Social Identity Complexity and Immigrants’ Attitude Toward the Host Nation

Abstract

Social identity complexity refers to individual differences in the interrelationships among multiple ingroup identities. The present research conducted in the Netherlands examines social identity complexity in relation to Muslim immigrants’ national identification and the attitude toward the host majority. Three studies are reported that focused on the interrelationship between ethnicity and religion and examined social identity complexity in different ways. Study 1 showed that lower social identity complexity is associated with lower national identification. Studies 2 and 3 examined the interaction between ethnic and religious group identification. For Muslim identifiers, higher ethnic identification was related to lower national identification and higher ingroup bias (Studies 2) and lower endorsement of national liberal practices (Study 3). In contrast, for those who did not strongly identify with Muslims, higher ethnic identification was associated with higher national identification, stronger endorsement of Dutch liberal practices, and more positive stereotypes about the Dutch outgroup (Study 3).

Keywords

identity complexity religion ethnicity immigrants

Social-psychological research is increasingly focusing on multiple identities. Individuals typically belong to many social groups and self-categorize in terms of multiple collective categories. Most immigrants¹ and ethnic minority group members struggle with the question of combining subgroup identities with commitments to the nation-state. For them, the prototypical case is that “an identity associated with the country of origin confronts the possibilities of a different American (or Canadian or French or other national) identity” (Deaux, 2006, p. 119). A growing number of studies examines the relationship between ethnic and national identification among immigrants and ethnic minorities (e.g., Citrin, Wong, & Duff, 2001; Fleischmann, 2011; Phinney, Berry, Vedder, & Liebkind, 2006; Sabatier, 2008; Staerklé, Sidanius, Green, & Molina, 2010).

Religious identity, however, has mostly been disregarded in this respect (Chaudhury & Miller, 2008; Sirin & Fine, 2008). This is unfortunate because religion is often of profound importance to people’s lives and religious groups are among the more salient buttresses of identity (Seul, 1999). Compared with identification with other social groups, religious group identification offers a distinctive “sacred” and unfalsifiable worldview, moral guidance for practices and behavioral choices, and “eternal” group membership (Ysseldyk, Matheson, & Anisman, 2010).

It is further unfortunate because questions of immigration and diversity are increasingly questions of religious diversity. In particular, Islam has emerged as the focus of immigration and diversity debates in Europe (Zolberg & Long, 1999) and is also increasingly discussed in the United States. Muslims have been defined as an “indigestible” minority (Huntington, 2004, p. 188), and it is argued that Islam forms a “bright boundary” separating immigrants from host societies (Alba, 2005). This is further illustrated by headscarf controversies in France and other countries, and by the national debates about Islamic schools and the place of other Islamic institutions, practices, and claims within the deeply embedded secularism of most liberal democracies.

For Muslim immigrants, ethnicity and religion might overlap to a large degree. What it means to be a member of their ethnic minority group can intersect with what it means to be a Muslim. One implication is that religious identity should be considered when focusing on the relationship between ethnic and national identification. Research on social identity complexity (Roccas & Brewer, 2002) suggests that the degree to which individuals combine or see their multiple identities as interrelated is associated with outgroup openness and tolerance. However, few studies have investigated the concept of social identity complexity empirically and no studies have investigated this concept among immigrant groups. Our research is one of the first attempts to examine the role of identity complexity for immigrants’ attitude toward the host nation, assessed with national identification and the attitude toward the native majority.

Our aim is to investigate whether the intersection between ethnic and religious identification is negatively associated to immigrants’ attitude toward the host society. The main prediction is that the interaction of relatively strong ethnic and strong religious group identifications implies a simplified or exclusive identity that is related to lower national identification and a less positive attitude toward the native majority. This prediction will be examined in three survey studies among Muslim immigrants of Turkish and Moroccan origin as members of the numerically two largest minority groups living in the Netherlands.

Social Identity Complexity

Social identity complexity refers to individual differences in the way in which different ingroup memberships are subjectively combined (Roccas & Brewer, 2002). An inclusive or complex identity structure implies that an individual accepts and acknowledges the distinctive memberships of his or her various ingroups. Alternatively, individuals with a relatively simplified structure perceive a strong overlap and interrelation among their identities. The concept of social identity complexity has similarities with the notion of intersectionality as put forward by critical race theorists (Cole, 2009). These theorists argue that social categories like race, gender, and social class often depend on one another for meaning. In particular, members of groups holding multiple disadvantaged statuses would experience these social categories in close association and simultaneously.

Similar to other West European countries, Dutch society and the majority group have become rather rejecting and assimilative in their orientation toward ethnic and cultural diversity. In public debates, assimilation has been proposed, and increasingly accepted, as the only viable option for a stable and cohesive Dutch society (Scholten & Holzhacker, 2009; Vasta, 2007). In addition, many immigrants are Muslims, and in the Netherlands, as in many Western countries, religious differences have come to the forefront with a focus on Islam. Islam has become the “negative other” and symbolic for problems related to ethnic minorities and immigration. Leading politicians have publicly described Islam as a “backward religion” that seriously threatens Dutch society, have defined Muslims as a “fifth column,” and argued for the need for a “cold war” against Islam (see Verkuyten & Zaremba, 2005). According to some commentators, there is an ongoing Dutch–Muslim cultural war (Scroggins, 2005).

From a social identity perspective, it can be argued that an emphasis on assimilation creates a distinctiveness threat to which minority members respond by reasserting their threatened minority identity (Brewer, 1991). Furthermore, research on the common ingroup identity model and on the mutual intergroup differentiation model has shown that neglecting a valued subgroup identity motivates higher levels of attachment to the subgroup (e.g., Dovidio, Gaertner, & Saguy, 2007; Hornsey & Hogg, 2000). Among immigrants, an emphasis on assimilation might also create uncertainty about oneself, others, and the social world. Uncertainty reduction theory argues that people feeling uncertain tend to identify more strongly with meaningful groups, like ethnic and religious ingroups (Castano, Yzerbyt, & Bourguignon, 2003; Hogg, 2000). In addition, minority group identification may be motivated by self-enhancement (Tajfel & Turner, 1979). Members of devalued groups can cope with identity threats by adopting group-based strategies involving increased ingroup identification and a distancing from the majority group. Threat perceptions can lead to increased self-stereotyping (Spears, Doosje, & Ellemers, 1997), and Roccas and Brewer (2002) argue that higher perceived threat lowers identity complexity. Evidence for this latter relationship is found in experimental research (Roccas & Brewer, 2002) and in two surveys in the context of Northern Ireland (Schmid, Hewstone, Tausch, Cairns, & Hughes, 2009). Based on these theories and findings, we expected relatively high levels of ethnic and Muslim identifications as well as a strong association between both identifications. For Muslim immigrants in the context of the Netherlands and Western Europe more generally, there probably is a relatively high degree of overlap between their meaningful and important ethnic and religious categories. The central question is whether this expected overlap, or low identity complexity, is associated with a more negative attitude toward the host nation.

