Self-Construal and the Intra- and Interethnic Social Interactions of Ethnic Minorities

Abstract

This study examined relationships between self-construal and the quality of daily interactions of three ethnic minority groups in Europe: ethnic Turks and Moroccans in the Netherlands and Chechens in Poland. They described the social interactions they had for 2 weeks and they completed measures of independent and interdependent self-construal. We expected that, regardless of whether individuals’ self-construals match with prevailing construals in the host society, interdependent self-construal would be positively related to the quality of intra- and interethnic contact. The results largely confirmed this expectation. Across the two samples, participants who were higher in interdependent self-construal had more positive (and less negative) interactions than participants who were lower in interdependent self-construal. Some of these relationships varied as a function of whether or not a majority group member was present, however. Persons with a more interdependent construal of self felt more liked, respected, accepted, and free to express opinion during interactions in which a majority group member was present, whereas no such relationships were found for intra-ethnic interactions. There were very few relationships between independent self-construal and the quality of either intra- or interethnic contact. The results suggest that for the quality of ethnic minorities’ daily interactions, their interpersonal orientation is more important than a match between their orientation and the dominant orientation of the majority culture.

Keywords

self-construal interethnic contact daily interactions minorities

One of the most influential frameworks for understanding and examining cultural differences in how individuals think, feel, and behave pertains to people’s self-construal–the extent to which their relationships with others are included in the self. Several investigators have argued that people in collectivistic societies tend to emphasize an interdependent construal of self: they define themselves in terms of their close relationships with others and value maintaining harmony with others and value fitting in. In most Western societies, however, people tend to construe the self as independent from others. They place less value on nurturing relationships with other people, and instead, they value their individual abilities, autonomy, and accomplishments (Markus & Kitayama, 1991; Singelis, 1994; Triandis, 1989).

Cross-cultural differences in self-construal can be relevant to understanding migration and interethnic contact, particularly when there are differences between the self-construal individuals have from their cultures of origins and the prevailing self-construals of the host society. It has been argued that if people’s self-construals do not match the orientation of the receiving society (e.g., when they are high in interdependent self-construal but have moved to an individualistic culture), they will experience less similarity between themselves and host majority group members and they will be less likely to enjoy interactions with them than with ethnic ingroup members (e.g., Cross, 1995; but see also Searle & Ward, 1990; Ward & Chang, 1997). This cultural-fit hypothesis has received mixed support, however. For example, although some researchers (e.g., Yang, Noels, & Saumure, 2006) have found that independent self-construal predicts better sociocultural adjustment among international students in Canada (an individualistic culture), others (e.g., Cross, 1995) did not find relationships between sojourners’ independent self-construal scores and their satisfaction with their relationships with host nationals in the United States (an individualistic culture as well).

The present study was conducted among members from ethnic minority cultures (first- and second-generation newcomers) in the Netherlands and in Poland. Our aim was to examine how individual differences in self-construal are related to ethnic minority members’ daily social interactions with host majority group members and with ethnic ingroup members. In contrast to the cultural-fit hypothesis, we argue that regardless of whether an ethnic minority member’s self-construal matches the dominant form of construal of the host society, ethnic minority members who have a stronger interdependent self-construal will have more rewarding interactions with both ethnic ingroup members and members of the host majority culture compared to minority members with a less interdependent construal of self. We label this the interpersonal style hypothesis.

Our primary rationale for this hypothesis is based on the relative centrality of “the other” and the interconnectedness between the self and others that characterizes interdependent self-construal, as opposed to the relative centrality of “the self” that characterizes independent self-construal. It is generally agreed that people with an interdependent self-construal derive their self-esteem from developing and maintaining close relationships with relevant others, whereas people with an independent self-construal derive their self-esteem from being unique and expressing their inner attributes (e.g., Markus & Kitayama, 1991; Singelis, 1994). Such processes may have important implications for how people approach and respond to social interactions. Most importantly, whereas people with an independent construal of self will be motivated to distinguish themselves from others, people with an interdependent construal of self are likely to think and behave in ways that strengthen their relationships with others (Cross, Bacon, & Morris, 2000).

More specifically, persons with an interdependent view of self are likely to have developed the skills and abilities that foster positive interactions with others (Cross & Madson, 1997; Markus & Kitayama, 1991). In this regard, there is evidence that they tend to be more agreeable and more attentive to the needs of others during social interactions than individuals with an independent construal of self and that they are more willing to adjust their behaviors accordingly. For example, Miller, Bersoff, and Harwood (1990) found that Indians (who are from a more collectivistic society) valued being responsive to the needs of others more than Americans (who are from a more individualistic society). In a similar vein, Cross et al. (2000) found that individuals who were high on relational-interdependent self-construal–a measure of self-construal that assesses the extent to which people include close relationships in their self-concepts–were viewed as more sensitive and responsive to the needs of their relational partners than were individuals who were low on this measure. There is also evidence that individuals with an interdependent construal of self are more motivated to view themselves and relevant others similarly and that they are more likely to attend to and remember information about relevant others (e.g., Cross & Morris, 2003; Cross, Morris, & Gore, 2002; Stapel & Koomen, 2001).

