Defining success in education: Exploring the frames of reference used by different voluntary migrant groups in Belgium

International Journal of Intercultural Relations 49 (2015) 1–12

Contents lists available at ScienceDirect

International Journal of Intercultural Relations journal homepage: www.elsevier.com/locate/ijintrel

Defining success in education: Exploring the frames of reference used by different voluntary migrant groups in Belgium Lore Van Praag a,∗ , Peter A.J. Stevens b,2 , Mieke Van Houtte a,1 a b

Department of Sociology, Ghent University, Korte Meer 5, 9000 Gent, Belgium Faculty of Political and Social Sciences, Ghent University, Universiteitsstraat 8, 9000 Gent, Belgium

a r t i c l e

i n f o

Article history: Received 2 September 2014 Received in revised form 4 June 2015 Accepted 11 June 2015 Available online 26 June 2015 Keywords: Ogbu Frame of reference Migration Belgium Secondary education

a b s t r a c t There has been little research, particularly in Europe, into who ethnic minority students compare themselves to in developing attitudes towards education and in making educational decisions, and into how their use of particular frames of reference can be explained. The analyses are derived from qualitative interviews and ethnographic observations involving students from Eastern European (N = 15), Turkish (N = 33), and Northern African descent (N = 18) from three secondary schools in Flanders (Belgium) providing academic, technical, and vocational tracks. This study finds that the generational status of immigrants, their experienced collective problems, the structural characteristics of the immigrant networks, and related access to resources shape students’ use of particular reference groups. This, in turn, seems to explain particular educational outcomes. The conclusion discusses the implications of this study for future research and social policy. © 2015 Elsevier Ltd. All rights reserved.

1. Introduction Since the first waves of labour migration to Belgium after the Second World War, immigrant children have underachieved in school compared to their Belgian peers. Scholars have studied a variety of factors to explain this underachievement, such as socioeconomic background, the experience of ethnic and cultural barriers, perceived discrimination, and the unequal treatment in school due to one’s linguistic or ethnic background (see overview Van Praag, Stevens, & Van Houtte, 2014). However, to fully capture the range of possible factors that influence students’ final educational achievement, scholars should not only look at the effort they put into school or into achieving a good grade for a particular test, but should also look at the factors that influence the development of the students’ aspirations and educational decision-making and actions (e.g. track or school choice and doing homework). These decisions and actions are developed within particular frames of reference (Azjen & Fishbein, 1980; Portes & Fernández-Kelly, 2008; Suls, Martin, & Wheeler, 2002) and depend on the groups with whom students compare themselves (Marsh, Trautwein, Lüdtke, & Köller, 2008; Trautwein, Lüdtke, Marsh, & Nagy, 2009). This idea is central to Ogbu’s (1987, 2008) cultural ecological theory, in which frames of reference are seen as thriving forces

∗ Corresponding author. Tel.: +32 09 264 68 01; fax: +32 09 264 67 96. E-mail addresses: [email protected] (L. Van Praag), [email protected] (P.A.J. Stevens), [email protected] (M. Van Houtte). 1 Tel.: +32 09 264 68 01; fax: +32 09 264 67 96. 2 Tel.: +32 09 264 97 33; fax: +32 09 264 38 52. http://dx.doi.org/10.1016/j.ijintrel.2015.06.007 0147-1767/© 2015 Elsevier Ltd. All rights reserved.