Social Identity Complexity and National Attitudes

There are cognitive and motivational reasons why lower identity complexity should be associated with less openness and lower outgroup tolerance (Roccas & Brewer, 2002). Low social identity complexity means that multiple identities are embedded in a single ingroup representation making an individual who is an outgroup member on one dimension also an outgroup member on another dimension. This lack of cross-cutting identities increases the ingroup–outgroup distinctions and thereby strengthens the distancing from outgroup members and increases the cognitive basis of ingroup bias (Crisp & Hewstone, 2006). In addition, overlapping group memberships strengthen the motivational bases for intergroup differentiation because of the increased self-evaluative significance of social comparisons and the importance of any one social identity for a sense of belonging (Brewer, 1991; Tajfel & Turner, 1979). There is some empirical evidence for the proposition that lower social identity complexity is indeed associated with lower openness, less outgroup tolerance, and more intergroup bias (Brewer & Pierce, 2005; Miller, Brewer, & Arbuckle, 2009; Roccas & Brewer, 2002; Schmid et al., 2009). However, the relationship between identity complexity and immigrants’ national attitudes has not been examined. In addition, identity complexity has been assessed indirectly and has not been operationalized in terms of group identifications.

Roccas and Brewer (2002) argue that social identity complexity “involves understanding what people mean when they say that I am both ‘A’ and ‘B’” (p. 93). This description implies a construal of the self in which an individual feels himself or herself a member of multiple ingroups. However, they notice that semantics make it difficult to know what the “and” in this self-description actually refers to. Therefore, they propose a measurement procedure that indirectly assesses how an individual perceives the degree of overlap between his or her different ingroup memberships. This indirect approach does not focus on the self-concept but rather, first, on perceived overlap between members of the various ingroups and, second, on representations of the different ingroups. On one hand, overlap complexity refers to the extent to which the members of one ingroup (e.g., Turks) are perceived to simultaneously share membership of another ingroup category (e.g., Muslims). Similarity complexity, on the other hand, denotes the degree to which an individual considers different categories as interrelated, based on the content of group representations (e.g., perceiving that being a Turk means the same as being a Muslim). Low similarity complexity implies that ingroup members are seen as highly similar to each other or that the ingroups are perceived to share the same basic attributes and values.

In Study 1, we examined the relationship between a direct measure of perceived social identity complexity and national identification. In Studies 2 and 3, we examined the interaction of ethnic identification and religious group identification as an indicator of social identity complexity. Although overlap complexity and similarity complexity focus on ingroup members and ingroup representations, an assessment of different group identifications (ethnicity and religion) implies a focus on the sense of self. For Turkish Muslim immigrants, a monolithic (Kelman, 2001) identification pattern in which ethnic and religious group identifications are strong and strongly interconnected implies high subjective correspondence between minority group memberships. This correspondence indicates low identity complexity and can be expected to be related to relatively low national identification and a less positive attitude toward the national majority group. We will examine this expectation in Studies 2 and 3 by focusing on the multiplicative product term derived from ethnic and religious identifications (Cole, 2009). More specifically, higher ethnic minority identification was expected to be associated with lower national identification and a less positive attitude toward the Dutch majority, but only for individuals with higher religious identification.

Overview

The first study presented here focused on the relationship between perceived social identity complexity (ethnicity and religion) and national Dutch identification. Lower identity complexity was expected to be associated with lower national identification, over and above the effects of measures of perceived incompatibility and perceived group equality.

The second study examined identity complexity in terms of ethnic and religious group identifications. A strong relationship between Turkish and Muslim identification was expected. In addition, our main prediction was that ethnic identification will be negatively related to national identification when religious group identification is relatively high. Thus, ethnic and religious identification were expected to interact in predicting national identification. In addition and following previous research on social identity complexity, we focused on ingroup bias as an additional outcome variable (e.g., Brewer & Pierce, 2005; Schmid et al., 2009). Ethnic identification was expected to be associated with higher ingroup bias but only for individuals with higher religious identification.

The third study focused on Dutch national identification and the content of Dutch national identity. In addition to national identification, we examined whether identity complexity was related to immigrants’ endorsement of Dutch self-defining liberal practices and to stereotypes of the Dutch. Ethnic and religious identification were expected to interact in predicting national identification, the endorsement of liberal practices, and outgroup stereotypes. More specifically, only for individuals with higher religious identification, ethnic identification was expected to be associated with lower national identification, lower endorsement of liberal practices, and less positive outgroup stereotypes.

In Studies 2 and 3, the role of gender, age, and passport nationality were also considered, and in Study 3, educational level was included as an additional control variable. Within the Turkish–Dutch community, people have a Turkish passport, a Dutch passport, or dual citizenship, all of which are legally possible in the Netherlands. Having a Dutch passport (with or without a Turkish one) might have more instrumental than symbolic meanings. It allows people, for example, to vote in national elections and to travel freely within the European Union. However, one’s passport identity might also be related to ethnic and national identification processes, and therefore this factor was included in these studies.

Study 1

In examining the relationship between similarity complexity and national identification, we included perceived identity incompatibility and perceived group equality as additional predictors of national identification. The purpose of Study 1 was to examine whether similarity complexity is related to national identification, above and beyond any effect of incompatibility and perceived group equality. We controlled for these factors because they might be related to similarity complexity and to national identification and therefore might be responsible for any relationship found.

First, identity incompatibility refers to the extent to which immigrants perceive their ethnoreligious identities and the host national identity as incommensurable rather than complementary. The perception of identity incompatibility has been found to be associated with lower national identification (Mähönen, Jasinskaja-Lahti, & Liebkind, 2010; Sindic & Reicher, 2009), and because of its threatening, exclusionary implications, it might also be related to lower social identity complexity. In addition, acculturation research has shown that not only the perceptions and preferences of immigrants themselves but also their perceptions of what the majority group beliefs or wants are important (see Brown & Zagefka, 2011, for a review). These metacognitions or perceptions of outgroup beliefs can have an independent impact on the attitudes of immigrants. Therefore, we assessed not only the own perception of identity incompatibility but also the extent to which the majority group is perceived to consider ethnic and religious minority identities as being incompatible with Dutch national identity.