These qualities may give rise to more rewarding social interactions. For example, Cross et al. (2000) found that individuals who interacted with a person who was high on relational-interdependent self-construal were more satisfied with the interactions and liked their partners more than those who interacted with a person who was low on that measure. Such positive responses may also affect how persons with a higher interdependent self-construal themselves react to interactions. For instance, according to Markus and Kitayama (1991), the other-oriented behaviors of persons with an interdependent view of self are likely to be intrinsically rewarding because they may invoke positive, other-focused emotions (e.g., feeling accepted or liked by others) while reducing negative emotions (e.g., feeling nervous or angry).

Nevertheless, some have also argued that persons with an interdependent construal of self may not be concerned with the needs of all others. For example, Markus and Kityama (1991) suggested that people with a strong interdependent sense of self will be more attentive and responsive during interactions with persons with whom they share a common fate, such as ingroup members, and will tend to draw a sharp distinction between in- and outgroup members. Thus, one could argue that ethnic minority members with an interdependent construal of self are likely to respond less positively (or more negatively) to interethnic interactions than to intra-ethnic interactions. Markus and Kitayama (1991) also point out, however, that people with an interdependent construal of self will generally be “motivated to find a way to fit in with relevant others” (p. 227, emphasis added). For ethnic minority group members, these relevant others may include dominant outgroup members as well, given that they often depend on members of the majority group for important outcomes (e.g., jobs, houses). Therefore, we expect that persons with an interdependent construal of self will tend to have more rewarding interactions with ethnic ingroup members as well as with majority group members.

To examine relationships between self-construal and the quality of naturally occurring daily inter- and intra-ethnic contact, we conducted a study in the Netherlands and in Poland. Participants were members of ethnic minorities in each country, individuals of Turkish and Moroccan descent in the Netherlands and Chechens in Poland. We chose these groups because they are clearly recognized as minority groups in their respective societies, and this provided a clear basis for distinguishing inter- and intra-ethnic contact.

In the Netherlands, we studied individuals of Turkish and Moroccan descent who had either emigrated to the Netherlands when very young or were children of immigrants. For several reasons, we felt that their social position is clearly that of an “outsider” relative to the majority culture. For example, the socioeconomic position of both groups is relatively unfavorable. Unemployment rates are relatively high and their position in the Dutch ethnic hierarchy is low (e.g., Verkuyten & Kinket, 2000). In addition, the frequency of social contacts between ethnic Turks and Moroccans and ethnic Dutch has been declining over the past few years (Gijsberts & Dagevos, 2005). Moreover, all our participants were Muslim (self-identified), in a country that has traditionally been identified as Christian.

In Poland, we studied Chechen refugees. Most Chechens who live in Poland are asylum seekers who live in or nearby reception centers. Similar to our Dutch participants, they were all Muslims (self-identified) who were living in a country traditionally identified as Christian. Unlike our Dutch participants, they were not citizens (or permanent residents), a status that probably contributed to their being viewed as “outsiders” by members of the majority culture. Poland also presents a mixed opportunity for Chechen refugees. On the one hand, it is the first safe “stop” on their escape from Chechnya. On the other hand, Poland may be only a “stop” in Chechens’ journey to Western Europe.

The two countries in which we conducted our study are generally considered to be more independent than interdependent in terms of the predominant pattern of self-construal (e.g., Hofstede, 2001), and this provided an ideal situation to compare the cultural-fit and interpersonal style hypotheses. If the dominant cultural pattern of self-construal in the countries in which we conducted our studies had been interdependent, than the two hypotheses would have made the same prediction. Within the context of both models, quality of contact with majority members would be expected to be positively related to how much minority members possessed an interdependent sense of self, and under such circumstances, if we found such relationships, this would not clearly support one model over the other.

Given however, that the dominant culture is independent/individualistic, the cultural fit hypothesis suggests that quality of contact with members of the majority culture would be positively related to how much minority members possessed an independent sense of self. In contrast, even when the dominant culture is independent/individualistic, the interpersonal style hypothesis would suggest that quality of contact with members of the majority culture would be positively related to how much minority members possessed an interdependent sense of self, which was our expectation. Therefore, the present situation (i.e., a dominant pattern of independence) provided a good basis to compare the explanatory power of the two approaches.

