intercultural friendship

Journal of Intercultural Communication Research Vol. 38, No. 2, July 2009, pp. 77–98 Influences of Culture on Self-Disclosure as Relationally Situated in Intercultural and Interracial Friendships from a Social Penetration Perspective Yea-Wen Chen & Masato Nakazawa Intercultural and interracial relationships face barriers, tensions, and challenges that are absent from intracultural and intraracial relationships. These challenges provide the impetus for this study to examine the influences of individualism-collectivism and relational intimacy on topics and dimensions of self-disclosure in intercultural/ interracial friendships from a social penetration perspective. A total of 252 participants responded to the instruments measuring the constructs of interest. This study found:

(1) relational intimacy was positively correlated with all six topics and four out of the five dimensions of self-disclosure; (2) individualism was a significant predictor of the five dimensions of self-disclosure as a set; and (3) one mirrors one’s intercultural/interracial friend in all six topics and the positive/negative dimension of self-disclosure. These results suggest that relational intimacy has a greater impact on close intercultural and interracial friendships than cultural variability.

Keywords: Self-disclosure; Social Penetration; Intercultural and Interracial Friendships; Relational Intimacy; Cultural Variability Introduction Communication scholars have paid limited attention to the increasingly impor- tant phenomenon of intercultural friendship (Chen, 2002; Gareis, 1999; Kudo & Yea-Wen Chen, (MA, University of North Texas, 2006) is a doctoral candidate at the University of New Mexico, Department of Communication and Journalism. Masato Nakazawa (MS, University of New Mexico, 2006) is a doctoral candidate at the University of New Mexico, Department of Psychology. Correspondence to Yea-Wen Chen, Department of Communication and Journalism, 1 University of New Mexico, MSC03 2240, Albuquerque, NM 87131-0001, USA. Fax: 505-277-4206. Tel: 505-277-2100. Email: [email protected] ISSN 1747-5759 (print)/ISSN 1747-5767 (online) 2009 World Communication Association DOI: 10.1080/17475750903395408 Simkin, 2003; Lee, 2006; Morgan & Arasaratnam, 2003). Morgan and Arasaratnam (2003) argue that ‘‘intercultural friendship is an area of research that is still in its early stages’’ (p. 176). As norms and ideas about what constitutes friendship are learned within racial, ethnic, and national group contexts, individuals learn early on to make in-group and out-group distinctions (Tajfel, 1978). Not surprisingly, intercultural and interracial relationships face barriers, tensions, and challenges that are absent from intracultural and intraracial relationships (e.g., Chen, 2002; Orbe & Harris, 2008). Studies suggest that cultural and racial/ethnic differences hinder the initiation and development of intercultural and interracial relationships (Diggs & Clark, 2002; Gareis, 1995, 2000; Kudo & Simkin, 2003). Kudo and Simkin (2003) conclude ‘‘many studies have shown a lack of contact and relationship development between international and host national students’’ (p. 92). The rarity and challenges of developing friendships across cultural boundaries provide the impetus for the present study to examine the communication process of developing intercultural and interracial friendships.

One of the key features of friendship formation and maintenance is self-disclosure (e.g., Matsushima & Shiomi, 2002; Samter, 2003). Self-disclosure has been identified as a critical indicator of intercultural friendship formation, third only to frequency of contact and similarity (Kudo & Simkin, 2003). Gareis (1999) contends that cultured self-disclosure practices are powerful factors influencing intercultural friendship formation. Additionally, Lee (2006) identifies self-disclosure as one of the seven strategies in the construction of relational identity in intercultural friendship.

Theoretically, Altman and Taylor’s (1973) social penetration theory provides the most commonly cited linkage between self-disclosure and relationship development.

However, the social penetrative process of self-disclosing in the development of intercultural and interracial relationships remains little researched (Chen, 2002; Gudykunst, 1985a; Hammer & Gudykunst, 1987).

Cross-cultural studies confirm that self-disclosure events differ across cultural groups such as between individuals from individualistic countries in the West and collectivistic countries in Asia (e.g., Barnlund, 1989; Chen, 1995; Gudykunst & Nishida, 1986; Lustig & Koester, 2006; Kito, 2005; Won-Doornink, 1985). Self- disclosure commonly refers to the ‘‘process of making the self known to other persons’’ (Jourard & Lasakow, 1958, p. 91). Intraculturally, Tardy and Dindia (2006) summarize that self-disclosure is often used strategically to regulate the development of a relationship and conclude that variables facilitating self-disclosure in personal relationships include gender, reciprocity, liking or affection, trust, and requests for disclosures. In comparison, although cross-cultural studies have been successful in demonstrating how depth, amount, breadth, valence, timing, targets, and social values regarding self-disclosure differ across cultures (Barry, 2003; Chen, 1995; Fitzpatrick et al., 2006; Lustig & Koester, 2006), little has been done to explore what variables facilitate or hinder self-disclosure in intercultural friendships.

Therefore, this study seeks to examine what factors facilitate or hinder self- disclosure in the development of intercultural and interracial friendships from a social penetration perspective. Specifically, this study investigates the influences 78Y.-W. Chen & M. Nakazawa of culture and relational intimacy on self-disclosure behaviors in intercultural and interracial friendships in the United States. First, the focus on both intercultural and interracial friendship has theoretical implications for understanding self- disclosure as interactionally/transactionally contextual and contingent in the larger processes of self and relationship development (Antaki, Barnes, & Leudar, 2005; Dindia, 1997). Also, such focus has implications for expanding social penetration theory to examine intercultural and interracial relationships. Second, the research proffered in this paper will answer the calls to build on the existing limited body of research on intercultural and interracial friendship (Diggs & Clark, 2002; Gudykunst, 1985a; Kudo & Simkin, 2003). Finally, the focus on intercultural and interracial friendships has social and practical implications for improving race relations (Aberson, Shoemaker, & Tomolillo, 2004; Orbe & Harris, 2001) and also for smoothing adaptation problems facing international students (Kudo & Simkin, 2003; Olaniran, 1996; Rohrlick & Martin, as cited in Gareis, 2000; Zimmermann, 1995).

Literature Review In the following sections, first key literature will be presented on intercultural and interracial friendship. Then, the guiding theoretical framework of social penetration theory will be discussed. Finally, self-disclosure research relevant to this study will be presented: (a) self-disclosure topics and dimensions; (b) self-disclosure and culture (i.e., individualism-collectivism); and (c) self-disclosure as dyadic and interactional.

Intercultural and Interracial Friendship Friendship is defined in this study as ‘‘a relationship involving voluntary interdependence in which the individuals respond to one another personalistically’’ (Wright, 1978, p. 199). However, what the termfriendmeans is culturally constructed and situated. Collier (1996) confirms that what characterizes friendship differs across ethnic groups in the United States: (a) Latinos stress relational support; (b) Asian-Americans emphasize positive exchanges of ideas; (c) African-Americans place emphasis on respect and acceptance; and (d) European-Americans value recognition of individual needs. Gareis (2000) found that both the different category width for the word ‘‘friend’’ and the different extent of public and private personality layers in German and U.S. American cultures caused confusion and misunderstand- ings in friendships between German students and the host U.S. students. Since friendships are enabled and constrained by cultures in terms of both nationality and race, research on intercultural and interracial friendship has identified the following communication issues: (a) identity negotiation and construction (Diggs & Clark, 2002; Lee, 2006); (b) cultural and communication competencies (Collier, 1996); (c) perceived similarities or commonalities (Gareis, 1999; Gudykunst, 1985a; Kudo & Simkin, 2003); (d) scope and depth of topics appropriate for discussion (Cahn, 1984; Gareis, 1999); (e) self-disclosure (Gareis, 1999; Kudo & Simkin, 2003); and (f) language proficiency (Gareis, 1999; Kudo & Simkin, 2003). Journal of Intercultural Communication Research79 Nonetheless, understandings of how cultures influence communication in inter- cultural and interethnic friendships are inconclusive.

