The Effects on Ethnic Identification of

Same- and Mixed- Ethnicity Marriage

A dissertation submitted in partial satisfaction of the requirements for the degree Doctor of Philosophy in Psychology

by

Karen Gei-men Cheng

University of California, Los Angeles

2002

Abstract

Does mixed-ethnicity marriage cause a decrease in one's level of ethnic identification? Using shared reality theory (Hardin & Conley, 2000; Hardin & Higgins, 1996), which posits that interpersonal relationships are built upon commonly held attitudes and beliefs, it was predicted that positive relationships with same-ethnicity others will elicit high ethnic identification but that positive relationships with different-ethnicity others will elicit lower ethnic identification. In this study, 112 Asian American women (66 from same-ethnicity marriages, 46 from mixed-ethnicity marriages) were instructed to think of the positive aspects of either their relationship with their spouse or their relationship with their mother. Women in mixed-ethnicity marriages thinking about their different-ethnicity spouse reported lower ethnic identification on the Multigroup Ethnic Identity Measure (Phinney, 1992), compared to women in mixed-ethnicity marriages thinking about their mother and to women in same-ethnicity marriages thinking about their mother or their spouse. Greater shared understanding of ethnicity with mother accounted for the higher levels of ethnic identification reported by women in mixed-ethnicity marriages thinking about their mother compared to women in mixed-ethnicity marriages thinking about their spouse. Greater shared understanding of ethnicity with spouse accounted for the higher levels of ethnic identification reported by women in same-ethnicity marriages compared to women in mixed-ethnicity marriages.

The Effect on Ethnic Identification of Same- and Mixed- Ethnicity Marriage

Have minorities who married outside of their ethnicity lost their connection to their ethnic background because of their spouse? “Love conquers all” - the old adage says, but does love conquer so much that it destroys the unique ethnic backgrounds and identities that each spouse brings to an interethnic marriage?

Whether and how ethnic identification is affected by mixed-ethnicity marriage is not only an interesting topic for scientific inquiry, it is practically relevant to the lives of many Americans. The 1990 Census showed that 25% of Asian American marriages and 19% of Latino marriages were mixed-ethnicity (Lee & Fernandez, 1998). Of American Indians surveyed in the same 1990 Census, 60% of those who were married were married to non-American Indian spouses. Though only 6% of African American marriages in 1990 were mixed-ethnicity, this stands as a marked increase from 3% in the 1980 Census. As the percentage of Americans who claim an ethnic minority background increases, from 16% in 1990 to 25% in 2000 (U.S. Census Bureau), it is likely that the numbers of mixed-ethnicity marriages too will increase.

However, the psychological and sociological literature reveals very little about whether and how one's ethnic identification is impacted by marriage to a different-ethnicity spouse. Though many published studies are concerned with the ethnic identity of children of mixed-ethnicity marriages (e.g., LaFromboise, Gerton, & Coleman, 1993), no published studies can appropriate address the impact of mixed-ethnicity marriage on the ethnic identification of the spouses themselves. Though a small number of studies show that people in mixed-ethnicity marriages report ethnicity to be an unimportant part of their relationship, these studies do not include comparison groups of same-ethnicity married couples (Imamura, 1986; Kouri & Laswell, 1993; Lewis, Yancey, & Bletzer, 1997; Porterfield, 1982).

Only two published studies compare the ethnic identification of college students in same-ethnicity and mixed-ethnicity dating relationships. These studies offer mixed results. Gurung and Duoung (1999) found that people in mixed-ethnicity dating relationships did not differ in level of ethnic identification from people in same-ethnicity dating relationships. But, Shibazaki and Brennan (1998) found that people in mixed-ethnicity dating relationships reported lower ethnic identification than people in same-ethnicity dating relationships. More work must be done to clarify these findings. Neither of these studies was designed to directly test whether mixed-ethnicity dating relationships affect ethnic identification; rather, the intended purpose of the studies was to compare the satisfaction of people in same- and mixed- ethnicity relationships. In addition, both studies included large proportions of Whites, making it difficult to apply their results to ethnic minorities. Roughly a third of Gurung and Duoung's sample were White, while over half of Shibazaki and Brennan's sample were White. While Whites may agree to answer questions on ethnic identification, American White ethnic identity may be very different from the ethnic identity of ethnic minorities (Helms, 1995). White ethnic identity is thought to be the degree to which one is aware that race is an important part of American daily affairs (Helms, 1995), while the ethnic identity of minorities is the degree to which one feels part of and connected to one's ethnic group (Atkinson, Morten, & Sue, 1989; Cross, 1978; Phinney, 1990; Sue & Sue, 1990). Because the meaning of ethnic identification differs between Whites and minorities, the data for ethnic minorities ought to be analyzed separately from the data for Whites.

Even if the ethnic identification findings from these studies are reliable for dating relationships, they may not generalize to marriage relationships. Dating relationships among college students differ qualitatively from marriages. They are generally shorter and involve less social and economic interdependence (i.e., living together, owning communal property, sharing parental responsibilities). They also tend to exist outside of the intervention of parents and other family members. Further evidence that mixed-ethnicity dating relationships differ from mixed-ethnicity marriages is a comparison of their proportions: while as many as 70% of American college students have dated someone of a different ethnicity (Tucker & Mitchell-Kernan, 1995), only 3% of American marriages are between people of different ethnicities (1998 Current Population Survey, U.S. Census Bureau).

Relevant Data and Psychological Theories

Though published studies do not address the impact of mixed-ethnicity marriage on ethnic identification, models of ethnic identity development, related data, and social psychological theories suggest that ethnic identification should indeed be affected by mixed-ethnicity marriage. Models of ethnic identity development stress changes over the lifespan and allude to interpersonal influences on the developmental process. Data show that ethnic identification shifts in response to situational influences and that one's level of ethnic identification is correlated with attributes of the interpersonal relationships one has. Several social psychological theories also suggest that ethnic identification is subject to external influences.

Models of ethnic identity development. Major theories of ethnic identification (e.g., Atkinson, Morten, & Sue, 1989; Cross, 1978; Helms, 1990; Phinney, 1989; Sue & Sue, 1990) depict the fluid nature of ethnic identification through developmental processes in which a minority person's understanding of her ethnicity changes and grows. The series of progressive stages describe the different ways in which a person may be identified with her ethnic background at different stages of life. For most of these theories, the stages are sequential and progress from less identified to more highly identified, though they often leave open the possibility that a person may recycle through earlier stages as she works through the development of her ethnic identification. Progression to the next stage or falling back into a previous stage is frequently brought on by interpersonal experiences.

For example, several aspects of Cross's psychological Nigresence model, originally created for African Americans, demonstrate the fluid response of ethnic identification to interpersonal relationships (Cross, 1978; 1995). First, interpersonal experiences prompt one to begin the developmental process of greater ethnic understanding. Encounters with racist people jostle an African American out of the Pre-Encounter stage, in which one either accepts the White standard or one is simply unaware of racial issues. Second, interpersonal relationships allow one to explore one's racial background. In the stage of Immersion, one surrounds oneself with African American people and culture in order to learn more about African American heritage. Third, interpersonal experiences elicit different facets of the achieved African American identity. Achieved African American identity allows an African American to bond with other African Americans, to buffer against racist interpersonal experiences, and to bridge with non-Blacks (Cross & Strauss, 1998).

Related data. Empirical data also show that ethnic identification is responsive to situational influences. Correlational studies find that ethnic identification is stronger when ethnic minorities attend school and live in neighborhoods where they are numerically few (Allen, Howard, & Grimes, 1997; Demo & Hughes, 1990; Jaret & Reitzes, 1999; McGuire, McGuire, Child, & Fujioka, 1978). Yip and Fuligni (in press) used a daily diary method and found fluctuations in Chinese-American adolescents' reports of “feeling Chinese” across 14 consecutive days. Experimental studies find that the degree of one's ethnic identification increases when one is induced to think of a conflict between one's own ethnic group and another ethnic group (Christian, Gadfield, Giles, & Taylor, 1976; Shelton & Sellers, 2000). That ethnic identification is affected by situational changes is consistent with extensive research showing that different aspects of the self-concept are made salient in different situations (e.g., McGuire, McGuire, & Cheever, 1986; Nurius & Markus, 1990; Oyserman & Markus, 1998).

Research also finds that ethnic identification appears to be tied to important relationships in one's life. In the Yip and Fuligni (in press) study, the direction of fluctuation was significantly correlated with the quality of their relationship with their parents, such that on days when adolescents felt like a good son or daughter, they reported feeling “more Chinese.” Similarly, Rosenthal and Hrynevich (1985) found that when Greek-Australian and Italian-Australian adolescents were asked in an interview which situations made them feel more ethnic, “being in the presence of family members” was the answer most frequently ranked first. The level of ethnic identification reported by children and adolescents is positively correlated with their parents' level of ethnic identification (Quintana and Vera, 1999; Knight, Bernal, Garza, Cota, & Ocampo, 1993; Rosenthal & Cichello, 1986; Sam, 1995; Thomas, 2000). It is also positively correlated with the degree to which the parents construct a home environment with traditional cultural values (Quintana and Vera, 1999; Quintana, Castaneda-English, & Ybarra, 1999; Sam, 2000). Across childhood, adolescence, and adulthood, higher ethnic identification is correlated with having a greater number of friends who share the same ethnicity (Demo & Hughes, 1990; Rosenthal & Feldman, 1992; Sanders Thompson, 1999; Saylor and Aries, 1998; Smith, Walker, Fields, Brookins, & Seay, 1999).

These data imply that ethnic identification can and should be influenced by a different-ethnicity marriage partner. If important relationships, such as parents and friends, influence one's ethnic identification, then it is likely a spouse will too. Indeed, extensive research has found that romantic partners have significant influence on people's views of themselves (e.g., Aron, Aron, Tudor, & Nelson, 1991; Murray, Holmes, & Griffin, 1996). For example, Aron, Aron, Tudor, and Nelson (1991) found that people were prone to confuse characteristics of their spouse with their own characteristics, suggesting that their self-representations were modified to include their spouse. And Murray, Holmes, and Griffin (1991) found that people's self-views shifted over time, to become more similar to their dating partner's positive views of them (also see Bosson & Swann, 2001).

Social psychological theories. Though not often applied to romantic relationships, several major social psychological theories suggest that a mixed-ethnicity marriage should influence one's ethnic identification. For example, self-categorization theory (Turner, 1987) states that in a given situation, the category of self-representation most appropriate to that situation will become salient and will guide the behavior of the individual. In the context of a mixed-ethnicity marriage, the relevant categorization may be ethnicity and in that way, ethnic identification may be influenced. However, it is difficult to ascertain from the theory when ethnicity will be the relevant category and in what direction ethnic identification will be influenced. Another major theory is balance theory (Heider, 1958), which postulates that people desire balance and harmony in their interpersonal relationships. In order for a relationship with a spouse to be in harmony, one should like or dislike an object to the same degree that one's spouse does. If one's spouse dislikes the ethnic group to which one belongs, it will be very difficult for the latter to maintain high ethnic identification. Social identity theory (Tajfel & Turner, 1986) implies that for someone from a low status ethnic group, marrying a spouse from a higher status ethnic group may lead one to identify with the higher status ethnic group instead of one's own ethnic group in order to gain self-esteem. However, not all of the previously mentioned theories are commonly applied to dyadic interpersonal relationships.

Shared Reality Theory

One theory that works at the interpersonal level of analysis and that makes clear predictions about how mixed-ethnicity marriage may impact ethnic identification is shared reality theory (Hardin & Conley, 2000; Hardin & Higgins, 1996). Shared reality theory posits that people's beliefs about themselves are held strongly to the extent that those beliefs are validated and shared with important others and to the extent that those relationships are strong. Relationships, in turn, are established and maintained on the basis of shared experience, that is, both people in the relationship have the same perspective on something or there is a mutual perception that they experience something the same way. Beliefs not shared with important others are held less strongly. A key element of the theory is that the desire to maintain a relationship activates the beliefs that are shared in that relationship.

