Social Comparison, Individualism-Collectivism, and Self-Esteem in China and the United States.

Possible differences in social comparison choices between individualistic and collectivistic cultures and individuals were explored among a sample of 235 student participants from the United States and China, focusing on self-esteem as a possible mediating variable. A self-report social comparison measure was developed and factor analyzed. Correlations revealed that higher collectivism scores were associated with an increased desire to compare in general, an increased desire to make upward comparisons, and a decreased desire to make downward comparisons. It is speculated that upward self-improvement comparisons for the sake of the group may be more common for those high in collectivism.

Research in social comparison has been primarily conducted in individualistic cultures (the United States in particular), and self-enhancement has been a central motive in social comparison choices. In this study, possible differences in social comparison choices between individualistic and collectivistic cultures and individuals will be explored in the context of the United States and China.

Individualism and Collectivism

Individuals' environments influence their patterns of living, including their perspective on social relationships (Zhou, 1993; Greenhalgh, 1994; Liping et al., 1995). Therefore, cultural values and norms (Triandis, 1989), political influences (Keping, 1995), family relations (Greenhalgh, 1994; Kim, 1994), and social relationships (Zhou, 1993; Liping et al., 1995) need to be considered when studying people from various cultural backgrounds (Malpass, 1977; Betancourt and Lopez, 1993). Individualism and collectivism are frequently used to operationalize these influences (Triandis et al., 1986; Triandis, 1990, 1994; Triandis et al., 1990; Cha, 1994; Ho and Chiu, 1994; Kagitcibasi and Cigdem, 1994; Yamaguchi, 1994; Kashima et al. 1995; Rhee et al., 1995; Yamaguchi et al., 1995; Fijineman, 1996).

The continuum from collectivism to individualism describes how people relate to others in the environment. Someone who is highly collectivist works for the group and not for his or her own personal gain. Family relations are extremely important in the collective family, and decisions are made with family or group members involved. Conformity and compliance to social or group norms are more frequently displayed by those with collective views of the self. People high in collectivism prefer to do things with others and believe that cooperation is the best way to achieve goals. For them, personal matters tend to involve friends or the members of a particular group. They also believe that the group is entitled to know and regulate what the individual does and thinks in private matters. In sum, any action of an individual within a group effects the group and may cause public criticism (Triandis et al., 1990; Ho and Chiu, 1994; Kagitcibasi, 1994; Fijneman, 1996). Eastern Asian cultures tend to be high in collectivism, and China, in particular, is based on a collectivist political system (Bond, 1988, 1994; Chance, 1991; Borthwick, 1992; Dikotter, 1992; Morris and Peng, 1994; Keping, 1995; Rhee, 1995).

In contrast to collectivism, individualism is associated with self-reliance and work towards the fulfillment of individual needs and interests. Those high in individualism strive for personal excellence and status (Ho and Chiu, 1994; Fijneman, 1996). Tasks tend to be done by the individual who takes responsibility for his or her own actions. Individualists tend to find their sense of worth and value through self-actualization or self-realization in order to develop to his or her fullest (Triandis et al., 1985; Yamaguchi et al., 1995). Democracy tends to be associated with individualism, with its emphasis on satisfying personal needs and protecting individual rights by law (Greenhalgh, 1994; Ho and Chiu, 1994; Kagitcibasi, 1994; Kashima, 1995).

Conceptually, a person may display or experience a combination of individualist and collectivist characteristics in different social settings. For example, a person may experience and display individualist thinking and behavior at work, such as trying to impress her or his boss in hopes of a wage increase. At home, on the other hand, she or he may display collectivist

behavior and attitudes, such as sacrificing his or her time for the betterment of a family member.

Individualism-collectivism (or similar constructs) has been examined in a wide variety of contexts, including values (Hofstede, 1980), social systems (Parsons and Shils, 1951), morality (Shweder, 1982), religion (Bakan, 1966), cognitive differentiation (Witkin and Berry, 1975), economic development (Adelman and Morris, 1967), modernity (Inkeles and Smith, 1974), the structure of constitutions (Massimini and Calegari, 1979), and cultural patterns (Hsu, 1983). Particularly relevant to this study, is the relationship between individualism-collectivism and global self-esteem; Tafarodi and Swann (1996) suggest that collectivist cultures promote the development of the self-liking aspects of self-esteem, whereas individualist cultures promote the development of the self-competence dimension of self-esteem. The relationship between individualism-collectivism and social comparison has received limited attention; the self-esteem variable may provide an important link between these two literatures.

Social Comparison Theory

Social comparison theory as proposed by Festinger (1954) states that people compare themselves with others who perform, in some manner, better than they do (upward comparisons) or with someone who is worse than they are (downward comparisons; for a reviews of social comparison literature, see Wood, 1989, 1996). Self-esteem tends to be correlated with social comparison choices, such that individuals with high self-esteem tend to make upward comparison choices, whereas low self-esteem individuals tend make upward comparisons only when a threat to their self-esteem is not present (Wood et al., 1994).

Predictions

We expect collectivism to be positively related to upward comparison choices as opposed to downward comparison choices. We suggest that individuals high in collectivism may make upward (self-improving or self-deprecating) comparisons in order to contribute to the betterment of their group. Because self-esteem seems to function differently in individualistic and collectivistic cultures, however, it is possible that the reverse relationship may be found. This study is, thus, somewhat exploratory.

METHOD

Two-hundred thirty-five participants (78 students from a small church-affiliated liberal arts university in California, and 157 students from East China University of Science and Technology, Shanghai), were administered a) an individual self-esteem scale (Rosenburg, 1965); b) a Collective Self-Esteem Scale (Luhtanen and Crocker, 1992); c) the shortened individualism-collectivism scale (INDCOL) (Hui and Yee, 1994); and d) a social comparison scale constructed for this study. The social comparison scale consisted of thirty-four questions designed to measure desire to compare in general (e.g., "I never look to see where my score lies within a listed distribution"), the desire to make upward comparisons (e.g., "I usually do not want to see how many people scored higher than I did."), and the desire to make downward comparisons (e.g., "If the professor was to list a general breakdown of how many people received which grade, I usually want to see how many people scored lower than me."). All of the questions focused on comparison in an academic setting. Participants responded on a six-point Likert-type scale. For the participants in China, the scale was translated into Mandarin Chinese, and back translation was used to verify each question's accuracy.

RESULTS

Because the social comparison scale was developed for this study, factor analyses were performed on the thirty-four items in this questionnaire. A scree plot based on principal components extraction suggested two- and five-factor solutions. Two- and five-factor solutions were computed using maximum likelihood extraction and Varimax rotation, accounting for 21.0 percent and 32.6 percent of the variance, respectively. Factor loadings are shown in Tables 1 and 2. The two-factor solution produced factors in which …