Sex, communication values, and cultural values: individualism-collectivism as a mediator of sex...

In an effort to understand how socialization shapes value orientations, the current study examined the mediating role of individualism-collectivism on the connection between sex and communication values in two cultures, American and Chinese. Participants (97 Americans, 39 men and 58 women; 105

Chinese, 44 men and 61 women) completed scales providing assessments of communication values and individualism-collectivism. Regression analyses showed that collectivism was an especially strong mediator of the sex-communication values association among Americans, but not among Chinese. In addition, collectivism made strong, independent contributions to the prediction of communication values. These results are interpreted in terms of how broad-scale value orientations (such as collectivism) shape more specific orientations to particular forms and functions of communication.

**********

* Recent research has found that the value placed on supportive forms of communication, including comforting and ego support, is associated with the kinds of messages used in support situations, as well as with the quality of people's interpersonal relationships (Samter & Burleson, 1990). Ego support generally refers to making people feel good about who they are or the things they have accomplished (Burleson & Samter, 1990). Two common forms of ego support include encouraging support (providing encouragement for undertaking a difficult or challenging task) and celebratory support (celebrating another's achievements or accomplishments). Comforting skill refers to the ability to alleviate another's emotional distress (Burleson, 1994). These two forms of communication play central roles in most close personal relationships, including friendships and romantic relationships (Buhrmester, Furman, Wittenberg, & Reis, 1988; Burleson & Samter, 1994; Westmyer & Myers, 1996). That is, in most close relationships, people look to their partners for ego support in both good and uncertain times, and for comfort in difficult times. Thus, it makes good sense that people who place a high value on these supportive forms of communication are themselves viewed as desirable relationship partners. Indeed, people who value these forms of communication highly have been found to have more friends and to be better liked by peers (Samter, 1992, 1994; Samter & Burleson, 1990).

Not surprisingly, reliable (if small) sex differences have been found in the value placed on comforting and ego support skills, with women placing somewhat greater value on these forms of communication than men (Burleson, Kunkel, Samter, & Werking, 1996). This pattern of sex differences is consistent with broader patterns of sex differences in orientations to friendships (see the summary by Fehr, 1996) and romantic relationships (see the summary by Winstead, Derlega, & Rose, 1997). For example, in contrast to those of men, women's friendships have been described as: affectively richer (Booth, 1972), focused on reciprocity rather than commonality and association (Reisman, 1990; Weiss & Lowenthal, 1975) and "expressive" rather than "instrumental" (Bakan, 1966). These differences suggest that women emphasize the importance of skills through which feelings and emotions are expressed and discussed. Thus, it makes sense that women generally place greater value than men on affectively oriented skills such as comforting and ego support. Men, in contrast, appear to emphasize the importance of communication skills through which activities are smoothly and enjoyably coordinated; thus, men tend to value instrumental forms of communication such as persuasion and narrative skill more highly than women.

Cross-cultural research suggests that such differences may be more a product of cultural orientation than sex. Research examining sex differences across cultures suggest that sex, as an organizing principle for the development of communication styles, does not often operate the same way in different cultures (Waldron & Di Mare, 1998). For example, several studies comparing Japanese and Americans suggest that cultural differences are stronger predictors than sex differences. An analysis of 16 quantitative studies assessing a variety of communication variables (Waldron & Di Mare) found that three studies detected unqualified main effects for sex, six studies found unqualified main effects for culture, and seven studies found that culture moderated the effects of sex, with sex differences appearing in one culture, but not the other. Waldron and Di Mare concluded that although "biological sex by itself is not an important factor in determining patterns of communication in Japan or the United States ... sex may combine with cultural factors to affect communication patterns" (p. 188).

At the same time, research examining cultural value orientations suggests a connection between value orientations and biological sex. In his initial study of fifty cultures, Hofstede (1980) reported that men varied significantly more in their value orientations across cultures than did women. Women tended to be more similar in value orientations across cultures. In particular, across different cultures, women tended to be more collectivist than their male counterparts. Subsequent research has generally confirmed Hofstede's initial observations regarding sex differences in individualism-collectivism (see the review by Fischer & Manstead, 2000).