Ethnic identification and mixed-ethnicity marriage fit well in the shared reality theory framework. From the perspective of shared reality theory, ethnic identification is a self-belief that is maintained to the extent that it is validated and shared with others, including one's spouse, parents, and friends. Hence, to the extent a different-ethnicity spouse is unable to fully understand feelings and events related to one's ethnicity in the same way that one understands them, and to the extent that one is close to one's spouse, it should be difficult for one to maintain high ethnic identification. By the same token, the theory predicts that one way to maintain the relationship with one's spouse is by de-emphasizing those self-beliefs that are not shared. Thus, people in mixed-ethnicity marriages, who are invested in and committed to their spouses, are likely to de-emphasize their ethnic identification to the degree that ethnic identification represents experiences that cannot be shared with their spouses.

However, people's social networks are complex and are often composed of relationships with both same- and different- ethnicity others. People in mixed-ethnicity marriages must also attend to their relationships with same-ethnicity others. For example, one's mother, who is generally from the same ethnic background, is likely to understand feelings and events related to one's ethnicity in the same way that one understands them. The motivation to maintain a strong, positive relationship with one's mother may encourage one to emphasize the common ethnicity they share by expressing high ethnic identification.

Do these two competing influences on one's ethnic identification result in high or low ethnic identification? According to shared reality theory, it depends on which relationship one is motivated to maintain at that moment. For example, when a Chinese American man is with his Chinese American mother, one way in which he may affirm his relationship with her is to see himself in terms of their common ethnic background. He may express high ethnic identification by speaking Mandarin and participating in traditional ways of preparing food. But when he is alone at home with his White wife, his high ethnic identification may be a challenge to the shared reality on which their relationship is based. To affirm his relationship with her, he may de-emphasize the ethnic background they do not share, expressing lower ethnic identification by not speaking Mandarin and by cooking in the way that is consistent with her cultural background. He may instead affirm the relationship by seeing himself in terms of an identity that they do share, for example, their common religion. According to shared reality theory, both situations reflect honest and authentic parts of this man, but his mother and spouse elicit different aspects of him.

Using shared reality theory, Cheng and Hardin (2002) have proposed a framework for understanding the existing research on how ethnic identification is related to interpersonal relationships. According to this framework, one's level of ethnic identification varies as a function of the ethnicity of the other person with whom one is interacting and the quality of the relationship with that person. Ethnic identification fluctuates in response to people's need to connect with or distance themselves from particular same- and different-ethnicity others. Put another way, ethnic identification is regulated by the need to manage interpersonal relationships. These relationships include relationships that one chooses (e.g., friends, romantic partner), as well as the many that one is obligated to maintain (e.g., parents, bosses, neighbors). Specifically, the model suggests that people assimilate their ethnic identification toward those with whom they have positive relationships and that people contrast their ethnic identification from those with whom they have negative relationships; whether this results in increased or decreased ethnic identification depends on the ethnicity of the other person (see Table 1).

Previous research supports the model's predictions. Consistent with the claim that positive relationships with same-ethnicity people increase ethnic identification, it has been found that positive relationships with same-ethnicity family and friends is associated with higher levels of ethnic identification (Carter, Sbrocco, Lewis, & Friedman, 2001; Demo & Hughes, 1990; Jasinskaja-Lahti & Liebkind, 1998). And consistent with the claim that negative relationships with different-ethnicity people increase ethnic identification, it has been found that negative feelings toward different-ethnicity people with whom one regularly interacts is associated with higher levels of ethnic identification (Branscombe, Schmitt, & Harvey, 1999; Duckitt & Mphuthing, 1998). Experimental studies have further shown that reminding people of the conflict between their ethnic group and another ethnic group results in increased ethnic identification (Christian, Gadfield, Giles, & Taylor, 1976; Shelton & Sellers, 2000). What has not been tested empirically, though, are the claims that positive relationships with different-ethnicity people decrease ethnic identification and that negative relationships with same-ethnicity people decrease ethnic identification. There have also been no experimental tests of the claim that positive relationships with same-ethnicity people increase ethnic identification. And finally, there have no experiments examining ethnic identification in the context of a marriage relationship.

The Present Study

The purpose of this dissertation was to experimentally investigate the impact of same- and mixed-ethnicity marriage on ethnic identification. Though researchers have not reached a consensus about the exact definition of ethnic identification, several dimensions recur in the literature: seeing one's ethnic group as a part of oneself, having positive attitudes toward one's ethnic group, having knowledge about the history and culture of one's ethnicity, and participating in its traditions. The focus here is on ethnic identification as seeing one's ethnic group as a part of oneself.

This study tested the part of Cheng and Hardin's (2002) model that concerns positive relationships. Previous research has not shown experimentally that positive relationships with same-ethnicity others increase ethnic identification or that positive relationships with different-ethnicity others decrease ethnic identification. This study aimed to experimentally test the effects of positive relationships on ethnic identification by comparing women in same-ethnicity marriages with women in mixed-ethnicity marriages. It was predicted that when motivated to maintain their relationship with their spouse, they will be more highly identified with their own ethnicity if he is of the same ethnicity but will be less identified with their own ethnicity if he is of a different ethnicity. When motivated to maintain a relationship with a close same-ethnicity person who is not their spouse, however, it was predicted that they will be highly identified with their ethnicity, regardless of whether their marriage is same- or mixed- ethnicity. Because the model suggests that only ethnic identification should change when thinking about a same- or different- ethnicity person, it was also predicted that self esteem will not be affected by the experimental manipulation or by the type of marriage.

These predictions were tested in an experimental study. People in same- and mixed- ethnicity marriages were induced to think either of their positive relationship with their mother or their positive relationship with their spouse. An ethnic identification measure and a self esteem measure followed. A religious identification measure was also given, to provide converging evidence that identities are tied to interpersonal relationships. It was expected that common religious beliefs with whomever one was thinking of would lead to higher religious identification. See Figures 1 and 2 for a conceptual overview of these predicted effects.

Shared reality theory suggests two possible mechanisms by which thinking about a same-ethnicity relationship may increase ethnic identification and thinking about a different-ethnicity relationship may decrease ethnic identification. The first is closeness to the other person; it may be that the more one feels close to the other person, the more that common identities and beliefs are elicited. The second possible mechanism is shared understanding. It is not simply being from the same ethnic group that elicits high ethnic identification; rather, it is a common understanding of what ethnicity means that elicits high ethnic identification. Different-ethnicity spouses do not understand what one's ethnicity means to the same degree that same-ethnicity people can understand. Closeness with mother and spouse and shared understanding of ethnicity with mother and spouse were measured in order to assess whether any of them mediated the effect of thinking about mother or spouse on ethnic identification.

Implications of This Study for Relationship Cognition

This study has several important implications for understanding human thought in the context of interpersonal relationships. Previous research has found that people's self-views alter to fit the views of the person they are thinking of (e.g., Baldwin, Carrell, & Lopez, 1990) and specifically, that romantic partners have significant influence on people's views of their own personality characteristics (Aron, Aron, Tudor, & Nelson, 1991; Murray, Holmes, & Griffin, 1996). This study is the first to experimentally examine whether group identification too can be influenced by romantic relationships. Moreover, this study is the first to compare the influence of two close and important relationships on the same aspect of the self. Though previous research has established that other people's views of oneself help one maintain those self views, no research has examined the effects of two different relationships offering competing influences.

Method

Overview

To examine whether people's spouses in mixed-ethnicity marriages elicit lower ethnic identification, Asian American women from same- and mixed- ethnicity marriages were induced to think of their relationship with their spouse or with their mother. It was expected that the level of ethnic identification will be lower when they are thinking of someone of a different ethnicity from themselves than when they are thinking of someone of the same ethnicity as themselves. Of those induced to think of their spouse, those in mixed-ethnicity marriages were expected to report lower ethnic identification than those in same-ethnicity marriages. Of those induced to think of their same-ethnicity mother, all were expected to report high ethnic identification, regardless of the ethnicity of the spouse. This study used a 2 X 2 between-subjects factorial design. See Figure 3 for a graph of the predicted pattern of results.

Participants

Participants were 118 married Asian American women, recruited through campus newspaper ads, flyers on campus, and snowballing. Participation was restricted to currently married women of Chinese (n=73), Japanese (n=21), or Korean (n=24) descent, who identified with only one ethnicity and whose mothers were alive. These restrictions were necessary for theoretical reasons: 1) shared reality theory stresses ongoing relationships, thus it was important that the participants' spouse and mother were, at the minimum, still living and 2) the measures of ethnic identification assume identification with only one ethnicity, thus it was better to sample people of one ethnic background to keep the construct clean. Four women were dropped for not completing the primes or for not completing the entire survey. Two additional women were not included in the analyses because they were adopted into White families as young children.

Of the remaining 112 women, 46 (41%) were in mixed-ethnicity marriages and 66 (59%) were in same-ethnicity marriages (see Table 2 for cross-tabulation of participant ethnicity and spouse ethnicity). Participants classified themselves into same- or mixed- ethnicity marriages by answering the question “Do you consider your marriage to be interethnic?” The average age of participants from same-ethnicity marriages was 32.9 years (range from 23 to 65), while the average age of participants from mixed-ethnicity marriages was 31.4 years (range from 23 to 51). At the time of the study, participants in same-ethnicity marriages had been married for an average 6.68 years (range from .42 to 44.5; median was 4.5) and participants in mixed-ethnicity marriages had been married for an average 5.27 years (range from .08 to 24.9; median was 2.41). Of the 66 women from same-ethnicity marriages, 32 were born outside of the U.S. and 34 were 2+ generations. Of the 46 women from mixed-ethnicity marriages, 21 were born outside of the U.S. and 25 were 2+ generations. Twenty-eight women from same-ethnicity marriages (42%) and 15 women from mixed-ethnicity marriages (32%) were mothers themselves; most having 1-2 children. Over half of the participants had no children, reflecting the fact that many of these women had been married for a relatively short time. Because the survey was available online, women from various regions on the United States could participate in this study. Of the participants in same-ethnicity marriages, 82% lived in California (n = 54) and 18% lived in other states (n = 12). Of the participants in mixed-ethnicity marriages, 85% lived in California (n = 39) and 15% lived in other states (n = 7). The only method of completing the survey was online, which probably limited the sample to relatively highly educated and high SES women who had internet access.

Materials

Experimental manipulation. To manipulate which relationship participants were motivated to maintain, participants were asked to think of either their relationship with their mother or their relationship with their spouse before completing the rest of the survey. They answered two short open-ended questions about the significance and positive impact of their relationship with their mother or with their spouse. The first question was “Think about your relationship with your spouse [mother].  In what ways is your relationship with him [her] unique and special?  What characteristics of your relationship make you and your spouse [mother] feel close to each other?” The second question was “The people who are important to us affect our lives in tremendous ways.  Think about how your spouse [mother] has shaped you as a person.  Please describe one way in which your spouse [mother] has positively affected the person that you are today.  For example, what positive personal characteristics do you have now because of your spouse [mother] and how did he [she] help you gain those characteristics?”

Manipulation check. The effectiveness of the prime was assessed by four items following the open-ended questions. Participants were asked to rate on a 7-point Likert scale how much they love their mother and their spouse and how close they feel to their mother and their spouse. Participants primed to think of their mother were expected to report feeling closer to their mother, compared to participants primed to think of their spouse. Participants primed to think of their spouse were expected to report feeling closer to their spouse, compared to participants primed to think of their mother.

Ethnic identification. Ethnic identification was measured by the 16-item Collective Self-Esteem Scale (CSES; Luhtanen & Crocker, 1992), modified for ethnic groups, and by the Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992). A short statement introducing the measures explained to participants that they should think of their own ethnic group whenever they read the phrase “ethnic group.”

The Collective Self Esteem Scale is composed of four subscales (see Appendix A). Because each subscale of four items assesses a different dimension of ethnic identification, all subscales were analyzed separately. The Membership subscale assessed the degree to which participants feel like worthwhile members of their ethnic groups (α = .76, .74) . A sample item of the Membership subscale was “I am a worthy member of the ethnic group I belong to.”   The Identity subscale assessed the degree to which the participants see their ethnic group as an important part of who they are (α = .74, .82). A sample item of the Identity subscale was “The ethnic group I belong to is an important reflection of who I am.” The Private subscale assessed how positively participants feel about their ethnic groups (α = .76, .90). A sample item of the Private subscale was “I feel good about the ethnic group I belong to.” The Public subscale assessed how positively participants perceive their ethnic group to be evaluated by others (α = .62, .75). A sample item of the Public subscale was “Overall, my ethnic group is considered good by others.”