On one hand, we concur with Ting-Toomey (2005) that culture essentially is ‘‘a learned system of meaning that fosters shared identity and community among its group members’’ (pp. 71–72). In this sense, culture means patterned values, norms, attitudes, beliefs, behaviors, and identities shared by members of the same group, even though it is not assumed that culture as a shared system is homogeneously distributed among group members, as Avruch (1998) argued.

As race and nationality are both types of group membership that delineates shared norms, values, and beliefs associated with friendships, culture in this study thus broadly encompasses both race and country of origin. Moreover, tensions and challenges in (a) friendships between individuals with perceived salient national differences and (b) friendships between individuals with perceived salient racial differences seem to mirror each other in terms of different cultural values and differently patterned communication processes. On the other hand, Collier (1996) argued that generalization from international/intercultural competence literature ‘‘may be of limited benefit in understanding interethnic friendships’’ (p. 334).

While it is acknowledged that specific sets of skills required to successfully develop and maintain friendship may differ between interracial friendships and intercultural friendships, this study attempts to examine the influence of culture as a broad and widely used construct, such as individualism-collectivism, on self-disclosure and does not intend to equate intercultural friendships with interracial friendships.

From a developmental perspective, differences in cultures are secondary in established intercultural friendships where personal differences are more significant than cultural differences (Gareis, 1999; Gudykunst, 1985a; Gudykunst, Nishida, & Chua, as cited in Gareis, 1999). However, it is yet to be examined in what way(s) personal differences weigh more than cultural differences in established friendships and when cultural differences do matter in intercultural and interethnic friendships.

Thus, research is needed to examine how cultures affect communication in the process of developing intercultural and interracial friendships.

Social Penetration Theory The theoretical framework guiding the present study is Altman and Taylor’s (1973) social penetration theory since it provides the linkage between self-disclosure and friendship development. The goals underlying Altman and Taylor’s (1973) develop- ment of social penetration theory were both to describe the course of growth or dissolution of interpersonal relationships and to address the issue of how people enact mutual exploration and formation of social bonds. In essence, Altman and Taylor (1973) explicated the roles of self-disclosure, intimacy, and communication in the development of interpersonal relationships. The theorists state that the basis of relationship development is levels of self-disclosure that progress gradually and orderly from peripheral layers to central layers of personality. Altman and Taylor (1973) conceptualize the term ‘‘social penetration’’ to refer to both overt 80Y.-W. Chen & M. Nakazawa interpersonal behaviors that take place in social interaction and internal subjective processes that precede, accompany, and follow overt exchange. Theoretically, social penetration theory assumes that (a) the social penetration process is orderly and progresses through stages over time; (b) the advancement of relationships is dependent on the amount and nature of the interpersonal rewards and costs that people possess the ability to assess; and (c) development (penetration) and dissolution (depenetration) follow the same principles of orderly progress and movement from either superficial to intimate or intimate to superficial areas of disclosure. Additionally, the foundation of social penetration theory lies in the assumption that making the self known or accessible to others through self-disclosure is intrinsically gratifying, and this, in turn, leads to the development of positive feelings for the others (Taylor & Altman, 1987). However, it has not been examined if the assumed gratification of self-disclosure is applicable to different cultures; thus, it warrants examination of the social penetration process in cross-cultural and intercultural relationships such as friendships.

Social penetration theory deconstructs and organizes personality structure into two general dimensions. The first is an area-based dimension of breath with not only breadth category but also breadth frequency. The second is a central-peripheral dimension of depth that starts at peripheral layers associated with biographical characteristics and moves toward central layers associated with less observable and more idiosyncratic characteristics. Additionally, social penetration theory emphasizes the views of interpersonal relationships from multiple behavioral perspectives such as perceptual, cognitive, verbal, nonverbal, and environmental levels of functioning or responses. In essence, the overlapping dimensions of verbal, nonverbal, and environmental behaviors result in eight generic dimensions of the social penetration process: richness-breadth of interaction; uniqueness of interaction; efficiency of exchange; substitutability and equivalency; synchronization and pacing; permeability and openness; voluntariness and spontaneity of exchange; and evaluation.

The empirical basis of Altman and Taylor’s (1973) social penetration theory is their ‘‘observations of social relationships in modern urban and suburban America’’ (p. 8). So, we do not yet know the extent to which principles of social penetration theory are applicable to the development of intercultural and interracial relationships except that the social penetration patterns in close intracultural and intercultural relationships are parallel (Gudykunst, 1985a). Thus, there is a need to examine how social penetration patterns occur in intercultural and interracial friendships.

Self-disclosure Topics and Dimensions Self-disclosure, the process of revealing personal information about oneself to another, is one of the most important factors in the development of intimate intercultural friendships (Barnlund, 1989; Gareis, 1999; Kudo & Simkin, 2003).

Of great importance to this study, Chen (1995) argues that ‘‘the comparative study of self-disclosure patterns among different cultures has gradually gained popularity in the field of intercultural communication’’ (p. 84). Embedded in this increasing Journal of Intercultural Communication Research81 popularity of cross-cultural comparisons in self-disclosure patterns is a fundamental assumption for this study: self-disclosure concepts mean the same thing across all cultures. In other words, self-disclosure patterns may differ across cultures, but the concept of self-disclosing is equivalent across cultures. For the purpose of this study, self-disclosure is operationalized in terms of (a) various topics of self-disclosure; and (b) various dimensions of self-disclosure such as intention, depth, amount, positivism-negativity, and honesty-accuracy. From the social penetration perspective, this means that, as relationships develop, relational partners will increase the breadth and depth of the personal information being volunteered.

Topics of self-disclosure With regard to topics of self-disclosure, one of the landmark studies is Jourard and Lasakow’s (1958) research on topic-based factors in self-disclosure. Their contribu- tion is the widely used Jourard–Lasakow Self-Disclosure questionnaire (SD-60), comprising 60 self-report items that ask respondents to report their disclosure to a specific person under six general subject areas: (a) attitudes and opinions; (b) tastes and interests; (c) work or studies; (d) money; (e) personality; and (f) body. Tardy (1988) confirms the validity of the SD-60 scale for measuring past disclosure to a specific targeted person.

Cross-cultural studies have found cultural differences in topics of self-disclosure (Cahn, 1984; Gudykunst and Nishida, 1983). Comparing across 10 topics of conversation among close same-sex Japanese and U.S. American friendships, Gudykunst and Nishida (1983) found that, in terms of the frequency of self- disclosure topics, the U.S. Americans had significantly higher ratings on four topics (relationship with others; love/dating and sex; interests/hobbies; and attitudes/ values), whereas the Japanese rated higher on one topic (physical condition).

In terms of intimacy ratings assigned to the topics, the U.S. Americans rated higher on three topics (own marriage and family; love/dating and sex; and emotions and feelings), whereas the Japanese rated higher on five topics (interests/hobbies; school/work; biographical information; religion; and money/property). Cahn (1984) found that the U.S. Americans discussed more intimate topics while Japanese discussed more superficial topics, and concluded it was of great importance to respect cultural differences in the depth and scope of conversational topics when establishing intercultural relationships. Thus, even though it is unknown whether these cultural differences found between Japanese and U.S. Americans can be generalizable to other cultures, the findings mentioned above support that members of different cultures do differ in patterns of self-disclosure.