Second, research has shown that perceived group discrimination is related to stronger minority group identification (e.g., Fleischmann, 2011; Verkuyten & Yildiz, 2007). Furthermore, perceived discrimination discourages minorities from interacting with the majority and decreases national identification over time (Jasinskaja-Lahti, Liebkind, & Solheim, 2009). In contrast, openness of the host society, public regard, and equal treatment have been found to be associated with higher national identification (Verkuyten & Reijerse, 2008). These positive perceptions tell people that they are equal members of society and that society itself is valuable, and they might also be related to social identity complexity. Therefore, we took perceived group equality into account for examining the relationship between similarity complexity and national identification.

Method

Sample

There were 131 Sunni Muslim adolescents who participated in the study.² The participants were between 15 and 18 years of age (M = 16.67, SD = .98), and 52.8% was female and 47.4% male. Of the participants, 48% had a Moroccan background and 62% a Turkish background. All participants were born in the Netherlands and lived in the city of Rotterdam. The adolescents who agreed to participate in the research filled in a questionnaire in Dutch.

Measures

National identification was measured with four items and using 7-point scales (1 = not al all, 7 = to a great extent). The participants were asked to indicate to what extent they feel Dutch, to what extent being Dutch is important to them, to what extent they feel strong ties with other Dutch people, and whether they feel personally criticized when someone who is not Dutch criticizes Dutch people. Cronbach’s alpha for these four items was .80.

Social identity complexity was measured with a single item that was comparable to the ones used by Schmid et al. (2009). On a 7-point scale, participants were asked “To what extent is it necessary for a person of your own ethnic group to also be a member of your own religious group?” A higher score on this question stands for lower identity complexity.

For identity incompatibility, two questions were asked: “To what extent do you feel that being Dutch is incompatible with being a member of your ethnic group?” and “To what extent do you feel that being Dutch is incompatible with being a member of your religious group?” Both questions were positively correlated (α = .59), and a mean score was computed with a higher score indicating higher incompatibility. For measuring the metaperception of identity incompatibility by the Dutch majority group, two similar questions were asked with the same 7-point scales: “To what extent do you think that the majority of people in the Netherlands feel that being Dutch is incompatible with being a member of your ethnic group” and “To what extent do you think that the majority of people in the Netherlands feel that being Dutch is incompatible with being a member of your religious group.” A higher score indicates higher metaperception of incompatibility (α = .65).

Five items (7-point scales) were used for measuring perceived group equality. These items were “People from my own ethnic group are treated fairly by officials in the Netherlands,” “People from my own ethnic group have the same opportunities as everyone else to succeed in the Netherlands,” “People from my own ethnic group have equal opportunity to do well in education in the Netherlands,” “People from my own ethnic group are free to express their political and religious views in the Netherlands,” and “People from my own ethnic group are free to live according to my group’s values and traditions in the Netherlands.” Cronbach’s alpha for these five items was .83, and a higher score indicates higher perceived group equality.

Results

Mean scores and intercorrelations

The mean scores and correlations are shown in Table 1. The mean score for national identification was at the neutral midpoint of the scale. The mean score for social identity complexity was a little above the midpoint of the scale, and the scores for the incompatibility questions were below the midpoint of the scales. Identity incompatibility as perceived by the participants themselves was significantly lower than the incompatibility that they thought the Dutch majority perceived, t(131) = 5.04, p < .001. Thus, participants considered their ethnic and religious identities as less incompatible with Dutch identity compared with the identity incompatibility that they believed that the Dutch majority perceived.

Table 1.

Means, Standard Deviations, and Intercorrelations for the Different Measures, Study 1

	1	2	3	4	M	SD
1. Similarity complexity	—				4.11	2.16
2. Incompatibility	.22*	—			2.96	1.66
3. Meta incompatibility	.11	.39**	—		3.77	1.51
4. Group equality	−.08	−.14	−.08	—	4.10	1.42
5. National identification	−.25**	−.19*	.05	.26**	3.85	1.48

p < .01. **p < .001.

The two incompatibility measures were significantly correlated, and participant’s perception of identity incompatibility was positively related to social identity complexity and negatively to national identification. Higher perception of group equality was associated with higher national identification. Lower identity complexity was related to lower national identification.

National identification

To examine these relationships more closely, national identification was subjected to a regression analysis with social identity complexity, the two perceptions of incompatibility, and perceived group equality as predictors. The model explains 12% of the variance, F(4, 131) = 3.76, p = 007. The two incompatibility measures were not independently related to national identification (ps > .10). However, higher perceived group equality in the Netherlands was associated with stronger national identification, β = .22, t = 2.39, p = .019. In addition, social identity complexity was significantly associated with national identification, β = −.19, t = 2.12, p = .036. The more the participants believed it was necessary for their ethnic group to be Muslim, the less they identified with Dutch identity.

These results indicate that the participants’ representation of the relationship between their ethnicity and their religion is a critical factor for their strength of identification with the national category. The more strongly they believed that their ethnicity and religion should be interconnected, the less they identified with the Dutch group. In other words, national identification was lower when they more strongly felt that it is necessary for a person of their ethnicity also to be a Muslim.

Study 2

In Study 1 and following previous research, we examined social identity complexity in terms of perceived differentiation. Another approach for examining the subjective intersection of both identities is to focus on the multiplicative combination of a person’s own psychological identification with the two groups. For Turkish Muslim immigrants, the interaction between a strong ethnic identification and a strong Muslim identification indicates a relatively simple identity structure: a structure in which both identities are more strongly intersected than when the two identifications are relatively independent. Therefore, in Study 2, we examined the role of social identity complexity for national identification by focusing on the product term derived from ethnic and religious group identification. Our main prediction was that ethnic identification will be negatively related to national identification when religious group identification is relatively high. In addition, we focused on ingroup bias as an outcome variable. The reason is that we wanted to investigate whether the use of both identifications for assessing social identity complexity replicates previous findings for the relationship between bias and overlap and similarity complexity (Brewer & Pierce, 2005; Schmid et al., 2009). A similar result would further support the validity of our approach. We predicted that ethnic identification interacts with religious identification in affecting ingroup bias. More specifically, ethnic identification was expected to be associated with higher ingroup bias but only for individuals with stronger religious identification.

