(1.64) 
F(1, 1008) = 85.49, p < .001, ?2p = .08 
 
 3.10 
(1.73) 
4.10 
(1.67) 
F(1, 772) = 64.41, p < .001, ?2p = .08 
 
  F(1, 1008) = 79.10, p < .001, ?2p = .07 
 
  F(1, 772) = 59.45, p < .001, ?2p = .07 
Biological 
hierarchy beliefs 
 
 2.42 
(1.58) 
3.66 
(1.46) 
F(1, 1008) = 167.54, p < .001, ?2p = .14 
 
 2.54 
(1.57) 
3.65 
(1.47) 
F(1, 772) = 99.14, p < .001, ?2p = .11 
 
  F(1, 1008) = 139.93, p < .001, ?2p = .12 
 
  F(1, 772) = 83.74, p < .001, ?2p = .10 
         
Individual 
achievements 2  
 4.67 
(1.80) 
5.40  
(1.63) 
F(1, 1008) = 46.38, p < .001, ?2p = .04 
 
 4.66 
(1.73) 
5.39 
(1.66) 
F(1, 772) = 33.60, p < .001, ?2p = .04 
 
  F(1, 1008) = 46.48, p < .001, ?2p = .04 
 
  F(1, 772) = 33.99, p < .001, ?2p = .04 
The self-ingroup 
overlap 
 
 4.63 
(1.53) 
5.23 
(1.35) 
F(1, 1008) = 44.20, p < .001, ?2p = .04 
 
 4.52 
(1.52) 
5.24 
(1.36) 
F(1, 772) = 46.46, p < .001, ?2p = .06 
 
  F(1, 1008) = 39.74, p < .001, ?2p = .04 
 
  F(1, 772) = 41.04, p < .001, ?2p = .05 
Roles/positions 
within ingroups 
 
 4.67 
(1.81) 
5.34 
(1.44) 
F(1, 1008) = 42.65, p < .001, ?2p = .04 
 
 4.45 
(1.76) 
5.33 
(1.45) 
F(1, 772) = 57.78, p < .001, ?2p = .07 
 
  F(1, 1008) = 45.05, p < .001, ?2p = .04 
 
  F(1, 772) = 59.36, p < .001, ?2p = .07 
 
Group-oriented 
norms/values 3 
 4.65 
(1.33) 
5.28 
(1.14) 
F(1, 1008) = 65.70, p < .001, ?2p = .06 
 
 4.49 
(1,35) 
5.28 
(1.15) 
F(1, 772) = 74.72, p < .001, ?2p = .09 
 
  F(1, 1008) = 69.44, p < .001, ?2p = .06 
 
  F(1, 772) = 74.26, p < .001, ?2p = .09 
 
1 Cultural differences based on the mean scores are stated in the top lines and cultural differences based on the latent factor (PAF) scores are stated in the bottom 
lines.   
2 Individual-oriented norms/values 
3 The latent factor (PAF) scores for group-oriented norms/values were computed based on the five group-oriented items (i.e., one-factor model).   
 
 
  91 
 
 
 
 
Table 4 
The Baseline Measurement Models in the US and South Korea  
 The United States (n = 508)  South Korea (n = 502) 
Unst. SE St. Model fit statistics 1 Unst. SE St. Model fit statistics 1 
Individual-oriented norms/values   
Saturated 
 
  
Saturated      IND -> ach1 1.63*** .08 .81 1.49*** .09 .78 
     IND -> ach2 1.58*** .08 .77 1.54*** .08 .82 
     IND -> ach3 1.78*** .08 .85 1.45*** .08 .78 
Group-oriented norms/values   
X2 (2) = .22, p = .90  
RMSEA = 0 (0, .039) 
Pclose-fit = .97 
CFI = 1.00 
SRMR = .004 
 
  
  
X2 (2) = .88, p = .64  
RMSEA = 0 (0, .070) 
Pclose-fit = .87 
CFI = 1.00 
SRMR = .007 
     GRP -> selfin1 1.20*** .15 .63 1.09*** .12 .66 
     GRP -> seflin2 .66** .19 .32 .80*** .16 .46 
     GRP -> selfin3  .96*** .13 .50 1.09*** .12 .59 
     GRP -> role1 .77*** .13 .39 .79*** .11 .49 
     GRP -> role2 .98*** .14 .43 .52*** .13 .28 
     selfin1-selfin2 (error cov) .66** .25 .17 .17 .19 .06 
     selfin2-selfin3 (error cov.) 1.45** .24 .36 .53* .21 .17 
     role1-role2 (error cov.) 1.25*** .25 .28 .72*** .16 .24 
Social hierarchy beliefs   
X2 (1) = .05, p = .83  
RMSEA = 0 (0, .070) 
Pclose-fit = .91 
CFI = 1.00 
SRMR = .001 
  
X2 (1) = .03, p = .86  
RMSEA = 0 (0, .065) 
Pclose-fit = .92 
CFI = 1.00 
SRMR = .001 
     SH -> sh1 1.63*** .10 .75 1.52*** .10 .69 
     SH -> sh2 1.68*** .08 .76 1.52*** .09 .70 
     SH -> sh3 1.64*** .10 .86 1.73*** .10 .88 
     SH -> sh4 2.11*** .08 .88 1.58*** .10 .76 
     SH -> sh5 1.79*** .08 .84 1.40*** .08 .70 
     sh1-sh3 (error cov.) -.15 .16 -.04 -.63** .21 -.15 
     sh1-sh4 (error cov.) .13 .19 .02 .12 .23 .03 
     sh2-sh3 (error cov.) .10 .17 .02 .14 .21 .03 
     sh4-sh5 (error cov.) -.47** .16 -.10 -.70*** .19 -.17 
Biological hierarchy beliefs   
X2 (3) = 2.69, p = .44  
RMSEA = 0 (0, .072) 
Pclose-fit = .82 
CFI = 1.00 
SRMR = .014 
  
X2 (3) = 2.97, p = .40  
RMSEA = 0 (0, .075) 
Pclose-fit = .79 
CFI = 1.00 
SRMR = .016 
     BH -> bh1 1.16*** .10 .66 1.26*** .10 .63 
     BH -> bh2 1.24*** .11 .72 1.37*** .10 .70 
     BH -> bh3 1.83*** .13 .70 .87*** .13 .39 
     BH -> bh4 1.47*** .12 .72 1.50*** .10 .72 
     BH -> bh5 1.46*** .11 .70 1.35*** .10 .64 
     bh2-bh3 (error cov.) -.37 .20 -.08 -.44** .16 -.10 
     bh3-bh5 (error cov.) .28 .29 .05 .42* .21 .09 
Notes.  IND = individual-oriented norms/values (latent factor); ach1-ach3 = items on individual achievements (indicators); GRP = group-oriented norms/values 
(latent factor); selfin1-selfin3 = items on the self-ingroup overlap (indicators); role1-role2 = items on roles/positions within ingroups (indicators); SH = social 
hierarchy beliefs (latent factor); sh1-sh5 = items on social hierarchy beliefs (indicators); BH = biological hierarchy beliefs (latent factor); bh1-bh5 = items on 
biological hierarchy beliefs (indicators); 1 The SB scaling method was applied; RMSEA was reported with 90% CI; * p < .05, ** p < .01, *** p < .001 
 
  92 
 
Table 5 
Configural and Metric Invariance in Measurement Models 
Models  
 
?X2 1 ?RMSEA ?CFI ?SRMR X2 2 df P RMSEA CFI SRMR 
Individual-
oriented 
norms/values 
All loadings are equal (H0) -- -- -- -- 6.92 5 .23 .028 1 .032 
All loadings are freed (vs. H0) 4.62 .008 0 .014 3.63 3 .30 .020 1 .018 
ach1 = 1 -- -- -- -- -- -- -- -- -- -- 
IND -> ach2 is freed (vs. H0) .50 .002 0 0 5.86 4 .21 .030 1 .032 
IND -> ach3 is freed (vs. H0) 4.54* .013 ? 0 .012 4.47 4 .35 .015 1 .020 
           
Group-
oriented 
norms/values 
All loadings/error covs are equal (H0) -- -- -- -- 28.10 11 .003 .056 .98 .050 
All loadings/error covs are freed (vs. H0) 26.93*** .056 ? .02 ? .043 ?  1.06 4 .90 0 1 .007 
All loadings are freed (vs. H0) 15.28** .013 (?) .01 ? .015 13.41 7 .06 .043 .99 .035 
selfin1 = 1 -- -- -- -- -- -- -- -- -- -- 
GRP -> selfin2 is freed (vs. H0) 4.68* .004 .01 ? .003 23.65 10 .01 .052 .99 .047 
GRP -> selfin3 is freed (vs. H0) 2.83 .001 .02 ?  0 25.17 10 .005 .055 .98 .050 
GRP -> role1 is freed (vs. H0) 1.60 .001 .02 ? 0 26.51 10 .003 .057 .98 .050 
GRP -> role2 is freed (vs. H0) 9.29*** .014 (?) .01 ? .009 19.03 10 .04 .042 .99 .041 
           
Social 
hierarchy 
beliefs 
All loadings/error covs are equal (H0) -- -- -- -- 19.12 10 .04 .043 1 .055 
All loadings/error covs are freed (vs. H0) 20.58** .043 ?  0 .054 ? .08 2 .96 0 1 .001 
All loadings are freed (vs. H0) 18.23*** .043  0 .041 ? 5.26 6 .51 0 1 .014 
SH -> sh1 is freed (vs. H0) .64 .002 0 .001 18.15 9 .03 .045 1 .056 
SH -> sh2 is freed (vs. H0) 1.20 0 0 .002 17.53 9 .01 .043 1 .053 
sh3 = 1 -- -- -- -- -- -- -- -- -- -- 
SH -> sh4 is freed (vs. H0) 9.45*** .017 ? 0 .017 12.15 9 .21 .026 1 .038 
SH -> sh5 is freed (vs. H0) 1.78 0 0 .003 17.50 9 .04 .043 1 .052 
           