Despite different cultural preferences in self-disclosure, the social penetration perspective would postulate that depth and frequency of topics of self-disclosure would be greater in friends with greater relational intimacy. Thus, the following hypothesis is proffered: H1: Levels of relational intimacy in intercultural and interracial friendships are positively correlated with depth and frequency of topics of self-disclosure. 82Y.-W. Chen & M. Nakazawa Following Emmers and Dindia’s (1995) conceptualization of relation intimacy as ‘‘how close one feels to one’s partner’’ (p. 230), relational intimacy in this study is defined as the closeness one feels and/or enacts toward one’s friend.

Dimensions of self-disclosure The concept of self-disclosure is also multidimensional, making it one of the most widely explored and extensively studied research areas in the communication discipline. In addition to Jourard and Lasakow’s (1958) work, another benchmark study on self-disclosure is Wheeless and Grotz’s Revised Self-disclosure Scale (RSDS), which comprises 31 items to reflect 5 major dimensions of self-disclosure (Wheeless, 1978): (a) intended disclosure; (b) amount of disclosure; (c) positive/negative disclosure; (d) control of depth in disclosure; and (e) honesty and accuracy in disclosure. As to the association between attributions and self-disclosure, Derlega, Winstead, Wong, & Greenspan’s (1987) study of experimental role-playing supported the belief that positive attributions and interest in initiating new relationships with strangers would foster and increase self-disclosing behaviors.

In this study, the social penetration perspective would postulate that dimensions of self-disclosure increase as relational intimacy increases. Thus, the following hypotheses are proffered: H2a: As the levels of relational intimacy in intercultural and interracial friendships increase, the intercultural and interracial friends have greater intent to disclose, disclose in greater amount and depth, and engage in more negative and honest-accurate disclosure.

H 2b: Intercultural and interracial friends with greater length of friendships have greater relational intimacy, resulting in stronger relationships among relational intimacy, self-disclosure, and culture (i.e., individualism- collectivism). Considering the association between self-disclosure and culture, Wheeless, Erickson, & Behrens’ (1986) examination of 411 U.S. American and international students revealed that students from non-Western cultures were correlated to greater depth of disclosiveness while studies from Western cultures were correlated to greater amount. Chen (2006) confirms that international students from East-Asian collectivistic cultures such as China, Japan, Korea, and Taiwan self-disclose in greater depth but in lesser amount in their intercultural friendships with U.S.

Americans. Kito’s (2005) study reveals that Japanese college students engage in lower levels of self-disclosure than their U.S. American counterparts in both romantic relationships and friendships.

Individualism-collectivism (I-C) For the purpose of this study, culture as a shared and learned system (Ting-Toomey, 2005) is operationalized in terms of collectivism-individualism (I-C) applied to cultural groups within national and ethnic or racial boundaries. Also, the individualism-collectivism dimension of culture is often used to understand the Journal of Intercultural Communication Research83 cross-cultural differences in self-disclosure between different nations as well as different groups in the United States (e.g., Chen, 1995; Fitzpatrick et al., 2006). 1 Although theorizing about culture at the national level has been critiqued as problematically overgeneralizing and essentializing, I-C remains one of the most used constructs to measure culture and is widely discussed in many contexts in the social sciences (e.g., Cai, Wilson, & Drake, 2000; Gudykunst et al., 1996; Hui & Yee, 1994; Triandis & Gelfand, 1998).

Conceptually, Hofstede’s (1991) seminal work postulates I-C as a conglomeration of values concerning the relationship betweenselfandothersand centers on the degrees of social connectedness among individuals (Azevedo, Drost, & Mullen, 2002; Earley, 1998). Individualism is defined by an emphasis on the separation of the self from others and draws upon self-regulated cognitions that privilege personal goals, self-reliance, autonomy, privacy, competitiveness, and aggressive creativity (Azevedo et al., 2002). Collectivistic cultures, on the other hand, are societies in which individuals think of themselves as parts of their collective groups: they place the goals of the group above their personal goals; they privilege obedience, duty, loyalty, obligation, respectfulness, hierarchy, and mutual dependence; and they distinguish between in-groups and out-groups (Triandis, 1995).

In terms of communication, cultural I-C has both direct and indirect effects on communication behaviors (Gudykunst et al., 1996). Gudykunst et al. (1996) summarized that members of individualistic cultures were found to be more affect oriented and more inclined to talk, and valued clarity and direct requests more than members of collectivistic cultures. For the purpose of this study, I-C is treated as a psychological variable measured within culture at the individual level, as Hui and Yee (1994) exemplified in their study. Based on the finding that members of individualistic cultures are more inclined to talk than members of collectivistic cultures, the following hypothesis is proposed. H3: The more individualistic a person is, the more s/he would self-disclose. Also, considering that I-C has a direct effect on communication behaviors such as self-disclosure and that self-disclosure is a construct with multiple dimensions, the following research question is proposed, as little is known about the direct effect of I-C on self-disclosure: RQ1: How does cultural variability of individualism-collectivism (I-C) influence dimensions of self-disclosure in intercultural and interracial friendships? Self-disclosure as Dyadic and Interactional Self-disclosure is believed to be a dyadic and dialectical process and a situated interactional practice that changes as individuals and relationships develop (Antaki et al., 2005; Dindia, 1997; Gudykunst, 1985b; Jourard, as cited in Won-Doornink, 1985). For instance, Gudykunst’s (1985b) study on 400 university students in the United States concluded that self-disclosure was influenced by (a) self-monitoring; (b) the degree of cultural similarity; and (c) the type of relationships.

84Y.-W. Chen & M. Nakazawa More importantly, the study concluded that the only differences between culturally similar and dissimilar friends were their attributional confidence and shared commu- nication networks. Moreover, as a dyadic and dialectical process and an interactional practice, self-disclosure was found to be guided by the norm of reciprocity (Pearce, Sharp, Wright, & Slama, 1974; Won-Doornink, 1985). Based on this body of research, the norm of reciprocity would predict that one’s self-disclosure increases with one’s relational partner’s disclosure. Thus, the following hypothesis is postulated: H4: As dyadic process, one’s self-disclosure behaviors mirror those of one’s intercultural or interracial friend’s. Methods Procedures An approval from the Institutional Review Board (IRB) at a large southwestern university was obtained prior to participant recruitment and data collection. Prior to their voluntary participation, the participants were informed of the topic and purpose of the study, of their right not to answer any statements, and of their right to withdraw from the study at any point. The questionnaire asked participants to select an interracial or intercultural friend who was deemed as more than an acquaintance.

In this study, intercultural friendship was defined for the participants as friendship between individuals from different countries of origin, while interracial friendship was defined as friendship between individuals from different racial/ethnic backgrounds in the United States. The participants were asked to respond to a series of items about the selected intercultural or interracial friendship. The questionnaire was laid out in the following format: (a) relational/social intimacy items; (b) topics of self-disclosure items measuring one’s own choices with topics of disclosure; (c) repeated topics of self-disclosure items measuring one’s perception of one’s friend’s choices with topics of disclosure; (d) Revised Self-disclosure Scale items measuring one’s choices with dimensions of disclosure; (e) selectively repeated RSDS items measuring one’s perception of one’s friend’s choices with dimensions of disclosure; (f) individualism- collectivism items; and (g) demographics. 2 Participants were recruited either through direct contact or indirectly through a snowball sampling method of referral from the first author’s colleagues, friends, and students. In cases where participants took part in this study as part of their experiential learning activities in communication courses, this study was offered either as one of the options to choose from or as an extra-credit activity at the discretion of individual course instructors. The questionnaire required approximately 20–25 minutes to complete. Participants either completed the questionnaires in class or on their own.