To date, communication values (and sex differences in these) have not been examined in relation to broader, culturally based values such as individualism-collectivism. That is, virtually all research examining communication values in close relationships has employed largely homogeneous samples of young, well educated, middle-class Americans. Thus, we do not currently know whether the sex differences in communication values observed in American samples generalize to other cultures. Nor do we currently have any understanding about how certain broad-based cultural value orientations, such as individualism-collectivism, might impact on people's communication values and explain sex-related differences in these communication values. In an effort to better understand how socialization shapes value orientations, the current study examined the mediating role of collectivism-individualism on the connection between sex and communication values in two cultures, American and Chinese.

Although there are many ways in which cultures differ, one dimension that has received considerable attention from both cross-cultural communication researchers and psychologists around the world is individualism-collectivism (Gaines, 1997; Gudykunst & Matsumoto, 1996). The importance of this dimension of culture can be seen in countless cross-cultural studies that have provided theoretical and empirical evidence of these value orientations in a wide range of cultures (see the review by Triandis, 1988).

The dimension of individualism-collectivism refers to the degree to which cultures encourage individual needs, wishes, desires, and values in relation to group and collective ones (Hofstede, 1980). Individualist cultures encourage their members to be unique; individual goals, values, behaviors, and self-expression take precedence over the collective needs of groups (Matsumoto, 1996). Individualist cultures foster an independent, autonomous sense of self that is clearly separated from others (Markus & Kitayama, 1991). Individuals with highly independent (individualist) self construals will have as a referent their own abilities, attributes, characteristics, or individual goals as opposed to the thoughts, feelings, or actions of others (Markus & Kitayama, 1994).

In contrast, collectivist cultures emphasize the needs of in-groups. Individual goals are subordinated to in-group goals. Individual identification in collectivist cultures occurs through in-group affiliations like family, classmates, or work colleagues. Conformity, compliance, and cooperation within in-groups are emphasized to a greater extent in collectivist cultures. Collectivist cultures foster an interdependent sense of self in which boundaries between oneself and others are less distinct (Markus & Kitayama, 1991). Individuals with highly interdependent (collectivist) self construals will have as a referent external, public features such as status, roles, relationships, and the occupation of one's proper place in the social landscape (Markus & Kitayama, 1991).

Different patterns in the communication behaviors of individualists and collectivists have been documented in a broad body of literature (Cai, Wilson, 8: Drake, 2000; Gudykunst & Matsumoto, 1996; Kim, 1993; Kim & Sharkey, 1995; Kim & Wilson, 1994). Of particular interest here is evidence that collectivist communicators demonstrate a greater concern for the feelings and face of others. In contrast, individualist communicators are more oriented to clarity and control rather than to feelings, and are influenced much less by issues of face.

Given that collectivist values place a premium upon relationships and the feelings of others, it makes sense to suggest that as a value orientation, collectivism should be a strong predictor of how people value supportive communication. Therefore it was hypothesized:

H1: Collectivism will be positively associated with the value placed on supportive communication skills (ego support and comforting).

Consistent with previous research (Burleson et al., 1996), it was expected that women would place a greater value on supportive communication skills than men. Thus, it was hypothesized:

H2: Women will place a higher value than men on supportive communication skills (ego support and comforting).

To date, no effort has been made to explain sex differences in communication values as a function of broader cultural value orientations. However, because collectivism and individualism have clear implications for patterns of communicative behavior, and are themselves sex-linked (Fischer & Manstead, 2000; Hofstede, 1980), it seemed possible that collectivism-individualism would mediate the effects of sex on supportive communication values. Hence, the following hypothesis:

H3: Collectivism-individualism will mediate the association between biological sex and the value placed on supportive communication skills.

To provide a generalizable test of these hypotheses, all three hypotheses were evaluated with both an American sample and a Chinese sample. On average, Americans are quite individualist while Chinese are quite collectivist (Hofstede, 1980; Markus & Kityama, 1994).

METHOD

Participants

Participants for this study were 203 students drawn from Euro-American (i.e., North American Caucasian cultural heritage) and Chinese cultural groups. Euro-American participants (39 males and 59 females) were recruited from students attending classes at a large Midwestern university. Euro-American participants were over 18 years old and U.S. citizens by birth. Chinese international students (44 males and 61 females) from the same university also served as participants. Cultural background was operationalized by having participants respond to a series of items that asked them to identify their major cultural/ethnic background (e.g., Euro-American, Latino, Asian, African, etc.). Chinese participants responded to questions that asked them to identify themselves as members of Taiwanese, Hong Kong, or Mainland Chinese nationalities. Chinese respondents averaged between 19-24 months of U.S. residency (range - 6 months to 72 months).