Biological 
hierarchy 
beliefs 
All loadings/error covs are equal (H0) -- -- -- -- 31.38 12 .002 .057 .99 .080 
All loadings/error covs are freed (vs. H0) 29.54*** .057 ? .01 ? .064 ? 5.59 6 .47 0 1 .016 
All loadings are freed (vs. H0) 29.14*** .057 ? .01 ? .063 ? 6.29 8 .62 0 1 .017 
BH -> bh1 is freed (vs. H0)  2.71 .001 0 .003 29.68 11 .002 .058 .99 .077 
BH -> bh2 is freed (vs. H0) .32 .001 0 .003 29.97 11 .002 .058 .99 .077 
BH -> bh3 is freed (vs. H0) 36.56*** .057 ? .01 ? .057 ? 7.76 11 .73 0 1 .023 
bh4 = 1 -- -- -- -- -- -- -- -- -- -- 
BH -> bh5 is freed (vs. H0) .36 .003 .01 ? .001 30.69 11 .001 .060 .99 .081 
Notes.  IND = individual-oriented norms/values (latent factor); ach1-ach3 = items on individual achievements (indicators); GRP = group-oriented norms/values 
(latent factor); selfin1-selfin3 = items on the self-ingroup overlap (indicators); role1-role2 = items on roles/positions within ingroups (indicators); SH = social 
hierarchy beliefs (latent factor); sh1-sh5 = items on social hierarchy beliefs (indicators); BH = biological hierarchy beliefs (latent factor); bh1-bh5 = items on 
biological hierarchy beliefs (indicators); 1 The SB scaling method was applied; 2 The SB scaled X2 difference computation procedure for LISREL 8 users (Bryant 
& Satorra, 2012) was used; * p < .05, ** p < .01, *** p < .001, ? Significant differences based on cutoff points suggested by Chen (2007)       
 
 
 
  93 
 
Table 6 
The Structural Models in the US and South Korea 
 All in the US 
(n = 508)  
Whites in the US 
(n = 292) 
 All in South Korea 
(n = 502) 
Koreans in South Korea 
(n = 482) 
Unst. SE St.  Unst. SE St.  Unst. SE St.  Unst. SE St.  
Two-Factor Model        
     IND -> SH .22** .07 .20  .18* .07 .18  -.13 .11 -.14  -.12 .11 -.12  
     IND -> BH .12* .06 .14  .05 .07 .05  -.09 .08 -.11  -.08 .08 -.11  
     GRP -> SH .59*** .18 .23  .51** .17 .25  1.59*** .31 .77  1.54*** .30 .76  
     GRP -> BH .52** .17 .26  .52** .17 .29  1.07*** .24 .62  1.04*** .23 .62  
     IND-GRP (cov.) .54*** .14 .47  .47** .16 .37  .68*** .13 .61  .70*** .14 .60  
     SH-BH (error cov.) 1.69*** .19 .64  1.42*** .18 .61  1.07*** .16 .59  1.09*** .17 .59  
     Total variance in SH 13% 13% 49% 48% 
     Total variance in BH 12% 10% 31% 31% 
     Model fit statistics 1 
 
X2 (117) = 188.40, p < .001  
RMSEA = .035 (.025, .044) 
Pclose-fit = 1.00 
CFI = .99, SRMR = .045 
X2 (117) = 161.22, p < .001  
RMSEA = .036 (.021, .049) 
Pclose-fit = .96 
CFI = .99, SRMR = .053 
X2 (117) = 228.76, p < .001  
RMSEA = .044 (.035, .052) 
Pclose-fit = .89 
CFI = .98, SRMR = .041 
X2 (117) = 215.16, p < .001  
RMSEA = .042 (.033, .050) 
Pclose-fit = .94 
CFI = .99, SRMR = .041 
  
Three-Factor Model     
     ACH -> SH .24*** .07 .21  .20** .07 .19  -.05 .09 -.05  -.04 .09 -.04  
     ACH -> BH .13* .07 .14  .04 .08 .04  -.03 .07 -.04  -.02 .07 -.03  
     SELFIN -> SH .25 .16 .16  .23 .26 .16  .20 .18 .14  .20 .17 .15  
     SELFIN -> BH .16 .15 .13  .05 .26 .04  .19 .14 .16  .19 .12 .17  
     ROLE -> SH .17 .15 .12  .16 .21 .13  .76*** .23 .58  .74*** .23 .56  
     ROLE -> BH .19 .14 .17  .29 .22 .28  .46** .17 .41  .45** .16 .41  
     ACH-SELFIN (cov.) .36* .14 .19  .33 .18 .18  .64*** .14 .40  .66*** .15 .39  
     ACH-ROLE (cov.) 1.09*** .17 .50  .85*** .19 .40  1.00*** .15 .58  1.01*** .15 .57  
     SELFIN-ROLE (cov.) .96*** .17 .61  1.14*** .22 .74  .74*** .13 .63  .76*** .13 .63  
     SH-BH (error cov.) 1.65*** .19 .63  1.41*** .18 .60  1.18*** .14 .65  1.21*** .15 .66  
     Total variance in SH 15% 14% 42%  41% 
     Total variance in BH 13% 11% 26%  26% 
     Model fit statistics 1 
 
X2 (115) = 184.14, p < .001  
RMSEA = .034 (.025, .043) 
Pclose-fit = 1.00 
CFI = .99, SRMR = .039 
X2 (115) = 158.67, p < .001  
RMSEA = .036 (.021, .049) 
Pclose-fit = .96 
CFI = .99, SRMR = .050 
X2 (115) = 227.35, p < .001  
RMSEA = .044 (.036, .053) 
Pclose-fit = .87 
CFI = .98, SRMR = .040 
X2 (115) = 213.64, p < .001  
RMSEA = .042 (.033, .051) 
Pclose-fit = .93 
CFI = .99, SRMR = .040 
Notes.  IND = individual-oriented norms/values; GRP = group-oriented norms/values; SH = social hierarchy beliefs; BH = biological hierarchy beliefs; ACH = 
individual achievements; SELFIN = the self-ingroup overlap; ROLE = roles/positions within ingroups  
1 The SB scaling method was applied and RMSEA was reported with 90% C.I. 
* p < .05, ** p < .01, *** p < .001 
 
 
 
  94 
 
Table 7 
Cultural Differences in Structural Models between the US and South Korea 
  
 
 
US vs. Korea 
  
US-W vs. KR-K 
 
?df 
 
df 
 
?X2 1 
 
X2 2 
 
RMSEA 
 
CFI 
 
SRMR 
 
 
 
?X2 1  
 
X2 2 
 
RMSEA 
 
CFI 
 
SRMR 
Two-Factor Model              
 All parameter estimates are equal (H0) -- 288 -- 631.38 .049 .98 .074  -- 578.31 .051 .97 .061 
All parameter estimated are freed (vs. H0) 54 234 230.28*** 416.16 .039 .99 .041  155.48*** 421.81 .046 .98 .041 
      
      
Metric 
invariance  
All paths are freed (vs. H0) 40 248 164.40*** 472.34 .042 .99 .059  106.70*** 465.94 .048 .98 .052 
IND -> SH is freed (vs. H0) 37 251 154.99*** 480.75 .043 .99 .062  96.24*** 476.80 .048 .98 .057 
IND -> BH is freed (vs. H0) 37 251 155.15*** 480.71 .043 .99 .061  96.66*** 476.22 .048 .98 .056 
GRP -> SH is freed (vs. H0) 37 251 154.82*** 481.55 .043 .99 .062  96.86*** 477.25 .048 .98 .057 
GRP -> BH is freed (vs. H0) 37 251 154.71*** 481.52 .043 .99 .061  95.89*** 477.68 .048 .98 .056 
Partial  
metric  
invariance  
All paths are freed (vs. H0) 47 241 210.32*** 434.78 .040 .99 .046  138.48*** 436.22 .046 .98 .045 
IND -> SH is freed (vs. H0) 44 244 199.45*** 443.32 .040 .99 .048  128.26*** 444.81 .046 .98 .047 
IND -> BH is freed (vs. H0) 44 244 199.31*** 443.97 .040 .99 .048  127.99*** 445.87 .046 .98 .048 
GRP -> SH is freed (vs. H0) 44 244 199.12*** 444.24 .040 .99 .049  128.43*** 446.30 .046 .98 .048 
GRP -> BH is freed (vs. H0) 44 244 199.00*** 444.28 .040 .99 .048  127.44*** 446.97 .046 .98 .048 
              