Participants Approximately 300 individuals were recruited for their voluntary participation in this study. The majority of them ( 280) were recruited through in-class announcements Journal of Intercultural Communication Research85 in communication courses and others ( 20) were referred by the first author’s friends/colleagues. Out of these 300 individuals, a total of 264 participated in the study, 12 of whom were excluded because they did not complete the questionnaire as directed. All remaining 252 participants (males¼102, females¼150) had at least one friend from a racial/ethnic or cultural background different from their own.

The 252 participants consisted of 234 students at a large southwestern university (227 undergraduate; 7 graduate students). The rest (18 participants) consisted of 3 high-school students, 4 high-school graduates, 10 college graduates, and 1 unidentified. The average age of the participants was 21.4 years (SD¼7.5). Of the 252 participants, 189 reported on an interracial friendship and 63 reported on an intercultural friendship. The self-reported racial/ethnic backgrounds included 115 Caucasians or Whites, 69 Hispanics, 19 multiracials who self-reported two or more racial backgrounds, 14 Mexican-Americans, 12 non-US citizens, 8 Native Americans, 7 Asian-Americans, 5 African-Americans, and 3 Spanish-Americans. The average length of intercultural/interracial friendship was 4.5 years (SD¼5.1).

Instrument The objective of this study was to determine the influence of individualism- collectivism, relational intimacy, and reciprocity on both topics and dimensions of self-disclosure in intercultural and interracial friendships. A questionnaire format was employed to investigate this objective. The independent variables (IVs) were relational intimacy and individualism-collectivism. The respondents were asked to select an intercultural or interracial friend who was more than an acquaintance and reflect on both their own and their selected friends’ self-disclosure choices.

The vast majority of the scales were shortened to prevent fatigue and to ensure that the developed questionnaire could be reasonably completed within 20–25 minutes.

Thus, to test the empirical validity and model fit of the shortened measurement, all shortened scales were submitted to confirmatory factor analysis (CFA), using the AMOS structural equation modeling program with maximum likelihood estimation (version 7.0). Two criteria were adopted to determine the inclusion of items and the improvement of model fit. The following criteria was used to retain items.

First, items needed to have a factor loading of .40. Second, items needed to be parallel to the other underlying constructs or have homogeneous content indicated by only having a single path to a latent variable. The following fit indices were used: (a) the ratio of 2to degrees of freedom ( 2/df); (b) the comparative fit index (CFI); (c) the goodness-of-fit index (GFI); (d) the incremental fit index (IFI); and (e) the root mean square residual (RMR). An acceptable model fit was defined as meeting at least four of the above criteria (Kline, 1998): ( 2/df) less than 3.0, RMR less than .08, and CFI/GFI/IFI greater than .90. The 2test was not used to judge model fit because it is sensitive to sample sizes and often mistakenly rejects a model (Kline, 1998).

Relational intimacy was measured with 8 items from a previously validated 17-item instrument of social intimacy (Miller & Lefcourt, 1982). Items included, for 86Y.-W. Chen & M. Nakazawa example, ‘‘How much do you like to spend time with him/her?’’ and ‘‘How much do you like to be encouraging and supportive to him/her when he/she is unhappy?’’ Nine items (e.g., ‘‘How often do you keep very personal information to yourself and do not share it with him/her?’’ or ‘‘How affectionate do you feel towards him/her?’’) were dropped from the original scale because they either overlapped with self- disclosure or were too difficult to be rewritten as to be relevant to friendship.

To ensure empirical validity of the shortened measurement, the 8-item Social Intimacy Scale was submitted to CFA to test the model fit. After having dropped three of the eight items, all five model fit indices were above the recommended benchmarks, 2 (5,N¼252)¼7.54,p4.1, 2/df¼1.51, GFI¼.99, IFI¼1.00, CFI¼1.00, RMR¼.07. Cronbach’s alpha for this revised 5-item shortened scale was .88. 3 Individualism-collectivism was measured with a 16-item scale, in which 8 items measured individualism and 8 items measured collectivism (Triandis & Gelfand, 1998). To confirm empirical validity of the selected measurement, the scale was submitted to CFA. After having dropped 9 of the 16 items as suggested by CFA, all five model fit indices reached the recommended benchmarks, 2 (13,N¼252)¼9.25,p4.1, 2/df¼0.71, GFI¼.99, IFI¼1.00, CFI¼1.00, RMR¼.03. Cronbach’s alphas for this study were (a) individualism .76 and (b) collectivism .76.

The dependent variables (DVs) were six topics of self-disclosure and five dimensions of self-disclosure. The topics of self-disclosure (attitudes and opinions, tastes and interests, work or studies, money, personality, and body) were measured with 30 items from a previously validated 60-item self-disclosure questionnaire (Jourard & Lasakow, 1958). The shortened 30-item self-disclosure questionnaire was repeated twice to measure both one’s own as well as one’s perception of one’s friend’s choices with topics of self-disclosure. To ensure empirical validity of the shortened measurement, both versions of the shortened self-disclosure questionnaire were submitted to CFA. One item was excluded from the one’s-own version, and 2 items were excluded from the one’s-perception-of-one’s-friend version. 4Both versions of the shortened self-disclosure questionnaire had a good model fit: 2(362, N¼252)¼658.46,p5.001, 2/df¼1.82, GFI¼.85, IFI¼.93, CFI¼.93, RMR¼.04 for one’s own self-disclosure; 2(362,N¼252)¼755.03,p5.001, 2/df¼2.09, GFI¼.84, IFI¼.92, CFI¼.92, RMR¼.04 for one’s perception of one’s friend’s topics of self-disclosure. Cronbach’s alphas for the six topics for both one’s own and one’s perception of one’s friend’s were all acceptable: (a) attitudes and opinions .70/.78, (b) tastes and interests .79/.84, (c) work or studies .86/.88, (d) money .91/.92, (e) personality .85/.88, and (f) body .91/.92.

Dimensions of self-disclosure were measured with 20 items from a previously validated 31-item Revised Self-Disclosure Scale (RSDS; Wheeless, 1978) to measure five dimensions of self-disclosure: intended disclosure, amount, positive/negative, control of depth, and honesty-accuracy. Ten of the 20-item RSDS were repeated to measure one’s perception of one’s friend’s choices with amount, control of depth, and positive/negative disclosure. Because it is difficult to know one’s friend’s choices Journal of Intercultural Communication Research87 with intent to disclose and honesty-accuracy of disclosure, these two dimensions were not repeated in the one’s-perception-of-one’s-friend version. To ensure empirical validity of the shortened instruments, both versions were submitted to CFA. As suggested by the CFA result, seven items were excluded from one’s-own version, and three items from one’s-perception-of-one’s-friend version.

For one’s own dimensions of self-disclosure, the shortened 13-item revised RSDS measuring had a good model fit: 2(62,N¼252)¼112.45,p5.001, 2/df¼1.81, GFI¼.94, IFI¼.90, CFI¼.89, RMR¼.06. Cronbach’s alphas for the five dimensions were as follows: (a) intended disclosure .70, (b) amount .60, (c) positive/negative .57, (d) control of depth .60, and (e) honesty-accuracy .64. For one’s perception of one’s friend’s dimensions of self-disclosure, the shortened self-disclosure questionnaire measuring two dimensions also had a good model fit: 2(14,N¼252)¼37.46, p¼.001, 2/df¼2.68, GFI¼.96, IFI¼.92, CFI¼.91, RMR¼.06. Cronbach’s alphas for the two dimensions were as follows: (a) amount .64, and (b) positive/negative .66.