Procedure

Euro-American participants attended one of four out-of-class data collection sessions. Chinese participants were recruited via fliers and announcements made at Chinese student groups on campus. Chinese participants were asked to come to a scheduled time and place for one of two data collection sessions. Chinese participants were informed that they would receive a monetary payment after completing their questionnaires. All materials for Chinese participants including instruments, consent forms, and debriefing statements were translated into Chinese to assure that participants had a full understanding of their rights and roles within the project. After signing the informed consent form, participants filled out a background information questionnaire designed to assess general information about sex, age, college status, major, and cultural background.

Measures

Two versions of all measures were employed in this study. One version, completed by Euro-American participants, was written in Standard American English. A second version completed by Chinese participants was written in Chinese. Linguistic equivalency was obtained via back-translation technique. This involved first translating the measures into Chinese using a professional bilingual translator. Then, another translator translated the measures back into English. The two translators then discussed any discrepancies and jointly agreed on any changes in the Chinese version to ensure cross-cultural equivalence in meaning, while striving to make the measures sound natural in Chinese. For this project, bilingual Chinese graduate students who were experienced in translating English texts were employed as translators.

Communication values. Participants' judgments of comforting and ego support as communicative skills were obtained by having them complete a version of the Communication Functions Questionnaire (CFQ; Burleson & Samter, 1990). This version of the CFQ is composed of 40 randomly ordered items written to reflect outcomes typically achieved through effective exercise of 8 different communication skills; five, seven-point items are used to tap each of the eight skills. Only those items pertaining to comforting skill (the ability to make others feel better when depressed, sad, or upset) and ego support (the ability to make another feel good about himself or herself) were retained for analysis in the current study.

Participants were asked to rate how important it was for a close, same-sex friend to embody each of these abilities. The CFQ has been used in a number of studies and has been shown to be a valid and reliable measure in cross-ethnic studies (Samter, Whaley, Mortenson, & Burleson, 1997). Communication values, as tapped by the CFQ, have been found associated with several interesting relationship outcomes such as interpersonal acceptance (Samter & Burleson, 1990), loneliness (Samter, 1994), and relationship satisfaction (Burleson, Kunkel, & Birch, 1994). In the entire sample, Cronbach's alpha for the five comforting items was .73, and was .75 for the five ego support items. For Americans, the reliabilities for the comforting and ego-support items were .81 and .80, respectively, while for the Chinese, the reliabilities for the comforting and ego-support items were .64 and .71, respectively.

Individualism and collectivism. Participants' personal levels of individualism and collectivism were obtained by having them complete a version of the Self-Construal (SCS, Singelis, 1994). The SCS is composed of 29 seven-point items divided into independent and interdependent subscales. Independent items focus on (a) a sense of uniqueness and (b) a sense of self-reliance. Interdependent items focus on (a) the importance of group goals and (b) connectedness to others (Singelis, 1994). Validity for the SCS has been established in interethnic comparisons (Singelis, 1994; Singelis & Sharkey, 1995) and by convergence with other measures of collectivism (Singelis & Brown, 1995).

A factor analysis (principal axis extraction with varimax rotation) forcing a two factor solution indicated that all items had their primary loadings on the appropriate factor (i.e., collectivist items loaded on the collectivism factor while individualist items loaded on the individualism factor). Secondary loadings were minimal (the highest secondary loading for any item was .25). In the entire sample, Cronbach's alpha was .81 for items loading on the individualism (independence) factor, and was .79 for the items loading on the collectivism (interdependence) factor. For Americans, the reliabilities for the individualism and collectivism items were .84 and .80, respectively, while for the Chinese the reliabilities for the individualism and collectivism items were .85 and .83, respectively.

RESULTS

Preliminary Analyses

Means and standard deviations for the variables in the study are summarized in Table 1. In order to ascertain that Chinese and Euro-American participants embodied the orientations of their respective cultures, respondent levels of individualism and collectivism were assessed. These data were examined with a 2 X 2 X 2 mixed-model ANOVA. The between-groups factors were participant gender and participant culture. The within-groups factor was value orientation (individualism vs. collectivism). The dependent variable was the relative level of identification with each value orientation.