Method

Sample

On an open-ended question, 204 participants of Turkish origin described themselves as Muslim (Sunni). All participants had a father and mother of Turkish background and were born in the Netherlands or had moved to this country more than 15 years ago. Of the participants, 30% were women and 70% were men. They were between 18 and 66 years of age, and their mean age was 30.2 (SD = 9.44). Of the participants, 32.7% had Turkish nationality and 67.3% had Dutch nationality or dual citizenship. The participants came from various cities in the middle of the Netherlands and were recruited through local contacts and associations. The questionnaire was in Dutch.

Measures

Turkish group identification was assessed by asking the participants to respond to six items (7-point scales) that were taken from previous studies in the Netherlands (see Verkuyten, 2005). These items measure the importance and value attached to one’s ethnic background and are similar to the items on Luhtanen and Crocker’s (1992) identity and private subscales. The six-item scale was internally consistent with a Cronbach’s alpha equal to .92.

Muslim group identification was assessed by items similar to those used for Turkish identification. The six items have been used in previous studies in the Netherlands (Verkuyten, 2007; Verkuyten & Yildiz, 2009). The items were “My Muslim identity is an important part of my self,” “I identify strongly with Muslims,” “I feel a strong attachment to Muslims,” “Being a Muslim is a very important part of how I see myself,” “I am proud of my Islamic background,” and “I feel a strong sense of belonging to Islam.” Cronbach’s alpha was .93.

Dutch group identification was measured using five items (7-point scales). Three sample items are “I identify with the Dutch,” “I feel Dutch,” and “I feel connected to the Dutch.” The alpha for these five items was .85.

Ingroup bias was measured using the well-known feeling thermometers. The participants were asked to indicate their overall feeling toward their own minority group (Turks) as well as toward the majority group (Dutch) on a scale ranging from 0 (extremely unfavorable) to 100 (extremely favorable). An index of ingroup bias was obtained by subtracting the outgroup rating from the ingroup rating. Higher scores indicate a higher degree of bias in favor of the minority ingroup.

Results

Mean scores and intercorrelations

Table 2 shows the mean scores and the correlations between the four measures. Similar to Study 1, the mean score for national identification was at the neutral midpoint of the scale, t(204) = −1.73, p > .05. The mean scores for ethnic identification and for religious identification were significantly above the neutral midpoint of the scales, t(204) = 12.07, p < .001 and t(204) = 14.69, p < .001. In addition, the participants showed ingroup bias, t(204) = 11.24, p < .001.

Table 2.

Means, Standard Deviations, and Intercorrelations for the Different Measures, Study 2

	1	2	3	M	SD
1. Ethnic identification	—			5.68	1.65
2. Religious identification	.75**	—		5.72	1.91
3. National identification	−.17*	−.13	—	3.86	1.68
4. Ingroup bias	.48**	.45**	−.38**	23.89	31.2

p < .01. **p < .001.

The correlations in Table 2 show that ethnic and religious identification were positively and strongly related. Ethnic identification was negatively related to national identification, and higher religious identification was marginally significantly related to lower national identification (p = .056). In addition, higher ethnic as well as religious identification were associated with higher ingroup bias, whereas national identification was associated with lower bias.

National identification

Because ethnic and religious identification were strongly correlated and we wanted to examine the effect of their interaction, we first examined whether there is a problem of multicollinearity. Using centered scores, the interaction term was significantly associated with ethnic identification (r = .61, p < .001) and with religious identification (r = .56, p < .001). A common method to detect multicollinearity uses the tolerance value and the variance inflation factor (VIF). According to Myers (1990) and Field (2000), a Tolerance value below 0.1 and a VIF value greater than 10 indicate a serious problem of multicollinearity. With our data, the lowest Tolerance value is .31 and the highest VIF statistic is 3.28. Thus, there is no problematic multicollinearity between the variables.

National identification was subjected to a hierarchical linear regression with the centered scores of ethnic identification and religious identification (Step 2) and their interaction (Step 3) as predictors. In Step 1, gender, age, and nationality were included as additional predictors. In Step 1, only nationality had a significant effect with participants having the (dual) Dutch nationality scoring higher on national identification compared with participants with Turkish nationality, β = .22, t = 3.19, p = .002. In Step 2, ethnic identification and religious identification were not significantly related to national identification (ps > .05). In Step 3, the interaction between both identifications was significant, β = −.30, t = −3.09, p = .002, and significantly increased the variance explained (ΔR² = .04, p = .004).³

As predicted, analysis of the simple slopes (Aiken & West, 1991) showed that ethnic identification negatively and significantly predicted national identification at one standard deviation above the mean of religious identification, β = −.63, t = −3.49, p = .001. At the mean of religious identification, there was also a negative but weaker association between ethnic and national identification, β = −.31, t = −2.51, p = .013. At one standard deviation below the mean of religious identification, there was no significant association between ethnic and national identification, β = .16, t = 1.74, p > .05. This pattern of findings indicates that there is a gradual decrease in the association between ethnic and national identification depending on the level of religious identification. The interaction for relatively high and low religious identification is shown in Figure 1.

Figure 1.

Associations between ethnic and national identification at religious group identification scores of mean ± 1 standard deviation, Study 2

Ingroup bias

Using the same regression procedure, age was negatively related to ingroup bias in Step 1, β = −.17, t = 2.41, p = .017, whereas gender and nationality had no significant effects. In Step 2, both ethnic and religious identification were positively associated with ingroup bias, β = .32, t = 3.62, p < .001 and β = .20, t = 2.25, p = .026. As predicted, these two associations were qualified in Step 3 by the significant interaction effect between ethnic and religious identification, β = .24, t = 2.55, p = .012. The full model accounted for 27% of the variance in ingroup bias.⁴

Simple slopes analysis showed that higher ethnic identification was strongly related to stronger ingroup bias at one standard deviation above the mean of religious identification, β = .66, t = 4.17, p < .001. At one standard deviation below the mean of religious identification, the association between ethnic and national identification was also significant but less strong, β = .23, t = 2.94, p < .01. At the mean of religious identification, the association was also positive β = .44, t = 3.89, p < .001. This pattern of findings indicates that depending on the level of religious identification, there is a gradual decrease in the association between ethnic identification and ingroup bias.

These findings show that the intersection of ethnic and religious identification is associated with lower national identification and higher ingroup bias. National identification is lower for participants with stronger ethnic and Muslim group identification, and these participants more strongly favor their Turkish ingroup over the Dutch outgroup. This further indicates that among immigrant groups, low social identity complexity can result in distancing from the host society.