The items for each variable were averaged to create a composite score. Social- intimacy items were measured with a 10-point Likert-type scale ranging from 1 (not much)to10(a great deal) as exemplified in Miller and Lefcourt’s (1982) original study. Topics of self-disclosure items were measured on a 4-point rating scale as demonstrated in Jourard and Lasakow’s (1958) original study: 0 (have told the other person nothing about this aspect of me), 1 (have talked in general terms about this item), 2(have talked in full and complete detail about this item to the other person), and X(have lied or misrepresented myself to the other person so that he has a false picture of me). The remaining items for dimensions of self-disclosure and individualism- collectivism were measured with a 5-point Likert-type scale ranging from 1 (strongly disagree)to5(strongly agree).

Results This study examined a set of IVs (relational intimacy and individualism-collectivism) and two sets of DVs (six topics of self-disclosure as well as one’s perceptions of one’s friend’s six topics of disclosure, and five dimensions of self-disclosure as well as one’s perception of one’s friend’s two dimensions of disclosure). Hypotheses concerning a relationship between specific variables were answered with Pearson correlations.

Other hypotheses and an RQ examining a relationship between a set of DVs and a set of IVs were answered by applying a multivariate analysis of covariance (MANCOVA) to prevent the inflation of type-I errors caused by multiple testing (Stevens, 2001).

When the MANCOVA’s multivariate F test was significant, a series of Pearson correlations (when the MANCOVA had only one IV) or partial s(when the MANCOVA had more than one IV) were computed to determine the strength of the linear relationships between a particular IV and each of the DV sets. The results are presented in the order of the hypotheses and RQ.

H 1predicated that depth and frequency of topics of self-disclosure would increase with the levels of relational intimacy in intercultural and interracial friendships. 88Y.-W. Chen & M. Nakazawa MANCOVA applied to six topics of self-disclosure as a set of DVs and relational intimacy as an IV indicated that relational intimacy significantly predicted the topics of self-disclosure (F(6,245)¼38.8,p5.001). A series of Pearson correlation coefficients were calculated examining the relationships between relational intimacy and the six topics of self-disclosure. Significant positive relationships were found between relational intimacy and each of the six topics (rs(250)4.38,ps5.01; see Table 1). As levels of relational intimacy in intercultural and interracial friendships increased, self-disclosure in the six topics also increased. Thus, H 1was supported.

H 2 predicated that as the levels of relational intimacy in intercultural and interracial friendships increased, friends would have greater intent to disclose, disclose in greater amount and depth, and engage in more negative and honest- accurate disclosure. MANCOVA applied to the dimensions of self-disclosure as a set of DVs and relational intimacy as an IV indicated that relational intimacy significantly predicted the dimensions of self-disclosure (F(6,245)¼5.5,p5.001).

A series of Pearson correlation coefficients were calculated examining the relation- ships between relational intimacy and each of the five dimensions of self-disclosure (see Table 2). Significant positive relationships were found between relational intimacy and four of the five dimensions (rs(250)4.14,ps5.05) but not positive/ negative self-disclosure (r¼ .04,p4.1). These results suggest that, as levels of relational intimacy in intercultural and interracial friendships increased, intercultural and interracial friends had greater intent to disclose, disclose in greater amount and depth, and engage in more honest-accurate self-disclosure. However, the increased levels of relational intimacy did not influence positive/negative self-disclosure as predicated. Thus, H 2was mostly supported.

To explore the possibility that as the friendship grows, H 3predicated that the more individualistic an individual was, the more the individual would engage in self- disclosure. In contrast to the hypothesis, individualism was not significantly related to amount of self-disclosure (r(250)¼.03,p4.1). Thus, H 3was not supported.

RQ1 inquired the influence of individualism-collectivism on dimensions of self-disclosure. To answer RQ1, MANCOVA was first applied to five dimensions Table 1.Pearson Correlation of Relational Intimacy and Six Topics of Self-disclosure. 1234567 Mean 7.94 1.23 1.46 1.32 0.67 1.13 0.99 Standard Deviation 1.58 0.51 0.47 0.61 0.73 0.66 0.72 (1) Relational Intimacy 1 (2) Attitudes and Opinions .51** 1 (3) Tastes and Interests .53** .51** 1 (4) Work or Studies .45** .52** .50** 1 (5) Money .38** .53** .44** .47** 1 (6) Personality .61** .59** .55** .41** .57** 1 (7) Body .62** .57** .51** .46** .53** .80** 1 *Correlation is significant at the .05 level (2-tailed).

**Correlation is significant at the .01 level (2-tailed). Journal of Intercultural Communication Research89 of self-disclosure as a set of DVs and individualism as an IV. It revealed that individualism significantly predicted the dimensions of self-disclosure as a set (F(6,245)¼6.3,p5.001). Then, another MANCOVA with the same set of DVs was applied to collectivism as an IV. It revealed that collectivism also significantly predicted the dimensions of self-disclosure as a set (F(6,245)¼2.8,p¼.02). These MANCOVAs were followed by a series of correlation coefficients (see Table 3).

Individualism was significantly correlated with intent, depth, and honesty-accuracy of self-disclosure (ps5.05) while collectivism was significantly correlated with intent and honesty-accuracy (ps5.05). These results suggest that individual differences in individualism/collectivism are related to certain dimensions of self-disclosure.

Since all individualism, collectivism, and relational intimacy IVs significantly predicted the set of five dimensional variables of self-disclosure, these IVs were included in a MANCOVA model to test which of the IVs would make the greatest contribution in predicting the dimension DV set. Furthermore, to test whether racial/ethnic backgrounds themselves influence dimensions of self-disclosure above and beyond the individual cultural differences, the ethnicity variable 5was also included in the model. The MANCOVA model revealed that, after controlling Table 3.Pearson Correlation of Individualism-Collectivism and Five Dimensions of Self-disclosure. 1234567 Mean 3.57 3.96 Standard Deviation 0.60 0.55 (1) Individualism 1 (2) Collectivism .32** 1 (3) Intended Self-disclosure .22** .13* 1 (4) Amount .03 .06 .02 1 (5) Positive/Negative .07 .08 .19** .07 1 (6) Control of Depth .13* .03 .00 .33** .03 1 (7) Honesty-Accuracy .25** .20** .32** .08 .13* .13* 1 *Correlation is significant at the .05 level (2-tailed).

**Correlation is significant at the .01 level (2-tailed).

Table 2.Pearson Correlation of Relational Intimacy and Five Dimensions of Self- disclosure. 1 2 3456 Mean 7.94 3.77 2.98 3.42 2.67 3.77 Standard Deviation 1.58 0.71 0.82 0.70 0.87 0.79 (1) Relational Intimacy 1 (2) Intended Self-disclosure .23** 1 (3) Amount .17* .01 1 (4) Positive/Negative .04 .25** .03 1 (5) Control of Depth .14* .00 .31** .14* 1 (6) Honesty-Accuracy .19** .39** .10 .19** .09 1 *Correlation is significant at the .05 level (2-tailed).

**Correlation is significant at the .01 level (2-tailed).

90Y.-W. Chen & M. Nakazawa for the other IVs, relational intimacy and individualism were significant predictors of the five dimensions of self-disclosure as a set (intimacy:F(5,243)¼4.7,p5.001; individualism:F(5,243)¼4.0,p5.01) while collectivism and ethnicity did not significantly contribute to predicting the DVs above and beyond the other IVs (ps4.1).