The ANOVA detected a quite large main effect for the factor of value orientation, F (1, 199) = 271.70, p < .001, [[eta].sup.2] = .578. On average, respondents from both cultures identified with an individualist value orientation significantly more (M= 5.79) than with a collectivist value orientation (M = 4.44).

The ANOVA detected a small, but significant main effect for the factor of participant culture, F(1, 199) = 6.25, p < .05, [[eta].sup.2] = 0.031. Overall, Chinese respondents identified with both value orientations significantly more (M = 5.21) than did Euro-Americans (M = 5.01). More important, the ANOVA detected a significant interaction between the factors of culture and value orientation, F(1, 199) = 21.03, p < .001, [[eta].sup.2] = .096. Decomposition of this interaction revealed that, as predicted, Chinese respondents identified significantly more (t [200] = 4.34, p < .001) with a collectivist value orientation (M = 4,74) than did Euro-Americans (M = 4.12). In contrast, Euro-Americans identified significantly more (t [200] = 1.98, p < .05) with an individualist value orientation (M = 5.85) than did Chinese (M = 5.67). Finally, the ANOVA detected a small, but significant interaction between the factors of sex and value orientation, F (1, 199) = 7.78, p < .05, [[eta].sup.2] = 0.038. Decomposition of this interaction revealed that women identified marginally more (t [2001 = 1.62, p < .11) with a collectivist orientation (M = 4.55) than did men (M = 4.36). Men in contrast, identified significantly more (t[200] = 2.46, p < .05) with an individualist orientation (M = 5.90) than did women (M = 5.66). No other effects were statistically significant. These results support the assumption that the Chinese sample of respondents in this study would embody a collectivist value orientation to a greater extent than the Euro-American respondents. Similarly, these results also support the assumption that the Euro-American respondents in this study would embody an individualist value orientation to a greater extent than the Chinese respondents.

Variation in the amount of time Chinese respondents had spent in the U.S. may potentially confound the results reported here. Chinese respondents averaged 19-24 months of U.S. residency, but the range extended from 6 months to 72 months of residency. To determine whether variation in length of residency influenced responses to the variables of interest in the study, a Pearson correlation analysis assessed the association of length of residency with the 4 dependent measures. No associations were statistically significant for the Chinese sample. It appears safe then, to discount differences in length of residence as a potential confound of culture-related differences in this study.

Tests of Hypotheses

Correlations among the variables included in the study were calculated separately for the American and Chinese samples. These correlations are reported in Table 2.

Hypothesis 1 predicted that collectivism would be positively associated with the value placed on supportive communication skills (ego support and comforting). As indicated by the correlations in Table 2, this hypothesis was supported for comforting skill by the data from both Euro-Americans (r = .29, p < .01) and Chinese (r = .21, p < .05). Hypothesis 1 was supported for ego support skill only by the data for Euro-Americans (r = .26, p < .01); collectivism and the value placed on ego support skill were not associated in the Chinese sample.

Hypothesis 2 predicted that women would place a higher value than men on supportive communication skills. As the correlations in Table 2 show, this hypothesis was supported only for Euro-Americans. American women placed a significantly higher value on both ego support skill and comforting skill than did men (also see the means in Table 1). In contrast, the correlational results indicated no significant association between gender and communication values for Chinese.

Hypothesis 3 predicted that collectivism-individualism would mediate the association between sex and the value placed on supportive communication skills. Regression analyses were used in testing this hypothesis; separate regressions were conducted for each of the two cultural groups (Euro-Americans and Chinese) and for each of the two communication variables (value placed on ego support and value placed on comforting). In each set of analyses, two regression equations were calculated. In the first equation, sex was entered at the initial step and individualism and collectivism were then entered as a block at a second step. The second equation reversed the order of entry for the two sets of predictors. Mediation effects were determined by comparing the increments in explained variance for each predictor set in the two equations (see Biddle & Marlin, 1987). The results of these analyses are summarized in Table 3.