Method

Samples

Participants in the Netherlands sample were recruited through a community organization. After inspection of their data, 12 participants were excluded from the analyses for failing to complete various forms properly, leaving a final sample of 18 persons of Turkish origin and 19 persons of Moroccan origin. A majority of participants (23) had been born in the Netherlands (with both parents having been born in Turkey or Morocco), whereas the remaining participants (14) had arrived in the Netherlands before their 8th birthday. Twenty-five participants were women. Participants’ ages ranged from 16 to 43 years (M = 26.7 years, SD = 7.7).

In Poland, potential participants were approached in meetings held either in or near refugee camps. It is important to note that potential participants were told that their decision to participate would have no bearing on their refugee status. In total, 89 Chechens agreed to participate. Eight participants were excluded from the analyses because they did not complete various forms properly, leaving 46 men and 35 women. Participants’ ages ranged from 18 to 67 years (M = 30.9 years, SD = 9.5). All participants lived in or near refugee camps, and none had a residence permit.

Instruments

The instruments used in the two studies were the same. The instruments were first prepared in English and were then translated and back-translated into Dutch and Russian following guidelines suggested by Van de Vijver and Leung (1997). Copies of the instruments are available from the authors.

Social Interaction Diary

Participants described their social interactions using a variant of the Rochester Interaction Record (Wheeler & Nezlek, 1977). Similar to previous research, a social interaction was defined as “an encounter with one or more other people in which the participants attend to one another and adjust their behavior in response to one another.” Our definition of a social interaction was similar to Goffman’s (1971) concept of a “social with,” and after they had been instructed, participants had no difficulties understanding what was and what was not a social interaction. Also similar to previous research, participants were told to record only face-to-face interactions that lasted 10 minutes or more. Participants were asked to maintain the diary for 2 weeks.

For each interaction, participants described the other people who were present, and they described their reactions to these interactions. For each of the others who were present, participants provided initials (unique for each person), gender, age, and ethnicity (the specific categories varied across the two studies). Participants rated each interaction in terms of how enjoyable the interaction was, how intimate it was, how influential they felt, how much the other people present respected them, how accepted they felt, how much the other people present liked them, and how free they felt to exchange ideas and opinions with others. These ratings were made using 9-point scales with endpoints labeled 1 = not at all and 9 = very much. Using 9-point scales, participants also described their affective reactions using four ratings with endpoints labeled sad-happy, relaxed-anxious/nervous, enthusiastic-bored, and content-angry. These ratings represented the affective circumplex (e.g., Barrett & Russell, 1998), a two-dimensional representation of emotional experience (a combination of positive/negative and active/deactive).

Measures of Self-Construal

Self-construal was measured with a subset of items taken from the Self-Construal Scale (Singelis, 1994). The Self-Construal Scale is a 24-item scale designed to measure the strength of individuals’ interdependent and independent self-construals. The interdependent self-construal subscale measures feelings of connectedness and relations to others, whereas the independent self-construal subscale measures feelings of separateness and uniqueness. Each of these subscales was measured with 5 of the original 12 items. We selected a subset of items from the original scale because pilot testing indicated that participants were unwilling to fill out lengthy questionnaires with seemingly repetitive items. Specific items were chosen on the basis of factor analyses and appropriateness for the samples being studied. Participants responded to each item using a 7-point scale (1 = strongly disagree, 7 = strongly agree). The specific items we used are presented in the appendix, and summary statistics for the scales are presented in Table 1. Independent and interdependent self-construal were not related in the Dutch sample (r = .03, ns), and they were positively related in the Polish sample (r = .29, p < .01).

Table 1.

Descriptive Statistics for Measures of Self-Construal

	Netherlands			Poland
	M	SD	α	M	SD	α
Independent	5.32	1.10	.71	5.22	1.10	.66
Interdependent	4.36	0.88	.51	5.42	1.05	.68

Procedure

Participants were introduced to the study individually or in small groups (10-15). They were told that the purpose of the study was to investigate daily social interactions and people’s responses to them. At the introductory session, participants were told how to use the diary form. Examples were provided to clarify what was an interaction (e.g., a conversation, dancing) and what was not an interaction (e.g., simply sitting next to someone on a bus). The various response categories were discussed until participants understood their definitions and felt comfortable with the forms and the procedure, and participants were given enough blank forms for the study. They were also told to complete the diary forms as soon as possible after each interaction, and at the latest, at the end of each day. At this session, participants also completed the measure of self-construal. During the study, a member of the research team maintained contact with participants to ensure that they were completing the diary forms as instructed. These contacts were too brief to be counted as interactions. The final Dutch sample of 37 participants described 662 interactions (M = 17.9, SD = 6.0), over an average of 12.9 days (SD = 1.8). The final Polish sample of 81 participants described 2,066 interactions (M = 25.9, SD = 14.5), over an average of 12.9 days (SD = 3.0).