These analyses were followed by a series of partial sto examine the extent to which each IV was correlated with the DVs after the other three IVs had been controlled for. Relational intimacy significantly predicted four of the dimension variables but not positive/negative self-disclosure (p4.1 for positive/negative; ps5.05 for the other DVs; see Table 4). It accounted for 9% of the variance of the 5 DVs as a set. Likewise, individualism significantly predicted intent, depth, and honesty-accuracy of self-disclosure (ps5.05), but not amount or positive/negative aspects of disclosure (ps4.1). It accounted for 8% of the variance of the DV set.

Neither collectivism nor ethnicity significantly predicted any of the 5 variables (ps4.05). These results indicate that, of the IVs examined, relational intimacy and individualism, but not collectivism, significantly influenced the dimensions of self-disclosure, and that ethnicity did not make any significant influence above and beyond the individual differences in relational intimacy and individualism.

H 4predicated that, as a dyadic and reciprocal process, one’s self-reported self- disclosure would largely mirror one’s perception of one’s intercultural/interracial friend’s self-disclosure. To answer the proposed hypothesis, a series of Pearson correlations were calculated between one’s own and one’s perception of one’s friend’s scores on six topics and two of the five dimensions (amount and positive/negative) of self-disclosure that were repeated to measure both one’s own and one’s friend’s self-disclosure (see Table 5). On the six topics of self-disclosure, all of the Pearson correlation coefficients were significant (rs(250)4.70,ps5.01). On the other hand, on the two dimensions self-disclosure, the correlation on the positive/negative was significant (p5.001), but not the amount (p4.1).

The correlation coefficient averaged across the six topics (r(250)¼.79) was much higher than the correlation averaged across the two dimensions (r(250)¼.20).

To test whether the magnitude of these coefficients significantly differed, Meng’s test was applied for correlated correlations (Meng, Rosenthal, & Rubin, 1992); it revealed Table 4.Partial Correlations of Four IVs of Culture and Five Dimensions of Self-disclosure. Relational Intimacy Individualism Collectivism Ethnicity Intended Self-disclosure .21** .15* .07 .03 Amount .16* .01 .04 .02 Positive/Negative .03 .05 .05 .01 Control of Depth .16* .15* .02 .03 Honesty-Accuracy .13* .18** .12 .06 *Correlation is significant at the .05 level (2-tailed).

**Correlation is significant at the .01 level (2-tailed). Journal of Intercultural Communication Research91 that these correlations significantly differed (Z¼8.59,p5.001). This result indicates that, on average, the extent to which one’s self-reported self-disclosure would mirror one’s perception of one’s friend’s self-disclosure may be stronger on topics than on dimensions. This suggests that one seems to mirror what one’s friend talks about more strongly than how the friend talks; however, future studies need to further investigate and validate this pattern.

Discussion and Conclusion This study sought to examine the effects of individualism-collectivism and relational intimacy on topics and dimensions of self-disclosure from a social penetration perspective. In general, findings in this study suggest that relational intimacy is positively correlated with all six topics and four of the five dimensions of self- disclosure. Also, individualism, independent of relational intimacy, is a significant predictor of the five dimensions of self-disclosure as a set while collectivism is not.

Moreover, one mirrors one’s intercultural/interracial friend in terms of all six topics of self-disclosure but only the dimension of positive/negative disclosure. In the section below, the findings organized around the proposed four hypotheses and one research question are discussed and then implications and limitations are noted.

H 1predicated that depth and frequency of topics of self-disclosure would increase with the levels of relational intimacy in intercultural and interracial friendships.

H 1was supported for all six topics under examination in this study. Consistent with the social penetration perspective (Altman & Taylor, 1973), levels of self-disclosure progressed from peripheral layers of personality to central layers associated with both breadth category and breadth frequency as relationships developed. Thus, levels of topics of self-disclosure progressed as intercultural and interracial friendships developed. More importantly, this finding partially supports the applicability of social penetration theory to intercultural and interracial friendships. Table 5.Correlation of One’s Own Self-disclosure and One’s Perception of One’s Friend’s Disclosure.

Variablerdf p Topics of Disclosure Attitudes and Opinions .86 250 .003** Tastes and Interests .74 2505.001** Work or Studies .73 2505.001** Money .73 2505.001** Personality .84 2505.001** Body .80 2505.001** Dimensions of Disclosure Amount of Disclosure .08 2504.20 Positive/Negative Disclosure .33 2505.001** *Correlation is significant at the .05 level (2-tailed).

**Correlation is significant at the .01 level (2-tailed). 92Y.-W. Chen & M. Nakazawa H2predicated that, as the levels of relational intimacy in intercultural and interracial friendships increased, friends would have greater intent to disclose, disclose in greater amount and depth, and engage in more negative and honest- accurate disclosure. The results were largely consistent with H 2except for positive/ negative self-disclosure. Predominantly consistent with the social penetration perspective (Altman & Taylor, 1973), self-disclosure exchanges progressed from public-outer areas of the selves to all public, immediate, and private areas of the selves as relationships developed. Thus, dimensions of self-disclosure increased as intercultural and interracial friendships developed. The lack of support for more negative than positive self-disclosure in intimate intercultural and interracial friendships might have been the result of (a) disposition to self-disclose positively interculturally (Allen, Long, O’Mara, & Judd, 2003) or (b) racial or cultural stereotypes that friends had of each other that prevented potentially problematic negative self-disclosure (Orbe & Harris, 2008).

H 3predicated that the more individualistic an individual was, the greater amount of self-disclosure the individual would engage in. The results did not support H 3, and there are three explanations for this lack of support. First, self-construals and individualistic values were found to mediate the influence of I-C on communication (Gudykunst et al., 1996). Thus, the insignificant finding of individualism on amount of self-disclosure in this study could be the result of individualism having indirect or mediated rather than direct effect on amount of self-disclosure. Second, it has been debated whether I-C is a unidimensional or multidimensional construct (Azevedo et al., 2002). Thus, the insignificant finding might have been the result of either methodological or conceptual measurement of the I-C scale utilized in this study. Third, the lack of support for H 3could potentially be due to communication accommodation in intercultural and interracial friendships where the individualistic friends modify self-disclosure choices to accommodate the other friends.

Future research needs to investigate when and how intercultural and interracial friends accommodate each other.

RQ1 inquired about the influence of individualism-collectivism on dimensions of self-disclosure. Individualism was found to positively correlate with intent, depth, and honesty-accuracy of self-disclosure, while collectivism was positively related to intended and honest-accurate self-disclosure. These results were consistent with the cultural values of sincerity, respectfulness, and propriety associated with collectivism (Gao & Ting-Toomey, 1998; Gareis, 1995; Triandis, 1995). The findings might also have been the result of treating close friends as insiders or in-group members with whom self-disclosure could be less risky (Triandis, 1995). Also, the positive correlation between individualism-collectivism and intent and honesty- accuracy of self-disclosure could be the outcome of intercultural friends mirroring each other’s choices of self-disclosure dimensions.

Follow-up analysis of RQ1 indicates that, after controlling for the other independent variables, individualism significantly predicated the five dimensions of self-disclosure as a set while collectivism did not. There are two possible explanations for this. First, this might have been influenced by the setting of this Journal of Intercultural Communication Research93 study, as it was conducted in the United States, a predominantly individualistic country. That is, most participants might have been more individualistic than collectivistic. Second, as Barnlund (1989) suggests that self-disclosure is a Western concept reflective of Eurocentric, self-centered cosmologies, the conceptualization of self-disclosure may have been a fundamentally individualistic construct.