All 20 items of the MEIM were administered (see Appendix B). Fourteen of the items combined to form the overall ethnic identification score (α = .85, .89). These items can be further broken down into subscales of Affirmation and Belonging (α = .84, .82), Ethnic Identity Achievement (α = .73, .81), and Ethnic Behaviors (α = .05, .17). A sample item of the Affirmation and Belonging subscale was “I am happy that I am a member of the group I belong to.” A sample item of the Ethnic Identity Achievement subscale was “I have a clear sense of my ethnic background and what it means for me.” A sample item of the Ethnic Behaviors subscale was “I am active in organizations or social groups that include mostly members of my own ethnic group.” Because the Ethnic Behaviors subscale consisted of only two items and was not reliable in this sample, the analyses with it will not be reported. The remaining six items formed the Other Group Orientation scale (α = .83, .69). Because this is not a measure of ethnic identification, the analyses with it will not be reported either. A sample item of the Other Group Orientation scale was “I like meeting and getting to know people from ethnic groups other than my own.”

Self esteem. Participants completed a 10-item measure of self-esteem (α = .90, .87; Rosenberg, 1965). A sample item was “I feel that I am a person of worth, at least on an equal basis with others.” See Appendix C.

Religious identification. Religious identification was also measured by the 16-item Collective Self-Esteem Scale (Luhtanen & Crocker, 1992) but modified for religious groups. Wherever the ethnic identification version of the Collective Self Esteem Scale read “ethnic group”, it was replaced with “religious group”; nothing else was changed. The Membership (α = .88, .78), Identity (α = .88, .92), Private (α = .90, .81), and Public (α = .78, .90) subscales were used. See Appendix D.

Closeness with spouse. Closeness with spouse was measured in five ways. Participants completed the 7-item Relationship Assessment Scale (RAS; Hendrick, 1988), α = .91, .92; the 6-item emotional intimacy subscale (α = .85, .87) and the 6-item intellectual intimacy subscale (α = .73, .86) of the PAIR inventory (Schaefer & Olson, 1981); the 1-item Including Other in the Self scale (IOS; Aron, Aron, & Smollan, 1992), and 18 items from the Dyadic Adjustment Scale (DAS; Spanier, 1976). A sample item from the RAS was “How well does your spouse meet your needs?” A sample item from the emotional intimacy subscale was “My spouse listens to me when I need someone to talk to.” A sample item from the intellectual intimacy subscale was “My spouse helps me clarify my thoughts.” See Appendices E, F, G, and H.

Closeness with mother. Closeness with mother was measured in four ways. Participants completed the RAS for their mother (α = .93, .94), the emotional intimacy (α = .85, .84) and intellectual intimacy (α = .83, .86) subscales of the PAIR inventory for their mother, and the IOS for their mother. Wherever the scales previously asked about “spouse”, it was replaced with “mother”; nothing else was changed. Participants were also asked how frequently they interact with their mother and how far away from their mother they live. They were also asked how responsible for their mother's day-to-day needs, to assess how personally taxing their relationship with their mother may be to them.

Shared understanding. The degree to which mothers and spouses share with participants an understanding of ethnicity and religion were assessed by scales that the author constructed. Seven items assessed the shared understanding of ethnicity with spouse (α = .73, .88) and seven items with mother (α = .66, .55). A sample item assessing shared understanding of ethnicity with spouse was “My spouse understands a lot about my ethnic group.” A sample item assessing shared understanding of ethnicity with mother was “My mother usually understands when I tell her my feelings about being a member of my ethnic group.” Four items assessed the shared understanding of religion with spouse (α = .89, .90) and four with mother (α = .89, .87). A sample item assessing shared understanding of religion with spouse was “My spouse thinks that my religious traditions are important.” A sample item assessing shared understanding of religion with mother was “My mother understands a lot about my religion.” See Appendices I and J.

Spouse's identifications. Participants reported their spouses' degrees of identification with their own ethnicity and religion on two close-ended items: “How important is your spouse's ethnic background to him/her?” and “How important is your spouse's religion to him/her?” Participants were also asked how similar their spouse's ethnicity was to their own and how similar their spouse's religion was to their own. These questions were answered on 7-point Likert scales.

Mother's identifications. Participants reported their mothers' degrees of identification with their own ethnicity and religion on two close-ended items: “How important is your mother's ethnic background to her?” and “How important is your mother's religion to her?” Participants were also asked how similar their mother's religion was to their own. These questions were answered on 7-point Likert scales.

Social Desirability. Participants completed the 6-item Conventionality subscale (α=.87) of the PAIR inventory as a measure of social desirability. This measure was chosen because its items pertain to romantic relationships and fit better with the other questions in the study. See Appendix K.

Procedure

Participants were recruited through three major means. First, an advertisement ran in a campus newspaper, the UCLA Daily Bruin, for a study on marriage. Second, flyers were posted around the UCLA campus, primarily near the medical center where more married women were likely to see the flyers. Third, in a large snowballing effort, friends of the author forwarded email flyers to their friends, who were asked to continue passing on the flyer. The ad and flyers invited interested married Asian American women to contact the author by phone or by email.

All potential participants were screened by telephone or email by the author or one of two research assistants. They were told that the purpose of the study was to understand how family relationships affect cultural and religious identity. The procedure of the study was explained and any questions answered. If potential participants met the criteria, they were given the link to the online survey and a password with which to access the survey (see http://www.karensurvey.com). Upon entering the site, participants were given an informed consent form, again explaining the purpose and the procedure of the study. They were then randomly assigned to receive the spouse prime webpage or the mother prime webpage. All subsequent pages after the prime webpage were identical for all participants. They completed the manipulation check questions, measures of ethnic and religious identification, measures of closeness with spouse, measures of closeness with mother, and demographics.

Participants were instructed to complete the survey in one sitting; however, the website did notify participants when they had completed half of the survey and that they could take a break at that point if necessary. The entire survey took approximately one hour to complete. Participants were given $10 compensation and entered in a drawing for a romantic dinner for two after they completed the survey.

Results

Overall, 62 participants received the spouse prime and 50 participants received the mother prime. Among those in same-ethnicity marriages, 36 (54%) completed the spouse prime and 30 (46%) completed the mother prime. Among those in mixed-ethnicity marriages, 26 (56%) completed the spouse prime and 20 (44%) completed the mother prime.

Participants who live outside of California (n = 19) did not differ from participants who live in California (n = 93) on demographic characteristics, such as age, length of marriage, and likelihood to be in a mixed-ethnicity marriage. After controlling for the effects of the prime and of type of marriage, participants living outside of California did not differ from participants living in California on any of the measures of ethnic identification. Because there were no significant differences, participants residing in any state were included in the analyses.

Covarying social desirability did not change the pattern of results, so the findings reported here do not control for social desirability.

Characteristics of the Sample

This sample differed in marked ways from the profile of average Asian American marriages. First, the proportion of mixed-ethnicity marriages was higher than the national average. According to Lee and Fernandez's (1998) analysis of the 1990 U.S. Census data, approximately 25% of married Asian Americans were married to someone of a different ethnicity. In this sample, 40% of the Asian American marriages were mixed-ethnicity. This high proportion was probably the result of using a snowballing method of recruitment: people in mixed-ethnicity marriages are more likely to have friends also in mixed-ethnicity marriages. However, this sample was similar to the national average in that half of the mixed-ethnicity marriages involve a spouse of a different Asian ethnicity and half of the mixed-ethnicity marriages involve a spouse of non-Asian ethnic background. Second, the age of marriage for people in mixed-ethnicity marriages may differ from the national average. Analysis of marriage license data from the 1980's suggests that Asian Americans who marry outside of their ethnic group are, on average, older at the time of marriage than Asian Americans who marry within their ethnic group (Sung, 1990). In this sample, the age of marriage for those in mixed-ethnicity marriages (M = 27.1) was almost identical to the age of marriage for those in same-ethnicity marriages (M = 26.4). An independent samples t-test indicated there was no difference between the two groups. A third way in which this sample differed from the national average of Asian American marriages was that the proportion of Christians was very high. The percentage of Asian Americans who are Christian is estimated to be 5% (K. J. U. Fong, personal communication, April 23, 2002). In this sample, nearly 80% of the participants claimed Christianity as their religion. This too was probably the result of using snowballing; people who are Christian are more likely to have Christian friends. While the differences between this sample and the average profile of Asian Americans may limit the ability to generalize from the descriptive aspects of the sample to the population of Asian Americans (e.g., overall level of self esteem or ethnic identification), it is still possible to infer the relationship between the priming manipulation and other variables in the study.

Self Esteem

As predicted, neither the priming manipulation nor the type of marriage affected participants' level of self esteem. The predicted contrast pattern for ethnic identification - that women primed with their different-ethnicity spouse will report lower ethnic identification than women primed with their same-ethnicity spouse or women primed with their mother - was not significant for self esteem, p = .95. Women reported high self esteem, regardless of the ethnicity of their spouse or with whom they were primed. See Figure 5. Women also reported high self esteem, regardless of the religious match with their mother or spouse.

Ethnic Identification

Manipulation Check

Two independent samples t-tests tested whether the priming manipulation was effective in making women feel closer to their mother when they were primed with their mother and in making women feel closer to their spouse when they were primed with their spouse. Results indicated that the mother prime was effective in making women feel closer to their mother, but the spouse prime did not make women feel closer to their spouse. Women who were primed to think of their positive relationship with their mother (M = 5.78) reported feeling significantly closer to their mother than women primed to think of their spouse (M = 5.27), t(110) = -1.97, p < .05, Cohen's d = .37. In contrast, women who were primed to think of their spouse (M = 6.03) did not report feeling closer to their spouse than women primed to think of their mother (M = 6.12), t(110) = -.41, p = .68. That the manipulation worked only for those in the mother condition and not for those in the spouse condition may have reflected participants' unwillingness to report lower levels of closeness to their spouse. The priming manipulation may still have effects on ethnic identification, though, aside from the explicit manipulation check.

Effects of Prime on Ethnic Identification

Though no separate predictions were made about the main effect of priming mother or spouse on ethnic identification, the predicted pattern of results - that women primed with their different-ethnicity spouse will report lower ethnic identification than women primed with their same-ethnicity spouse or women primed with their mother - implies that those primed with their spouse will report lower ethnic identification than those primed with their mother. Consistent with that implication, it was found that, collapsing across participants in both same- and mixed- ethnicity marriages, those primed to think of their mother reported higher scores than those primed to think of their spouse on the Multigroup Ethnic Identity Measure (MEIM; M = 5.38 vs. M = 5.07, F(1,108) = 5.22, p < .05, Cohen's d = .44) and the Ethnic Identity Achievement subscale of the MEIM (M = 5.19 vs. M = 4.78, F(1,108) = 7.38, p <. 01, d = .52).

Effects of Type of Marriage on Ethnic Identification

In addition, a main effect of type of marriage on ethnic identification was consistent with the theoretical predictions of shared reality theory. Because married people may be chronically motivated to maintain a relationship with their spouse, those married to a different-ethnicity spouse may report lower ethnic identification overall compared to those married to a same-ethnicity spouse. Analysis of variance tests, controlling for the effect of the prime and the interaction between the prime and the type of marriage, suggested that indeed women in mixed-ethnicity marriages were less identified with their ethnicity than women in same-ethnicity marriages. Women in same-ethnicity marriages scored significantly higher than women in mixed-ethnicity marriages on the overall MEIM (M = 5.38 vs. M = 4.95, F(1,108) = 6.26, p <. 05, Cohen's d = .48) and the Affirmation and Belonging subscale of the MEIM (M = 5.81 vs. M = 5.36, F(1,108) = 6.29, p < .05, d = .48). Women in same-ethnicity marriages also reported more positive evaluations of their own ethnic group than women in mixed-ethnicity marriages. Compared to those in mixed-ethnicity marriages, those in same-ethnicity marriages felt more positively about their own ethnic group, as measured by the Private subscale of the Collective Self Esteem Scale (CSES) (M = 6.29 vs. M = 5.98, F(1,108) = 4.44, p < .05, d = .40), and perceived others to feel more positively about their own ethnic group, as measured by the Public subscale of the CSES (M = 5.94 vs. M = 5.61, F(1,108) = 5.13, p < .05, d = .43).

