Structural and normative conditions for interethnic friendships in multiethnic classrooms

Abstract

Interethnic friendships between students are important for harmonious intercultural relations at school. Drawing on research on intergroup contact and cultural distance between immigrant and non-immigrant groups, we examined how structural and normative conditions in the classroom context are associated with friendships between early adolescents with and without an immigrant background in ethnically heterogeneous schools. The sample comprised 842 students (M_age = 11.50 years, SD_age = .71; 53% male) attending multiethnic schools in Southwest Germany.

Results revealed that perceived positive contact norms in class and perceived cultural distance predicted friendships between immigrant and non-immigrant students in both groups, even when the ethnic composition of the classroom was taken into account. The associations were largely the same for immigrants and non-immigrants. We conclude that interventions to foster interethnic friendships should aim to reduce perceptions of cultural distance and monitor and improve contact norms in intergroup settings.

Introduction

In Western Europe, the population is becoming increasingly multicultural. In Germany about one third of school-aged children have an immigrant background (Statistisches Bundesamt, 2012). However, members of different ethnic groups are often separated from the mainstream society and intercultural encounters restricted to the public, formal domain. Schools can provide valuable opportunities for interethnic contact and the formation of interethnic friendships. Such relationships have been shown to promote positive intergroup attitudes (Pettigrew & Tropp, 2006) and are an important acculturation outcome for immigrant youth (e.g., Schachner, Van de Vijver, & Noack, 2014a). Yet, the opportunity for contact is a necessary but often insufficient condition for such friendships. The preference for friends who are similar, including culturally similar, is strong, a phenomenon called homophily (McPherson, Smith-Lovin, & Cook, 2001). Interethnic friendships are less prevalent, less stable over time, and often less intimate than intraethnic friendships (Aboud et al., 2003, Jugert et al., 2011).

We examine structural and normative conditions for German-immigrant friendships in multiethnic classrooms in order to increase our understanding about the mechanisms behind the formation of such friendships. Research on intergroup contact (Allport, 1954, Blau and Schwartz, 1997) and perceived cultural distance (Galchenko and Van de Vijver, 2007, Hagendoorn, 1995) specifies conditions that affect the relationship between different ethnic groups in a society. Research on intergroup contact is mainly concerned with characteristics of the contact situation and how they impact intergroup relations, whereas research on perceived cultural distance is concerned with characteristics of the groups involved and the effect of perceived differences between ethnic groups. We test the effect of these conditions on intergroup friendships by employing the Theory of Planned Behaviour (Ajzen, 1991, Ajzen, 2012), which distinguishes between conditions, intentions, and actual behaviour.

The contact hypothesis by Allport (1954), one of the most cited theories on majority–minority relations, proposes that contact between members of different groups leads to more positive intergroup attitudes. Contact is proposed to be most effective when both groups have an equal status and common goals, there is intergroup cooperation, contact is supported by the authority, and there is the possibility to establish intergroup friendships (Allport, 1954, Pettigrew and Tropp, 2006). Hence, intergroup friendships, a context in which these conditions are met at an interpersonal level, are regarded as a highly intimate and effective form of intergroup contact. Intergroup friendships, however, can also be a potential outcome of optimal conditions at the community level. The numerous classroom interventions that have been designed and implemented, generally lead to better interethnic relations amongst students (for a review see Slavin & Cooper, 1999). Remarkably, only a handful of studies have measured perceived normative conditions for interethnic contact, particularly in classrooms where no intervention takes place. It was found that these normative conditions had a significant impact on children's level of prejudice (Molina & Wittig, 2006) and native children befriending immigrant children (Jugert et al., 2011).

Research on intergroup contact has focussed primarily on studying contact effects on intergroup attitudes of majority group members, while studies including a minority perspective are still rare. Studies that do investigate effects for both groups often find that effects are stronger for the majority group (Tropp & Pettigrew, 2005). This may be rooted in a different evaluation of the contact situation, with low status groups being more sensitive to subtle status differences or discrimination in the contact situation than members of the majority (Dixon et al., 2005, Tropp and Pettigrew, 2005). It is therefore important to consider individual perceptions of such normative conditions and how they facilitate intergroup friendships, even in classrooms without a specific contact-focussed intervention. We expect that immigrant children perceive these normative conditions as less favourable than non-immigrant children. Yet, we expect the effect of these perceived conditions on interethnic friendships to be the same in both groups.

Beyond these normative conditions, there are also structural aspects of the contact situation that may affect intergroup outcomes, such as the proportion of in-group and out-group members in the class (Blau & Schwartz, 1997) – in our case immigrant and non-immigrant students – and their diversity in terms of the number and relative size of subgroups (Putnam, 2007). Findings concerning the association between ethnic composition and interethnic relations in the classroom context are mixed (for a review see Thijs & Verkuyten, 2014). On the one hand some studies have found that a higher share of immigrant students was associated with indicators of more positive interethnic relations, such as lower levels of ethnic victimisation (Agirdag, Demanet, Van Houtte, & Van Avermaet, 2011) and more out-group friends amongst majority group members (Ttzmann et al., 2015, Van Houtte and Stevens, 2009, Wilson and Rodkin, 2011). On the other hand, some studies have found negative effects of the proportion of immigrants in the school population on interethnic relations. In classrooms with more immigrants, higher levels of perceived discrimination are reported by immigrants (Brenick, Titzmann, Michel, & Silbereisen, 2012), and both groups hold more negative out-group attitudes (Vervoort, Scholte, & Scheepers, 2011). Yet, Brenick et al. (2012) suggest that differential effects of the proportion of immigrants on interethnic relations may be related to normative conditions in these schools. We expect that in classrooms with more immigrants there will be more interethnic friendships between immigrants and non-immigrants after controlling for norms around intergroup contact.

