Close Relationships Between Asian American and European American College Students

Abstract The authors examined attitudes and behaviors regarding close relationships between European and Asian Americans, with a particular emphasis on 5 major subgroups of Asian Americans (Chinese, Japanese, Korean, Vietnamese, and Filipino Americans). Participants were 218 Asian American college students and 171 European American college students attending a culturally diverse university. The European Americans did not differentiate among the various subgroups of Asian Americans. Their attitudes regarding close relationships were less positive toward Asian Americans than toward Mexican and African Americans, a finding contrary to the prediction of social exchange theory (H. Tajfel, 1975). In contrast to the European Americans' view of homogeneity among Asian Americans, the 5 major subgroups of Asian Americans expressed a distinctive hierarchy of social preference among themselves. Results are discussed in terms of their implications for future research on interethnic relations involving Asian Americans.


Issues Addressed
Specifically, we addressed two issues in this exploratory study. First, we examined whether the Asian Americans and European Americans would view the five major subgroups of Asian Americans as one group or as several distinctive groups when evaluating them for potential close relationships (best friends, romantic partners, or both). Empirical research is scant with respect to how European Americans view different subgroups of Asian Americans and how Asians of different ethnic groups view one another.
Distinctions among the Asian groups can be made in terms of variables such as the geographical region of their home countries, length of their immigration history, and their present economic status in the United States. For example, Chinese, Japanese, and Koreans are all East Asians, whereas Vietnamese and Filipinos are Southeast Asians. In terms of immigration history, the Chinese and Japanese have had a longer presence in the United States than have the Koreans, Vietnamese, and Filipinos. There has also been a large variation in their socioeconomic status. For example, more Japanese American and Chinese American men (40%) have held managerial or professional occupations than have Filipino American (22%) and Vietnamese American men (1 8%), with Korean American men (31%) in the middle (Mar & Kim, 1994). Japanese Americans and Filipino Americans had higher household incomes (Mdn = $43,000) than Koreans and Vietnamese (Mdn income = $3 1,OOO); Chinese Americans' household incomes (Mdn = $38,000) were somewhere in between (Ong & Hee, 1994). Therefore, we expected that both Asian Americans and European Americans would differentiate among Asian Americans in the context of interethnic relationships, most likely favoring the Japanese because of their longer presence in the United States and their high economic status. This hypothesis is also consistent with the literature on social distance, in which Japanese Americans were ranked ahead of the other Asian groups (Owen, Eisner, & McFaul, 1981). We also expected that Asian Americans would reveal a greater differentiation among themselves than would European Americans because of the former's more intimate knowledge about the various subgroups of Asian Americans.
Second, we examined the relative social standing of Asian Americans vis-8-vis other ethnic groups. To this end, we explored European Americans' attitudes toward and friendships with Asian, African, and Mexican Americans. Although Asian Americans have traditionally occupied a lower status along with other minority groups such as African Americans (Smith & Dempsey, 1983), the recent economic and scholastic achievements of Asian Americans have singled them out as a "model minority" and have placed them in an intermediate tier between the top tier of White Protestants and the bottom tier of African Americans, Hispanic Americans, and Native Americans (Kitano, 1997). It is not known, however, whether such particular realms of accomplishment have changed Asian Americans' status in the social hierarchy, espe-cially in terms of desired partners for close social relationships. According to social exchange theory (Tajfel, 1975), Asian Americans would become a favored ethnic minority because of their high levels of education and economic achievement-one of the desirable attributes in a social exchange. Liu et al. (1995), however, found that Asian Americans, although rated by other ethnic groups as having high social status, were not necessarily preferred as targets of romantic relationships. In the present study, we examined both the attitudes and the behaviors of European and Asian Americans in regard to their close relationships with Asian Americans versus non-Asian groups. On the basis of the findings of Liu et al. on attitudes, we expected that Asian Americans as a whole would not be a favored group either as best friends or dating partners, both attitudinally and behaviorally.