Effects of Prime and Type of Marriage on Ethnic Identification

Planned contrasts tested the prediction that people primed to think of their different-ethnicity spouse will report lower ethnic identification than people primed to think of their same-ethnicity spouse or people primed to think of their mother. See Table 3 for the contrast weights used in testing the predicted pattern. The data were found to match the predicted pattern of results for the Multigroup Ethnic Identity Measure (MEIM) but not for the Collective Self Esteem Scale (CSES).

Ethnic identification, as assessed by the MEIM, was higher when women were experimentally induced to think of a positive relationship with someone of the same ethnicity than when they were experimentally induced to think of a positive relationship with someone of a different ethnicity. The predicted contrast pattern was significant when testing the overall MEIM score, t(108) = 3.83, p < .001, d = .74. See Figure 6. Post-hoc t-tests confirmed that women in mixed-ethnicity marriages who were primed to think of their different-ethnicity spouse reported significantly lower MEIM scores (M = 4.66) than women in same-ethnicity marriages who were primed to think of their same-ethnicity spouse (M = 5.36, t(60) = 3.43, p < .01, d = .88), women in mixed-ethnicity marriages who were primed to think of their mother (M = 5.32, t(44) = 2.5, p < .01, d = .75), and women in same-ethnicity marriages who were primed to think of their mother (M = 5.42, t(54) = 3.33, p < .01, d = .90). The MEIM scores of the women in the latter three conditions did not differ significantly from one another. The predicted pattern was significant for the Ethnic Identity Achievement subscale of the MEIM, t(108) = 3.73, p < .01, d = .72. See Figure 7. Post-hoc t-tests showed that women in mixed-ethnicity marriages primed with their different-ethnicity spouse reported significantly lower Ethnic Identity Achievement scores (M = 4.39) than women in same-ethnicity marriages primed with their same-ethnicity spouse (M = 5.06, t(60) = 2.84, p < .01, d = .73), women in mixed-ethnicity marriages primed to think of their mother (M = 5.25, t(44) = 2.95, p<.01, d = .89), and women in same-ethnicity marriages primed to think of their mother (M = 5.16, t(54) = 3.14, p < .01, d = .85). The contrast pattern was also significant for the Affirmation and Belonging subscale of the MEIM, t(108) = 2.75, p < .05, d = .53. See Figure 8. Post-hoc t-tests showed that women primed with their different-ethnicity spouse reported significantly lower Affirmation and Belonging scores (M = 5.19) than women in same-ethnicity marriages primed to think of their same-ethnicity spouse (M = 5.82, t(60) = 3.04, p < .01, d = .78) and women in same-ethnicity marriages primed to think of their mother (M = 5.79, t(54) = 2.47, p < .05, d = .67), but not significantly lower than women in mixed-ethnicity marriages primed to think of their mother (M = 5.58). These findings provide experimental evidence to support the theoretical claim that positive relationships with same-ethnicity people increase ethnic identification while positive relationships with different-ethnicity people decrease ethnic identification.

Planned contrasts testing the aspects of ethnic identity assessed by the Collective Self Esteem Scale provided limited support for the hypothesis. The Membership and the Identity subscales, which measure sense of belonging to one's ethnic group and degree of identification with one's ethnic group, respectively, were expected to vary as a function of relationships with same- and different- ethnicity people and thus were expected to show the predicted pattern. The means for the Membership subscale were consistent with the predicted pattern of women in mixed-ethnicity marriages reporting lower ethnic identification when thinking about their different-ethnicity spouse (M = 5.28), compared to women in mixed-ethnicity marriages thinking about their mother (M = 5.57) and women in same-ethnicity marriages thinking about their mother (M = 5.50) or their spouse (M = 5.72). The contrast test, though, was not significant, p = .12. Surprisingly, the predicted contrast pattern was not significant for the Identity subscale nor were the means consistent with the predicted pattern. It may be that the wording of these items conveyed too much of a “right” answer for subtle decreases or increases to occur. In contrast to the Ethnic Identity Achievement subscale which emphasizes cultural behaviors and clear understanding of the meaning of one's ethnicity, the Identity subscale emphasizes how important to one's self-concept one considers one's ethnic group to be. Perhaps the importance that one attaches to one's ethnicity does not easily fluctuate. The Private and the Public subscales, which measure one's own and others' affective evaluation of one's ethnic group, do not measure ethnic identification as the Membership and the Identity subscales do and thus were not expected to show the predicted pattern. Consistent with that expectation, the contrast tests on the Private and Public subscales were not statistically significant. See Table 4 for the means of the Collective Self Esteem subscales as a function of prime and type of marriage.

Mediational Analyses

Mediational analyses were conducted to test whether women's closeness to their mother or spouse or shared understanding of ethnicity with their mother or with their spouse mediated the effects of the prime on ethnic identification and the effect of type of marriage on ethnic identification. Because the measures of closeness were highly correlated with one another, each scale was standardized and averaged together to create one closeness score for mother and one closeness score for spouse. The method to test mediation recommended by Baron and Kenny (1986) was followed, including the modified Sobel test to assess the statistical significance of the effect of the mediator.

Effects of type of marriage mediated by shared understanding of ethnicity with spouse. Tests of mediation suggest that the higher ethnic identification reported by people in same-ethnicity marriages compared to people in mixed-ethnicity marriages is mediated by shared understanding of ethnicity with spouse. That is, women in same-ethnicity marriages perceive their spouses to better understand their ethnicity than women in mixed-ethnicity marriages, and this accounts for their own higher ethnic identification. All regression analyses reported in this section control for the independent effect of the prime and the interaction between the prime and the type of marriage.

Differences between women in same- and mixed- ethnicity marriages on their levels of ethnic identification, as measured by the Multigroup Ethnic Identity Measure (MEIM), were mediated by the degree of shared understanding of ethnicity with spouse. The first regression confirmed that the type of marriage significantly predicted women's MEIM score, such that women in mixed-ethnicity marriages had lower MEIM scores than women in same-ethnicity marriages, β = -.25, p < .01. The second regression showed that type of marriage significantly predicted the degree to which women perceive their spouse to have a shared understanding of ethnicity, such that women in mixed-ethnicity marriages reported lower shared understanding of ethnicity than women in same-ethnicity marriages, β = -.50, p < .001. When type of marriage and shared understanding of ethnicity with spouse were entered together into the third regression, type of marriage no longer predicted the MEIM score, β = .05, ns, but shared understanding of ethnicity with spouse did, β = .61, p < .001. The modified Sobel test indicated that the mediation effect was significant, z = -4.56, p < .001. See Figure 9 for a graphical depiction.

Similarly, differences on the Affirmation and Belonging subscale of the MEIM between women in same- and mixed- ethnicity marriages were mediated by the degree of shared understanding of ethnicity with spouse. The first regression showed that women in mixed-ethnicity marriages reported lower Affirmation and Belonging scores than women in same-ethnicity marriages, β = -.25, p < .01. Because it was already found that type of marriage predicted shared understanding of ethnicity with spouse, this regression was not repeated. In the last regression, when type of marriage and shared understanding of ethnicity with spouse were entered together, the effect of type of marriage dropped to nonsignificance, β = -.04, but the effect of shared understanding of ethnicity with spouse was significant, β = .41, p < .001. The modified Sobel test was significant, z = -3.37, p < .001. See Figure 10.

Interestingly, the degree of shared understanding of ethnicity with spouse also mediated the difference between women in same- and mixed- ethnicity marriage on their personal feelings about their own ethnic group, as assessed by the Private subscale of the Collective Self Esteem Scale (CSES). The first regression confirmed that women in mixed-ethnicity marriages reported lower Private scores than women in same-ethnicity marriages, β = -.19, p < .05. When both type of marriage and shared understanding of ethnicity with spouse are entered into the regression, type of marriage no longer predicted Private scores, β = -.07, ns, but shared understanding of ethnicity with spouse did, β = .23, p < .05. The modified Sobel indicated that the mediation effect was significant, z = -2.04, p < .05. See Figure 11.

In addition, the degree of shared understanding of ethnicity with spouse mediated the effect of the type of marriage on how women perceive others to evaluate their own ethnic group, as measured by the Public subscale of the CSES. Women in mixed-ethnicity marriages reported lower Public scores than women in same-ethnicity marriages, β = -.20, p < .05. When type of marriage and shared understanding of ethnicity with spouse are both entered in the regression, type of marriage no longer predicted Public scores, β = -.10, ns, but shared understanding of ethnicity with spouse did, β = .21, p < .05. However, the modified Sobel test indicated that this mediation effect was only marginally significant, z = -1.88, p = .06. See Figure 12.

No other dimensions of women's relationships with their mother or their spouse were found to explain the difference in level of ethnic identification between those in same-ethnicity marriages and those in mixed-ethnicity marriages. The degree of shared understanding of ethnicity with mother, closeness to mother, and closeness to spouse did not mediate the effect of type of marriage on ethnic identification. Shared understanding of religion with mother or spouse, and the perceived importance of ethnicity to mother or spouse also did not mediate the effect of type of marriage on ethnic identification. These findings suggested that among the relationship dimensions measured in this study, only the degree of shared understanding of ethnicity with spouse explains why women in mixed-ethnicity marriages reported lower ethnic identification overall than women in same-ethnicity marriages.

Effect of prime for women in mixed-ethnicity marriage may be mediated by shared understanding of ethnicity with mother. Tests of mediation suggested that the effect of being primed to think of one's mother on the ethnic identification of women in mixed-ethnicity marriages may be mediated by their perceived shared understanding of ethnicity with their mother. These regression analyses focused only on women in mixed-ethnicity marriages.

Differences between women in mixed-ethnicity marriages who were primed with mother and who were primed with spouse on ethnic identification, as measured by the MEIM, may be mediated by shared understanding of ethnicity with mother. The first regression confirmed that those primed with mother reported higher MEIM scores than those primed with spouse, β = .35, p < .05. The second regression showed that those primed with mother reported higher levels of shared understanding of ethnicity with mother, β = .33, p < .05. When both prime and shared understanding of ethnicity with mother were included in the regression, the effect of being primed with mother on women's MEIM scores dropped to β = .22 , ns, but shared understanding of ethnicity with mother did significantly predict MEIM scores, β = .39, p < .01. The modified Sobel test indicated that the mediation effect was marginally significant, z = 1.80, p=.07. See Figure 13.

The effect of thinking about their mother versus thinking about their spouse on the Ethnic Identity Achievement scores of women in mixed-ethnicity marriage may also be mediated by shared understanding of ethnicity with mother. The first regression confirmed that those primed with mother reported higher Ethnic Identity Achievement scores than those primed with spouse, β = .41, p < .01. When both prime and shared understanding of ethnicity with mother were included in the next regression, the effect of being primed with mother dropped in its predictive ability but remained significant, β = .27, p < .05; shared understanding of ethnicity with mother also significantly predicted Ethnic Identity Achievement scores, β = .40, p <.01. The modified Sobel test indicated that the mediation effect was marginally significant, z = 1.85, p = .06. See Figure 14.

Other dimensions of relationships with mother and spouse did not explain why women in mixed-ethnicity marriages primed with their mother reported higher ethnic identification than women in mixed-ethnicity marriages primed with their spouse. Closeness with mother, shared understanding of ethnicity with spouse, and closeness with spouse did not mediate the effect of being primed with mother on any measures of ethnic identification. Shared understanding of religion with mother or spouse, and perceived importance of ethnicity to mother or spouse also did not mediate the effect of the prime on ethnic identification.

For the women in mixed-ethnicity marriages, shared understanding of ethnicity with their mother may explain why thinking about her results in higher ethnic identification compared to thinking about their spouse. Though the Sobel test indicated that the mediation effects were only marginally significant, these findings were similar to the mediational analyses on the effects of type of marriage on ethnic identification. In both sets of analyses, it was shared understanding of ethnicity and not closeness that mediated the effects. That the mediation effects in the latter analyses were only marginally significant may be due to the low reliability of the measure assessing shared understanding of ethnicity with mother.