For the analysis on ego support among Euro-Americans, the effect for sex was significant at the first step of the analysis, [beta] = . 19, p < .05, [R.sup.2] = .036. When individualism and collectivism were entered at a second step, they uniquely explained a significant 9.8% of the variance in ego support evaluations. Both individualism ([beta] = .21, p < .05) and collectivism ([beta] = .25, p < .05) contributed significantly to the prediction of the criterion. When sex was entered in the regression after collectivism and individualism, it dropped only slightly in predictive value (from explaining 3.6% of the variance to explaining 3.0% of the variance). This result supports (albeit, weakly) the predicted mediating effect of individualism-collectivism.

For the analysis of the value placed on comforting skill by Euro-Americans, the effect for sex was significant at the first step of the analysis, [beta] = .19, p < .05, [R.sup.2] = .037. The inclusion of individualism-collectivism at the second step of the analysis added significantly to the prediction of comforting skill value ([R.sup.2] change = .10, F change = 5.33, p < .006), with only collectivism ([beta] = .28, p < .01) contributing to the prediction of the criterion. When sex was entered second in the regression, it declined noticeably in predictive value (from explaining 3.7% of the variance to explaining 2.7% of the variance). Thus, about 27% of the sex-related variance in evaluations of comforting skills was explained by the cultural values of individualism and collectivism (with collectivism doing virtually all of the mediating work). These results provide some support for the predicted mediating effect of individualism-collectivism.

Among the Chinese, since sex was not significantly related to the value placed on either ego support skill or comforting skill, there was no association for cultural values to mediate (see Table 3). For the Chinese, value placed on ego support skill was not predicted significantly by any regression model, [R.sup.2] = .041, F (3, 101) = 1.45, ns. However, the value placed on comforting skill was predicted at a marginal level of statistical significance, [R.sup.2] = .061, F(3, 101) = 2.18, p < .10. Although sex did not explain significant variance in comforting skill at the first step of the analysis ([beta] = .09, ns), the inclusion of individualism-collectivism accounted for a marginally significant increment in explained variance ([R.sup.2] change = .052, p < .06). Collectivism ([beta] = .20, p < .04), but not individualism ([beta] = .11, ns), contributed significantly to the prediction of the criterion.

DISCUSSION

All three of the study's hypotheses received some support. At the same time, complex relationships among cultural values, national culture, sex, and the value placed on supportive communication emerged from the data. These relationships can be summarized in terms of four specific trends. First, different cultural values were associated with valuing different kinds of supportive communication. Second, the cultural value orientations of individualism and collectivism appeared to be sex linked. Third, cultural values have a small, but reliable, mediating effect on sex differences in the valuing of supportive communication. Fourth, the valuing of supportive communication appears to be sex linked within Euro-American culture but not in Chinese culture. In other words, national culture appears to moderate the mediating effect of cultural value systems on the link between sex and communication values.

With regard to national culture and cultural orientations, Euro-Americans and Chinese conformed to their respective descriptions as individualists and collectivists. At the same time, the results indicated that across both cultures, men were more individualist than women, and women were more collectivist than men. This suggests that dimensions like individualism-collectivism can be understood both in terms of culture-related differences (Americans and Chinese differ with regard to average levels of individualism), as well as similarities (among both Americans and Chinese, women are, on average, more collectivist than men while men, on average, are more individualist than women). Thus, as with some other social and communication variables, there are sex differences that span cultures (see Fischer & Manstead, 2000).

However, consistent with patterns reported by Waldron and Di Mare (1998), the current study found that some sex-related differences were specific to a particular culture. Hypothesis 2 predicted that women would place a higher value than men on supportive communication skills. This hypothesis was supported for Euro-Americans, but not for Chinese. While Euro-American women valued supportive communication significantly more than Euro-American men, there were no sex differences among Chinese participants. Sex differences in the value placed on supportive communication seemed to be driven by Euro-American men who showed the lowest values for supportive communication among all four groups. Perhaps the fact that collectivism is higher among both Chinese men and women than Americans accounts for the uniform value placed by these participants on the value of supportive communication. Given that collectivist values are geared toward social harmony and tightly knit in-group relations, it seems reasonable that supportive communication would be valued equally by both men and women in more collectivist cultures.