Participants
Participants were 389 college students (2 18 Asian Americans and 17 1 European Americans). We recruited them from a culturally diverse U.S. university campus, where 37% of the student population were European Americans; 43%, Asian Americans; 1 1 %, Mexican Americans; 3%, African Americans; and 6%, others. We recruited most of the participants from social science classes. They participated in the study for extra course credit. The mean age of the Asian American participants was 19.6 years (SD = 1.3) and that of the European Americans was 20.4 years (SD = 1.6). Among the Asian American participants, 65% were women; among the European Americans, 74%. These proportions reflect the gender distribution of students in the academic departments from which we drew the research participants, where female students constituted between 56% and 72% of the total.
The Asian American sample included 73 Chinese, 55 Korean, 21 Japanese, 23 Vietnamese, 19 Filipino, and 27 other Asians (including those of mixed Asian descent). In terms of generational status, 62% of the Asian American students and 90% of their parents had been born outside the United States. On average, the Asian American adolescents born overseas had lived in the United States for 12.8 years. Because of the nature of immigration patterns, there were variations among the five groups of Asian Americans in terms of their generational status: The Japanese Americans were most likely (91%) to have been born in the United States, and the Vietnamese Americans, the least likely (9%). The other groups fell in between (24%-56%).
As reported by the students, the European American parents in the present sample had a higher level of education than did the Asian American parents. The European American fathers, on average, had a college education (M = 4.9, SD = 1.1; 5 = college graduate), whereas the average level of education for Asian American fathers was between some college (4) and a college degree (M = 4.5, SD = 1.4), F( 1, 383) = 1 1.72, p < .01. Similarly, the European American mothers had a higher level of education (M = 4.5, SD = 1.1) than did the Asian American mothers (M = 3.9, SD = 1.4), F(1, 386) = 22.99, p < .001. Five of the six groups of Asian Americans were similar in their fathers' education level (Ms = 4.4-4.9, SDs = 0.9-1.3), but the Chinese Americans' fathers had somewhat less education attainment (M = 3.9, SD = 1.8) than did the other fathers, F(5, 209) = 4.39, p < .001. The education levels of both the Chinese and Vietnamese mothers were lower (both Ms = 3.3, SDs = 1.5 and 1.6, respectively) than those of the other groups of mothers (Ms = 4.1-4.9, SDs = 0.7-1.2), F(5, 212) = 7.43, p c .001.

Measures
As part of a larger project on the lives of late adolescents, we asked the students the following sets of questions regarding interethnic attitudes and behaviors: 1. "How easy is it for you to imagine that you might have a best friend who was (a) Caucasian (White), (b) African American (Black), (c) Mexican American, (d) Chinese American, (e) Japanese American, (f) Korean American, (g) Vietnamese American, and (h) Filipino American?" In rating each ethnic group, students were given four choices: 1 = hard to imagine, very unlikely; 2 = easy to imagine, but not very likely; 3 = easy to imagine, quite likely; and 4 = I don't know. We excluded respondents who answered "I don't know" (6%) from analyses that used these variables as an ordinal scale.
2. "How would you feel about going out on a date with a person who was [the same eight ethnic categories as in the preceding paragraph]?" Students were given four choices: 1 = would not date; 2 = would consider; but not sure would date; 3 = would date; and 4 = I don't know. Again, we excluded the respondents who answered "I don't know" (7%) from analyses that used these variables as an ordinal scale.
3. We asked the respondents to provide information about the ethnic composition of their friendship networks. They accomplished that task by writing down their friends' initials (up to 10 friends) and indicating, among other demographic attributes, the ethnic background of each friend. We also asked the respondents to indicate from their lists their best friends and current romantic partners (if any).
We also asked the students about their family demographics. They provided information on parental education, family structure, and years of residence in the United States.