Results

The data collected in this study constituted what is called a hierarchically nested or multilevel data structure: interactions were treated as nested within individuals, and the data were analyzed with a series of multilevel modeling analyses (MLM). Within the nomenclature of MLM, interactions were the Level 1 observations and persons were the Level 2 observations. The data were analyzed with the program HLM (HLM6; Raudenbush, Bryk, Cheong, & Congdon, 2004). Multilevel modeling analyses of social interaction diary data are discussed in Nezlek (2003).

Relationships Between Self-Construal and Reactions to Interactions Involving and Not Involving a Member of the Majority Cultural Group

The primary focus of the analyses was the extent to which self-construal was related to reactions to interactions that involved and did not involve a member of the ethnic majority culture. For the Netherlands study, on average, 38% of participants’ interactions involved at least one ethnic Dutch person. For the Poland study, approximately 13% of participants’ interactions involved at least one ethnic Pole. Although theoretically interesting, it was not possible to examine interactions between members of different minority groups because very few interactions involved a member of an ethnic minority other than the participant’s group. The mean percentages for such interactions were 5% for the Turkish participants, 7% for Moroccan participants, and 2% for Chechen participants. This meant that we were limited to classifying interactions as a function of whether or not a member of the ethnic majority culture was present during the interaction. To simplify the presentation and discussion of our results, we sometimes refer to interactions that involved a member of the majority culture as “interethnic” and interactions that did not involve a member of the majority culture as “intraethnic,” and we address this issue in more detail in the Discussion section.

As discussed by Nezlek (2003), mutually exclusive categories such as we had in our study provided a basis to estimate means separately for interactions in each category. This was done using a Level 1 model with two dummy-coded predictors and no intercept (sometimes called a zero-intercept model). One predictor (Majority) was coded 1 if a member of the majority culture was present and 0 if a member was not present. The second predictor (NoMajority) was coded 1 if a member of the majority culture was not present, and 0 if a member was present.¹ The second predictor is the mirror image of the first. Both are entered into the model uncentered. At Level 1 (the interaction level) of the analyses, this scheme is represented below:

Level 1: y_{ij} = β_{1 j} (Majority) + β_{2 j} (NoMajority) + r_{ij} .

In this model, there were i observations (measures of interactions) treated as nested within j participants. For each participant, a mean was estimated for interactions that involved a member of the majority culture and interactions that did not involve a member of the majority culture. How such a scheme works can be understood by estimating predicted values for interactions when the Level 1 predictors took on certain values. The first coefficient estimated reactions to interactions when a member of the majority culture was present (Majority = 1, NoMajority = 0) because the NoMajority coefficient dropped out (was 0), and the second coefficient estimated reactions to interactions when a member of the majority culture was not present (Majority = 0, NoMajority = 1) because the Majority coefficient dropped out.

Within the multilevel framework, coefficients from Level 1 can be thought of as being “passed up” to analyses at Level 2. In essence, two means were calculated for each person, one representing interactions when a Majority member was present and the other representing interactions when no Majority member was present, and individual (between-person) differences in these means were examined at Level 2. Relationships between the Level 1 means and the Level 2 measures of self-construal were examined with the model below, which estimated separate coefficients for each sample. This model also relied on a dummy-coded set of predictors. One predictor (NL) was coded 1 for participants from the Netherlands and 0 for participants from Poland. Another predictor (PL) was coded 1 for participants from Poland and 0 for participants from the Netherlands. Self-construal was included by multiplying these dummy codes by participants’ self-construal scores (standardized within each sample). All predictors were entered into the model uncentered, and the intercept was dropped.

Similar to the Level 1 model, how such a scheme works can be understood by estimating predicted values. For participants in the Netherlands, the PL and PL-SC coefficients were 0 because for these participants PL = 0, whereas for participants in Poland, the NL and NL-SC coefficients were 0 because NL = 0. This meant that the NL-SC coefficient represented relationships between reactions to interactions and a measure of self-construal for participants in the Netherlands, and the PL-SC coefficient represented relationships between reactions to interactions and a measure of self-construal for participants in Poland. See Nezlek (2010) for a discussion of using MLM to analyze cross-cultural data.

\begin{array}{l} Level 2: β_{1 j} = γ_{11} (NL) + γ_{12} (NL-SC) + γ_{13} (PL) + γ_{14} (PL-SC) + u_{1 j} . \\ β_{2 j} = γ_{21} (NL) + γ_{22} (NL-SC) + γ_{23} (PL) + γ_{24} (PL-SC) + u_{2 j} . \end{array}

The coefficients estimated by these models can then be compared using what are called “tests of fixed effects” (e.g., Raudenbush & Bryk, 2002), which are constraints on a model. For example, to determine if the relationship between a measure of self-construal and a reaction to interactions with a majority member was the same in the Dutch and Polish samples, the γ₁₂ and γ₁₄ coefficients would be constrained to be equal. If a constraint leads to a poorer fit, it can then be concluded that the coefficients are significantly different. Similarly, relationships between self-construal and reactions to interactions involving and not involving a member of the majority culture could be compared (e.g., γ₁₂ vs. γ₂₂ and γ₁₄ vs. γ₂₄).