Study 3

A third study was conducted to examine whether these findings were reliable and could be generalized to another sample of Turkish Muslim immigrants. Again, we expected that ethnic identification will be negatively related to national identification when religious group identification is relatively high. In addition, we focused on the content of the national identity and on outgroup stereotypes as two additional outcomes variables. First, the conceptualization of similarity complexity refers to the content of group representations. Social identity complexity is low when individuals see their ingroups as sharing the same basic attributes and values. This similarity in identity content can be expected to make the ethnoreligious minority identity rather difficult to reconcile with some of the self-defining values and practices of the national majority group. The Netherlands was the first country in the world that legalized euthanasia and allowed gay marriage, and the country has been a strong defender of the women’s right on abortion. For many native Dutch, the legalization of these practices forms the expression of Dutch liberal identity and culture and represents a continuation of traditional Dutch tolerance (Jaspers, 2008). We examined the endorsement of these national self-defining practices, and we expected for high but not for low Muslim identifiers that ethnic identification is negatively associated to the endorsement of these practices.

Second, in Study 3, we focused on outgroup stereotypes rather than affective ingroup bias. Stereotype ratings assess particular contents, and the assessment of these contents might differ from warmth-like feelings. A similar finding for outgroup stereotypes as for thermometer-like feelings would further support the generality of the finding that the intersection between ethnic and religious group identification is associated with less openness and acceptance of the national majority outgroup. We predicted that ethnic identification interacts with religious identification in affecting the stereotypical evaluation of the Dutch. More specifically, ethnic identification was expected to be associated with less positive stereotypes, but only for individuals with higher religious identification.

In Study 3, educational level was considered as an additional individual background characteristic. Information on the level of education was not collected in Study 2, but it is likely that education reduces ethnic and religious identification and increases the orientation of Turkish–Dutch people toward Dutch society, which may lead to a stronger commitment to the nation-state.

Method

Sample

On an open-ended question, 249 participants of Turkish origin described themselves as Muslim (Sunni). All participants had a father and mother of Turkish background and were born in the Netherlands or had moved to this country more than 15 years ago. Of the participants, 44.6% were women and 53.4% were men. They were between 18 and 70 years of age, and their mean age was 38.8 (SD = 13.4). Of the participants, 63.8% had Turkish nationality, 10.5% had Dutch nationality, and 25.7% had dual citizenship. The participants came from various cities in the middle and eastern part of the Netherlands and were, again, recruited through local contacts and associations. The questionnaire was in Dutch.

Measures

Turkish group identification as well as religious and national identification were assessed by the same items and scales used in Study 2. For this sample, the alphas were .89, .90, and .81, respectively.

For assessing outgroup stereotypes, the participants were presented with nine positive stereotypical traits, partly taken form Leach, Ellemers, and Barreto (2007). They were asked to indicate the extent to which the native Dutch were honest, reliable, trustworthy, competent, efficient, productive, friendly, hospitable, and warm. Responses were given by using 7-point scales, ranging from “not at all” to “very much.” Maximum likelihood estimation with oblique rotation was used to determine the underlying dimensions of the stereotype ratings. A clear one-factor structure emerged, and this factor explained 58.1% of the variance.⁵ The alpha for the nine traits was .92, and a higher score indicated more positive outgroup stereotypes.

The endorsement of national liberal values was assessed with three items. These items refer to three practices that were first legalized in the Netherlands and that are considered to represent Dutch liberal values. On 7-point scales, the participants were asked whether they agreed with euthanasia, abortion, and gay marriage as acceptable and legalized practices. Alpha for these three items was .71.

Results

Mean scores and intercorrelations

Table 3 shows the means, standard deviations, and intercorrelations for the different measures. Similar to Studies 1 and 2, the mean score for national identification was at the neutral midpoint of the scale. The mean scores for outgroup stereotypes and for the endorsement of national liberal values were also around the midpoint of the scales. The mean scores for ethnic identification and for religious identification were significantly above the neutral midpoint of the scales, t(249) = 2.46, p = .014 and t(249) = 4.03, p < .001.

Table 3.

Means, Standard Deviations, and Intercorrelations for the Different Measures, Study 3

	1	2	3	4	M	SD
1. Ethnic identification	—				4.22	1.42
2. Religious identification	.72**	—			4.34	1.35
3. National identification	.19*	.23**	—		3.96	1.23
4. Outgroup stereotypes	.24**	.23**	.38**	—	4.19	1.34
5. Liberal practices	.13*	.08	.45**	.07	3.94	1.29

p < .01. **p < .001.

Similar to Study 2, the correlations show that ethnic and religious identification were strongly and positively related (Table 3). In addition and in contrast to Study 2, ethnic and religious identification were positively related to national identification and to the stereotypical evaluation of the Dutch. Higher ethnic identification, but not religious identification, was associated with a stronger endorsement of the liberal practices. Dutch identification and the stereotype ratings were positively associated, but the two measures were relatively independent because they shared 14% of their variance. In addition, the endorsement of the liberal practices was positively associated with national identification but not with outgroup stereotypes.

Considering the strong correlation between ethnic and religious identification, we also examined the associations of these two measures with their interaction term (centered scores). The interaction term was significantly but not strongly associated with ethnic identification (r = .24, p < .01) and with religious identification (r = .26, p < .01). In addition, the lowest tolerance value is .44 and the highest VIF statistic is 2.39, indicating that there is no problem of multicollinearity between the variables (Field, 2000; Myers, 1990).⁶

National identification

Using the same steps as in Study 2, national identification was subjected to a hierarchical linear regression with the centered scores of ethnic identification and religious identification (Step 2) and their interaction (Step 3) as predictors. In Step 1, gender, age, nationality, and educational level were included as additional predictors. Education and nationality had a significant and positive effect on national identification (Table 4). In Step 2, ethnic identification was not significantly related to national identification, whereas stronger religious identification was a significant predictor. Similar to Study 2, in Step 3, the interaction between both identifications was significant and significantly increased the variance explained.

Table 4.

Hierachical Regression Analyses With National Identification, Liberal Practices, and Outgroup Stereotypes as Dependent Variables: Standardized Regression Coefficients (Beta), Study 3

	National identification			Liberal practices			Outgroup stereotypes
	Step 1	Step 2	Step 3	Step 1	Step 2	Step 3	Step 1	Step 2	Step 3
Gender	.01	.01	.01	−.03	−.02	−.01	.03	.03	.03
Age	.05	.01	.01	.06	.05	.03	.09	.05	.04
Dutch nationality	.16*	.09	.14*	.11	.11	.16*	.12	.08	.11
Education	.19*	.13	.07	.18*	.16*	.11	.16*	.10	.06
Ethnic identification		.04	.04		.12	.11		.15	.15
Religious identification		.22*	.10		−.08	−.20*		.08	.02
Ethnic × Religious identification			−.35***			−.37***			−.20**
R² change	.07	.05	.12	.05	.01	.12	.04	.03	.03
F change	4.05**	6.67**	32.72***	3.07*	0.88	32.14***	2.55*	4.56*	8.25**

p < .05. **p < .01. ***p < .001.