One or Many Groups of Asian Americans?
European Americans' perspective. Table 1 contains the results of a factor analysis (principal component analysis with varimax rotation) of students' attitudes toward befriending and dating the five groups of Asian Americans. Clearly, the European Americans did not differentiate among the five groups of Asian Americans in either their attitudes toward befriending (Factor 1 ) or their attitudes toward dating Asian Americans (Factor 2). Furthermore, repeated measures analysis of variance (ANOVA) revealed no significant difference among the mean ratings of European Americans' attitudes toward the five subgroups of Asian Americans (Ms = 2.19-2.27 and 2.44-2.53, SDs = .92-.95 and .77-.84 for dating and friendship, respectively). Nor were there gender differences in the European American students' attitudes toward befriending and dating the five groups of Asian Americans, with one exception: The European American men were more positive than the women toward dating Japanese Americans, F( 1,149) = 3 . 9 7 ,~ < .05.
In terms of behavioral aspects of the European Americans' relations with the five subgroups of Asian Americans, there was again little differentiation. Each of the Asian American groups accounted for between 1% and 3% (a total of 9%) of the European Americans' friends, whereas the five groups of Asian Americans accounted for between 4% and 12% (a total of 35%) of the student population. Of the 64% (109) of European Americans who were dating, only 5 were dating a person from one of the five subgroups of Asian Americans (1 Vietnamese, 1 Korean, and 3 Filipino Americans). Given the low proportion of each of the five Asian American groups in the European Americans' close relationship networks, there is no basis for suggesting that European Americans differentiated among these groups.
Asian Americans ' perspective. Factor analyses revealed a three-factor structure for the Asian Americans. The first two factors were similar to those for the European Americans: one for attitudes toward friendships (Factor 1) and the other for dating relationships (Factor 2). In addition, a third factor (Factor 3), which accounted for an additional 12% of the variance, appeared to distinguish Vietnamese and Filipino Americans from the other three groups. The most obvious interpretation of this factor is that the Asian Americans tended to distinguish between those of East Asian ancestry and those of Southeast Asian ancestry. Perhaps the geographic distance among these groups' countries of origin was directly associated with cultural distance among them.
Another indicator of the large variations in Asian Americans' attitudes was that the three factors noted in the preceding paragraph accounted for a lower percentage of variance (75%; 63% for the first two factors) than did the two factors of attitudes for the European Americans (92% for the first two factors).
A further examination of the attitudes of different groups of Asian Americans toward one another revealed two reasons for the heterogeneity in their attitudes. First, there was a clear in-group preference (i.e., within ethnic subgroup; Figure 1). The difference in mean ratings for attitudes toward befriending one's own group versus other Asian Americans ranged from .I3 to .74 (M = .37, SD = .14). Effect sizes (ds) for those values ranged from .25 to 1.30 (M = .65, SD = .25). The results for dating were similar (averaged d = .55), with the exception that the Vietnamese and Filipino respondents were about equally positive toward dating Japanese as toward dating within their own ethnic groups.
In-group preference was more clearly demonstrated by a correspondence analysis (Weller & Romney, 1990), which allowed u s to reduce the 5 (respondent group) x 5 (target group) data matrix to its basic structure through a process called singular value decomposition. The results of the analysis are a spatial representation of the data, in which the distance among data points can be meaningfully interpreted. The analyses revealed the patterning of respondent groups and target groups (Figure 2). Clearly, each respondent group was located near itself as the target group, indicating consistent in-group preference. Also clear from the graphs is that, overall, interethnic attitudes were most dissimilar (with greatest distance) among the Korean, Vietnamese, and Filipino Americans. This partially reflects that fact that these groups tended to rate each other as the least favorable in close relationships (cf. Figure 1).

1.5-
In addition to in-group preference, each Asian subgroup also seemed to show a gradient of preference among the four other Asian American subgroups (Figure 1). In general, the most favored group was Japanese or Chinese; the least favored, Vietnamese.
There were few gender differences in the Asian American subgroups' attitudes toward one another. None of the tests were statistically significant for attitudes about befriending one another. In terms of attitudes toward dating each The Vietnamese men were more positive than the Vietnamese women toward dating Filipinos and Koreans, and the Korean men were more positive than the Korean women toward dating Filipinos and Vietnamese.
The results of analyses of the ethnic composition of students' friendship networks and dating partners further supported the aforementioned conclusions about the distinctions among Asian American groups. A majority of the Asian Americans' friends (70%) and best friends (72%) were Asian Americans. Of those close relationships with Asian Americans, between 57% and 77% were from the same specific ethnic group, a clear indication of in-group preference. The one exception was Japanese Americans, for whom only a third or fewer of their friends were also Japanese Americans. Of their out-group Asian friends, Chinese Americans accounted for the largest share (12%-29%), with the remainder spread across other groups. The number of cases in interethnic dating was too small for analyses at the subgroup level. At the panethnic level, 65% of the Asian Americans who were currently involved in romantic relationships reported that their partners were Asian Americans.