Three-Factor Model              
 All parameter estimates are equal (H0) -- 286 -- 624.38 .048 .98 .072  -- 575.07 .051 .97 .060 
All parameter estimates are freed (vs. H0) 56 230 231.02*** 410.44 .039 .99 .040  157.04*** 418.07 .046 .98 .040 
 Metric 
invariance  
All paths are freed (vs. H0) 43 243 171.59*** 461.83 .042 .99 .055  111.25*** 460.07 .048 98 .050 
ACH -> SH is freed (vs. H0) 38 248 153.54*** 479.76 .043 .99 .064  94.67*** 478.45 .049 .98 .061 
ACH -> BH is freed (vs. H0) 38 248 154.14*** 479.20 .043 .99 .062  95.53*** 477.14 .049 .98 .059 
SELFIN -> SH is freed (vs. H0) 38 248 154.63*** 479.30 .043 .99 .062  95.37*** 477.70 .049 .98 .059 
SELFIN -> BH is freed (vs. H0) 38 248 154.37*** 479.22 .043 .99 .062  95.39*** 478.17 .049 .98 .059 
ROLE -> SH is freed (vs. H0) 38 248 155.88*** 477.66 .043 .99 .062  97.61*** 475.51 .049 .98 .058 
ROLE -> BH is freed (vs. H0) 38 248 153.73*** 479.90 .043 .99 .062  94.66*** 478.72 .049 .98 .059 
Partial  
metric 
invariance  
All paths are freed (vs. H0) 49 237 211.09*** 428.56 .040 .99 .045  140.08*** 432.12 .046 .98 .044 
ACH -> SH is freed (vs. H0) 44 242 194.50*** 441.08 .040 .99 .053  125.66*** 445.40 .047 .98 .053 
ACH-> BH is freed (vs. H0) 44 242 194.92*** 440.94 .040 .99 .052  126.14*** 444.47 .047 .98 .052 
SELFIN -> SH is freed (vs. H0) 44 242 195.80*** 441.16 .040 .99 .052  126.28*** 445.27 .047 .98 .052 
SELFIN -> BH is freed (vs. H0) 44 242 195.53*** 440.77 .040 .99 .052  126.35*** 445.22 .047 .98 .052 
ROLE -> SH is freed (vs. H0) 44 242 196.56*** 439.82 .040 .99 .052  128.08*** 442.96 .046 .98 .051 
ROLE -> BH is freed (vs. H0) 44 242 194.48*** 441.60 .040 .99 .052  125.40*** 445.92 .047 .98 .052 
Notes.  IND = individual-oriented norms/values; GRP = group-oriented norms/values; SH = social hierarchy beliefs; BH = biological hierarchy beliefs; ACH = 
individual achievements; SELFIN = the self-ingroup overlap; ROLE = roles/positions within ingroups 
1 The SB scaled X2 difference computation procedure for LISREL 8 users (Bryant & Satorra, 2012) was used. 
2 The SB scaling method was applied.  
*** p < .001 
 
 
 
 
 
  95 
 
Table 8  
Means (SDs) and Cultural Differences in the Degree of Perceived Social Distance to Four Target Groups and Individual- and Group-oriented Norms/Values 
between the US and South Korea 
   
The United States 
 
 
South Korea 
 
 
Cultural differences 1 
Low SES groups 
 
 6.17 (1.54) 5.73 (1.06) F(1, 606) = 16.76, p < .001, ?2p = .03 
 
  F(1, 606) = 21.48, p < .001, ?2p = .03 
 
Non-normative groups 
 
 5.60 (1.81) 6.23 (1.44) F(1, 606) = 22.18, p < .001, ?2p = .04 
 
  F(1, 606) = 20.74, p < .001, ?2p = .03 
 
Racial/ethnic outgroups  4.87 (1.94) 5.15 (1.21) F(1, 606) = 4.60, p = .03, ?2p = .01 
 
  F(1, 606) = 4.68, p = .03 ?2p = .01 
 
Value-based outgroups  5.65 (1.35) 5.69 (1.20) F(1, 606) = .17, p = .68, ?2p < .001 
 
  F(1, 606) = .04, p = .85, ?2p < .001 
 
Essentialism  5.60 (1.68) 6.13 (1.29) F(1, 606) = 19.29, p < .001, ?2p = .03 
  F(1, 606) = 19.50, p < .001, ?2p = .03 
Individual uniqueness  6.85 (1.09) 6.01 (1.16) F(1, 606) = 84.10, p < .001, ?2p = .12 
 
  F(1, 606) = 87.96, p < .001, ?2p = .13 
Conformity with 
norms/conventions 
 4.72 (1.49) 5.70 (1.24) F(1, 606) = 77.93, p < .001, ?2p = .11 
  F(1, 606) = 90.54, p < .001, ?2p = .13 
1 Cultural differences based on the mean scores are stated in the top lines and cultural differences based on the latent factor (PAF) scores are stated in the bottom 
lines.   
 
 
 
 
 
 
 
 
  96 
 
Table 9 
The Baseline Measurement Models in the US and South Korea 
 The United States (n = 306) 
 
 South Korea (n = 302) 
Unst. SE St. Model fit statistics 1 Unst. SE St. Model fit statistics 1 
Individual uniqueness 2     
X2 (1) = .34, p = .56  
RMSEA = 0 (0, .13) 
Pclose-fit = .69 
CFI = 1.00 
SRMR = .007 
     
X2 (1) = 4.21, p = .04  
RMSEA = .10 (.018, .21) 
Pclose-fit = .12 
CFI = .98 
SRMR = .025 
     UNIQUE -> unique1 .69*** .12 .44 .41** .14 .23 
     UNIQUE -> unique2 .82*** .12 .53 .95*** .17 .55 
     UNIQUE -> unique3  1.13*** .11 .76 1.42*** .22 .86 
     UNIQUE -> unique4 .77*** .11 .55 .55*** .14 .35 
     unique1-unique4 (error cov.) .62*** .14 .28 .96*** .20 .34 
Conformity with norms/conventions   
X2 (1) = 3.21, p = .07  
RMSEA = .085 (0, .20) 
Pclose-fit = .18 
CFI = 1.00 
SRMR = .014 
  
X2 (1) = 2.06, p = .15  
RMSEA = .059 (0, .18) 
Pclose-fit = .30 
CFI = 1.00 
SRMR = .011 
     CONFORM-> conform1 1.49*** .12 .80 1.37*** .11 .82 
     CONFORM -> conform2 1.18*** .12 .60 .94*** .09 .67 
     CONFORM -> conform3 1.20*** .12 .64 .82*** .11 .48 
     CONFORM -> conform4 1.23*** .12 .65 .89*** .10 .56 
     conform3-conform4 (error cov.) .95*** .20 .27 1.24*** .16 .46 
Social distance    
 
X2 (23) = 33.17, p = .08  
RMSEA = .038 (0, .065) 
Pclose-fit = .75 
CFI = .99 
SRMR = .031 
  
 
X2 (23) = 28.41, p = .20  
RMSEA = .026 (0, .057) 
Pclose-fit = .89 
CFI = .99 
SRMR = .031 
     LOWSES -> poor 1.48*** .14 .80 1.13*** .11 .77 
     LOWSES -> uneduc 1.36*** .11 .77 .80*** .08 .64 
     LOWSES -> lowcoun 1.34*** .12 .70 .86*** .08 .70 
     NONNOR -> homo 1.20*** .15 .53 .72*** .13 .36 
     NONNOR -> disabi 1.50*** .16 .72 1.29*** .14 .83 
     ETHOUT -> noname 1.72*** .12 .85 1.00*** .09 .74 
     ETHOUT -> immigr 1.93*** .12 .92 1.03*** .10 .73 
     VALOUT -> difper 1.23*** .12 .75 1.21*** .14 .83 
     VALOUT -> difsoc 1.05*** .12 .65 .77*** .11 .55 
     VALOUT -> difrel .96*** .12 .58 .71*** .12 .45 
     poor-uneduc (error cov.) -.01 .24 -.00 .10 .12 .05 
     poor-lowcoun (error cov.) -.24 .17 -.07 -.26* .11 -.14 
     lowcoun-homo (error cov.) -.52** .18 -.12 -.31* .13 -.13 
     lowcoun-noname (error cov.) .87*** .21 .22 .15 .08 .09 
     lowcoun-immigr (error cov.) 1.03*** .22 .26 .09 .09 .05 
     difsoc-difrel (error cov.) .75*** .19 .28 .88*** .16 .40 
     LOWSES-NONNOR (cov.) .67*** .08 .67 .86*** .10 .86 
     LOWSES-ETHOUT (cov.) .49*** .06 .49 .41*** .08 .41 
     LOWSES-VALOUT (cov.) .50*** .07 .50 .23** .07 .23 
     NONNOR-ETHOUT (cov.) .42*** .07 .42 .41*** .08 .41 
     NONNOR-VALOUT (cov.) .46*** .08 .46 .25** .08 .25 
     ETHOUT-VALOUT (cov.) .19** .07 .19 .49*** .08 .49 
 
Notes.  UNIQUE = individual uniqueness (latent factor); unique1-unique4 = items on individual uniqueness (indicators); CONFORM = conformity with 
norms/conventions (latent factor); conform1-conform4 = items on conformity with norms/conventions (indicators); LOWSES = distance to low SES groups 
(latent factor); poor, uneduc, & lowcoun  = items on low SES groups (indicators, poor/unemployed/homeless, uneducated, & born in a low-status country, 
  97 
 
respectively); NONNOR = distance to non-normative groups (latent factor); homo & disabi = items on non-normative groups (indicators, homosexuals & born 
with physical/mental disability, respectively); ETHOUT = distance to racial/ethnic outgroups (latent factor); noname & immigr = items on racial/ethnic 
outgroups (indicators, non-Americans/Koreans & immigrants to the US/Korea, respectively); VALOUT = distance to value-based outgroups (latent factor); 
difper, difsoc, & difrel = items on value-based outgroups (indicators, distance to people with different personal interests/tastes, social/political/economic issues, 
& religious issues, respectively)   
 
1 RMSEA was reported with 90% CI 
2 Satorra-Bentler (SB) scaling method was applied 
* p < .05, ** p < .01, *** p < .001 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  98 
 
Table 10  
Configural and Metric Invariance in Measurement Models 
Models  
 
?X2 ?df ?RMSEA ?CFI ?SRMR X2  Df P RMSEA CFI SRMR 
Individual 
uniqueness1 2 
All loadings/error covs are equal (H0) -- -- -- -- -- 32.44 11 .001 .080 .95 .087 
All loadings/error covs are freed (vs. H0) 27.55*** 9 .017? .040? .062? 4.39 2 .11 .063 .99 .025 
unique1 = 1 -- -- -- -- -- -- -- -- -- -- -- 
UNIQUE -> unique2 is freed (vs. H0) 24.86*** 7 .030? .040? .051? 7.04 4 .13 .050 .99 .036 
UNIQUE -> unique3 is freed (vs. H0) 24.05*** 7 .024? .040? .042? 7.78 4 .10 .056 .99 .045 
UNIQUE -> unique4 is freed (vs. H0) 24.12*** 7 .025? .040? .048? 7.61 4 .11 .055 .99 .039 
            