Consistent with recommendations to use forward-stepping procedures in multilevel modeling (e.g., Raudenbush & Bryk, 2002), separate analyses were done for independent and interdependent self-construal. If both were significant predictors of the same Level 1 coefficient, then an analysis was conducted in which both were included as predictors. Also, a series of analyses were done that included participant sex at Level 2. Including sex had no influence on the results.

The results of the analyses of interdependent and independent self-construal are summarized in Tables 2 and 3, respectively. The mean values for reactions (the γ₁₁, γ₁₃, γ₂₁, and γ₂₃ coefficients) are not presented because they are not the focus of this article. When interpreting the coefficients presented in the tables, keep in mind that the self-construal measures were standardized. This means that the coefficients represent the change in the dependent measure associated with a 1 SD increase in a measure of self-construal.

Table 2.

Relationships Between Interdependent Self-Construal and Reactions to Inter- and Intraethnic Interactions

	Interethnic		Intraethnic
	Netherlands	Poland	Netherlands	Poland
Enjoyable	.34*	.37**	.23^†	.37**
Intimate	.46**	.41**	.34**	.33**
Influential		.35*		.24^†
Liked	.28^†	.62**		.21^†
Respected	.36*	.41*
Accepted	.34^†	.21
Free to express opinions	.34*	.40**
Happy	.32^†	.32	.31*	.23
Anxious/nervous		−.21	−.21	−.22
Bored/sluggish		−.53**	−.30^†	−.28
Angry	−.38^†	−.71**	−.31*	−.46**

Note: Coefficients with an absolute value of less than .20 are not tabled.

p < .05. **p < .01. ^†p < .10.

Table 3.

Relationships Between Independent Self-Construal and Reactions to Inter- and Intraethnic Interactions

	Interethnic		Intraethnic
	Netherlands	Poland	Netherlands	Poland
Enjoyable
Intimate		.29
Influential
Liked	.24
Respected	.22
Accepted	.23
Free to express opinions	.26^†		.30*	−.20*
Happy
Anxious/nervous	−.42*
Bored/sluggish	−.32*	−.23	−.27^†
Angry	−.26

Note: Coefficients with an absolute value of less than .20 are not tabled.

p < .05. **p < .01. ^†p < .10.

The results were clear and consistent across the two samples. In both samples, there were numerous significant relationships between reactions to interactions and interdependent self-construal, but there were few significant relationships between reactions and independent self-construal. Consistent with the “interpersonal style hypothesis,” participants who were higher in interdependent self-construal tended to have more positive (and less negative) interactions than participants who were lower in interdependent self-construal. Moreover, there were no significant (or near significant) differences between the two samples in these relationships (for both interdependent and independent self-construal).

In addition, relationships between interdependent self-construal and a group of reactions to interactions varied as a function of whether a member of the majority ethnic culture was present or not. This difference occurred for reactions that emphasized the interpersonal aspects of interaction: feeling liked, respected, and accepted, and feeling free to express one’s opinion. As can be seen in Table 2, across both samples, when a member of the majority culture was present (interethnic), seven of eight of the coefficients representing relationships between interdependent self-construal and these measures were significant. In contrast, when a member of the majority culture was not present (intraethnic), only one coefficient was (marginally) significant.²

The differences between intra- and interethnic contact in these relationships were confirmed by a series of analyses that constrained the coefficients representing these relationships to be equal. For all four measures, this constraint led to a poorer fitting model, indicating that the coefficients were significantly different: feeling liked, χ²(1) = 8.09, p < .001; feeling respected, χ²(1) = 5.51, p = .01; feeling accepted, χ²(1) = 2.56, p = .10; feeling free to express opinion, χ²(1) = 31.3, p < .001.

Although there were few relationships between independent self-construal and reactions to interaction, these relationships were consistent with the “cultural fit hypothesis.” In the Dutch sample only, participants who were higher in independent self-construal felt less nervous and less bored/sluggish (or alternatively, more relaxed and more enthusiastic) when a member of the majority culture was present compared to participants who were lower in independent self-construal. Moreover, and only for the Dutch sample, other coefficients that were not statistically significant also suggested that participants higher in independent self-construal had more positive interethnic interactions than participants who were lower in independent self-construal. Nevertheless, due to the fact that these coefficients were not statistically significant, these results need to be considered cautiously.