Close Relationships With Asian Americans Versus Non-Asians
European Americans'perspective. Figure 3 shows European Americans' attitudes toward African Americans, Mexican Americans, and the five groups of Asian Americans. To examine whether the European Americans' attitudes toward Asian Americans were significantly different from their attitudes toward the other two groups, we conducted repeated measures ANOVAs. Because the European Americans did not differentiate in their attitudes toward the five major groups of Asian Americans (as reported earlier), we combined those five variables for ease of comparison to form a single score for attitude toward Asian Americans. According to the results of the repeated measures ANOVAs, the European Americans' attitudes toward the three minority groups differed significantly: for friendships, F(2, 137) = 12.25, p < .001, and for dating, F(2, 136) = 17.01, p c 301. The results of additional t tests showed that the European Americans were less positive about dating Asian Americans than they were about dating either African Americans, t(139) = -2.69, p < .01, or Mexican Americans, r(144) = -5.83, p < .001. In terms of best friendships, the European Americans' attitudes toward Asian Americans were less positive than were their attitudes toward Mexican Americans, t(144) = 4 0 7 , p < .001, but did not differ from their attitudes toward African Americans.
The results from analyses of students' friendship networks were generally consistent with the foregoing findings. On the basis of demographics for the university setting where we collected the data, among the five major subgroups, the chance of an Asian American's becoming a best friend to a European American was .23 (8% divided by 35%; Table 2). The corresponding values for African Americans' and Mexican Americans' becoming best friends to a European American were .33 and .73, respectively. In terms of becoming romantically involved with European Americans, the probabilities were .14 (Asian Americans), .67 (African Americans), and .55 (Mexican Americans). Those results paralleled our findings on attitudes toward close relationships: The European Americans preferred Mexican Americans and African Americans to Asian Americans in terms of dating and preferred Mexican Americans to the other two groups in terms of best friendship. Asian Americans ' perspective. Before examining the Asian Americans' attitudes and behaviors toward close relationships among themselves as compared with

12%
"or the purpose of comparisons, this column includes only the five major subgroups of Asian Americans. The other Asian Americans are included in the "Others" category.  We also found significant variations among the five subgroups of Asian Americans in their attitudes toward all three non-Asian groups (European, African, and Mexican Americans) in both types of relationships (dating and friendship), Fs(5, 190-203) = 2.82-6.63, ps c .05. Post hoc Scheffi contrasts showed that the Korean Americans were generally less positive than the Vietnamese Americans and Filipino Americans toward non-Asians, whereas the Chinese Americans' and Japanese Americans' attitudes were somewhere in between (Figure 4). There were no significant gender differences in Asian Americans' attitudes toward the three out-groups in terms of either best friends or dating.
By comparing the data in Figure 4 with those in Figure 1, we found that the Asian American subgroups generally rated Asian American out-groups higher than the lowest rated non-Asians (i.e., African Americans). The mean ratings for the most highly rated non-Asians (i.e., European Americans) were usually located between the lowest ratings for an Asian out-group (typically Vietnamese) and the highest ratings for Asian groups (i.e., own group and Japanese). In sum, there was not a clear Asian versus non-Asian distinction.
In terms of friendship networks, European Americans accounted for 15% to 24% of the Asian Americans' friends, best friends, and romantic partners, whereas African Americans and Mexican Americans each accounted for 5% or less, ts(214) = 4.92-9.10, ps c .001 (Table 2). Those results are consistent with Asian Americans' attitudes as reported earlier.