Conformity 
with norms/ 
conventions 
All loadings/error covs are equal (H0) -- -- -- -- -- 72.49 11 .001 .130 .93 .150 
All loadings/error covs are freed (vs. H0) 67.22*** 9 .057? .070? .139? 5.27 2 .07 .073 1.00 .011 
conform1 = 1 -- -- -- -- -- -- -- -- -- -- -- 
CONFORM -> conform2 is freed (vs. H0) 64.91*** 7 .076? .070? .119? 7.58 4 .11 .054 1.00 .031 
CONFORM -> conform3 is freed (vs. H0) 65.65*** 7 .082? .070? .126? 6.84 4 .14 .048 1.00 .024 
CONFORM -> conform4 is freed (vs. H0) 64.92*** 7 .076? .070? .119? 7.57 4 .11 .054 1.00 .031 
            
Social 
Distance 
All loadings/covs are equal (H0) -- -- -- -- -- 307.18 78 .001 .092 .91 .170 
All loadings/covs are freed (vs. H0) 245.60*** 32 .060? .080? .139? 61.58 46 .06 .032 .99 .031 
poor = 1 -- -- -- -- -- -- -- -- -- -- -- 
LOWSES -> uneduc is freed (vs. H0) 241.58*** 27 .062? .080? .136? 65.60 51 .08 .030 .99 .034 
LOWSES -> lowcoun is freed (vs. H0)  239.24*** 27 .060? .080? .135? 67.94 51 .06 .032 .99 .035 
homo = 1 -- -- -- -- -- -- -- -- -- -- -- 
NONNOR -> disabi is freed (vs. H0) 241.18*** 27 .062? .080? .137? 66.00 51 .08 .030 .99 .033 
noname = 1 -- -- -- -- -- -- -- -- -- -- -- 
ETHOUT -> immigr is freed (vs. H0) 239.49*** 27 .060? .080? .135? 67.69 51 .06 .032 .99 .035 
difper = 1 -- -- -- -- -- -- -- -- -- -- -- 
VALOUT -> difsoc is freed (vs. H0) 239.34*** 27 .060? .080? .135? 67.84 51 .06 .032 .99 .035 
VALOUT -> difrel is freed (vs. H0) 239.21*** 27 .060? .080? .135? 67.97 51 .06 .032 .99 .035 
Notes.  ESSEN = essentialism (latent factor); essen1-essen2 = items on essentialism; UNIQUE = individual uniqueness (latent factor); unique1-unique4 = items 
on individual uniqueness (indicators); CONFORM = conformity with norms/conventions (latent factor); conform1-conform4 = items on conformity with 
norms/conventions (indicators); LOWSES = distance to low SES groups (latent factor); poor, uneduc, & lowcoun  = items on low SES groups (indicators, 
poor/unemployed/homeless, uneducated, & born in a low-status country, respectively); NONNOR = distance to non-normative groups (latent factor); homo & 
disabi = items on non-normative groups (indicators, homosexuals & born with physical/mental disability, respectively); ETHOUT = distance to racial/ethnic 
outgroups (latent factor); noname & immigr = items on racial/ethnic outgroups (indicators, non-Americans/Koreans & immigrants to the US/Korea, 
respectively); VALOUT = distance to value-based outgroups (latent factor); difper, difsoc, & difrel = items on value-based outgroups (indicators, distance to 
people with different personal interests/tastes, social/political/economic issues, & religious issues, respectively)   
1 Satorra-Bentler (SB) scaling method was applied; 2 The SB scaled X2 difference computation procedure for LISREL 8 users (Bryant & Satorra, 2012) was used 
* p < .05, ** p < .01, *** p < .001; ? Significant differences based on cutoff points suggested by Chen (2007)       
 
 
 
 
  99 
 
Table 11 
Structural Models in the US and South Korea 
 The United States (n = 306)  
 
 South Korea (n = 302) 
Unst. SE St. Model fit statistics 1, 2 Unst. SE St. Model fit statistics 1, 2 
Essentialism      
X2 (38) = 44.24, p = .22  
RMSEA = .023 (0, .048) 
Pclose-fit = .96 
CFI = 1.00 
SRMR = .033 
     
X2 (38) = 31.36, p = .77  
RMSEA = 0 (0, .029) 
Pclose-fit = 1.00 
CFI = 1.00 
SRMR = .031 
     ESSEN -> LOWSES .35** .14 .26 .61** .22 .39 
     ESSEN -> NONNOR .24 .14 .23 .42* .17 .42 
     ESSEN -> ETHOUT .17 .13 .11 .51** .19 .38 
     ESSEN -> VALOUT .02 .11 .02 .26 .18 .16 
     LOWSES-NONNOR (error cov.) 1.08*** .24 .60 .60*** .15 .71 
     LOWSES-ETHOUT (error cov.) 1.16*** .22 .45 .29* .12 .26 
     LOWSES-VALOUT (error cov.) .91*** .19 .49 .23 .12 .17 
     NONNOR-ETHOUT (error cov.) .79*** .22 .39 .19* .08 .26 
     NONNOR-VALOUT (error cov.) .67*** .19 .45 .17 .09 .19 
     ETHOUT-VALOUT (error cov.) .38* .17 .18 .52*** .14 .43 
     Total variance in LOWSES 7% 15% 
     Total variance in NONNOR 6% 17% 
     Total variance in ETHOUT 1% 14% 
     Total variance in VALOUT .02% 3% 
    
Uniqueness-Conformity     
X2 (112) = 172.93, p < .001  
RMSEA = .042 (.029, .054) 
Pclose-fit = .85 
CFI = .98 
SRMR = .059 
     
X2 (112) = 130.10, p = .12  
RMSEA = .023 (0, .039) 
Pclose-fit = 1.00 
CFI = .99 
SRMR = .054 
     UNIQUE -> LOWSES .02 .15 .01 -.26 .27 -.10 
     UNIQUE -> NONNOR -.36* .17 -.22 -.09 .16 -.05 
     UNIQUE -> ETHOUT -.08 .17 -.03 .04 .21 .02 
     UNIQUE -> VALOUT .001 .15 .00 -.19 .26 -.07 
     CONFORM -> LOWSES .27** .08 .26 .20** .06 .25 
     CONFORM -> NONNOR .29** .09 .33 .12* .06 .23 
     CONFORM -> ETHOUT .07 .09 .06 .13* .06 .18 
     CONFORM -> VALOUT .18* .08 .21 .19* .08 .24 
     UNIQUE-CONFORM (cov.) -.22* .10 -.20 .-.05 .06 -.08 
     LOWSES-NONNOR (error cov.) 1.11*** .22 .58 .67*** .16 .81 
     LOWSES-ETHOUT (error cov.) 1.20*** .22 .47 .42** .13 .37 
     LOWSES-VALOUT (error cov.) .82*** .18 .44 .22 .12 .17 
     NONNOR-ETHOUT (error cov.) .88*** .21 .40 .28** .10 .38 
     NONNOR-VALOUT (error cov.) .63*** .18 .39 .17 .10 .21 
     ETHOUT-VALOUT (error cov.) .37* .17 .17 .52*** .14 .47 
     Total variance in LOWSES 7% 8% 
     Total variance in NONNOR 18% 6% 
     Total variance in ETHOUT 1% 3% 
     Total variance in VALOUT 4% 7% 
 
  100 
 
Notes.  ESSEN = essentialism (latent factor); UNIQUE = individual uniqueness (latent factor); CONFORM = conformity with norms/conventions (latent factor); 
LOWSES = distance to low SES groups (latent factor); NONNOR = distance to non-normative groups (latent factor); ETHOUT = distance to racial/ethnic 
outgroups (latent factor); VALOUT = distance to value-based outgroups (latent factor) 
 
1 Satorra-Bentler (SB) scaling method was applied 
2 RMSEA was reported with 90% CI 
* p < .05, ** p < .01, *** p < .001 
    
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  101 
 
 
 
Table 12  
Cultural Differences in the Structural Models between the US and South Korea   
 
 
?df 
 
?X2 df X2 RMSEA CFI SRMR 
Essentialism        
     All parameter estimates are equal (H0) -- -- 109 261.04 .068 .95 .110 
     All parameter estimates are freed (vs. H0) 31 178.36*** 78 74.55 .000 1.00 .031 
     ESSEN -> LOWSES is freed (vs. H0) 28 172.24*** 81 77.50 .000 1.00 .035 
     ESSEN -> NONNOR is freed (vs. H0) 28 172.23*** 81 77.51 .000 1.00 .035 
     ESSEN -> ETHOUT is freed (vs. H0) 28 172.50*** 81 76.63 .000 1.00 .034 
     ESSEN -> VALOUT is freed (vs. H0) 28 173.90*** 18 76.45 .000 1.00 .033 
     LOWSES-NONNOR is freed (vs. H0 22 144.60*** 87 105.98 .027 .99 .063 
     LOWSES-ETHOUT is freed (vs. H0) 22 146.38*** 87 101.25 .023 .99 .049 
     LOWSES-VALOUT is freed (vs. H0) 22 145.38*** 87 103.04 .025 .99 .046 
     NONNOR-ETHOUT is freed (vs. H0) 22 143.20*** 87 105.16 .026 .99 .055 
     NONNOR-VALOUT is freed (vs. H0) 22 141.36*** 87 107.92 .028 .99 .056 
     ETHOUT-VALOUT is freed (vs. H0) 22 150.71*** 87 97.58 .020 1.00 .064 
        