H 4 predicated that, as a dyadic and reciprocal process, one’s self-reported self-disclosure would largely mirror one’s perception of one’s intercultural or interracial friend’s self-disclosure. The results mostly supported this fourth hypoth- esis. Self-reported self-disclosure and perceived friend’s self-disclosure were signif- icantly correlated in all six topic areas and also in positive/negative disclosure. As the norm of reciprocity in self-disclosure indicated (Pearce et al., 1974; Won-Doornink, 1985), intercultural and interracial friends mutually engaged in equivalent levels of self-disclosure in all six topics areas and one dimension. The lack of support for reciprocal amount of self-disclosure might have been the result of the subjective nature of self-reporting or the influences of social desirability. Also, this study did not take into account the different types of intercultural and interracial friendships in terms of the social avenues through which the friendships were formed, such as at work, at school, or through mutual friends.

Implications The findings in this study largely support examining intercultural and interracial friendships from a social penetration perspective in that levels of self-disclosure topics and dimensions increased as intercultural and interracial friendship developed.

Also, this study highlights the norm of reciprocity in intercultural and interracial friendships. However, it is important not to assume that reciprocal self-disclosure implies gratification for both intercultural and interracial friends in self-disclosing.

More importantly, this study indicates that relational intimacy has greater influence than culture on close intercultural and interracial friendships. This suggests that the key to overcoming challenges implicated by cultural differences in intercultural relationships is to develop personal relationships, as exemplified in Lee’s (2006) study on the importance of co-constructing shared relation identity in intercultural friendships. Finally, this study raises the question of whether self-disclosure is fundamentally an individualistic construct that may not be conceptually equivalent from a cultural standpoint of collectivism.

Limitations Several methodological limitations in this study should be noted. First, this study did not distinguish between intercultural and interracial friendships, which may face similar yet different challenges. Second, this study did not adequately account for ethnicity that might have different culturally specific individualistic or collectivistic values. Third, the use of self-reporting may have inherent biases that could not be accounted for. Fourth, the results concerning one’s own as well as one’s perception of one’s friend’s dimensions of self-disclosure should be interpreted with caution 94Y.-W. Chen & M. Nakazawa because of relatively low reliabilities (50.7): it is possible that low reliabilities may have attenuated the statistical power to detect the relationship between one’s own and one’s perception of one’s friend’s amount of self-disclosure. Fifth, the CFA models fit the data, providing the evidence of construct validity of the shortened instruments. Future studies should provide convergent and discriminant validities of the instruments. Sixth, the design of this study did not allow us to examine the effect of time on the relationship between the relational intimacy and self-disclosure.

To test this, as an anonymous reviewer pointed out, future studies should collect data at several points in a relationship to show that an increase in intimacy over time would be accompanied by a parallel increase in self-disclosure.

Future research needs to further investigate other aspects of self-disclosure in intercultural friendships to fill the gap between self-disclosure and culture. Also, this study highlights the urgent need to further examine the conceptual equivalence of self-disclosure in individualistic and collectivistic cultures. Future research should further validate if culture indeed has less impact on intimate intercultural friendships.

Finally, this study encourages future research to investigate what functional self-disclosure norms or rules exist in intercultural friendships.

Notes [1] The self-construal scale is another widely used measurement of culture in cross-cultural and intercultural communication research. However, recent debates regarding the validity of self-construal scales prevent the authors from adopting it (see Levine et al., 2003).

[2] One may wonder whether the set of data on participants’ friends’ self-disclosure should have been obtained actually from their friends. Instead, it was decided to obtain both sets of data (i.e., participants’ own and their friends’ self-disclosure) from the participants themselves to maximize the magnitude of the relationship between these sets of data (monotrait- monomethod correlations; Furr & Bacharach, 2007). Future studies will obtain both types of data sets from both participants and their friends to see how similar or different their perception would be from their friends’ opinions (monotrait-heteromethod correlations).

[3] Factor analyses tables are available from the first author.

[4] ‘‘Inconsistency’’ between the two versions (i.e., Item 16 was excluded from one’s own version but not from the other version) was allowed because of an interest in selecting items that would optimally capture the constructs of self-disclosure. Making items consistent between the two versions (i.e., both versions have exactly the same items) would have resulted in excluding items with high factor loadings.

[5] To test whether racial/ethnic backgrounds themselves, independent of individual differences in intercultural/racial friendship or individualism/collectivism, would affect self-disclosure, a new independent variable was created—ethnicity—by combining Caucasians/White with Spanish-Americans (European White,n¼118), and combining the other groups (non-White,n¼134).

References Aberson, C. L., Shoemaker, C., & Tomolillo, C. (2004). Implicit bias and contact: The role of interethnic friendships.Journal of Social Psychology,144, 335–347. Journal of Intercultural Communication Research95 Allen, J. L., Long, K. M., O’Mara, J., & Judd, B. B. (2003). Verbal and nonverbal orientations toward communication and the development of intracultural and intercultural relationships.

Journal of Intercultural Communication Research,32, 129–160.

Altman, I., & Taylor, D. (1973).Social penetration: The development of interpersonal relationships.

New York: Holt, Rinehart & Winston.

Antaki, C., Barnes, R., & Leudar, I. (2005). Self-disclosure as a situated interactional practice.British Journal of Social Psychology,44, 181–199.

Avruch, K. (1998).Culture and conflict resolution. Washington, DC: United States Institute of Peace.

Azevedo, A., Drost, E. A., & Mullen, M. R. (2002). Individualism and collectivism: Toward a strategy for testing measurement equivalence across culturally diverse groups.Cross-Cultural Management,9, 19–29.

Barnlund, D. C. (1989).Communicative styles of Japanese and Americans: Images and realities.

Belmont, CA: Wadsworth Publishing.

Barry, D. T. (2003). Cultural and demographic correlates of self-reported guardedness among East Asian immigrants in the U.S.International Journal of Psychology,38, 150–159.

Cahn, D. D. (1984). Communication in interpersonal relationships in two cultures: Friendship formation and mate selection in the U.S.and Japan. Communication,13, 31–37.

Cai, D. A., Wilson, S. R., & Drake, L. E. (2000). Culture in the context of intercultural negotiation:

Individualism-collectivism and paths to integrative agreements.Human Communication Research,26, 591–617.

Chen, G. -M. (1995). Differences in self-disclosure patterns among Americans versus Chinese:

A comparative study.Journal of Cross-Cultural Psychology,26, 84–91.

Chen, L. (2002). Communication in intercultural relationships. In W. B. Gudukunt & B. Mody (Eds.),Handbook of international and intercultural communication(2nd ed., pp. 241–257).

Thousand Oaks, CA: Sage.

Chen, Y. -W. (2006). Intercultural friendship from the perspectives of East Asian international students.China Media Research,2(3), 43–58.

Collier, M. J. (1996). Communication competence problematic in ethnic friendships.

Communication Monographs,63, 314–336.

Derlega, V. J., Winstead, B. A., Wong, P. T. P., & Greenspan, M. (1987). Self-disclosure and relationship development: An attributional analysis. In M. E. Roloff & G. R. Miller (Eds.), Interpersonal processes: New directions in communication research(pp. 172–187). Newbury Park, NJ: Sage.

Diggs, R. C., & Clark, K. D. (2002). It’s a struggle but worth it: Identifying and managing identities in an interracial friendship.Communication Quarterly,50, 368–390.

Dindia, K. (1997). Self-disclosure, self-identity, and relationship development: A transactional/ dialectical perspective. In S. Duck (Ed.),Handbook of personal relationships: Theory, research and interventions(pp. 411–426). Chichester, UK: John Wiley & Sons.

Earley, P. C. (1998). Taking stock in our progress on individualism-collectivism: 100 years of solidarity and community.Journal of Management,24(3), 265–304.