Differences between Asian ethnic groups

Further analyses were done to test whether Chinese Americans, Japanese Americans, and Korean Americans differed from one another in the effect of the priming manipulation on their ethnic identification. An overall analysis of variance comparison of the three groups, controlling for the effect of the prime, revealed that the level of ethnic identification did not differ among the three ethnic groups. However, they did differ in the effect of the prime on their ethnic identification. Japanese American women did not show the predicted pattern on any measure of ethnic identification, but Chinese American women and Korean American women did.

Similar to the overall sample, planned contrasts showed that Chinese American women primed with their different-ethnicity spouse reported lower ethnic identification than Chinese American women primed with their same-ethnicity spouse and Chinese American women primed to think of their mother on the Multigroup Ethnic Identity Measure (t(66) = 3.05, p < .05, Cohen's d = .75), the Affirmation and Belonging subscale (t(66) = 2.52, p < .05, d= .62), the Ethnic Identity Achievement subscale (t(66) = 2.94, p < .01, d = .72), and the Membership subscale of the Collective Self Esteem Scale (t(66) = 2.04, p < .05, d = .50). See Table 5 for the means. Perhaps because the number of Chinese Americans in mixed-ethnicity marriages was small (n = 28), shared understanding of ethnicity with mother or spouse and closeness to mother or spouse were not found to mediate the effect of the prime on ethnic identification.

Even though the number of Korean Americans in this sample was small (n = 22), they nevertheless showed the predicted contrast pattern on two measures of ethnic identification. Korean American women primed with their different-ethnicity spouse reported lower ethnic identification than Korean women in any of the three other conditions on the MEIM (t(18) = 2.28, p < .05, d = 1.07) and Ethnic Identity Achievement (t(18) = 2.94, p < .01, d = 1.38). See Table 5 for the means.

Interestingly, though the number of Japanese Americans (n = 20) was similar to the number of Korean Americans in the sample, Japanese American women did not show the predicted contrast effect on any measure of ethnic identification. This may be due to low power, though that seems unlikely given the effects that nearly the same number of Korean Americans showed. A more likely explanation is that these Japanese American women differed from the Chinese American and Korean American women in this sample on one important dimension - place of birth. All 20 of the Japanese American women were born in the United States and 85% of them were either 3^rd or 4^th generation. In contrast, 19 of the 22 Korean American women were born outside of the U.S. Approximately half of the Chinese American sample were born outside the U.S. and half were born in the U.S. However, among the Chinese Americans born in the U.S., 83% were 2^nd generation. This suggests that the Japanese American women in this study were more assimilated than the Korean American and the Chinese American women. Greater assimilation may mean that there is less perceived cultural difference between their different-ethnicity husband and their mothers, thus being primed with either produces equivalent effects on their ethnic identification.

Differences among women in mixed-ethnicity marriages: when her spouse is not Asian vs. when her spouse is of a different Asian ethnicity

Analyses were done to test the possibility that Asian American women who were primed with their non-Asian different-ethnicity spouse would report a greater decrease in ethnic identification than Asian American women primed with their Asian different-ethnicity spouse. It may be that the amount of the decrease in ethnic identification depends on how different the spouse's culture is from one's own. After controlling for the priming manipulation, it was found that women with non-Asian different-ethnicity spouses rated their spouse's culture more different from their own culture than women with Asian different-ethnicity spouses did (M = 5.13 vs. M = 3.91, F(1, 42) = 8.72, p < .01, d = .17). However, other data did not support the contention that women in mixed-ethnicity marriages experience a differential effect on ethnic identification as a function of their spouse's ethnicity. Independent samples t-tests revealed no differences on any measure of ethnic identification between women primed with their different-ethnicity non-Asian spouse and women primed with their different-ethnicity Asian spouse. When planned contrasts compared women in same-ethnicity marriages with women whose different-ethnicity spouse is non-Asian, the predicted pattern was significant for the MEIM score (t(85) = 2.08, p < .05, Cohen's d = .45) and the Ethnic Identity Achievement subscale (t(85) = 1.98, p < .05, d = .43). The pattern was marginally significant for the Membership subscale of the Collective Self Esteem Scale, p = .07. When planned contrasts compared women in same-ethnicity marriages with women whose different-ethnicity spouse is Asian, the predicted patterns were significant and even stronger for the MEIM (t(83) = 3.70, p < .001, d = .81), the Ethnic Identity Achievement subscale (t(83) = 3.62, p < .001, d = .79), and the Affirmation and Belonging subscale (t(83) = 2.75, p < .01, d = .60). Again, perhaps because the number of women with a non-Asian different-ethnicity spouse (n = 23) and the number of women with an Asian different-ethnicity spouse (n = 22) were small, no mediation effects were found for closeness to mother or spouse, or for shared understanding of ethnicity with mother or spouse.

Religious Identification

In this sample, there was a significant difference between the religious groups on strength of religious identification, even after controlling for the priming manipulation. Participants who identified as Christian reported significantly higher religious identification (M=6.17) than participants who identified as Agnostic (M=4.10, F(1,92)=102.22, p<.001), Atheist (M=2.00, F(1,86)=53.11, p<.001), Buddhist (M=4.65, F(1,88)=16.54, p<.001), or Spiritual (M=4.34, F(1,94)=84.31, p<.001). There were also many more participants who identified as Christian (n = 88) than participants who identified as Agnostic (n = 8), Atheist (n = 1), Buddhist (n = 4), or Spiritual (n = 11). Among the participants in this study, women who were not Christian were much more likely to be in mixed-religion marriages (67%) than women who were Christian (9%). Because level of religious identification may be confounded with type of religion and type of marriage and in order to simplify the presentation of the results, the analyses on religious identification reported here include only Christians.

Effect of Type of Marriage on Religious Identification

Consistent with the previous finding that women in mixed-ethnicity marriages reported lower ethnic identification then women in same-ethnicity marriages, Christian women in mixed-religion marriages reported lower religious identification than Christian women in same-religion marriages, after controlling for the effect of the priming manipulation and the interaction between the prime and the type of marriage. Women whose spouse is of a different religion scored lower than women whose spouse is of the same religion on the Membership subscale of the Collective Self Esteem Scale (CSES) measure modified for religious groups (M = 4.62 vs. M = 6.17, F(1, 84) = 29.16, p < .001, d = 1.18) and the Identity subscale (M = 5.19 vs. M = 6.44, F(1, 84) = 23.75, p < .001, d = 1.06). They also felt less positively about Christianity than women married to same-religion spouses, as assessed by the Private subscale of the CSES (M = 6.09 vs. M = 6.63, F(1, 84) = 9.10, p < .01, d = .66).

Mediational analyses suggested that shared understanding of religion with spouse and perceived importance of religion to spouse mediated the main effect of type of marriage on religious identification. All mediation analyses control for the effect of the priming manipulation and the interaction between the priming manipulation and the type of marriage.

That Christian women with same-religion spouses reported a greater sense of belonging to the Christian religion than Christian women with mixed-religion spouses was mediated by shared understanding of religion with spouse and perceived importance of religion to spouse. The first regression confirmed that women in mixed-religion marriages reported lower scores on the Membership subscale then women in same-religion marriages, β = -.49, p <.001. The second regression showed that women in mixed-religion marriages reported lower shared understanding of religion with spouse than women in same-religion marriages, β = -.72, p < .001. When both type of marriage and shared understanding of religion with spouse were entered in the third regression, type of marriage became a nonsignificant predictor, β = -21, ns, but shared understanding of religion with spouse was significant, β = .39, p < .01. The modified Sobel test indicated that this mediation is statistically significant, z = -2.75, p < .01. See Figure 15. In another series of regressions testing the mediating effect on the Membership subscale, the first regression showed that women in mixed-religion marriages reported lower perceived importance of religion to spouse than women in same-religion marriages, β = .65, p < .001. When type of marriage and perceived importance of religion to spouse were entered into the regression together, type of marriage remained significant as a reduced predictor, β = .34, p < .01, and perceived importance of religion to spouse was significant, β = .30, p < .05. The modified Sobel test indicated that this mediation effect was significant, z = -2.37, p < .05. See Figure 16.

The main effect of type of marriage on how identified Christian women were with their religion was also mediated by shared understanding of religion with spouse and by perceived importance of religion to spouse. The first regression confirmed that women in mixed-religion marriages reported lower scores on the Identity subscale than women in same-religion marriage, β = -.46, p <.001. The second regression showing that women in mixed-religion marriages reported lower shared understanding of religion with spouse than women in same-religion marriages was already reported. When type of marriage and shared understanding of religion with spouse were both entered into the regression, type of marriage became a nonsignificant predictor, β = .002, ns, while shared understanding of religion with spouse was significant, β = .64, p < .001. The modified Sobel test indicated that the mediation was significant, z = -4.48, p < .001. See Figure 17. When type of marriage and perceived importance of religion to spouse were entered into the regression together, type of marriage became nonsignificant as a predictor, β = -.05, ns, and perceived importance of religion to spouse was significant, β = .58, p < .001. The modified Sobel test indicated that this mediation effect was significant, z = -4.22, p < .001. See Figure 18.

Similar to the ethnic identification results, the effects on women's feelings about their religious group were mediated by the same variables as the effects on their religious identification. The main effect of type of marriage on the Private subscale was mediated by shared understanding of religion with spouse and by perceived importance of religion to spouse. The first regression showed that women in mixed-religion marriages reported lower Private scores than women in same-religion marriage, β = -.32, p <.01. The second regression showing that women in mixed-religion marriages reported lower shared understanding of religion with spouse than women in same-religion marriages was already reported. When type of marriage and shared understanding of religion with spouse were both entered into the regression, type of marriage became a nonsignificant predictor, β = .03, ns, while shared understanding of religion with spouse was significant, β = .49, p < .01. The modified Sobel test indicated that the mediation was significant, z = -3.17, p < .01. See Figure 19. When type of marriage and perceived importance of religion to spouse were entered into the regression together, type of marriage became nonsignificant as a predictor, β = -.02, ns, and perceived importance of religion to spouse was significant, β = .39, p < .01. The modified Sobel test indicated that this mediation effect was significant, z = -2.76, p < .01. See Figure 20.

No other relationship dimensions were found to explain the difference between the level of religious identification of Christian women in mixed-religion marriages and Christian women in same-religion marriages. Closeness with mother or with spouse, shared understanding of religion with mother, and perceived importance of religion to mother were not found to mediate the main effect of type of marriage on religious identification. Rather, among the relationship dimensions assessed in this study, only the degree of shared understanding of religion with spouse and the perceived importance of religion to spouse explain why Christian women in same-religion marriages identify more strongly with their religion than Christian women in mixed-religion marriages do.

The religious identification results provide converging evidence that identification is tied to specific relationships, though the extreme difference in sample size between the two groups (80 Christian women in same-religion marriages, 8 Christian women in mixed-religion marriages) makes these findings less statistically reliable than the ethnic identification findings.

Effect of Mother's Religion on Religious Identification

Christian women whose mother is of a different religion did not report lower religious identification on any of the CSE subscales compared to Christian women whose mother is of the same religion, after controlling for the priming manipulation and the interaction between the prime and the type of marriage in an analysis of variance.

Effect of Prime on Religious Identification

Christian women primed with their mother did not report higher or lower religious identification on any of the CSE subscales compared to Christian women primed with their spouse, after controlling for the type of marriage and the interaction between the type of marriage and the prime in an analysis of variance.

Effect of Prime and Mother's Religion on Religious Identification

Because an overwhelming portion of the Christians in this study were married to same-religion spouses (80 in same-religion marriages, 8 in mixed-religion marriages), the interactive effects of prime and mother's religion were tested on religious identification of women in same-religion marriages only.

Planned contrasts tested the prediction that women primed to think of their different-religion mother would report lower religious identification than women primed to think of their same-religion mother and women primed to think of their same-religion spouse. None of the contrasts were significant for any of the measures of religious identification.