Uniqueness-Conformity        
     All parameter estimates are equal (H0) -- -- 271 536.37 .057 .94 .084 
     All parameter estimates are equal (H0) 45 217.58*** 226 299.93 .033 .98 .054 
     UNIQUE -> LOWSES is freed (vs. H0) 38 208.80*** 233 304.08 .032 .98 .056 
     UNIQUE -> NONNOR is freed (vs. H0) 38 209.27*** 233 305.46 .032 .98 .054 
     UNIQUE -> ETHOUT is freed (vs. H0) 38 209.67*** 233 306.90 .032 .98 .055 
     UNIQUE -> VALOUT is freed (vs. H0) 38 208.85*** 233 307.27 .032 .98 .055 
     CONFORM -> LOWSES is freed (vs. H0) 38 209.37*** 233 307.58 .033 .98 .055 
     CONFORM -> NONNOR is freed (vs. H0) 38 210.69*** 233 305.09 .032 .98 .055 
     CONFORM -> ETHOUT is freed (vs. H0) 38 209.47*** 233 307.85 .033 .98 .055 
     CONFORM -> VALOUT is freed (vs. H0) 38 208.62*** 233 307.83 .033 .98 .055 
     UNIQUE-CONFORM is freed (vs. H0)  31 177.56*** 240 340.69 .037 .98 .065 
     LOWSES-NONNOR is freed (vs. H0) 32 184.07*** 239 337.28 .037 .98 .069 
     LOWSES-ETHOUT is freed (vs. H0) 32 185.68*** 239 332.17 .036 .98 .061 
     LOWSES-VALOUT is freed (vs. H0) 32 184.57*** 239 332.29 .036 .98 .060 
     NONNOR-ETHOUT is freed (vs. H0) 32 179.52*** 239 338.28 .037 .98 .065 
     NONNOR-VALOUT is freed (vs. H0) 32 178.38*** 239 341.10 .038 .98 .065 
     ETHOUT-VALOUT is freed (vs. H0) 32 186.83*** 239 330.01 .035 .98 .069 
Notes. ESSEN = essentialism (latent factor); UNIQUE = individual uniqueness (latent factor); CONFORM = conformity with norms/conventions (latent factor); 
LOWSES = distance to low SES groups (latent factor); NONNOR = distance to non-normative groups (latent factor); ETHOUT = distance to racial/ethnic 
outgroups (latent factor); VALOUT = distance to value-based outgroups (latent factor) 
1 Satorra-Bentler (SB) scaling method was applied; 2 The SB scaled X2 difference computation procedure for LISREL 8 users (Bryant & Satorra, 2012) was used 
* p < .05, ** p < .01, *** p < .001
  102 
 
Figure 1 
Proportions of Mean Negative Responses (Social Distance) to Groups Perceived as Personally 
Responsible and Essentially Different in the US (Whites) and South Korea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  103 
 
Figure 2 
Social and Biological Hierarchy Beliefs in the US and South Korea  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  104 
 
Figure 3 
Standardized Parameter Estimates of the Two-Factor Structural Models in the US and South Korea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  105 
 
Figure 4 
Standardized Parameter Estimates of the Three-Factor Structural Models in the US and South Korea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  106 
 
Figure 5 
Mean Social Distance to Four Types of Target Groups in the US and South Korea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  107 
 
Figure 6 
Standardized Parameter Estimates of the Structural Models in the US and South Korea 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
  108 
 
Figure 7 
Standardized Parameter Estimates of the Structural Models in the US and South Korea 
 
 
 
 
  109 
 
Appendices 
Appendix A 
 
 Items 
 
Individual-
oriented 
norms/values 
Individual 
achievements 
ach1.  When I think about a person, I usually think first of what he or she has achieved or accomplished in the past.   
ach2.  When I think about a person, I usually think first of what he or she is currently trying to achieve or accomplish.   
ach3.  When I think about a person, I usually think first of what he or she will achieve or accomplish in the future.  
  
Group- 
oriented 
norms/values 
The self-ingroup 
overlap 
selfin1.  What is good for the group I belong to (e.g., family, a group of friends, community, religious group, school, 
company, etc.) is also mostly what is good for me.  1   
selfin2.  Even when I would rather do something different, I usually go along with what the group I belong to (e.g., family, 
a group of friends, community, religious group, school, company, etc.) wants to do.   
selfin3.  Once the group I belong to (e.g., family, a group of friends, community, religious group, school, company, etc.) 
makes a decision, I usually accept the decision without any disagreement. 
Roles/positions 
within ingroups 
role1.  A person?s role or position in the groups he or she belongs to is very important to me in defining who that person is 
 
role2.  A person?s role or position in the family (e.g., the head or eldest son/daughter of the family) is very important to me 
in defining who that person.   
 
Hierarchically-
based 
intergroup 
biases 
Social  
hierarchy beliefs 
sh1.  In general, people who graduate from a high-ranked school are superior. 
 
sh2.  People who have a job that is socially recognized are superior to people who have a job that is not socially 
recognized.   
sh3.  People who have low income are inferior to people whose income is high.   
 
sh4.  People who have a professional degree (e.g., a doctoral degree) are superior to people who do not. 
 
sh5.  People who have not attended a college are inferior to people who have attended.   
 
Biological 
hierarchy beliefs 
bh1.  In general, the physically disabled are inferior. 
 
bh2.  In general men are superior to women. 
 
bh3.  There are people who are genetically superior or inferior from birth. 
 
bh4.  Certain races or nations are superior to other races or nations.  
 
bh5.  People who are genetically superior should be in more important positions within a society or group.   
 
1 Modified from an item in the individualism-collectivism scale implemented by Oyserman (1993). 
 
  110 
 
Appendix B 
 
 Items 
 
Essentialism essen1.  Some people are smarter that others by birth. 
essen2.  There are people who are good or bad by nature; such nature (the innate disposition) doesn?t change easily during 
their lifetime.  
 
Individual uniqueness unique1.  I enjoy being unique and different from others in many aspects.  
unique2.  I usually display my personal preferences when I?m in a group.  
unique3.  I like those who express their own personal preferences when they are in a group.   
unique4.  I like those who have their own unique life styles.   
 
Conformity with 
norms/conventions 
conform1.  It is important to follow the social conventions or customs most people of the society follow.    
conform2.  It is important to follow the moral standards most people in the society follow.  
conform3.  It is better to follow what the majority of the society does.     
conform4.  It is better to follow what most people normally do.  
 
 
  111 
 
Bibliography 
 
Aiken, L. S., & West S. G. (1991). Multiple regression: Testing and interpreting 
interactions. Newbury Park, CA: Sage. 
Allport, G. W. (1954/1979). The nature of prejudice. Cambridge, MA: Perseus Books. 
Ashmore, R. D., Deaux, K., & McLaughlin-Volpe, T. (2004). An organizing framework 
for collective identity: Articulation and significance of multidimensionality. 
Psychological Bulletin 130 (1), 80-114.  
Bentler, P. M. (1990). Comparative fit indexes in structural models. Psychological 
 Bulletin, 107, 238-246.   
Bentler, P. M., & Wu, E. J. C. (2002). EQS for Windows user?s guide. Encino, CA:  
 Multivariate Software, Inc.   
Berry, J. W. (1994). Ecology of individualism and collectivism. In U. Kim, H. C. 
Triandis, C. Kagit?ibasi, S. C. Choi, & G. Yoon (Eds.), Individualism and 
collectivism:  Theory, method, and applications (pp. 77-84). Thousand Oaks, CA:  
Sage.  
Bobocel, D. R., & Hafer, C. L. (2007). Justice motive theory and the study of justice in 
 work organizations: A conceptual integration. European Psychologist, 12, 283- 
 289. 
Bogardus, E. S. (1933). A social distance scale. Sociology and Social Research, 17, 265- 
 271.  
Breivik, E., & Olsson, U. H. (2001). Adding variables to improve fit:  The effect of 
 model size on fit assessment in LISREL. In R. Cudeck, S. Du Toit, & D. S?rbom 
  112 
 
 (Eds.), Structural equation modeling:  Present and future. A festschrift in honor of 
 Karl J?reskog (pp. 169-194). Lincolnwood, IL: Scientific Software International.    
Brett, J. M., & Gelfand, M. J. (2006). A cultural analysis of the underlying assumptions 
 of negotiation theory. In L. Thompson (Ed.), Frontiers of negotiation research 
 (pp. 173-201). Philadelphia, PA: Psychology Press. 
Brewer, M. B., & Chen, Y.-R. (2007). Where (who) are collectives in collectivism?  
Toward conceptual clarification of individualism and collectivism. Psychological 
Bulletin 114 (1), 133-151.  
Browne, 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 
Bryant, F. B., & Satorra, A. (2012). Principles and practice of scaled difference chi-
square testing. Structural Equation Modeling 19 (3), 372-398 
Byrne, B. M. (2008). Testing for multigroup equivalence of a measuring instrument: A 
walk through the process. Psicothema 20(4), 872-882. 
Byrne, B. M., Shavelson, R. J., & Muthen, B. (1989). Testing for the equivalence of 
 factor covariance and mean structures: The issue of partial measurement 
invariance. Psychological Bulletin 105 (3), 456-466.  
Campbell, D. T. (1965). Ethnocentric and other altruistic motives. In D. Levine (Ed.), 
Nebraska symposium on motivation (Vol. 13, pp. 283-311). Lincoln, NE: 
University of Nebraska Press.    
Cha, J.-H. (1994). Aspects of individualism and collectivism in Korea. In U. Kim, H. C. 
Triandis, C. Kagit?ibasi, S.-C. Choi, & G. Yoon (Eds.), Individualism and 
  113 
 
collectivism:  Theory, method, and applications (pp. 157-174). Thousand Oaks, 
CA: Sage. 
Chen, F. F. (2007). Sensitivity of goodness of fit indices to lack of measurement 
invariance. Structural Equation Modeling, 14, 464-504. 
Chen, F. F. (2008). What happens if we compare chopsticks with forks? The Impact of 
making inappropriate comparisons in cross-cultural research. Journal of 
Personality and Social Psychology, 95(5), 1005-1018. 
Chiao, J. Y., & Ambady, N. (2007). Cultural neuroscience: Parsing universality and  
 diversity. In Kitayama, & D. Cohen (Eds.), Handbook of cultural psychology (pp. 
237-254). New York, NY: The Guilford Press.     
Cho, H. (1998). Male dominance and mother power: The two sides of Confucian 
patriarchy in Korea. In W. Slote & G. DeVos (Eds.), Confucianism and the family 
(pp. 187-207). Albany, NY:  State University of New York Press.   
Cho, K. (2007). ???? ????? ????? ?? [Confucian background of East  
 Asian collectivism]. Seoul, Korea: Jisik-Sanup Publishing Co., Ltd.  
Choi, I., & Nisbett, R. (1998). Situational salience and cultural differences in the 
Correspondence bias and actor-observer bias. Personality and Social Psychology  
Bulletin, 9, 949?960. 
Choi, I., Nisbett, R. E., & Norenzayan, A. (1999). Causal attribution across cultures:   
 Variation and universality. Psychological Bulletin, 125, 47-63.     
Cohen, D., & Kitayama, S. (2007). Cultural psychology: This stanza and the next. In S. 
Kitayama, & D. Cohen (Eds.), Handbook of cultural psychology (pp. 847-851).  
New York, NY: The Guilford Press.      
  114 
 