Different cultural values were associated with valuing different kinds of supportive communication. Hypothesis 1 predicted that collectivism would be positively associated with the value placed on supportive communication skills (ego support and comforting). This hypothesis found support in relation to comforting but not ego support. In both national cultures, collectivism was positively associated with valuing comforting skill. Within the American sample, valuing ego support skill was positively associated with both individualism and collectivism, but within the Chinese sample, there were no associations between ego support and either collectivism or individualism. In short, ego support appears associated with individualism (for Euro-Americans only) and comforting with collectivism (for both national groups). This may be due to the way that ego support is perceived within Chinese culture by its more collectivist members.

Ego support, especially celebratory communication, is not seen as especially important among collectivists. Among individualists, individual achievements are seen as one of the most joy-producing experiences one can have (Matsumoto, 1991). Collectivists on the other hand, rarely perceive their achievements as individual achievements. One's successes are understood in connection to the support and efforts made by other in-group members. As such, collectivists do not perceive achievement as an experience producing as much happiness as do individualists (Matsumoto, 1991). Collectivists are also encouraged to downplay their prideful emotions in order to avoid inciting jealousy among in-group members (Hall, 1976). It makes sense then, that comforting, which helps to ameliorate negative emotions among in-group members, would be valued more by Chinese than ego support, which might actually cause problems.

Hypothesis 3 predicted that collectivism-individualism would mediate the association between biological sex and the value placed on supportive communication skills. This hypothesis found mixed support across cultures and skill types. Individualism-collectivism could not mediate sex differences in the Chinese sample because there were no significant sex differences to mediate. However, among Euro-Americans, individualism-collectivism was found to have a small mediating effect on sex differences in the valuing of supportive communication, especially the value placed on comforting skill. Although sex accounted for relatively small amounts of variance in the value Americans placed on ego support and comforting (3.6% and 3.7%, respectively), the regression analyses showed that the broader values of individualism and collectivism mediated the effects of sex, slightly for ego support and more substantially for comforting. Aside from their mediating effects, both individualism and collectivism were good predictors of the value accorded ego support by Americans, as well as the value accorded to comforting by both Americans and Chinese (providing additional support for Hypothesis 1).

Taken together, these results suggest that broad-scale value orientations (such as individualism-collectivism) play some role in shaping more specific orientations to particular forms and functions of communication. For example, while Euro-Americans were broadly more individualist and Chinese broadly more collectivist, within each culture men and women were socialized to embody more specific collectivist or individualist orientations based on sex. These orientations, in turn, shaped the ways that people valued specific communication behaviors like comforting and ego support, The results of this study suggest that cultural orientations, sex differences, and communication phenomena are linked in multiple ways that require taking a complex view of potential associations rather than looking for simple, dichotomous differences.

A current debate in communication studies centers around the value of studying sex differences in communicative behavior. Sex differences in communication studies are often statistically small and contradictory, leading many researchers to question the value of examining them at all (Canary & Hause, 1993). The results of this study suggest that sex differences and similarities need to be taken in context with cultural membership and broad-scale value orientations. Many studies assessing sex differences (and so-called "gender" differences, for that matter) attempt to simplify the effects of sex by focusing on differences within a single culture. Future research needs to provide a more sophisticated portrait of male and female communication styles across both cultures and domains of behavior.

TABLE 1
Cell Means and Standard Deviations for Variables
Included in the Study

                               Americans                Chinese

Variables                  Men         Women        Men       Women

Collectivism              4. 00         4.31       4.67        4.79
                          (1.02)       (0.89)     (0.90)      (0.82)
Individualism              5.95         5.80       5.87        5.53
                          (0.74)       (0.61)     (0.59)      (0.76)
Ego Support Skill          5.40         5.77       5.69        5.52
                          (0.98)       (0.91)     (0.92)      (0.96)
Comforting Skill           5.21         5.60       5.37        5.51
                          (1.07)       (0.94)     (0.78)      (0.74)

NOTE: Figures in parentheses are standard deviations.
TABLE 2
Correlations Among Variables Included in the Study for Americans
and Chinese

Variable                   Sex     Collectivism    Individualism

Sex                        --           .07           -.24 **
Collectivism              .16 *          --            .00
Individualism              -.11        -.06             --
Ego-Support Skill         .19 *         .26 **         .17 *
Comforting Skill          .19 *         .29 **         .13

Variable                  Ego-Support Skill     Comforting Skill

Sex                              -.09                 .09
Collectivism                      .11                 .21 *
Individualism                     .16 *               .08
Ego-Support Skill                 --                  .51 ***
Comforting Skill                  .77 ***              --