Discussion

We examined relationships between self-construal and the quality of daily intra- and interethnic interactions of ethnic Turks and Moroccans in the Netherlands and Chechens in Poland. Previously, it has been argued that members from minority cultures will be more likely to establish positive interactions with host majority group members if their self-construal is similar to the normative type of construal in the host country (the “cultural fit” hypothesis). In contrast, we expected that ethnic minority group members with stronger interdependent self-construal would have more rewarding social interactions (the “interpersonal style” hypothesis)–regardless of the match between their self-construal and prevailing norm in the host society–because they are likely to think and behave in ways that foster positive interactions with others (e.g., Cross & Madson, 1997; Markus & Kitayama, 1991). We expected that they would have more positive (or less negative) interactions with both ethnic ingroup and host majority group members.

The results largely supported our interpersonal style hypothesis. We found few significant relationships between independent self-construal and the quality of intra- or interethnic contact, whereas we found numerous significant relationships between interdependent self-construal and quality of contact. Across the two samples, we found that individuals with a higher interdependent self-construal reported more positive (and less negative) social interactions. For example, in the Dutch sample, participants with a more interdependent construal of self found their interactions to be more enjoyable and intimate, and they also felt happier and less angry during social contact. In the Polish sample, individuals who were higher on interdependent self-construal rated their interactions as more enjoyable and intimate, and they felt more influential, more liked, and less angry. These results suggest that regardless of the specific context, interactions tend to be more rewarding for individuals with a higher interdependent self-construal.

Our expectation that individuals with a higher interdependent self-construal would, compared to persons who were low on this measure, have more rewarding interactions with both ethnic ingroup members and with majority group members was partially supported. For more intrapersonally focused ratings (e.g., enjoyable, intimate) and for some more affectively focused ratings (e.g., anger, liked), there were no differences between interethnic and intra-ethnic interactions. In contrast, for the more interpersonally focused ratings (feeling liked, respected, free to express opinion, accepted), we found that persons with a higher interdependent construal of self tended to rate interactions with majority group members as more positive than individuals with a lower interdependent construal of self. In contrast, for the intra-ethnic interactions, there was only one marginally significant relationship between interdependent self-construal and the interpersonal ratings (i.e., liked). The differences between intra- and interethnic interactions in these relationships remained after constraining the coefficients representing these relationships to be equal.

These results are noteworthy because it has been argued that individuals with a higher interdependent self-construal are more likely to make a sharp distinction between in- and outgroup members and should therefore have less rewarding interactions with outgroup members (e.g., Markus & Kitayama, 1991). Nevertheless, and consistent with some parts of Markus and Kitayama’s argument, we think that individuals who are high in interdependent self-construal are motivated to fit in with relevant others and that members of the majority group are likely to be considered “relevant others” by ethnic minority group members. That is, members of minorities are often dependent on majority group members in many aspects of their everyday lives (e.g., work, school, housing). Moreover, individuals with a higher interdependent self-construal may pay closer attention to the beliefs and values of relevant others during interactions than those who are lower on interdependent self-construal or high on independent self-construal (e.g., Cross & Morris, 2003). Given that minority members are often expected to fit into the majority culture, this ability to attend to others may lead to more rewarding interactions with majority group members.

In addition, the interpersonal qualities that are usually associated with interdependent self-construal (e.g., being sensitive to other people’s needs) were probably more relevant during contact with majority group members than during contact with ethnic ingroup members. When people are interacting with ingroup members, they are probably more familiar with the norms and expectations for interaction than when they are interacting with outgroup members. Therefore, when interacting with ingroup members, how sensitive an individual is to the needs of others (i.e., how interdependent a person is) may be less important than when interacting with outgroup members because the greater familiarity an individual has with ingroup norms reduces the importance of being sensitive to others’ needs. People know more about the needs of their fellow ingroup members because of their experience as members of the ingroup. In contrast, when interacting with outgroup members, people may be unfamiliar with the norms, and so how attuned they are to others will matter more. Moreover, it is possible that greater interdependence on the part of minority members is seen as a positive, approach orientation by members of the majority culture. If this is the case, greater interdependence may help reduce the anxiety that is often present in interethnic contact (Stephan & Stephan, 1985).

Limitations, Implications for Future Research, and Conclusions

Certainly, the results of the present study need to be considered within some limitations. First, our samples were relatively small and were not random. Although it is not clear how the manner in which we obtained our samples could have biased the results in favor of our hypotheses, this may have occurred. In addition, we focused on specific aspects of interaction, and it is possible that other measures may have led to different conclusions. For example, we treated sadness and happiness as opposite ends of a bipolar continuum, whereas some have argued that in some instances happiness and sadness should conceptualized as two, separate, unipolar continua (e.g., Larsen, McGraw, & Cacioppo, 2001). It is also important to acknowledge the possibility that cultural fit may be a more important influence for ongoing relationships with majority group members than it is for social interactions, such as the ones we measured.