Beyond the proportion of immigrants in class, some studies have found a positive relation between the ethnic diversity of the immigrant group and the formation of interethnic friendships between minority and majority students (Moody, 2001, Van Houtte and Stevens, 2009). Other studies did not find a link between classroom diversity and interethnic relations when the proportion of immigrants in the classroom was controlled (Agirdag et al., 2011, Vervoort et al., 2011). This suggests that more diversity in the immigrant group may be associated with more contact and friendships with members of the majority group. So, the mere ratio of immigrants to non-immigrants in class seems to be a more important predictor of friendships occurring between the groups.

The cultural distance between ethnic groups (i.e., perceived differences in values, attitudes, and beliefs) moderates the relationship between those groups (Galchenko and Van de Vijver, 2007, Hagendoorn, 1995). The perception of cultural distance between the culture of origin and the host country has been shown to negatively affect adaptation to a new country and the relationship with members of the mainstream society for adult (Galchenko & Van de Vijver, 2007) and adolescent immigrants (Phalet & Hagendoorn, 1996). At the same time, cultural distance underlies evaluations of different immigrant groups. In the so-called ethnic hierarchy (Hagendoorn, 1995), different ethnic groups are ranked in terms of the status they hold in a particular society, with the national group having the top rank. Usually, groups that are considered to be more similar to the national group rank higher in terms of status. It has been shown that there is agreement on this order between members of different ethnic groups. Groups at the bottom are typically rejected by higher ranking minority groups (Verkuyten, Hagendoorn, & Masson, 1996). The notion of such a hierarchy can already be found amongst minority and majority early adolescents (Verkuyten et al., 1996) and even children from lower ranking groups report experiencing more discrimination by peers (Verkuyten, 2002). Baerveldt, Zijlstra, De Wolf, Van Rossem, and Van Duijn (2007) found that in a classroom context with a higher share of immigrants from countries that are culturally more similar, the likelihood of friendships between immigrant and non-immigrant children was higher than in a context with a high share of immigrants from culturally more distant backgrounds. We expect that greater perceived cultural distances amongst immigrant and mainstream children will be linked with less motivation to engage with out-group members and fewer intergroup friendships. Also, social identity theory suggests that individuals are motivated to improve their own status by establishing relationships with members of higher status groups (Tajfel & Turner, 1986); therefore, we expect that immigrant children will be more motivated to engage in intergroup friendships than mainstream children.

The Theory of Planned Behaviour (TPB; Ajzen, 1991, Ajzen, 2012) is one of the most cited social psychological theories in explaining the link between individual attitudes and behaviour. It proposes that attitudes towards a certain behaviour, subjective norms, and perceived behavioural control predict behaviour via the intention to perform the behaviour. Perceived behavioural control can also serve as a moderator of the link between behavioural intention and actual behaviour and can have a direct effect on behavioural outcomes. The theory has been supported with empirical evidence from a wide range of areas, including social behaviour (Armitage & Conner, 2001). In recent years, there has been increasing interest in effects of normative antecedents. As a consequence, descriptive norms have been added to the theory as an antecedent (Rivis & Sheeran, 2003). Unlike prescriptive norms, which directly state – or prescribe – how individuals should behave in a certain situation or context, descriptive norms describe how individuals normally behave in a particular context. Especially in younger (non-adult) samples, descriptive norms have a high predictive power beyond the originally specified antecedents and should therefore be included in studies with this age group.

We designed a conceptual model in which perceived (descriptive) norms around intergroup contact in the classroom and perceived cultural distance serve as normative and attitudinal antecedents to the intention to establish intergroup friendships. This intention, in turn, is expected to predict the actual intergroup friendships. We did not include a measure of perceived behavioural control. Yet, conceptually speaking, the proportion of out-group members in a class can be seen as a measure of behavioural control as it facilitates or restricts an individual's opportunity to establish friendships with members of the out-group. We therefore also included a direct path from the proportion of out-group members in a class to the proportion of out-group friends. In our case we used the proportion of immigrants as a positive indicator of opportunity for mainstream children and as a negative indicator of opportunity for immigrant children (see Fig. 1 for our full conceptual model).

Pulling together the findings from the different lines of research presented above, we expect that the context surrounding a contact situation (i.e., structural composition, normative condition, and perceived differences between the groups involved) will be associated with the occurrence of friendships between immigrant and non-immigrant children as specified in our conceptual model. Concerning specific patterns and relationships for immigrant and non-immigrant children we have the following expectations:

Hypothesis 1

Immigrant children will perceive contact norms in the classroom as more negative than non-immigrant children.

Hypothesis 2

Immigrant children will have a stronger intention to engage with out-group members than German children.

Hypothesis 3

The relationships between variables will be the same in both groups, except for a stronger effect of contact opportunity amongst majority children (Hypothesis 3a).