NOTE: Sex was coded 1 = Male, 2 = Female. Correlations below the
diagonal are for Americans (N = 97); those above the diagonal are for
Chinese (N = 105). * p < .05; ** p < .01; *** p < .001.
TABLE 3
Summary of Regression Analyses for Two Cultural Groups
Assessing the Mediational Effects of Collectivism and
Individualism on the Association Between Sex and the Value
Placed on Ego Support Skill and Comforting Skill

                                            Americans

                                    Sex                    Sex
                                Entered First         Entered Second

 Dependent                   [R.sup.2]      F       [R.sup.2]     F
 Variable    Predictor        Change      Change     Change     Change

Ego          Sex               .036      3.59 (+)     .030     3.18 (+)
Support      Collectivism      .098      5.24 **      .104     5.48 **
Skill        Individualism
Comforting   Sex               .037      3.61 (+)     .027     2.87 (+)
Skill        Collectivism      .099      5.33 **      .109     5.75 **
             Individualism

                                             Chinese

                                     Sex                    Sex
                                Entered First          Entered Second

Dependent                    [R.sup.2]      F       [R.sup.2]     F
 Variable    Predictor        Change      Change     Change     Change

Ego          Sex               .008        0.85       .004      .043
Support      Collectivism      .033        1.75       .037     1.98
Skill        Individualism
Comforting   Sex               .009       0.089       .010     1.09
Skill        Collectivism      .052        2.80 (+)   .051     2.72 (+)
             Individualism

NOTE: For Americans, N = 97; for Chinese, N = 105. (+) p < .10;
* p < .05; ** p < .01; *** p < .001.

REFERENCES

Bakan, D. (1966). The duality of human existence. Boston: Beacon Press.

Biddle, B. J., & Marlin, M. M. (1987). Causality, confirmation, credulity, and structural equation modeling. Child Development, 58, 4-17.

Booth, A. (1972). Sex and social participation. American Sociological Review, 37, 183-192.

Buhrmester, D., Furman, W., Wittenberg, M. T., & Reis, H. T. (1988), Five domains of interpersonal competence in peer relationships. Journal of Personality and Social Psychology, 55, 991-1008.

Burleson, B. R. (1994). Comforting messages: Features, functions, and outcomes. In J. A. Daly & J. M. Wiemann (Eds.), Strategic interpersonal communication (pp. 135-161). Hillsdale, NJ: Erlbaum.

Burleson, B. R., Kunkel, A. W., & Birch, J. D. (1994). Thoughts about talk in romantic relationships: Similarity makes for attraction (and happiness, too). Communication Quarterly, 42, 259-273.

Burleson, B. R., Kunkel, A. W., Samter, W., & Werking, K. J. (1996). Men's and women's evaluations of communication skills in personal relationships: When sex differences make a difference--and when they don't. Journal of Social and Personal Relationships, 13, 201-224.

Burleson, B. R., & Samter, W. (1990). Effects of cognitive complexity on the perceived importance of communication skills in friends. Communication Research, 17, 165-182.

Burleson, B. R., & Samter, W. (1994). A social skills approach to relationship maintenance: How individual differences in communication skills affect the achievement of relationship functions. In D. J. Canary & L. Stafford (Eds.), Communication and relational maintenance (pp. 61-90). San Diego, CA: Academic Press.

Cai, D. A., Wilson, S. R., & Drake, L. E. (2000). Culture in the context of intercultural negotiation: Individualism-collectivism and paths to integrative agreements. Human Communication Research, 26, 591-617.

Canary, D., & Hause, K. (1993). Is there any reason to research sex differences in communication? Communication Quarterly, 41, 129-144.

Fehr, B. (1996). Friendship processes. Thousand Oaks, CA: Sage.

Fischer, A. H., & Manstead, A. S. R. (2000). The relation between gender and emotions in different cultures. In A. H. Fischer (Ed.), Gender and emotion: Social psychological perspectives (pp. 71-94). Cambridge, UK: Cambridge University Press.

Gaines, S. O., Jr. (1997). Culture, ethnicity, and personal relationship processes. New York: Routledge.