Crandall, C. C., & Martinez, R. (1996). Culture, ideology, and antifat attitudes.  
 Personality and Social Psychology Bulletin, 22, 1165-1176.    
Cuddy, A. J. C., Fiske, S. T., Kwan, V. S. Y., Glick, P., Demoulin, S, Leyens, J.- 
 P.,?Ziegler, R. (2009). Stereotype content model across cultures:  Towards 
universal similarities and some differences. British Journal of Social Psychology,  
48, 1-33.  
Curran, P. J., West, S. G., & Finch, J. F. (1996). The robustness of test statistics to non- 
 Normality and specification error in confirmatory factor analysis. Psychological 
Methods, 1, 16-29.  
Dovidio, J. F., Major, B., & Crocker, J. (2000). Stigma:  Introduction and overview. In T. 
F. Heatherton, R. E. Kleck, M. R. Hebl, & J. G. Hull (Eds.), The social 
psychology of stigma (pp. 1-28). New York, NY:  The Guilford Press.  
Dovidio, J. F., & Gaertner, S. L. (2010).  Intergroup bias.  In S. T. Fiske, D. Gilbert, & G. 
Lindzey (Eds.), Handbook of social psychology (Vol. 2, 1084-1121).  New York: 
Wiley. 
Dovidio, J. F., Hewstone, M., Glick, P., & Esses, V. M. (2010). Prejudice, stereotyping, 
And discrimination:  Theoretical and empirical overview. In J. F. Dovidio, M. 
Hewstone, P. Glick, & V. M. Esses (Eds.), The SAGE handbook of prejudice,  
stereotyping and discrimination (pp. 3-28). Thousand Oaks, CA: Sage. 
Dovidio, J. F., Major, B., & Crocker, J. (2000). Stigma: Introduction and overview. In T. 
F. Heatherton, R. E. Kleck, M. R. Hebl, & J. G. Hull (Eds.), The social  
psychology of stigma (pp. 1?28). New York, NY: The Guilford Press. 
Duckitt, J. (2005). Personality and prejudice. In J. F. Dovidio, P. Glick, & L. A. Rudman 
  115 
 
(Eds.), On the nature of prejudice:  Fifty years after Allport (pp. 227-243).  
Malden, MA:  Blackwell. 
Esses, V. M., Dovidio, J. F., Jackson, L. M., & Armstrong, T. L. (2001). The immigration 
dilemma:  The role of perceived group competition, ethnic prejudice, and national 
identity. Journal of Social Issues, 57, 389-412.  
Esses, V. M., Jackson, L. M., & Armstrong, T. L. (1998). Intergroup competition and 
Attitudes toward immigrants and immigration: An instrumental model of group 
conflict. Journal of Social Issues, 54, 699-724.   
Esses, V. M., Jackson, L. M., Dovidio, J. F., & Hodson, G. (2005). Instrumental relations 
among groups:  Group competition, conflict, and prejudice. In J. F. Dovidio, P. 
Glick, & L. A. Rudman (Eds.), On the nature of prejudice:  Fifty years after 
Allport (pp. 227-243). Malden, MA: Blackwell. 
Finney, S. J., & DiStefano, C. (2006). Nonnormal and categorical data in structural 
Equation modeling. In G. R. Hancock, & R. O. Mueller (Eds.). Structural 
equation modeling: A second course (pp. 269-314). Greenwich, CT: Information 
Age Publishing, Inc.   
Fiske, S. T. (1998). Stereotyping, prejudice, and discrimination. In D. T. Gilbert, S. T.  
 Fiske, & G. Lindzey (Eds.), The handbook of social psychology (4th ed., Vol. 2,  
 pp. 357-411). New York: McGraw-Hill.   
Fiske, A., Kitayama, S., Markus, H. R., & Nisbett, R. E. (1998). The cultural matrix of 
social psychology. In D. T. Gilbert, S. T. Fiske, & G. Lindzey (Eds.), The 
handbook of social psychology (pp. 915-981). New York: McGraw-Hill.  
Gaertner, S. L., & Dovidio, J. F. (2012). Reducing intergroup bias: The Common Ingroup 
  116 
 
Identity Model. In Van Lange, P. A. M., Kruglanski, A. W., & Higgins, E. T.  
(Eds), Handbook of Theories of Social Psychology. Thousand Oaks, CA: Sage. 
Gardner, W. L., & Seeley, E. A. (2001). Confucius, ?Jen,? and the benevolent use of  
 power:  The interdependent self as psychological contract preventing exploitation. 
In A. Y. Lee-Chai & J. A. Bargh (Eds.), The use and abuse of power (pp. 263-280). 
Philadelphia, PA: Psychology Press.  
Gilbert, D. T., & Malone, P. S. (1995). The correspondence bias. Psychological Bulletin, 
117, 21?38.  
Goffman, E. (1963). Stigma:  Notes on the management of spoiled identity. Englewood 
Cliffs, NJ:  Prentice-Hall.  
Greenwald, A. G., McGhee, D. E., & Schwartz, J. K. L. (1998). Measuring individual 
differences in implicit cognition: The Implicit Association Test. Journal of  
Personality and Social Psychology, 74, 1464-1480. 
Hamilton, D. L., & Sherman, S. J. (1996). Perceiving persons and groups. Psychological 
Review,103, 336-355. 
Haslam, N., Bastian, B., Bain, P., & Kashima, Y. (2006). Psychological essentialism, 
Implicit theories, and intergroup relations. Group Processes & Intergroup 
Relations, 9, 63?76. 
Hayduk, L. A., Pazderka-Robinson, H., Cummings, G. C., Levers, M. J. D., & Beres, M. 
A. (2005). Structural equation model testing and the quality of natural killer cell 
Activity measurements. Medical Research Methodology, 5(1).   
Heaven, P. C. (1990). Human values and suggestions for reducing unemployment.  
 British Journal of Social Psychology, 29, 257?264.  
  117 
 
Heine, S. J. (2010). Cultural psychology. In D. T. Gilbert, S. Fiske, & G. Lindzey (Eds.), 
Handbook of Social Psychology (pp. 1423-1464). New York: Wiley. 
Heine, S. J., Lehman, D. R., Markus, H. R., & Kitayama, S. (1999). Is there a universal 
need for positive self-regard? Psychological Review, 106, 766-794.   
Henrich, J. Heine, S. J., & Norenzayan, A. (2010). The weirdest people in the world? 
Behavioral and Brain Science, 1-75. doi:10.1017/S0140525X0999152X 
Henry, P. J., & Sears, D. O. (2002). The symbolic racism 2000 scale. Political 
Psychology, 23(2), 253-283.  
Horn, J. L., & McArdle, J. J. (1992). A practical and theoretical guide to measurement 
invariance in aging research. Experimental Aging Research, 18, 117-144.  
Hofstede, G. (1980). Culture?s consequences: International differences in work-related 
values.  London: Sage. 
Hofstede, G. (1991). Cultures and organizations: Software of the mind.  London: 
McGraw-Hill. 
Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure 
analysis: Conventional criteria versus new alternatives. Structural Equation 
Modeling, 6, 1-55.    
Im, H. (2011, February 21). ?Graduating from a good school is the most important factor 
to succeed in the Korean society.? The Korea Times. Retrieved from  
http://www.hankooki.com  
Jackman, M. R. (1994). The velvet glove. Berkeley, CA: University of California Press.  
Jordan, D. K. (1998). Filial piety in Taiwanese popular thought. In W. Slote & G. DeVos 
(Eds.) Confucianism and the family (pp. 267-283). Albany, NY: State University 
  118 
 
of New York Press. 
J?reskog K. G. (1971). Simultaneous factor analysis in several populations.  
 Psychometrika, 36, 409-426.  
J?reskog K. G., & S?rbom D. (2006). LISREL (Version 8.80) [Computer software]. 
Lincolnwood, IL: Scientific Software International 
Jones, J. M. (1986). Racism: A cultural analysis of the problem.  In J. F. Dovidio & S. L. 
Gaertner (Eds.), Prejudice, discrimination, and racism (pp. 279-314). Orlando, 
FL: Academic Press. 
Jones, E. E., Farina, A., Hastorf, A. H., Markus, H., Miller, D. T., & Scott, R. A. (1984). 
Social stigma:  The psychology of marked relationships. New York:  Freeman.  
Jost, J. T., & Banaji, M. R. (1994). The role of stereotyping in system-justification and 
The production of false consciousness. British Journal of Social Psychology, 33, 
1-27.   
Jost, J. T., Banaji, M. R., & Nosek, B. A. (2004). A decade of system justification theory: 
Accumulated evidence of conscious and unconscious bolstering of the status quo. 
Political Psychology, 25, 881-920.  
Jost, J. T., Kivetz, Y., Rubini, M., Guermandi, G., & Mosso, C. (2005). System-justifying 
functions of complementary regional and ethnic stereotypes:  Cross-national 
evidence. Social Justice Research, 18, 305-333.  
Kaiser, C. R., & Miller, C. T. (2001). Reacting to impending discrimination: 
Compensation for prejudice and attribution to discrimination. Personality and 
Social Psychology Bulletin, 27, 1357-67. 
Kashima, Y., Kashima, E., Chiu, C.-Y., Farsides, T., Gelfand, M., Hong, Y.- 
  119 
 