Discussion

The results of this study provide experimental evidence to support shared reality theory's prediction that ethnic identification varies as a function of the ethnicity of the other person with whom one is in a positive relationship; if the other person is of the same ethnicity, ethnic identification is high, but if the other person is of a different ethnicity, ethnic identification is lower. People in mixed-ethnicity marriages who thought about their positive relationship with their spouse reported lower ethnic identification compared to people in same-ethnicity marriages who thought about their positive relationship with their spouse and people in either type of marriage who thought about their positive relationship with their mother. The size of this effect was large and consistent across the subscales of the Multigroup Ethnic Identity Measure (MEIM; Phinney, 1992), a well-known measure of ethnic identification used by many ethnic identity researchers. This effect was also found among Chinese American women on the Membership subscale of the Collective Self Esteem Scale (CSES). Surprisingly, however, the effect was not found for the Identity subscale of the CSES. More research must be done to clarify the differences between the Identity subscale and other measures of ethnic identification, and to understand how the identification with one's ethnic group might differ from the evaluation of one's ethnic group (i.e., Private and Public subscales).

In support of shared reality theory's claim that shared perspectives strengthen identification, the mechanism explaining the differences in levels of ethnic identification was found to be shared understanding of ethnicity and not closeness. Greater shared understanding of ethnicity with one's spouse accounted for the higher ethnic identification of people in same-ethnicity marriages compared to people in mixed-ethnicity marriages. Greater shared understanding of ethnicity with mother accounted for the higher ethnic identification of people in mixed-ethnicity marriages thinking about their mother compared to people in mixed-ethnicity marriages thinking about their spouse. The impact of shared understanding on ethnic identification required an understanding specific to ethnicity and not a general sense of being understood by one's spouse or mother. Neither the intimacy subscales, which were part of the closeness score, nor the shared understanding of religion were able to explain the differences in ethnic identification.

The specificity of shared identities was further supported by the finding that only ethnic identification changed in response to the ethnicity of the person one was thinking of. Religious identification and self esteem did not vary as a function of the ethnicity of the person one was thinking of. That self esteem was equally high in all experimental conditions is consistent with the sociometer theory of self esteem (Leary, Tambor, Terdal, & Downs, 1995), which states that self esteem is high when one's interpersonal relationships are positive. According to the sociometer theory, then, self esteem should be high in all conditions because people were only instructed to think of the positive aspects of their relationship with their mother and of their relationship with their spouse. Interestingly, these results also demonstrate an instance when ethnic identification and self esteem diverge. Previous research concluded that high ethnic identification was essential for ethnic minorities because it was positively correlated with self esteem (for a meta-analysis of the relationship between self esteem and ethnic identification, see Allen, Howard, & Grimes, 1997). The present study found that lower ethnic identification is not accompanied by lower self esteem when one's interpersonal relationship with a different-ethnicity person is positive. This finding is consistent with recent developments in the conceptualization of ethnic identity, suggesting that ethnic minorities can receive high self esteem from strong identification with any group (e.g., religious group, gay community) and not necessarily strong identification with their ethnic group (Cross & Phagen-Smith, 1996).

The data on religious identification provide some converging evidence that the strength of identification is related to having a shared perspective. Greater shared understanding of religion with spouse accounted for the higher religious identification of people in same-religion marriages compared to people in mixed-religion marriages. However, religious identification did not vary with mother's religion or with shared understanding of religion with mother. One possible reason may be that because specific identities are tied to specific relationships, the women in this sample tie their religious identity more strongly to their relationship with their spouse than to their relationship with their mother. For example, one woman, when asked what makes her relationship with her spouse unique and special, wrote, “Most importantly, we share the same faith. We pray together and that draws us very close. We worship and serve and value the same things in life.” Another woman wrote, “We've gone through ups and downs over 25 years, but our commitment to each other is solid because of our commitment to God. I would say our spiritual base and purpose is what holds us together.” Why identities may be tied to some relationships and not to other relationships is not yet clear, and more research is needed.

Limitations

There are several limitations to the findings reported here. First, the sample consisted only of Asian American women. Further research should test whether the ethnic identification of African Americans and Latinos similarly respond to positive relationships with same- and different-ethnicity people. Men should be included in future studies also, in order to see whether their greater power in the marriage relationship reduces their likelihood to change their ethnic identification to accommodate their wife. Second, the religious identification data did not fully replicate the findings from the ethnic identification data. More research needs to be done to clarify how religious identification may differ from ethnic identification and why religious identification was not stronger when people's mothers adhered to the same religion. Third, the shared understanding measures only assessed the perceived shared understanding by one person in the relationship. A more powerful test of shared reality theory's claim that shared understanding strengthens identification would be to assess the other person's level of shared understanding and to see, for example, whether the husband's reported understanding of his wife's ethnicity predicts her level of ethnic identification. Fourth, further research must be done to explain why people in mixed-ethnicity marriages evaluate their own ethnic group less positively than people in same-ethnicity marriages, whether this is the cause or the result of mixed-ethnicity marriage, and whether there are long-term psychological consequences to having a less positive (but not negative) evaluation of one's ethnic group.

Conclusion

Taken together, the results of this study provide strong support for the predictions made by shared reality theory and the predictions for positive relationships made by Cheng and Hardin's (2002) model. This study demonstrated that mixed-ethnicity marriage leads to lower ethnic identification. Importantly, however, it also showed that people in mixed-ethnicity marriages do not have lower ethnic identification in every situation. They express high ethnic identification when they are with same-ethnicity family and friends. In those situations, their level of ethnic identification is just as high as the ethnic identification of people in same-ethnicity marriages. For people in a mixed-ethnicity marriage who value high ethnic identification, the results of this study suggest two ways to maintain high ethnic identification: keep up positive relationships with same-ethnicity family and friends, and encourage one's different-ethnicity spouse to learn more about one's ethnicity so that there will be a greater shared understanding of ethnicity in the marriage.

References

Allen, M., Howard, L., & Grimes, D. (1997). Racial group orientation and self-concept: examining the relationship using meta-analysis. Howard Journal of Communications, 8(4), 371-386.

Aron, A., Aron, E. N., & Smollan, D. (1992). Inclusion of other in the self scale and the structure of interpersonal closeness. Journal of Personality & Social Psychology, 63(4), 596-612.

Aron, A., Aron, E. N., Tudor, M., & Nelson, G. (1991). Close relationships as including other in the self. Journal of Personality and Social Psychology, 60, 241-253.

Atkinson, D. R., Morten, G., & Sue, D. W. (1989). A minority identification development model. In D. R. Atkinson, G. Morten, & D. W. Sue (Eds.), Counseling American minorities (3^rd ed., pp. 35-47). Dubuque, IA: William C. Brown.

Baldwin, M. W., Carrell, S. E., & Lopez, D. F. (1990). Priming relationship schemas: My advisor and the pope are watching me from the back of my mind. Journal of Experimental Social Psychology, 26, 435-454.

Bosson, J. K., & Swann, W. B., Jr. (2001). The paradox of the sincere chameleon: Strategic self-verification in close relationships. In J. Harvey & A. Wenzel (Eds.), Close Romantic Relationships: Maintenance and Enhancement. (pp. 67-86). Mahwah, NJ: Lawrence Erlbaum Associates.

Branscombe, N. R., Schmitt, M. T., & Harvey, R. D. (1999). Perceiving pervasive discrimination among African Americans: implications for group identification and well-being. Journal of Personality & Social Psychology, 77(1), 135-149.

Carter, M. M., Sbrocco, T., Lewis, E. L., & Friedman, E. K. (2001). Parental bonding and anxiety: Differences between African American and European American students. Journal of Anxiety Disorders, 15, 555-569.

Cheng, K. G. & Hardin, C. D. (2002). In search of missing persons: An interpersonal relationships framework for understanding ethnic identification. Manuscript submitted for publication.

Christian, J., Gadfield, N. J., Giles, H., & Taylor, D. M. (1976). The multidimensional and dynamic nature of ethnic identity. International Journal of Psychology, 11(4), 281-291.

Cross, W. E., Jr. (1978). The Thomas and Cross models of psychological nigrescence: a review. Journal of Black Psychology, 5(1), 13-31.

Cross, W. E., Jr. (1995). The psychology of nigrescence: Revising the Cross model. In J. G. Ponterotto, J. M. Cases, L. A. Suzuki, & C. M. Alexander (Eds.), Handbook of Multicultural Counseling. Thousand Oaks, CA: Sage Publications.

Cross, W. E. Jr., & Fhagen-Smith, P. (1996). Nigrescence and ego identity development: Accounting for differential Black identity patterns. In P. B. Pedersen & J. G. Draguns (Eds.), Counseling Across Cultures (4th Ed.). (pp. 108-123). Thousand Oaks, CA: Sage Publications.

Cross, W. E., Jr., & Strauss, L. (1998). The everyday functions of African American identity. In J. K. Swim & C. Stangor (Eds.), Prejudice: the Target's Perspective. (pp. 267-279). San Diego, CA: Academic Press.

Demo, D. H., & Hughes, M. (1990). Socialization and racial identity among black americans. Social Psychology Quarterly, 53(4), 364-374.

Duckitt, J., & Mphuthing, T. (1998). Group identification and intergroup attitudes: a longitudinal analysis in South Africa. Journal of Personality & Social Psychology, 74(1), 80-85.

Gurung, R. A. R., & Duong, T. (1999). Mixing and matching: assessing the concomitants of mixed-ethnic relationships. Journal of Social & Personal Relationships, 16(5), 639-657.

Hardin, C. D. & Conley, T. D. (2000). A relational approach to cognition: Shared experience and relationship affirmation in social cognition. In G. B. Moskowitz (Ed.) Cognitive social psychology: The Princeton Symposium on the Legacy and Future of Social Cognition (pp 3-17). Mahwah, NJ: Lawrence Erlbaum Associates.

Hardin, C. D., & Higgins, E. T. (1996). Shared reality: How social verification makes the subjective objective. In R. M. Sorrentino & Tory Higgins (Eds), Handbook of motivation and cognition, Vol. 3: The interpersonal context (pp. 28-84). New York: Guilford Press.

Heider, F. (1958). The psychology of interpersonal relations. Hillsdale, NJ: Lawrence Erlbaum Associates.

Helms, J. E. (1990). Black and White racial identity: Theory, research, and practice. New York, NY: Greenwood Press.

Helms, J. E. (1995). An update of helm's white and people of color racial identity models. In J. G. Ponterotto, J. M. Cases, L. A. Suzuki, & C. M. Alexander (Eds.), Handbook of Multicultural Counseling. Thousand Oaks, CA: Sage Publications.

Hendrick, S. S. (1988). A generic measure of relationship satisfaction. Journal of Marriage and the Family, 50, 93-98.

Imamura, A. E. (1986). Ordinary couples? Mate selection in international marriage in nigeria. Journal of Comparative Family Studies, 17(1), 33-42.

Jaret, C., & Reitzes, D. C. (1999). The importance of racial-ethnic identity and social setting for blacks, whites, and multiracials. Sociological Perspectives, 42(4), 711-737.

Jasinskaja-Lahti, I., & Liebkind, K. (1998). Content and predictors of the ethnic identification of Russian-speaking immigrant adolescents in Finland. Scandinavian Journal of Psychology, 39(4), 209-219.

Knight, G. P., Bernal, M. E., Garza, C. A., Cota, M. K., & Ocampo, K. A. (1993). Family socialization and the ethnic identity of Mexican American children. Journal of Cross-Cultural Psychology, 24(1), 99-114.

Kouri, K. M., & Lasswell, M. (1993). Blackhite marriages: social change and intergenerational mobility. Marriage & Family Review, 19(3-4), 241-255.

LaFromboise, T., Coleman, H. L. K., & Gerton, J. (1993). Psychological impact of biculturalism: Evidence and theory. Psychological Bulletin, 114, 395-412.

Leary, M. R., Tambor, E. S., Terdal, S. K., & Downs, D. L. (1995). Self-esteem as an interpersonal monitor: The sociometer hypothesis. Journal of Personality & Social Psychology, 68(3), 518-530.

Lee, S. M., & Fernandez, M. (1998). Trends in asian american racial/ethnic intermarriage: a comparison of 1980 and 1990 census data. Sociological Perspectives, 41(2), 323-342.

Lewis, R. J., Yancey, G., & Bletzer, S. S. (1997). Racial and nonracial factors that influence spouse choice in black/white marriages. Journal of Black Studies, 28(1), 60-78.

Luhtanen, R., & Crocker, J. (1992). A collective self-esteem scale: self-evaluation of one's social identity. Personality & Social Psychology Bulletin, 18(3), 302-318.

McGuire, W. J., McGuire, C. V., & Cheever, J. (1986). The self in society: effects of social contexts on the sense of self. British Journal of Social Psychology, 25(3), 259-270.