Gudykunst, W. B., & Matsumoto, Y. (1996). Cross-cultural variability of communication in personal relationships. In W. B. Gudykunst, S. Ting-Toomey, & T. Nishida (Eds.), Communication in personal relationships across cultures (pp. 19-56). Thousand Oaks, CA: Sage.

Hofstede, G. (1980). Culture's consequences.. International differences in work-related values. Beverly Hills, CA: Sage.

Kim, M. S. (1993). Culture-based conversational constraints in explaining cross-cultural strategies competence. In R. L. Wiseman & J. Koester (Eds.), Intercultural communication competence (pp. 132-150). Newbury Park, CA: Sage.

Kim, M. S., & Sharkey, W. F. (1995). Independent and interdependent construals of self: Explaining cultural patterns of interpersonal communication in multi-cultural organizational settings. Communication Quarterly, 43, 20-38.

Kim, M. S., & Wilson, S. R. (1994). A cross cultural comparison of implicit theories of requesting. Communication Monographs, 61, 210-235.

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. M. Markus (Eds.), Culture, self, and emotion (pp. 89-130). Washington, DC: American Psychological Association.

Matsumoto, D. (1991). Cultural influences on facial expressions of emotion. The Southern Communication Journal, 56, 128-137.

Matsumoto, D. (1996). Unmasking Japan: Myths and realities about the emotions of the Japanese. Palo Alto, CA: Stanford University Press.

Reisman, J. M. (1990). Intimacy in same-sex friendships. Sex Roles, 23, 65-82.

Samter, W. (1992). Communicative characteristics of the lonely person's friendship circle. Communication Research, 19, 212-239.

Samter, W. (1994). Unsupportive relationships: Deficiencies in the support-giving skills of the lonely person's friends. In B. R. Burleson, T. L. Albrecht, & I. G. Sarason (Eds.), The communication of social support: Messages, interactions, relationships, and community (pp. 195-214). Thousand Oaks, CA: Sage.

Samter, W., & Burleson, B. R. (1990). Evaluations of communication skills as predictors of peer acceptance in a group living situation. Communication Studies, 41, 311-326.

Samter, W., Whaley, B. B., Mortenson, S. R., & Burleson, B. R. (1997). Ethnicity and emotional support in same-sex friendship: A comparison of Asian-Americans, African-Americans, and Euro-Americans. Personal Relationships, 4, 413-430.

Singelis, T. M. (1994). The measurement of independent and interdependent self-construals. Personality & Social Psychology Bulletin, 20, 580-590.

Singelis, T. M., & Brown, W. J. (1995). Culture, self, and collectivist communication: Linking culture to individual behavior. Human Communication Research, 21, 354-389.

Singelis, T. M., & Sharkey, W. F. (1995). Culture, self construal, and embarrassability. Journal of Cross-Cultural Psychology, 26, 622-644.

Triandis, H. (1988). Collectivism vs. individualism: A reconceptualization of a basic concept in cross-cultural psychology. In G. K. Verma & C. Bagely (Eds.), Cross-cultural studies of personality, attitudes and cognition (pp. 60-95). London: Macmillan.

Waldron, V. R., & Di Mare, L. (1998). Gender as a culturally determined construct: Communication styles in Japan and the United States. In D. J. Canary & K. Dindia (Eds.), Sex differences and similarities in communication. (pp. 179-200). Mahwah, NJ: Erlbaum.

Weiss, L., & Lowenthal, M. F. (1975). Life course perspectives on friendship. In M. F. Lowenthal, M. Turner, & D. Chirboga (Eds.), Four stages of life (pp. 48-61). San Francisco: Jossey-Bass.

Westmyer, S. A., & Myers, S. A. (1996). Communication skills and social support messages across friendship levels. Communication Research Reports, 13, 191-197.

Winstead, B. A., Derlega, V. J., & Rose, S. (1997). Gender and close relationships. Thousand Oaks, CA: Sage.

Steven T. Mortenson (Ph.D., Purdue University, 1999) is an assistant professor in the Department of Communication Arts, Georgia Southern University Statesboro, GA 30460 (email: mortenso@gasou.edu). This article is based on a doctoral dissertation completed under the direction of Brant R. Burleson at Purdue University. The research reported in this paper was supported, in part, by a grant from the Charles T. Kinley Trust awarded to Brant Burleson. A version of this article was presented at the annual meeting of the International Communication Association, Washington, DC, May 25-28, 2001.