In our study, we assumed that self-construal was a relatively stable individual difference (as it is discussed in the literature) that led to individual differences in social interactions. Such an assumption is consistent with the results of panel studies (across time) in which causal relationships from dispositional measures to social interaction have been found (e.g., Neyer & Asendorpf, 2001; Nezlek, 2001). Nevertheless, it is possible that causal relationships between self-construal and social interaction are bi-directional. Positive social interactions with majority group members may lead to changes in interdependent self-construal, which in turn may result in more rewarding interactions. Unfortunately, our data did not allow us to examine such possibilities.

Perhaps the most important limitation of our study is the specific nature and status of the ethnic minority and majority cultures we studied. We referred to interactions that involved a member of the majority culture as interethnic. Although interactions that involve a member of the majority culture and a member of a minority culture are certainly a type of interethnic interaction, the results we found beg questions about other types of interethnic interactions such as those between members of different minority cultures, interactions that could be described as “inter-minority.” Would the same differences occur when members of a minority group were interacting with members of another minority group with which they were unfamiliar, or must one group be a majority group with the attendant power divisions, social hierarchy, and the like? Noting this, in our samples, there were precious few interminority interactions (less than 5% across all samples), suggesting that interminority interactions are not common. See also Nezlek and Schaafsma (2010), who discuss studying interethnic contact and the implications of low levels of contact between groups for such research.

We also studied a specific majority-minority distinction–Muslims living within traditionally Christian European countries. Important questions exist regarding how the results from this combination would characterize other majority-minority distinctions. For example, would the results be similar if the interactions of Christians living in a traditionally Islamic country were studied? What is the relative importance of different ways of conceptualizing minority status in situations in which individuals can be considered as minorities in terms of more than one dimension (e.g., Asian Buddhists living in traditionally Christian Great Britain)? Perhaps most important, what roles do power and status play? The ethnic minority groups we studied did not have high social status or power. Would our results have differed if we had studied the interactions of a high-status minority interacting with members of a lower status majority? More research is needed to examine the influence of minority status per se versus the influence of a being a specific type of minority within a specific majority culture/setting. By studying two samples, we hoped to address this issue in a preliminary fashion, but answering such a broad question will require multiple studies across multiple cultural settings involving a variety of minority/majority combinations.

Nevertheless, we think the present study makes a meaningful contribution to our understanding of the roles self-construal play in interethnic contact. To our knowledge, our study is the first to examine relationships between ethnic minority members’ self-construal and their daily intra- and interethnic interactions using diary measures. There is a growing realization that such methods can provide more accurate descriptions of people’s everyday interactions than single assessment measures such as surveys or questionnaires that ask people to retrospect over lengthy or (undefined) periods of time (e.g., Reis & Gable, 2000), and that such methods may be particularly useful for studies of interethnic contact (Nezlek & Schaafsma, 2010). Moreover, above and beyond highlighting the potential benefits of naturalistic methods per se to the study of identity, we think the present results suggest that self-construal may function as more than an individual-level manifestation of cultural norms.

Footnotes

Appendix

Acknowledgements

Different analyses of some of the data presented in this article appear in Schaafsma, Nezlek, Krejtz, and Safron (2010). We are grateful for the cooperation of our participants. We are also grateful to Francis Koelemaij, Snjezana Matijevic, and the intermediaries from the Center for Foreign Women (Centrum Buitenlandse Vrouwen) in Tilburg for their help collecting the data for the Netherlands study, and to Hanna Pasierska and Joanna Konikiewicz, the Polish Office for Repatriation and Aliens, and the Family Support Centre in Warsaw for their help collecting the data in the Poland study.

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

References

Barrett

L. F.

Russell

J. A.

(1998). Independence and bipolarity in the structure of current affect. Journal of Personality and Social Psychology, 74, 967-984.

Cross

S. E.

(1995). Self-construals, coping, and stress in cross-cultural adaptation. Journal of Cross-Cultural Psychology, 26, 673-697.

Cross

S. E.

Bacon

Morris

(2000). The relational-interdependent self-construal and relationships. Journal of Personality and Social Psychology, 78, 791-808.

Cross

S. E.

Madson

(1997). Models of the self: Self-construals and gender. Psychological Bulletin, 122, 5-37.

Cross

S. E.

Morris

(2003). Getting to know you: The relational self-construal, relational cognition, and well-being. Personality and Social Psychology Bulletin, 29, 512-523.