As expected, analysis of the simple slopes, again, showed that ethnic identification negatively and significantly predicted national identification at one standard deviation above the mean of religious identification, β = −.30, t = −2.80, p = .006. At one standard deviation below the mean of religious identification, there was a significant positive association between ethnic and national identification, β = .29, t = 2.65, p = .009. In addition, there was no significant association between ethnic and national identification at the mean of religious identification β = .04, t = 0,76, p > .010. This pattern of findings indicates that depending on religious identification, there is a gradual change in the association between ethnic and national identification. The interaction for relatively low and high religious identification is shown in Figure 2.

Figure 2.

Associations between ethnic and national identification at religious group identification scores of mean ± 1 standard deviation, Study 3

Liberal practices

We conducted another hierarchical regression analysis, using the same steps to predict the endorsement of Dutch liberal practices. In Step 1, the higher educated endorsed the liberal practices more strongly than the lower educated (Table 4). Gender, age, and nationality had no independent significant effects. Ethnic and religious identification did not predict liberal practices in Step 2. Importantly, the interaction between ethnic and religious identification was significant in Step 3.

As predicted, analysis of the simple slopes showed that ethnic identification was negatively related to the endorsement of liberal practices at one standard deviation above the mean of religious identification, β = −.24, t = −2.20, p = .029. In contrast, ethnic identification was positively related to the endorsement of these practices at one standard deviation below the mean of religious identification, β = .32, t = 3.12, p = .002. In addition, there was no significant association between ethnic and national identification at the mean of religious identification, β = .11, t = 1.24, p > .05. In other words, for higher religious identifiers, stronger ethnic identification was related to lower endorsement of liberal practices, whereas for lower religious identifiers, there was a positive association between ethnic identification and liberal practices.

Outgroup stereotypes

We conducted a further hierarchical regression analysis, using the same steps to predict stereotypical evaluations of the Dutch majority group. In Step 1, the higher educated were more positive toward the Dutch than the lower educated. Gender, age, and nationality had no independent significant effects (Table 4). Ethnic and religious identification were not significantly related to outgroup stereotypes in Step 2. Importantly and as predicted, the interaction between ethnic and religious identification was significant in Step 3.

Analysis of the simple slopes showed that ethnic identification was negatively but not significantly related to outgroup stereotypes at one standard deviation above the mean of religious identification, β = −.13, t = −.92, p > .05. In contrast, ethnic identification was positively related to stereotypical evaluations of the Dutch at one standard deviation below the mean of religious identification, β = .32, t = 2.85, p = .005. The association between ethnic identification and outgroup stereotypes was marginally significant at the mean of religious identification, β = .15, t = 1.84, p = .068. In other words, for higher religious identifiers, stronger ethnic identification tended to be related to a less positive evaluation of the Dutch, whereas for lower religious identifiers, there was a positive association between ethnic identification and outgroup stereotypes.

The findings of Study 3 further support the role of identity complexity for immigrants’ attitudes toward the host nation. The mean levels of ethnic and religious identification were high, and both identifications were strongly associated. Furthermore, for Muslim identifiers, higher ethnic identification was related to lower national identification and lower endorsement of national liberal practices. In contrast, for those who did not strongly identify with Muslims, higher ethnic identification was associated with higher national identification, stronger endorsement of Dutch liberal practices, and more positive stereotypes about the Dutch outgroup.

General Discussion

Our focus on immigrants highlights the importance of investigating multiple identities and ethnic, religious, and national identification in particular. Examining the relationships between these identifications adds significantly to an understanding of acculturation and adaptation processes. Our research is the first attempt to examine the role of identity complexity in immigrants’ attitude toward the host society. Although research tends to focus on the relationship between ethnic and national identification (Phinney et al., 2006; Staerklé et al., 2010), we have tried to show that religious group identification can be important to consider. There is a relatively high degree of overlap between ethnic and religious categories among Muslim immigrants in the Netherlands and in other West European countries (Fleischmann, 2011; Modood et al., 1997; Verkuyten & Yildiz, 2007). In these countries, what it means to be a Turk, Moroccan, or Pakistani often intersects with what it means to be a Muslim. When these identities are subjectively combined, there is a relatively simplified and exclusive identity structure. As a result, host nationals, who are perceived to be outgroup members on one group dimension (ethnicity) are also perceived to be outgroup members on the other dimension (religion). This increases the importance of the ingroup in intergroup comparisons and the motivation to favor one’s own group.

The findings of the three studies show that lower social identity complexity was associated with a more negative attitude toward the host society. For the three different samples, national identification was lower among participants with relatively high social identity complexity (Study 1) and with strong ethnic and religious group identification (Studies 2 and 3). In contrast, ethnic identification was positively associated (Study 3) or not associated (Study 2) with national identification for lower religious identifiers. This pattern of findings contributes to an understanding of the conditions under which ethnic and national identifications are compatible or rather contradictory. Previous research has shown that the particular national context affects this relationship (e.g., Phinney et al., 2006; Staerklé et al., 2010) and also that perceived discrimination (Verkuyten & Yildiz, 2007) and the ability to live by one’s minority identity (Sindic & Reicher, 2009) are important conditions. The current findings add to this by showing that the subjective combination of minority identities is important. Muslim immigrants who identify strongly with their ethnic and religious ingroups tend to distance themselves more from the host society than immigrants with a more complex identity structure. Furthermore, social identity complexity was found to be associated with the attitude toward national identity content, ingroup bias, and stereotypes of the majority group. Participants with simpler identities endorsed Dutch self-defining liberal practices less, showed higher ingroup feeling bias, and tended to have less positive stereotypes about the Dutch.

The results of the three studies indicate that the interaction of strong ethnic and religious identification is a useful indicator of social identity complexity among Muslim immigrants. The findings for national identification were the same in Study 1 in which social identity complexity was measured in terms of beliefs about social category membership and overlap (Roccas & Brewer, 2002), and in Studies 2 and 3 in which we used the multiplicative product term of the individual’s own group identifications. In addition, in Studies 2 and 3, we successfully replicated the results of previous survey research showing a relationship between social identity complexity and intergroup feeling bias and outgroup attitudes (e.g., Brewer & Peirce, 2005; Miller et al., 2009; Schmid et al., 2009).