Discussion
First, in this empirical study, we assessed the similarities and differences among five major subgroups of Asian Americans in terms of their interethnic attitudes and behaviors in the domain of close relationships. Second, it may also be the first study of how European Americans view Asian Americans of different ethnic origins. Several of the present findings have important implications for an understanding of interethnic relations and for the design of future research. Our first major finding was that, among the five Asian American groups, there were significant variations in their attitudes toward one another as well as in their attitudes toward non-Asians. In addition to the obvious in-group preference (at the specific ethnic level, rather than the panethnic group of Asian Americans), we found a hierarchy of preference among Asian Americans that favored the Japanese and Chinese over the other three groups, but especially over the Vietnamese. This type of hierarchy may have been related to the former groups' longer history of presence in the United States and their higher socioeconomic status. Future researchers should explore these and other variables that might explain the aforementioned hierarchy.
The five major subgroups of Asian Americans also differed in their attitudes toward non-Asians. The Koreans were the least positive about close relationships with non-Asians. This finding is consistent with archival data on out-marriages.
For example, Kitano (1997), using data from Los Angeles County, reported that Korean Americans had the lowest rate of out-marriages (1 1%) compared with the other four groups included in our study (27%-51%).
Given such differences in interethnic attitudes and behaviors among Asian Americans, the common practice of grouping all Asian Americans into one group for research involving interethnic relations may not be warranted. Because each subgroup of Asian Americans defined the in-group in terms of a specific ethnic group, researchers using photographs of "Asians" or the general verbal label of "Asians," or an "Asian" confederate may have confounded the results if the Asian research participants viewed those photos or confederates as being out-group Asians.
On the other hand, the present European American students perceived all five subgroups of Asian Americans similarly, a finding that, at first glance, appears consistent with the out-group homogeneity effect (Tajfel, 1978). However, the European Americans viewed the five subgroups of Asian Americans as significantly different from the other two out-groups-the Mexican Americans and African Americans. Perhaps the homogeneity effect about Asian Americans was attributable to the fact that three of the five subgroups (i.e., Koreans, Vietnamese, and Filipinos) were more or less recent arrivals. Therefore, for studies of European Americans' attitudes toward Asian Americans, "Asian Americans" seems to be an appropriate category or group for now, although its fitness may change with the passage of time and increased familiarity with members of different Asian subgroups.
A second major finding is that the common recognition of Asian Americans as a "model" minority seemed to have no effect on their social attractiveness with respect to interethnic relations among the present college students. The European Americans' attitudes toward Asian Americans (all five major subgroups) were less positive than were their attitudes toward Mexican Americans and African Americans. Liu et al. (1995) showed that Latino Americans and African Americans also rated Asian Americans low in their dating preferences. Furthermore, in terms of actual behaviors, the present European American students dated far fewer Asian Americans than the proportion expected on the basis of the student population. In sum, the Asian Americans' status as a "model" for other minority groups was, perhaps, limited to particular areas (e.g., school, work) and did not appear to extend into areas important to intimate social relationships (best friends and dating partners). That finding represents a major challenge to the social exchange theory (Tajfel, 1975), according to which the high socioeconomic status of Asian Americans vis-A-vis other ethnic minority groups would lead them to enjoy a preferred status as dating partners. Future researchers should explore why social exchange theory seems unable to account for interethnic relations involving Asian Americans. Perhaps there are specifk conditions under which social exchange serves as a guiding principle in interethnic relations.
Data at the behavioral level supported conclusions based on the attitudinal measures. Our third major finding was that actual friendships and dating relationships showed clear in-group preference at the specific ethnic level, a hierarchy of preference among Asian Americans toward various Asian American ethnic groups, and a disproportionally low representation of Asian Americans in European Americans' friendship and dating networks. Finally, three limitations of the study should be noted so that its results are not overgeneralized. First, we based the study exclusively on students' self-reports of attitudes and behaviors. The common problems of social desirability in attitudinal measures and inaccuracy in self-report of behaviors may have introduced biases into the study. Future researchers should include a wider variety of attitudinal measures and obtain behavioral data from other sources (e.g., students' friends or roommates). Second, college students represent a selected portion of the population. One should not generalize the results of our study to other populations (e.g., older adults, young children, non-college-educated adults) that may have completely different interethnic experiences. Third, interethnic relations are forever evolving. Even though we tried to update the literature on interethnic relations by studying the often ignored diversity within Asian American groups, our update may be quickly outdated if not followed up by relevant research.