Cross

S. E.

Morris

Gore

(2002). Thinking about oneself and others: The relational-interdependent self-construal and social cognition. Journal of Personality and Social Psychology, 82, 399-418.

Gijsberts

Dagevos

(2005). Uit elkaars buurt: De invloed van etnische concentratie op integratie en beeldvorming [Love thy neighbor: Effects of ethnic concentration on integration and stereotyping]. Den Haag: Sociaal & Cultureel Planbureau.

Goffman

(1971). Relations in public. New York: Basic Books.

Hofstede

(2001). Culture’s consequences: Comparing values, behaviors, institutions, and organizations across nations. Thousand Oaks, CA: Sage Publications.

10.

Larsen

J. T.

McGraw

A. P.

Cacioppo

J. T.

(2001). Can people feel happy and sad at the same time? Journal of Personality and Social Psychology, 81, 684-696.

11.

Markus

Kitayama

(1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224-253.

12.

Miller

J. G.

Bersoff

D. M.

Harwood

R. L.

(1990). Perceptions of social responsibilities in India and in the United States: Moral imperatives or personal decisions? Journal of Personality and Social Psychology, 58, 33-47.

13.

Neyer

F. J.

Asendorpf

J. B.

(2001). Personality-relationship transition in young adulthood. Journal of Personality and Social Psychology, 81, 1190-1204.

14.

Nezlek

J. B.

(2001). Causal relationships between perceived social skills and day-to-day social interaction: Extending the sociometer hypothesis. Journal of Social and Personal Relationships, 18, 387-404.

15.

Nezlek

J. B.

(2003). Using multilevel random coefficient modeling to analyze social interaction diary data. Journal of Social and Personal Relationships, 20, 437-469.

16.

Nezlek

J. B.

(2010). Multilevel modeling and cross-cultural research. In Matsumoto

van de Vijver

A. J. R.

(Eds.), Cross-cultural research methods in psychology (pp. 299-347). Oxford, UK: Oxford University Press.

17.

Nezlek

J. B.

Schaafsma

(2010). Understanding the complexity of everyday interethnic contact: Recommendations for researchers. Social and Personality Psychology Compass, 4/10, 795-806.

18.

Raudenbush

S. W.

Bryk

A. S.

(2002). Hierarchical linear models (2nd ed.). Newbury Park, CA: Sage Publications.

19.

Raudenbush

S. W.

Bryk

A. S.

Cheong

Y. F.

Congdon

(2004). HLM 6: Hierarchical linear and nonlinear modeling. Lincolnwood, IL: Scientific Software International.

20.

Reis

H. T.

Gable

S. L.

(2000). Event sampling and other methods for studying daily experience. In Reis

H. T.

Judd

C. M.

(Eds.), Handbook of research methods in social and personality psychology (pp. 190-222). New York: Cambridge University Press.

21.

Schaafsma

Nezlek

J. B.

Krejtz

Safron

(2010). Ethnic identification and naturally occurring interethnic social interactions: Muslim minorities in Europe. European Journal of Social Psychology, 40, 1010-1028.

22.

Searle

Ward

(1990). The prediction of psychological and sociocultural adjustment during cross-cultural transitions. International Journal of Intercultural Relations, 14, 449-464.

23.

Singelis

T. M.

(1994). The measurement of independent and interdependent self-construals. Personality and Social Psychology Bulletin, 20, 580-591.

24.

Stapel

D. A.

Koomen

(2001). I, we, and the effect of others on me: How self-construal moderates social comparison effects. Journal of Personality and Social Psychology, 80, 766-781.

25.

Stephan

W. G.

Stephan

C. W.

(1985). Intergroup anxiety. Journal of Social Issues, 41, 157-175.

26.

Triandis

H. C.

(1989). The self and social behavior in different cultural contexts. Psychological Review, 96, 269-289.

27.

Van de Vijver

F. J. R.

Leung

(1997). Methods and data analysis for cross-cultural research. Newbury Park, CA: Sage Publications.

28.

Verkuyten

Kinket

(2000). Social distances in a multi ethnic society: The ethnic hierarchy among Dutch preadolescents. Social Psychology Quarterly, 63, 75-85.

29.

Ward

Chang

W. C.

(1997). “Cultural fit”: A new perspective on personality and sojourner adjustment. International Journal of Intercultural Relations, 21, 525-533.

30.

Wheeler

Nezlek

(1977). Sex differences in social participation. Journal of Personality and Social Psychology, 35, 742-754.

31.

Yang

R. P.

Noels

Saumure

K. D.

(2006). Multiple routes to cross-cultural adaptation for international students: Mapping the paths between self-construals, English language confidence, and adjustment. International Journal of Intercultural Relations, 30, 487-506.