McGuire, W. J., McGuire, C. V., Child, P., & Fujioka, T. (1978). Salience of ethnicity in the spontaneous self-concept as a function of one's ethnic distinctiveness in the social environment. Journal of Personality & Social Psychology, 36(5), 511-520.

Murray, S. L., Holmes, J. G., & Griffin, D. W. (1996). The self-fulfilling nature of positive illusions in romantic relationships: love is not blind, but prescient. Journal of Personality & Social Psychology, 71(6), 1155-1180.

Nurius, P. S., & Markus, H. (1990). Situational variability in the self-concept: appraisals, expectancies, and asymmetries. Journal of Social & Clinical Psychology, 9(3), 316-333.

Oyserman, D., & Markus, H. R. (1998). Self as social representation. In E. Flick (Ed.) The Psychology of the Social (pp. 107-125). New York: Cambridge University Press.

Phinney, J.S. (1989). Stages of ethnic identity development in minority group adolescents. Journal of Early Adolescence, 9(1-2), 34-49.

Phinney, J.S. (1992). The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research, 7(2), 156-176.

Porterfield, E. (1982). Black-american intermarriage in the united states. Marriage & Family Review, 5(1), 17-34.

Quintana, S. M., & Vera, E. M. (1999). Mexican american children's ethnic identity, understanding of ethnic prejudice, and parental ethnic socialization. Hispanic Journal of Behavioral Sciences, 21(4), 387-404.

Quintana, S. M., Castaneda-English, P., & Ybarra, V. C. (1999). Role of perspective-taking abilities and ethnic socialization in development of adolescent ethnic identity. Journal of Research on Adolescence, 9(2), 161-184.

Rosenberg, M. (1965). Society and the adolescent self-image. Princeton, NJ: Princeton University Press.

Rosenthal, D. A., & Cichello, A. M. (1986). The meeting of two cultures: Ethnic identity and psychosocial adjustment of Italian-Australian adolescents. International Journal of Psychology, 21, 487-501.

Rosenthal, D. A., & Feldman, S. S. (1992). The nature and stability of ethnic identity in Chinese youth: Effects of length of residence in two cultural contexts. Journal of Cross-Cultural Psychology, 23(2), 214-227.

Sam, D. L. (1995). Acculturation attitudes among young immigrants as a function of perceived parental attitudes toward cultural change. Journal of Early Adolescence, 15(2), 238-258.

Sam, D. L. (2000). Psychological adaptation of adolescents with immigrant backgrounds. Journal of Social Psychology, 140(1), 5-25.

Sanders Thompson, V. L. (1999). Variables affecting racial-identity salience among african americans. Journal of Social Psychology, 139(6), 748-761.

Saylor, E. S., & Aries, E. (1999). Ethnic identity and change in social context. Journal of Social Psychology, 139(5), 549-566.

Shelton, J. N., & Sellers, R. M. (2000). Situational stability and variability in african american racial identity. Journal of Black Psychology, 26(1), 27-50.

Shibazaki, K., & Brennan, K. A. (1998). When birds of different feathers flock together: a preliminary comparison of intra-ethnic and inter-ethnic dating relationships. Journal of Social & Personal Relationships, 15(2), 248-256.

Smith, E. P., Walker, K., Fields, L., Brookins, C. C., & Seay, R. C. (1999). Ethnic identity and its relationship to self-esteem, perceived efficacy and prosocial attitudes in early adolescence. Journal of Adolescence, 22(6), 867-880.

Spanier, G. B. (1976). Measuring dyadic adjustment: New scales for assessing the quality of marriage and similar dyads. Journal of Marriage and the Family, 38(1), 15-28.

Sue, D. W. & Sue, D. (1990). Counseling the culturally different. New York: John Wiley.

Sung, B.L. (1990). Chinese American intermarriage. Journal of Comparative Family Studies, 21(3 ), 337-352.

Tajfel, H., & Turner, J. C. (1986). The social identity theory of intergroup behavior. S. Worchel, & W. G. Austin (Eds.), Psychology of Intergroup Relations . Chicago, IL: Nelson-Hall.

Thomas, A. J. (2000). Impact of racial identity on African American child-rearing beliefs. Journal of Black Psychology, 26(3), 317-329.

Tucker, M. B., & Mitchell-Kernan, C. (1990). New trends in Black American interracial marriage: The social structural context. Journal of Marriage and the Family, 52(2), 209-218.

Turner, J. (1987). Rediscovering the social group: A self-categorization theory. Oxford: Basil Blackwell.

Yip, T. & Fuligni, A. J. (in press). Daily variation in ethnic identity, ethnic behaviors, and psychological well-being among american adolescents of chinese descent. Child Development.

Appendix A

Collective Self Esteem measure for ethnic groups

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate your agreement or disagreement with each statement.

Membership subscale

1. I am a cooperative participant in the ethnic group I belong to.

2. I am a worthy member of the ethnic group I belong to.

3. I feel I don't have much to offer to the ethnic group I belong to.

4. I often feel I'm a useless member of my ethnic group.

Identity subscale

1. In general, belonging to my ethnic group is an important part of my self-image. 

2. Overall, my ethnic group has very little to do with how I feel about myself.

3. The ethnic group I belong to is an important reflection of who I am.

4. The ethnic group I belong to is not important to my sense of what kind of person I am

Private subscale

1. I feel good about the ethnic group I belong to.

2. I often regret that I belong to the ethnic group that I do.

3. In general, I'm glad to be a member of the ethnic group I belong to.

4. Overall, I often feel that the ethnic group of which I am a member is not worthwhile.

Public subscale

1. In general, others respect the ethnic group I am a member of.

2. In general, others think that my ethnic group is unworthy.

3. Most people consider my ethnic group, on the average, to be more ineffective than

4. Overall, my ethnic group is considered good by others.

Appendix B

Multigroup Ethnic Identity Measure

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate your agreement or disagreement with each statement.

Ethnic Identity

Affirmation & Belonging subscale

1. I am happy that I am a member of the group I belong to.

2. I have a strong sense of belonging to my own ethnic group.

3. I have a lot of pride in my ethnic group and its accomplishments.

4. I feel a strong attachment towards my own ethnic group.

5. I feel good about my cultural or ethnic background.

Ethnic Identity Achievement subscale

1. I have spent time trying to find out more about my own ethnic group, such as its history, traditions, and customs.

2. I have a clear sense of my ethnic background and what it means for me.

3. I think a lot about how my life will be affected by my ethnic group membership.

4. I am not very clear about the role of my ethnicity in my life.

5. I really have not spent much time trying to learn more about the culture and history of my ethnic group.

6. I understand pretty well what my ethnic group membership means to me, in terms of how to relate to my own group and other groups.

7. In order to learn more about my ethnic background, I have often talked to other people about my ethnic group.

Ethnic Behaviors subscale

1. I am active in organizations or social groups that include mostly members of my own ethnic group.

2. I participate in cultural practices of my own group, such as special food, music, or customs.

Other Group Orientation

1. I like meeting and getting to know people from ethnic groups other than my own.

2. I sometimes feel it would be better if different ethnic groups didn't try to mix together.

3. I often spend time with people from ethnic groups other than my own.

4. I don't try to become friends with people from other ethnic groups.

5. I am involved in activities with people from other ethnic groups.

6. I enjoy being around people from ethnic groups other than my own.

Appendix C

Rosenberg Self Esteem

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate your agreement or disagreement with each statement.

1. I feel that I am a person of worth, at least on an equal basis with others.

2. I feel that I have a number of good qualities.

3. All in all, I am inclined to feel that I am a failure.

4. I am able to do things as well as most other people.

5. I feel I do not have much to be proud of.

6. I take a positive attitude toward myself.

7. On the whole, I am satisfied with myself.

8. I wish I could have more respect for myself.

9. I certainly feel useless at times.

10. At times I think I am no good at all.

Appendix D

Collective Self Esteem measure for religious groups

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate your agreement or disagreement with each statement.

Membership subscale

1. I am a cooperative participant in the religious group I belong to.

2. I am a worthy member of the religious group I belong to.

3. I feel I don't have much to offer to the religious group I belong to.

4. I often feel I'm a useless member of my religious group.

Identity subscale

1. In general, belonging to my religious group is an important part of my self-image. 

2. Overall, my religious group has very little to do with how I feel about myself.

3. The religious group I belong to is an important reflection of who I am.

4. The religious group I belong to is not important to my sense of what kind of person I am

Private subscale

1. I feel good about the religious group I belong to.

2. I often regret that I belong to the religious group that I do.

3. In general, I'm glad to be a member of the religious group I belong to.

4. Overall, I often feel that the religious group of which I am a member is not worthwhile.

Public subscale

1. In general, others respect the religious group I am a member of.

2. In general, others think that my religious group is unworthy.

3. Most people consider my religious group, on the average, to be more ineffective than

4. Overall, my religious group is considered good by others.

Appendix E

Relationship Assessment Scale

The sentences below reflect different aspects of your relationship your spouse. Please use the scales below each question to answer that question.

1. How well does your spouse meet your needs?

Not well at all 1

2

3

Pretty well 4

5

6

Extremely well 7

2. In general how satisfied are you with your marriage?

Very un-satisfied 1

2

3

Pretty satisfied 4

5

6

Extremely satisfied 7

3. How good is your marriage compared to most other marriages?

Very poor 1

2

3

Pretty good 4

5

6

Extremely good 7

4. How often do you wish you had NOT gotten into this marriage?

Never 1

2

3

Sometimes 4

5

6

Often 7

5. To what extent has your marriage met your original expectations?

Not at all 1

2

3

Pretty much 4

5

6

Very much 7

6. How much do you love your spouse?

Not at all 1

2

3

Quite a lot 4

5

6

Extremely 7

7. How many problems are there in your marriage?

None 1

2

3

Some 4

5

6

Many 7

Appendix F

PAIR Inventory

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate how much you agree or disagree with each statement about your relationship with your spouse.

Emotional intimacy

1. My spouse listens to me when I need someone to talk to.

2. I can state my feelings without him getting offensive.

3. I often feel distant from my spouse.

4. My spouse can really understand my hurts and joys.

5. I feel neglected at times by my spouse.

6. I sometimes feel lonely when we're together.

Intellectual intimacy

1. My spouse helps me clarify my thoughts.

2. When it comes to having a serious discussion, it seems that we have little in common.

3. I feel “put-down” in a serious conversation with my spouse.

4. I feel it is useless to discuss some things with my spouse.

5. My spouse frequently tries to change my ideas.

6. We have an endless number of things to talk about.

Appendix G

Including Other in the Self

Please select the number associated with the picture below that best represents your relationship with your spouse.

Appendix H

Dyadic Adjustment Scale

The following topics are some things that married couples frequently discuss. Using the scale below, please rate how much you and your spouse agree on these things.

Never agree Sometimes agree Always agree

1 2 3 4 5 6 7

1. Philosophy of life (Aims, goals, things believed important in life)

2. Friends

3. Right, good, or proper conduct

4. Matters of recreation

5. Ways of dealing with in-laws/each other's parents

6. Demonstrations of affection

7. Religious matters

8. Amount of time spent together

9. Major decision-making

10. Leisure time interests and activities

Use the scale below to answer the next set of questions about your relationship with your spouse.

All Most Of More Often Occasionally Rarely Never

The Time The Time Than Not

1 2 3 4 5 6

1. How often do you discuss or have you considered divorce, separation, or termination of your marriage?

2. How often do you think that things between you and your spouse are going well?

3. How often do you confide in your spouse?

4. How often do you regret that you married?

5. How often do you and your spouse quarrel?

6. How often do you and your spouse get on each others' nerves?

Please circle the number that best describes the degree of happiness you feel in your relationship with your spouse. The middle point, 4 “happy”, represents the average degree of happiness of most people's relationships with their spouse.

Extremely Fairly A Little Happy Very Extremely Perfectly

Unhappy Unhappy Unhappy Happy Happy Happy

1 2 3 4 5 6 7

Appendix H (cont.)

Which of the following statements best describes how you feel about the future of your marriage? Fill in only one circle.