 Y.,?Yzerbyt, V. (2005). Culture, essentialism, and agency: Are individuals  
 universally believed to be more real entities than groups? European Journal of  
 Social Psychology, 35, 147-169.     
Kashima, Y., Siegal, M., Tanaka, K., & Isaka, H. (1988). Universalism in lay conceptions 
Of distributive justice:  A cross-cultural examination. International Journal of 
Psychology, 23, 51-64.  
Kashima, Y., Siegal, M., Tanaka, K., & Kashima, E. S. (1992). Do people believe 
behaviors are consistent with attitudes?  Toward a cultural psychology of 
attribution processes. British Journal of Social Psychology, 331, 111-124.   
Katz, I., & Hass, R. G. (1988). Racial ambivalence and American value conflict:   
 Correlational and priming studies of dual cognitive structures. Journal of  
 Personality and Social Psychology, 55, 893-905.  
Kazin, M, Edwards, R, & Rothman, A. (Eds.) (2010). The Princeton encyclopedia of 
American political history. Princeton, NJ: Princeton University Press.  
Keller, J. (2005). In genes we trust: The biological component of psychological 
essentialism and its relationship to mechanisms of motivated social cognition. 
Journal of Personality and Social Psychology, 88, 686?702. 
Kim, H., & Markus, H. R. (1999). Deviance or uniqueness, harmony or conformity? A 
cultural analysis. Journal of Personality and Social Psychology, 77(4), 785-800.    
Kim, U. (1994). Individualism and collectivism: Conceptual clarification and elaboration. 
In U. Kim, H. C. Triandis, C. Kagit?ibasi, S.-C. Choi, & G. Yoon (Eds.), 
Individualism and collectivism. Theory, method, and applications (pp. 19-40). 
Thousand Oaks, CA: Sage. 
  120 
 
Kim, U., Triandis, H. C., Kagit?ibasi, C., Choi, S.-C., & Yoon, G. (Eds.) (1994). 
Individualism and collectivism. Theory, method, and applications. Thousand 
Oaks, CA: Sage. 
Kitayama, S., & Cohen, D. (Eds.) (2007). Handbook of cultural psychology. New York,  
 NY: The Guilford Press.   
Kline, R. B. (2011). Principles and practice of structural equation modeling (3rd ed.).  
 New York, NY: The Guild Press.    
Kluckhohn, C. (1951). Values and value-orientations in the theory of action: An  
 exploration in definition and classifications. In T. Parsons & E. Shils (Eds.),  
 Toward a general theory of action (pp. 388?433). Cambridge, MA: Harvard  
 University Press. 
Lebra, T. S. (1998). Confucian gender role and personal fulfillment for Japanese women.  
 In W. Slote & G. DeVos (Eds.), Confucianism and the family (pp. 209-227). 
Albany, NY: State University of New York Press. 
Leung, K., Au, Y., Fernandez-Dols, J. M., & Iwawaki, S. (1992). Preference for methods 
Of conflict processing in two collectivist cultures. International Journal of  
Psychology, 27, 195-209.    
Leung, K., Bond, M. H., Carment, D. W., Krishman, L., & Liebrand, W. B. G. (1990). 
Effects of cultural femininity on preference for methods of conflict processing: A 
cross-cultural study. Journal of Experimental Social Psychology, 26, 373-388.   
LeVay, S. (1991). A difference in hypothalamic structure between heterosexual and 
homosexual men. Science, New Series, 253(5023), 1034-1037.   
  121 
 
Leyens, J.-P., Cortes, B., Demoulin, S., Dovidio, J. F., Fiske, S. T., Gaunt, R., et al. 
(2003). Emotional prejudice, essentialism, and nationalism. European Journal of 
Social Psychology, 33, 703?717. 
Little, R. J. A. (1988). A test of missing completely at random for multivariate data with 
missing values, Journal of the American Statistical Association, 83, 1198-1202.  
Little, R. J. A., & Rubin, D. B. (2002). Statistical analysis with missing data (2nd ed.). 
Hoboken, NJ:  Wiley.  
Locke K., & Baik, K-D. (2009). Does an acquiescent response style explain why Koreans 
are less consistent than Americans?  Journal of Cross-Cultural Psychology, 40(2), 
319-323.    
Major, B., Kaiser, C. R., O?Brien, L. T., & McCoy, S. K. (2007). Perceived  
 discrimination as worldview threat or worldview confirmation: Implications for 
self-esteem.  Journal of Personality and Social Psychology, 92, 1068-1086. 
Mahler, I. (1974). A comparative study of locus of control. Psychologia 17(3), 135?139. 
Marcus, E. (2002). Making gay history:  The half century fight for lesbian and gay equal 
rights. New York, NY: HarperCollins Publishers, Inc.  
Mardia, K. V. (1970). Measures of multivariate skewness and kurtosis with applications.   
 Biometrika, 57, 519-530.  
Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, 
emotion, and motivation. Psychological Review, 98, 224-253. 
Markus, H. R. & Kitayama, S. (1994). The cultural construction of self and emotion: 
Implications for social behavior. In S. Kitayama & H. R. Markus (Eds.), Emotion 
  122 
 
and culture: Empirical studies of mutual influence (pp. 89?132). Washington, 
DC: American Psychological Association. 
Markus, H. R., Kitayama, S., & Heiman, R. J. (1998). Culture and ?basic? psychological 
principles. In E. T. Higgins & A. W. Kruglanski (Eds.), Social psychology:  
Handbook of basic principles (pp. 857-913). New York, NY: The Guilford Press.    
Markus, H. R., Mullally, P., & Kitayama, S. (1997). Selfways: Diversity in modes of 
Cultural participation. In U. Neisser & D. A. Jopling (Eds.), The conceptual self  
in context: Culture, experience, self-understanding (pp. 13?61). Cambridge, 
England: Cambridge University Press. 
Masuda, T., & Nisbett, R. E. (2001). Attending holistically versus analytically: 
Comparing the context sensitivity of Japanese and Americans, Journal of 
Personal Social Psychology, 81, 922-934. 
McConahay, J. B. (1986). Modern racism, ambivalence, and the Modern Racism Scale. 
In J. F. Dovidio & S. L. Gaertner (Eds.), Prejudice, discrimination, and racism 
(pp. 99?125). Orlando, FL: Academic Press.  
Meade, A. W., & Lautenschlager, G. J. (2004). A Monte-Carlo study of confirmatory 
factor analytic tests of measurement equivalence/invariance. Structural Equation 
Modeling, 11, 60?72.  
Menon, T., Morris, M. W., Chiu, C.-Y., & Hong, Y. Y. (1999). Culture and construal of 
agency: Attribution to individual versus group dispositions. Journal of 
Personality and Social Psychology, 76(5), 701-717.  
Milfont, T. L., & Fischer, R. (2010). Testing measurement invariance across groups:   
 Applications in cross-cultural research. International Journal of Psychological 
  123 
 
 Research, 3(1), 111-121.    
Millsap, R. E., & Kwok, O.-M. (2004). Evaluating the impact of partial factorial 
invariance on selection of multiple populations. Psychological Methods, 9, 93-15.   
Morris, M. W., & Peng, K. (1994). Culture and cause: American and Chinese attributions 
For physical and social events. Journal of Personality and Social Psychology, 67,  
949?971. 
Morrison, M. A., & Morrison, T. G. (2003). Development and validation of a scale 
Measuring modern prejudice toward gay men and lesbian women. Journal of 
Homosexuality, 43(2), 15-37.  
Naoi, A., & Schooler, C. (1985). Occupational conditions and psychological functioning 
in Japan. American Journal of Sociology, 90, 729-752.   
Nevitt, J., & Hancock, G. R. (2000). Improving the root mean square error of  
approximation for nonnormal conditions in structural equation modeling. Journal 
of Experimental Education, 68, 251-268.    
Nisbett, R. E. (2003). The geography of thought:  How Asians and Westerners think 
differently and why.  New York:  The Free Press. 
Nisbett, R. E., Peng, K., Choi, I. & Norenzayan, A. (2001). Culture and system of  
thought: Holistic versus analytic cognition, Psychological Review, 108, 291-310. 
Olsson, U. H., Foss, T., & Breivik, E. (2004). Two equivalent discrepancy functions for  
 maximum likelihood estimation:  Do their test statistics follow a non-central chi- 
square distribution under model misspecification?  Sociological Methods and 
Research, 32, 453-500.  
Oyserman, D. (1993). The lens of personhood:  Viewing the self and others in  
  124 
 
multicultural society. Journal of Personality and Social Psychology, 65, 993- 
1009.   
Parsons, O. A., & Schneider, J. M. (1974). Locus of control in university students from 
Eastern and Western societies. Journal of Consulting and Clinical Psychology 
42(3), 456?461. 
Pettigrew, T. F. (1959). Regional differences in anti-Negro prejudice. The Journal of 
Abnormal and Social Psychology, 59, 28-36.   
Pettigrew, T. F. (2001). Personality and sociocultural factors in intergroup attitudes: A 
cross-national comparison. In M. A. Hogg & D. Abrams (Eds.), Intergroup 
relations: Essential reading. Key readings in social psychology (pp. 18-29). New 
York: Psychological Press.  
Pettigrew, T. F. (2003). People under threat: Americans, Arabs, and Israelis. Peace and 
Conflict: Journal of Peace Psychology, 9, 68-90.   
Pettigrew, T. F., & Tropp, L. (2006). A meta-analytic test of intergroup contact theory. 
Journal of Personality and Social Psychology, 90, 751-783.  
Pratto, F., Sidanius, J., Stallworth, L. M., & Malle, B. F. (1994). Social dominance 
orientation: A personality variable predicting social and political attitudes. 
Journal of Personality and Social Psychology, 67, 741-763. 
Pratto, F., Liu, J.H., Levin, S., Sidanius, J., Shih, M., Bachrach, H., & Hegarty, P. (2000). 
Social dominance orientation and the legitimization of inequality across cultures.  
Journal of Cross-Cultural Psychology, 31, 369-409. 
Pratto, F., Saguy, T., Stewart, A. L., Morselli, D., Foels, R., Aiello, A.,?Sweetman, J. 
(2013). Social dominance in context and in individuals:  Contextual moderation 
  125 
 