The similar findings in Studies 2 and 3 further supports the generality of the role of social identity complexity for immigrant’s national identification and outgroup attitudes. This is because the two samples were not only similar in level of national identification and in the associations between ethnic and religious identification but also differed in notable ways. For example, the simple correlations between ethnic and religious identification, on one hand, and national identification, on the other hand, were negative in Study 2 and positive in Study 3. In addition, for lower religious identifiers, ethnic identification was positively related to national identification in both studies but the association was only significant in Study 3. A possible reason for these diverging findings is that the sample in Study 2 is more traditional and religious orthodox than the one in Study 3. In Study 2, the mean levels of ethnic and religious identification were clearly higher than in Study 3. For Turkish Muslims in the Netherlands, a stronger commitment to their ethnic and religious group might make it more difficult to develop a sense of belonging to the host nation (Verkuyten & Yildiz, 2007). In addition, in Study 2, relatively low religious identification will still mean a sense of religious group commitment, which might explain why for the lower religious identifiers ethnic identification was not significantly related to national identification. So there are notable differences between the two samples, but despite these differences, the findings for the interaction between ethnic and religious identification are similar. This supports the construct of social identity complexity and further indicates that the multiplicative combination of group identifications can offer a useful measure of the degree to which people’s subjective sense of their collective self has a simple and exclusive or rather a more complex and inclusive structure.

Future research should examine whether the interaction of different group identifications is a useful indicator of social identity complexity among other groups and other social identities. We focused on two categories known to be central to many Muslim immigrants in Western Europe: ethnicity and religion. A combination of two strong identities is probably less indicative of social identity complexity for categories that relate to very different social domains, such as one’s profession and one’s political party membership. Furthermore, one may also consider using different dimensions of group identification (see Ashmore, Deaux, & McLaughlin-Volpe, 2004) and test, for example, whether some of these dimensions (e.g., sense of interdependence and mutual fate) are better indicators of social identity complexity than others (e.g., importance). Another limitation that should be acknowledged is that we cannot draw causal inferences because the three studies were cross-sectional. In principle, it is possible that the more strongly immigrants’ identify with the host society, the less strong the relationship between their ethnic and religious group identification becomes. Experimental and longitudinal designs are needed to examine the causal order. There are a few studies that have used a longitudinal design and that have shown, for example, that ethnic identification and perceived discrimination affect national identification, rather than vice versa (e.g., Jasinskaje-Lahti et al., 2009). However, these studies were not concerned with social identity complexity.

In conclusion, the three studies presented in this article make a valuable contribution to the growing body of research on multiple identities and on social identity complexity in particular. The findings add to this research by using a different approach for assessing social identity complexity and by focusing on immigrant’s national identification and attitudes. In addition, the research makes a contribution to the understanding of immigrants’ adaptation processes. It shows that it is important to go beyond the association between ethnic and national identification by including other category memberships, like religion, and by examining the ways in which these category memberships are subjectively interconnected. This allows us to better understand when and why a sense of belonging toward one’s ethnic group and toward the host nation are perceived and experienced as complementary or rather contradictory. Given the findings, we believe that it is useful and important that future studies investigate other immigrant groups in other countries. In doing so, different approaches to the assessment of social identity complexity can be used, and various antecedents and consequences of identity complexity should be considered.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This research was funded by the Netherlands Organization for Scientific Research (NWO). The first author received a grant from the NWO Conflict and Security program.

Notes

References

Aiken

L. S.

West

S. G.

(1991). Testing and interpreting interactions. Newbury Park, CA: SAGE.

Alba

(2005). Bright vs. blurred boundaries: Second-generation assimilation and exclusion in France, Germany and the United States. Ethnic and Racial Studies, 28, 20-49.

Ashmore

R. D.

Deaux

McLaughlin-Volpe

(2004). An organizing framework for collective identity: Articulation and significance of multidimensionality. Psychological Bulletin, 130, 80-114.

Brewer

M. B.

(1991). The social self: On being the same and different at the same time. Personality and Social Psychology Bulletin, 17, 475-482.

Brewer

M. B.

Pierce

K. P.

(2005). Social identity complexity and outgroup tolerance. Personality and Social Psychology Bulletin, 31, 428-437.

Brown

Zagefka

(2011). The dynamics of acculturation: An intergroup perspective. Advances in Experimental Social Psychology, 44, 129-184. doi:10.1016/B978-0-12-385522-0.000003-2

Castano

Yzerbyt

Bourguignon

(2003). We are one and I like it: The impact of ingroup entitativity on ingroup identification. European Journal of Social Psychology, 33, 735-754.

Chaudhury

S. R.

Miller

(2008). Religious identity formation among Bangladeshi American Muslim adolescents. Journal of Adolescent Research, 23, 383-410.

Citrin

Wong

Duff

(2001). The meaning of American national identity: Patterns of ethnic conflict and consensus. In Ashmore

R. J.

Jussim

Wilder

(Eds.), Social identity, intergroup conflict and conflict reduction (pp. 71-100). New York, NY: Oxford University Press.

10.

Cole

E. R.

(2009). Intersectionality and research in psychology. American Psychologist, 64, 170-180.

11.

Crisp

R. J.

Hewstone

(Eds.). (2006). Multiple social categorization: Processes, models and applications. New York, NY: Psychology Press.

12.

Deaux

(2006). To be an immigrant. New York, NY: Russell Sage.

13.

Dovidio

J. F.

Gaertner

S. L.

Saguy

(2007). Another view of “we”: Majority and minority group perspectives on a common in-group identity. European Review of Social Psychology, 18, 296-330.

14.

El-Dereny

Rashwan

N. I.

(2011). Solving multicollinearity problem using ridge regression models. International Journal of Contemporary Mathematical Sciences, 6, 585-600.

15.

Field

(2000). Discovering statistics using SPSS. London, England: SAGE.

16.

Fiske

S. T.

Cuddy

A. J.

Glick

(2002). A model of (often mixed) stereotype content: Competence and warmth respectively follow from perceived status and competition. Journal of Personality and Social Psychology, 82, 878-902.

17.

Fleischmann

(2011). Second-generation Muslims in European societies: Comparative perspectives on education and religion. Utrecht, Netherlands: Ercomer.

18.

Hoerl

A. E.

Kennard

R. W.

(1970). Ridge regression: Biased estimation for nonorthogonal problems. Technometrics, 42, 55-67.