O	I want desperately for my relationship to succeed, and would go to almost any length to see that it does.
O	I want very much for my relationship to succeed, and will do all I can to see that it does.
O	I want very much for my relationship to succeed, and will do my fair share to see that it does.
O	It would be nice if my relationship succeeded, but I can't do much more than I am doing now to keep the relationship going.
O	It would be nice if it succeeded, but I refuse to do any more than I am doing now to keep the relationship going.
O	My relationship can never succeed, and there is no more that I can do to keep the relationship going.

Appendix I

Use the scale below to indicate how much you agree or disagree with each statement.

Strongly Strongly

disagree agree

1 2 3 4 5 6 7

Shared Understanding of Ethnicity with Spouse

1. My spouse thinks that my ethnic group's traditions are important.

2. My spouse listens but does not add much to the conversation when I tell her/him about my experiences as a member of my ethnic group.

3. My spouse is able to relate to the experiences I have had as a member of my ethnic group.

4. My spouse sometimes makes hurtful jokes about my ethnicity.

5. My spouse doesn't always understand me when I talk about my feelings about my ethnic background.

6. My spouse is always willing to do with me the things that are important to people of my ethnic background.

7. My spouse understands a lot about my ethnic group.

Shared Understanding of Religion with Spouse

1. My spouse thinks that my religious traditions are important.

2. My spouse doesn't always understand when I talk about my religious experiences.

3. My spouse is always willing to do with me the things that are of religious importance to me.

4. My spouse understands a lot about my religion.

Appendix J

Use the scale below to indicate how much you agree or disagree with each statement.

Strongly Strongly

disagree agree

1 2 3 4 5 6 7

Shared Understanding of Ethnicity with Mother

1. My mother shares with me the physical appearance of being from the same ethnic background.

2. Most of the things I know about my ethnic group I learned from my mother.

3. It is difficult for me to explain to my mother my feelings about my ethnic group.

4. My mother enjoys doing those things that are also important to other people of my ethnic background.

5. My mother usually understands when I tell her my feelings about being a member of my ethnic group

6. My mother and I don't see eye-to-eye on how I should be connected to my ethnic group.

7. My mother thinks I don't have a great enough appreciation for my ethnic group.

Shared Understanding of Religion with Mother

1. My mother thinks that my religious traditions are important.

2. My mother doesn't always understand when I talk about my religious experiences.

3. My mother is always willing to do with me the things that are of religious importance to me.

4. My mother understands a lot about my religion.

Appendix K

Conventionality

Strongly disagree 1

Disagree 2

Somewhat Disagree 3

Neither agree nor disagree 4

Somewhat agree 5

Agree 6

Strongly agree 7

Please use the scale above to indicate how much you agree or disagree with each statement about your relationship with your spouse.

1. My spouse has all of the qualities I've always wanted in a mate.

2. There are times when I do not feel a great deal of love and affection for my spouse.

3. Every new thing I have learned about my spouse has pleased me.

4. My spouse and I understand each other completely.

5. I don't think anyone could possibly be happier than my spouse and I when we are with one another.

6. I have some needs that are not being met by my relationship.Table 1

Ethnic Identification as a Function of Ethnicity of Interaction Partner and Quality of Relationship

with Interaction Partner

Ethnicity of Interaction Partner Quality of Relationship

(in relation to self) Positive Negative

Same-Ethnicity High Ethnic Lower Ethnic

Identification Identification

Different-Ethnicity Lower Ethnic High Ethnic

Identification Identification

Table 2

Cross-Tabulation of Participant Ethnicity and Spouse Ethnicity

Participant Ethnicity

Spouse Ethnicity Chinese Japanese Korean

Chinese 45 4 5

Japanese 4 10 1

Korean 1 2 11

Filipino 1 - -

½ same Asian

½ diff Asian 1 1 -

½ same Asian

½ White 2 1 -

White 13 1 5

Black 1 - -

Latino 1 1 -

Native American 1 - -

Table 3

Contrast Weights for the Predicted Effect of Type of Marriage and Person for Whom Relationship Motivation was Primed on Ethnic Identification

Relationship Motivation is Primed for

Type of Marriage Mother Spouse

Same-Ethnicity High Ethnic High Ethnic

Identification Identification

(+1) (+1)

Mixed-Ethnicity High Ethnic Low Ethnic

Identification Identification

(+1) (-3)

Table 4

Mean Ethnic Identification Scores on the Collective Self Esteem (CSE) Subscales as a Function of Type of Marriage and Prime

Same-ethnicity Marriage Mixed-ethnicity Marriage

Spouse Mother Spouse Mother

CSE Subscale Prime Prime Prime Prime

Membership 5.72_a 5.50_b 5.28 5.57

Private 6.21 6.40 6.04 5.90

Public 5.95_a 5.93 5.74 5.44_b

Identity 5.44 5.20 5.07 5.04

Note: Means with different subscripts were significantly different from one another in independent samples t-test comparisons. Comparisons were only made within each row.Table 5

Mean Ethnic Identification Scores as a Function of Participant Ethnicity, Type of Marriage, and Prime

Same-ethnicity Marriage Mixed-ethnicity Marriage

Spouse Mother Spouse Mother

Prime Prime Prime Prime

Chinese Americans (n = 73)

MEIM 5.34_a 5.28_a 4.58_b 5.36_a,b

Affirmation & Belonging 5.77_a 5.78_a 5.06_b 5.63_a,b

Ethnic Identity Achievement 5.07_a 5.01_a 4.30_b 5.23_a

Membership Subscale 5.71_a,b 5.50_a,b 5.16_b 5.79_a

Korean Americans (n = 22)

MEIM 5.70_a 5.44_a,b 4.61_b 5.21_a,b

Ethnic Identity Achievement 5.59_a 5.32_a,b 4.33_b 5.23_a,b

Note: All means reported here were part of a statistically significant contrast for the predicted pattern that women in mixed-ethnicity marriages primed with their different-ethnicity spouse report lower ethnic identification scores than women in same-ethnicity marriages primed with their same-ethnicity spouse, women in mixed-ethnicity marriages primed with their mother, and women in same-ethnicity marriages primed with their mother. Means with different subscripts were significantly different from one another in independent samples t-test comparisons. Comparisons were only made within each row.

Figure 1. Predicted outcomes on ethnic and religious identification when people are thinking of


their mother.

































Figure 2. Predicted outcomes on ethnic and religious identification when people are thinking of



their spouse.Figure 3. Predicted pattern for ethnic identification as a function of prime and type of marriage.

Figure 4. Predicted pattern for religious identification as a function of prime and type of marriage.

Figure 5. Self Esteem as a Function of Prime and Type of Marriage

Figure 6. Multigroup Ethnic Identity Measure (MEIM) as a Function of Prime and Type of Marriage

Figure 7. Ethnic Identity Achievement as a Function of Prime and Type of Marriage

Figure 8. Affirmation and Belonging as a Function of Prime and Type of Marriage

Figure 9. Mediational Analysis of Effect of Type of Marriage on the Multigroup Ethnic Identity Measure

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.Figure 8. Mediational Analysis of Effect of Type of Marriage on the Affirmation and Belonging Subscale of the Multigroup Ethnic Identity Measure

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

Figure 11. Mediational Analysis of Effect of Type of Marriage on the Private Subscale of the Collective Self Esteem Measure

All numbers are beta coefficients.
( ) indicates


the beta coefficient when the mediator is not included in the regression.

Figure 12. Mediational Analysis of Effect of Type of Marriage on the Public Subscale of the Collective Self Esteem Measure

All numbers are beta coefficients.
( ) indicates


the beta coefficient when the mediator is not included in the regression.

Figure 13. Mediational Analysis of Effect of Prime on the MEIM Scores of Women in Mixed-Ethnicity Marriages

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.Figure 14. Mediational Analysis of Effect of Prime on the Ethnic Identity Achievement Scores of Women in Mixed-Ethnicity Marriages

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.Figure 15. Mediational Analysis of Effect of Type of Marriage on the Membership Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

Figure 16. Mediational Analysis of Effect of Type of Marriage on the Membership Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.Figure 17. Mediational Analysis of Effect of Type of Marriage on the Identity Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

Figure 18. Mediational Analysis of Effect of Type of Marriage on the Identity Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

Figure 19. Mediational Analysis of Effect of Type of Marriage on the Private Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

Figure 20. Mediational Analysis of Effect of Type of Marriage on the Private Subscale for Religious Identification

All numbers are beta coefficients.
( ) indicates



the beta coefficient when the mediator is not included in the regression.

The author used email to ask her friends to participate in the study if they met the eligibility criteria and to forward the email announcement about the study to their friends who might qualify for the study. In addition, flyers were also handed out at several Asian American churches and an announcement was posted to the Princeton alumni listserve.

Independent samples t-test revealed no significant differences on age and marriage length between participants from same-ethnicity marriages and participants from mixed-ethnicity marriages.

Chi-square analysis revealed no significant differences in place of birth, number of children, and place of residence between participants from same-ethnicity marriages and participants from mixed-ethnicity marriages.

Alpha coefficients are reported separately for people in same- and mixed- ethnicity marriages to show that measures were reliable for both groups. The first number is the reliability for people in same-ethnicity marriages, and the second number is the reliability for people in mixed-ethnicity marriages.

Shared reality theory did not predict a crossover interaction between the priming manipulation and the type of marriage. There was, however, a significant crossover interaction for the Ethnic Identity Achievement subscale, F(1,108) = 4.68, p < .05, Cohen's d = .41 and a marginal interaction for the overall MEIM, p = .06.

Participants were instructed to label themselves Spiritual if they believed in God but did not adhere to any particular religion.

Mixed-Ethnicity Marriage and Ethnic Identity 47

6

7

5

4

2

3

1

You

Your

Partner

You

Your

Partner

You

Your

Partner

You

Your

Partner

You

Your

Partner

You

Your

Partner

You

Your

Partner

Self's

Religion

Self's

Religion

Self's

Ethnicity

Self's

Ethnicity

Spouse

Mother

Self

Same as Self

Self's

Religion

Self's

Religion

Self's

Ethnicity

Self's

Ethnicity

Spouse

Mother

Self

High ethnic identification because Mother is same ethnicity

High religious identification if Mother is same religion; Low if Mother is different religion

May be same or different from Self

Self's

Religion

Self's

Religion

Self's

Ethnicity

Self's

Ethnicity

Spouse

Mother

Self

Self's

Religion

Self's

Religion

Self's

Ethnicity

Self's

Ethnicity

Spouse

Mother

Self

High religious identification if Spouse is same religion; Low if Spouse is different religion

High ethnic identification if Spouse is same ethnicity; Low if Spouse is different ethnicity

May be same or different from Self

May be same or different from Self

Shared Understanding of Ethnicity with Spouse

.61*

-.50*

(-.25*)

.05, ns

Multigroup Ethnic Identity Measure

Type of Marriage

Shared Understanding of Ethnicity with Spouse

.41*

-.50*

(-.25*)

-.04, ns

Affirmation and Belonging

Type of Marriage

Shared Understanding of Ethnicity with Spouse

.23*

-.50*

Private Subscale

(-.19*)

-.07, ns

Type of Marriage

Shared Understanding of Ethnicity with Spouse

.21*

-.50*

Public Subscale

(-.20*)

-.10, ns

Type of Marriage

Shared Understanding of Ethnicity with Mother

.39*

.33*

(.35*)

.22, ns

Multigroup Ethnic Identity Measure

Prime

Shared Understanding of Ethnicity with Mother

.40*

.33*

(.41*)

.27*

Ethnic Identity Achievement

Prime

Shared Understanding of Religion with Spouse

.39*

-.72*

(-.49*)

-.21, ns

Membership Subscale

Type of Marriage

Perceived Importance of Religion to Spouse

.30*

-.65*

(-.49*)

-.34*

Membership Subscale

Type of Marriage

Shared Understanding of Religion with Spouse

.64*

-.72*

(-.46*)

.002, ns

Identity Subscale

Type of Marriage

Perceived Importance of Religion to Spouse

.58*

-.65*

(-.46*)

-.05, ns

Identity Subscale

Type of Marriage

Shared Understanding of Religion with Spouse

.49*

-.72*

(-.32*)

.03, ns

Private Subscale

Type of Marriage

Perceived Importance of Religion to Spouse

.30*

-.65*

(-.32*)

-.02, ns

Private Subscale

Type of Marriage

---