of robust effects of social dominance orientation in 15 language and 20 countries. 
Social Psychological Personality Science, 4(5), 587-599.  
Quinn, D. M., & Crocker, J. (1999). When ideology hurts: Effects of belief in the 
Protestant ethic and feeling overweight on the psychological well-being of  
women. Journal of Personality and Social Psychology, 77, 402-414. 
Rokeach, M. (1973). The nature of human values. New York: Free Press.  
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the 
attribution process. In L. Berkowitz (Ed.), Advances in experimental social 
psychology (vol. 10, pp. 173?220). New York: Academic Press. 
Rotter, J. (1966). Generalized expectancies for internal versus external control of 
reinforcements. Psychological Monographs, 80, 1-28. 
Satorra, A., & Bentler, P. M. (1988). Scaling corrections for chi-square statistics in 
covariance structure analysis. Proceedings of the Business and Economic 
Statistics Section (pp. 308?313). Alexandria, VA: American Statistical  
Association. 
Satorra, A., & Bentler, P. M. (1994). Corrections to test statistics and standard errors in 
covariance structure analysis. In A. von Eye & C. C. Clogg (Eds.), Latent 
variable analysis: Applications to developmental research (pp. 399-419). 
Thousand Oaks, CA: Sage Publications, Inc. 
Satorra, A., & Bentler, P. M. (2010). Ensuring positiveness of the scaled chi-square test  
 statistic. Psychometrika, 75, 243?248.  
Schaberg, L. (2002). Toward a cultural broadening of feminist theory:  A comparison of 
  126 
 
everyday outlooks on social roles, dissent, and feminism among the United States 
and Japanese adults (Unpublished doctoral dissertation). University of Michigan, 
Ann Arbor.  
Schiefer, D., Mollering, A., Daniel, E., Benish-Weisman, M., & Boehnke, Klaus. (2010). 
Cultural values and outgroup negativity: A cross-cultural analysis of early and 
late adolescents. European Journal of Social Psychology, 40 (4), 635-651.  
Schwartz, S. H. (1992). Universals in the content and structure of values: Theory and 
empirical tests in 20 countries. In M. Zanna (Ed.), Advanced in experimental 
social psychology (pp. 1-65). New York: Academic Press.  
Schwartz, S. H. (1994). Beyond individualism/collectivism: New cultural dimensions of  
 values. In U. Kim, H. C. Triandis, C. Kagit?ibasi, S.-C. Choi, & G. Yoon (Eds.), 
Individualism and collectivism. Theory, method, and applications (pp. 85-119).  
Thousand Oaks, CA: Sage. 
Shin, H., Dovidio, J. F., & Napier, J. L. (2013). Cultural differences in targets of 
stigmatization between individual- and group-oriented cultures. Basic and 
Applied Social Psychology 35(1), 98-108. 
Sidanius, J., & Pratto, F. (1999). Social dominance:  An intergroup theory of social 
hierarchy and oppression. New York: Cambridge University Press.  
Simmel, G. (1917). Grundfragen der Soziologie [Fundamental questions of sociology]. 
Berlin, Germany: G?schen 
Slote, W, & DeVos, G. (1998). Confucianism and the family. Albany, NY:  State 
University of New York Press. 
  127 
 
Smith, P. B. & Bond, M. H. (1993). Social psychology across cultures: Analysis and 
perspectives. Hemel Hempstead, England: Harvester Wheatsheaf.  
Smith, E. R., & Henry, S. (1996). An in-group becomes part of the self:  Response time 
evidence. Personality and Social Psychology Bulletin, 22, 635-642.  
Spector, P. E., Cooper, C. L., Sanchez, J. I., O?Driscoll, M., Sparks, K., Bernin, P.,  
 Bossing, A.,?Yu, S. (2002). Locus of control and well-being at work: How 
generalizable are Western findings? Academy of Management Journal, 45, 453- 
466. 
Stangor, C. (2004). Social groups in action and interaction. New York, NY: Psychology 
Press.  
Steele, C. M., Spencer, S. & Aronson, J. (2002). Contending with group image: The 
Psychology of stereotype and social identity threat. In M. Zanna (ed.), Advances 
in experimental social psychology, 34. New York NY: Academic Press.  
Steiger, J. H. (1990). Structural model evaluation and modification:  An interval  
 Estimation approach. Multivariate Behavioral Research, 25, 173-180.   
Stephan, W. G. & Stephan, C. W. (2000). An integrated threat theory of prejudice. In S. 
Oskamp (Ed.), Reducing prejudice and discrimination (pp. 23-46). Hillsdale, NJ: 
Lawrence Erlbaum.  
Tajfel, H., & Turner, J. C. (1979). An integrative theory of intergroup conflict. In W. G. 
Austin & S. Worchel (Eds.), The social psychology of intergroup relations (pp. 
33-48). Monterey, CA: Brooks/Cole.  
Triandis, H. C. (1994). Theoretical and methodological approaches to the study of 
collectivis and individualism. In U. Kim, H. C. Triandis, C. Kagit?ibasi, S.-C.  
  128 
 
Choi, & G. Yoon (Eds.), Individualism and collectivism. Theory, method, and  
applications (pp. 85-119). Thousand Oaks, CA: Sage. 
Triandis, H. C. (1995). Individualism and collectivism. Boulder, CO: Westview Press, 
Inc.  
Triandis, H. C. (2000). Dialectics between cultural and cross-cultural psychology. Asian  
Journal of Social Psychology, 3, 185-195.   
Triandis, H. C. (2007). Culture and psychology: A history of the study of their  
 relationship. In S. Kitayama, & D. Cohen (Eds.), Handbook of cultural 
psychology (pp. 59-76). New York, NY: The Guilford Press.       
Triandis, H. C., McCusker, C., & Hui, C. H. (1990). Multimethod probes of 
individualism and collectivism. Journal of Personality and Social Psychology, 59, 
1006-1020. 
Triandis, H. C., & Triandis, L. M. (1960). Race, social class, religion, and nationality as 
eterminants of social distance. The Journal of Abnormal and Social Psychology, 
61(1), 110-118.  
Tu, W.-M. (1998a). Confucius and Confucianism. In W. Slote & G. DeVos (Eds.),  
 Confucianism and the family (pp. 3-36). Albany, NY:  State University of New 
York Press. 
Tu, W.-M. (1998b). Probing the ?three bonds? and ?five relationships? in Confucian 
humanism. In W. Slote & G. DeVos (Eds.), Confucianism and the family (pp. 
121-136). Albany, NY:  State University of New York Press. 
van de Vijver, F. J. R., & Leung, K. (1997). Methods and data analysis for cross-cultural 
research. Thousand Oaks, CA: Sage.  
  129 
 
van de Vijver, F. J. R., & Watkins, D. (2006). Assessing similarity of meaning at the 
individual and country level: An investigation of measure of independent and  
interdependent self. European Journal of Psychological Assessment, 22(2), 69-77.  
Weber, M. (1905/2008). The Protestant ethic and the spirit of capitalism. Digireads.com 
publishing.  
Weiner, B. (1995). Judgments of responsibility: A foundation for a theory of social 
conduct. New York: Guilford 
Weisz, J. R., Rothbaum, F. M., & Blackburn, T. G. (1984). Standing out and standing in: 
The psychology of control in America and Japan. American Psychologist, 39, 
955-969. 
Whiting, B. B., & Whiting, J. W. M. (1975). Children of six cultures: A psycho-cultural 
analysis. Cambridge, MA: Harvard University Press. 
Williams, R. M., Jr. (1968). Values. In E. Sills (Ed.), International encyclopedia of the 
social sciences. New York, NY: Macmillan. 
Williams, M. J., & Eberhardt, J. L. (2008). Biological conceptions of race and motivation 
to cross racial boundaries. Journal of Personality and Social Psychology, 94 (6), 
1033-1047. 
Williams, T. P., & Sogon, S. (1984). Group composition and conforming behavior in 
Japanese students. Japanese Psychological Research, 26, 231-234.   
Winter, J. C. F., & Dodou, D. (2012).  Factor recovery by principal axis factoring and  
maximum likelihood factor analysis as a function of factor pattern and sample 
size. Journal of Applied Statistics, 39 (4), 695-710.   
  130 
 
World Values Survey (n.d.). World Values Survey Association 
(www.worldvaluessurvey.org).   
Yim, D. (1998). Psychocultural features of ancestor worship in modern Korean society. 
In W. Slote & G. DeVos (Eds.), Confucianism and the family (pp. 163-186). 
Albany, NY:  StatemUniversity of New York Press. 
Young, M. (1958, unpublished). The rise of the meritocracy. Britain Cambridge, UK:  
Blackwell.  
Yzerbyt, V., Corneille, O., & Estrada, C. (2001). The interplay of subjective essentialism 
and entitativity in the formation of stereotypes. Personality and Social  
Psychology Review, 5, 141?155. 
Yzerbyt, V.Y., Rocher, S., & Schadron, G. (1997). Stereotypes as explanations: A 
subjective essentialistic view of group perception. In R. Spears, P. J. Oakes, N. 
Ellemers, & S. A. 
Haslam (Eds.), Thesocial psychology of stereotyping and group life (pp. 20?50). 
Oxford, England: Blackwell. 
Zick, A., Pettigrew, T. F., & Wagner, U. (2008). Ethnic prejudice and discrimination in 
Europe. Journal of Social Issues. 64, 233-251.