19.

Hogg

M. A.

(2000). Subjective uncertainty reduction through self-categorization: A motivational theory of social identity processes. European Review of Social Psychology, 11, 223-255.

20.

Hornsey

M. J.

Hogg

M. A.

(2000). Subgroup relations: A comparison of mutual intergroup differentiation and common ingroup identity models of prejudice reduction. Personality and Social Psychology Bulletin, 26, 242-256.

21.

Huntington

S. P.

(2004). Who are we: The challenges to America’s national identity. New York, NY: Simon & Schuster.

22.

Jasinskaja-Lahti

Liebkind

Solheim

(2009). To identity or not to identity? National disidentification as an alternative reaction to perceived ethnic discrimination. Applied Psychology, 59, 105-128.

23.

Jaspers

(2008). Intolerance over time: Macro- and micro-level questions on attitudes towards euthanasia, homosexuality and ethnic minorities. Nijmegen, Netherlands: ICS.

24.

Kelman

H. C.

(2001). The role of national identity in conflict resolution: Experiences from Israeli-Palestinian problem-solving workshops. In Ashmore

R. J.

Jussim

Wilder

(Eds.), Social identity, intergroup conflict and conflict reduction (pp. 187-212). New York, NY: Oxford University Press.

25.

Leach

C. W.

Ellemers

Barreto

(2007). Group virtue: The importance of morality (vs. competence and sociability) in the positive evaluation of ingroups. Journal of Personality and Social Psychology, 93, 234-279.

26.

Luhtanen

Crocker

(1992). A Collective Self-Esteem Scale: Self-evaluation of one’s social identity. Personality and Social Psychology Bulletin, 18, 302-318.

27.

Mähönen

T. A.

Jasinskaja-Lahti

Liebkind

(2010). Cultural discordance and the polarization of identities. Group Processes & Intergroup Relations. Advance online publication. doi:10.1177/136843021037900610.1177/1368430210379006

28.

Miller

K. P.

Brewer

M. B.

Arbuckle

N. L.

(2009). Social identity complexity: Its correlates and antecedents. Group Processes & Intergroup Relations, 12, 79-94.

29.

Modood

Berthoud

Lakey

Nazroo

Smith

Bvirdee

Beishon

(1997). Ethnic minorities in Britain: Diversity and disadvantage. London, England: Policy Studies Institute.

30.

Myers

(1990). Classical and modern regression with applications (2nd ed.). Boston, MA: Duxbury.

31.

Phinney

J. S.

Berry

J. W.

Vedder

Liebkind

(2006). The acculturation experience: Attitudes, identities, and behaviors of immigrant youth. In Berry

J. W.

Phinney

J. S.

Sam

D. L.

Vedder

(Eds.), Immigrant youth in cultural transition: Acculturation, identity, and adaptation across national contexts (pp. 71-116). Mahwah, NJ: Lawrence Erlbaum.

32.

Roccas

Brewer

(2002). Social identity complexity. Personality and Social Psychology Review, 6, 88-109.

33.

Sabatier

(2008). Ethnic and national identity among second-generation immigrant adolescents in France: The role of social context and family. Journal of Adolescence, 31, 185-205.

34.

Schmid

Hewstone

Tausch

Cairns

Hughes

(2009). Antecedents and consequences of social identity complexity: Intergroup contact, distinctiveness threat, and outgroup attitudes. Personality and Social Psychology Bulletin, 35, 1085-1098.

35.

Scholten

Holzhacker

(2009). Bonding, bridging and ethnic minorities in the Netherlands: Changing discourse in a changing nation. Nations and Nationalism, 15, 81-100.

36.

Scroggins

(2005, June 27). The Dutch-Muslim cultural war. The Nation, pp. 21-25.

37.

Seul

J. R.

(1999). “Ours is the way of God”: Religion, identity, and intergroup conflict. Journal of Peace Research, 36, 553-569.

38.

Sindic

Reicher

S. D.

(2009). ‘Our way of life is worth defending’: Testing a model of attitudes towards superordinate group membership through a study of Scots’ attitudes towards Britain. European Journal of Social Psychology, 39, 114-129.

39.

Sirin

S. R.

Fine

(2008). Muslim American youth: Understanding hyphenated identities through multiple methods. New York, NY: New York University Press.

40.

Spears

Doosje

Ellemers

(1997). Self-stereotyping in the face of threats to group status and distinctiveness: The role of group identification. Personality and Social Psychology Bulletin, 23, 538-553.

41.

Staerklé

Sidanius

Green

E. G. T.

Molina

L. E.

(2010). Ethnic minority-majority asymmetry in national attitudes around the world: A multilevel analysis. Political Psychology, 31, 491-519.

42.

Tajfel

Turner

(1979). An integrative theory of intergroup conflict. In Austin

W. G.

Worchel

(Eds.), The social psychology of intergroup relations (pp. 33-47). Monterey, CA: Brooks/Cole.

43.

Vasta

(2007). From ethnic minorities to ethnic majority policy: Multiculturalism and the shift to assimilationism in the Netherlands. Ethnic and Racial Studies, 30, 713-740.

44.

Verkuyten

(2005). The social psychology of ethnic identity. Hove, UK: Psychology Press.

45.

Verkuyten

(2007). Religious group identification and inter-religious relations: A study among Turkish-Dutch Muslims. Group Processes & Intergroup Relations, 10, 341-357.

46.

Verkuyten

Reijerse

(2008). Intergroup structure and identity management among ethnic minority and majority groups: The interactive effects of perceived stability, legitimacy, and permeability. European Journal of Social Psychology, 38, 106-127.

47.

Verkuyten

Yildiz

A. A.

(2007). National (dis)identification and ethnic and religious identity: A study among Turkish-Dutch Muslims. Personality and Social Psychology Bulletin, 33, 1448-1462.

48.

Verkuyten

Yildiz

A.A.

(2009). Muslim immigrants and religious group feelings: A study among Sunni and Alevi Turkish-Dutch. Ethnic and Racial Studies, 32, 1121-1142.

49.

Verkuyten

Zaremba

(2005). Interethnic relations in a changing political context. Social Psychology Quarterly, 68, 375-386.

50.

Ysseldyk

Matheson

Anisman

(2010). Religiosity as identity: Toward an understanding of religion from a social identity perspective. Personality and Social Psychology Review, 14, 60-71.

51.

Zolberg

A. R.

Long

L. W.

(1999). Why Islam is like Spanish: Cultural incorporation in Europe and the United States. Politics & Society, 27, 5-38.