Emmers, T. M., & Dindia, K. (1995). The effect of relational stage and intimacy on touch:

An extension of Guerrero and Andersen.Personal Relationships,2(3), 225–236.

Fitzpatrick, J., Liang, S., Feng, D., Crawford, D., Tsorell, G., & Morgan-Fleming, B. (2006).

Social values and self-disclosure: A comparison of Chinese native, Chinese residents (in U.S.) and North American spouses.Journal of Comparative Family Studies,37, 113–127.

Furr, R. M., & Bacharach, V. R. (2007).Psychometrics: An introduction. Beverly Hills, CA: Sage.

Gao, G., & Ting-Toomey, S. (1998).Communicating effectively with the Chinese. Thousand Oaks, CA: Sage.

Gareis, E. (1995).Intercultural friendship: A qualitative study. Lanham, MD: University Press of America.

Gareis, E. (1999). Rhetoric and intercultural friendship formation.International and Intercultural Communication Annual: Rhetoric in Intercultural Contexts,22, 91–117. 96Y.-W. Chen & M. Nakazawa Gareis, E. (2000). Intercultural friendship: Five cases studies of German students in the USA.

Journal of Intercultural Studies,21, 67–91.

Gudykunst, W. B. (1985a). An exploratory comparison of close intracultural and intercultural friendships.Communication Quarterly,33, 270–283.

Gudykunst, W. B. (1985b). The influence of cultural similarity, type of relationship, and self- monitoring on uncertainty reduction progress.Communication Monographs,52, 203–217.

Gudykunst, W. B., Matsumoto, Y., Ting-Toomey, S., Nishida, T., Kim, K., & Heyman, S. (1996).

The influence of cultural individualism-collectivism, self-construals, and individual values on communication styles across cultures.Human Communication Research,22, 510–543.

Gudykunst, W. B., & Nishida, T. (1983). Social penetration in Japanese and American close friendships. In R. N. Bostrom & B. H. Westley (Eds.),Communication yearbook(Vol. 7, pp. 592–610). Beverly Hills, CA: Sage.

Gudykunst, W. B., & Nishida, T. (1986). Attributional confidence in low- and high-context cultures.Human Communication Research,12, 525–549.

Hammer, M. R., & Gudykunst, W. B. (1987). The influence of ethnicity and sex on social penetration in close friendships.Journal of Black Studies,17(4), 418–437.

Hofstede, G. (1991). Levels of culture.Cultures and organizations: Software of the mind(pp. 3–19).

Maidenhead, UK: McGraw-Hill.

Hui, C. H., & Yee, C. (1994). The shortened individualism-collectivism scale: Its relationship to demographic and work-related variables.Journal of Research in Personality,28, 409–424.

Jourard, S. M., & Lasakow, P. (1958). Some factors in self-disclosure.Journal of Abnormal and Social Psychology,56, 91–98.

Kito, M. (2005). Self-disclosure in romantic relationships and friendships among American and Japanese college students.Journal of Social Psychology,145, 127–140.

Kline, R. B. (1998).Principles and practice of structural equation modeling. New York: Guilford Press.

Kudo, K., & Simkin, K. A. (2003). Intercultural friendship formation: The case of Japanese students at an Australian university.Journal of Intercultural Studies,24, 91–114.

Lee, P.-W. (2006). Bridging cultures: Understanding the construction of relational identity in intercultural friendship.Journal of Intercultural Communication Research,35, 3–22.

Levin, T. R., Bresnahan, M. J., Park, H. S., Lapinski, M. K., Wittenbaum, G. M., Shearman, S. M., et al.(2003). Self-construal scales lack validity.Human Communication Research,29(2), 210–252.

Lustig, M. W., & Koester, J. (2006).Intercultural competence: Interpersonal communication across cultures(5th ed.). Boston, MA: Pearson.

Matsushima, R., & Shiomi, K. (2002). Self-disclosure and friendship in junior high school students.

Social Behavior and Personality,30, 515–526.

Meng, X. L., Rosenthal, R., & Rubin, D. B. (1992). Comparing correlated correlation coefficients.

Psychological Bulletin,111, 172–175.

Miller, R. S., & Lefcourt, H. M. (1982). The assessment of social intimacy. Journal of Personality Assessment,46(5), 514–518.

Morgan, S. E., & Arasaratnam, A. (2003). Intercultural friendships as social excitation: Sensational seeking as a predictor of intercultural friendship seeking behavior.Journal of Intercultural Communication Research,3, 175–186.

Olaniran, B. A. (1996). Social skills acquisition: A closer look at foreign students on college campuses and factors influencing their levels of social difficulty in social situations.

Communication Studies,47, 72–88.

Orbe, M., & Harris, T. M. (2001).Interracial communication: Theory into practice. Belmont, CA:

Wadsworth/Thomson Learning.

Orbe, M., & Harris, T. M. (2008).Interracial communication: Theory into practice(2nd ed.).

Los Angeles, CA: Sage. Journal of Intercultural Communication Research97 Pearce, W. B., Sharp, S. T., Wright, P. H., & Slama, K. M. (1974). Affection and reciprocity in self-disclosing communication.Human Communication Research,1(1), 5–14.

Samter, W. (2003). Friendship interaction skills across the life span. In J. O. Greene & B. R. Burleson (Eds.),Handbook of communication and social interaction skills. Mahwah, NJ:

Lawrence Erlbaum Associates, Publishers.

Stevens, J. P. (2001).Applied multivariate statistics for the social sciences(4th ed.). Mahwah, NJ:

Lawrence Erlbaum Associates, Publishers.

Tajfel, H. (1978). Interindividual and intergroup behaviour. In H. Tajfel (Ed.),Differentiation between social groups(pp. 27–60). London: Academic Press.

Tardy, C. H., & Dindia, K. (2006). Self-disclosure: Strategic revelation of information in personal and professional relationships. In W. Hargie (Ed.),The handbook of communication skills (pp. 229–266). New York: Routledge.

Taylor, D. A., & Altman, I. (1987). Communication in interpersonal relationships: Social penetration processes. In M. E. Roloff & G. R. Miller (Eds.),Interpersonal processes:

New directions in communication research(pp. 257–277). Newbury Park, NJ: Sage.

Ting-Toomey, S. (2005). The matrix of face: An updated face-negotiation theory.

In W. B. Gudykunst (Ed.),Theorizing about intercultural communication(pp. 71–92).

Thousand Oaks, CA: Sage.

Triandis, H. C. (1995).Individualism and collectivism. Boulder, CO: Westview.

Triandis, H. C., & Gelfand, M. J. (1998). Converging measurement of horizontal and vertical individualism and collectivism.Journal of Personality and Social Psychology,74(1), 118–128.

Wheeless, L. R. (1978). A follow-up study of the relationships among trust, disclosure, and interpersonal solidarity.Human Communication Research,4, 143–157.

Wheeless, L. R., Erickson, K. V., & Behrens, J. S. (1986). Cultural differences in disclosiveness as a function of locus of control.Communication Monographs,53, 36–46.

Won-Doornink, M. J. (1985). Self-disclosure and reciprocity in conversation: A cross-national study.Social Psychology Quarterly,48, 97–107.

Wright, P. H. (1978). Toward a theory of friendship based on a conception of self.Human Communication Research,4, 196–207.

Zimmermann, S. (1995). Perceptions of intercultural communication competence and international student adaptation to an American campus.Communication Education,44, 321–335.

98Y.-W. Chen & M. Nakazawa Copyright of Journal of Intercultural Communication Research is the property of World Communication Association and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.