2

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

affecting ethnic minority students’ motivations. More specifically, students’ frames of reference refer to the group in the specific country that students use as a yardstick to compare their educational opportunities and benefits. In his cultural ecological theory, Ogbu tries to understand ethnic minority students’ achievement results by relating those results to their context. Ogbu predicts that minorities’ school orientation and educational outcomes are the result of interactions between ‘the system’ and ‘community forces’ that influence minority educational strategies. The system consists of educational policies at different levels, the treatment of minority children in school, and the societal rewards that minority students receive when they achieve. Community forces refer to minority community factors, and springs from the attitudes, beliefs, and behaviours regarding education within immigrant communities. These forces include ‘frames of reference for comparison’, ‘instrumental beliefs about schooling’, ‘a relational domain’ (e.g. the degree of trust and mistrust towards school and school personnel), and ‘expressive factors’ (e.g. collective identity). Ogbu argues that, based on the shared collective history of different racial/ethnic minority groups, immigrant and involuntary or caste-like minorities can be distinguished. Immigrant minorities migrated more or less voluntarily to the receiving country to obtain better opportunities. Involuntary minorities entered the immigrant country through slavery, colonization, or conquest, and, by contrast, perceive their presence as forced by the immigrant country. Due to their initial migration motivation, ‘voluntary immigrants’ are expected to have better educational outcomes than ‘involuntary immigrants’ ones as they compare their educational opportunities and benefits to those in their country of origin and involuntary minority students with those in the country of residence. Building on Ogbu’s framework may add to the existing knowledge, as there is little research investigating how structural factors inform the frames of reference and reference groups ethnic minority students use (Ainsworth-Darnell & Downey, 1998). However, it remains unclear to what extent students actually use these frames of reference, how they relate to structural factors, and whether they really affect their school results in the way predicted by Ogbu. There is to date, particularly in Europe, little research on this topic in relationship to ethnic minorities, for whom the choice of particular reference groups is often even more complicated (e.g. Gibson, 1997; Oyserman, Brickman, & Rhodes, 2007; Portes & Fernández-Kelly, 2008). For instance, reference groups may consist of the social networks that students are a part of – usually considered students’ social capital – but the reference groups and social networks are not necessarily the same. This could be because students do not necessarily identify themselves with all members of their social networks (Terry & Hogg, 1996), or find themselves in particular situations in which they are more likely to opt for reference groups outside their social networks. Merton (1968) argued that this could be the case when students have aspirations of social upward mobility, when they are looking for prestige, or when they are isolated within their own networks. Furthermore, Ogbu’s theory has been widely criticized because it fails to incorporate in-group variations within the voluntary/involuntary typology, does not consider specific historical relations sufficiently, and cannot be replicated or applied to ‘non-settler’ European countries (e.g. Foster, 2004; Gibson, 1997). To obtain a more in-depth understanding of the frames of reference students use to evaluate their own situations, we incorporated previously formulated criticism of Ogbu’s work by studying variations in the frames of reference of three distinct ethnic minority groups – all categorized in Ogbu’s voluntary immigrant group – in a relatively new immigrant country, namely, Belgium. Focusing on these voluntary immigrant groups is theoretically relevant, as variations exist between them in terms of their migration histories, settlement patterns, and relationships with the dominant society. Immigrant groups vary in their specific historical relations with the dominant society (being stigmatized or confronted with ethnic stereotypes, or by being relatively unknown for the ethnic majority, see De Witte, 1999; Hermans, 2004; D’hondt, Van Praag, Stevens, & Van Houtte, 2015; Siongers, 2011, 2013), length of stay (migrated during their school career or born in the immigrant country, see Paspalanova, 2006; Verhaeghe, Van der Bracht, & Van de Putte, 2012) and immigrant community characteristics (dense community network vs. lack of immigrant networks or unifying features, see Van Kerckem, Van de Putte, & Stevens, 2013; Verhaeghe et al., 2012). More specifically, starting from a grounded theory approach (Glaser & Strauss, 1967), we will distinguish distinct voluntary minority groups based on the immigrant group characteristics and the frames of reference they use. By doing so, attention will be given to the structural and cultural factors that relate to such frames of reference. The central research question of this study is threefold. First, we want to understand the factors and processes that help to shape students’ choices of particular frames of reference and the comparison groups that students use with respect to their educational careers. Second, we want to study how these frames of reference relate to students’ educational trajectories. Third, we intend to incorporate previously formulated criticism of Ogbu’s work by studying variations in the frames of reference of three distinct ethnic minority groups categorized in Ogbu’s voluntary immigrant group, in a non-settler and relatively new immigrant country, namely, Belgium.

2. Method The initial objective of the data collection for this study was to explore success determinants of students of immigrant descent in secondary school through ethnographic observations and semi-structured interviews with students. Ethnographic research was conducted in three schools in a large Flemish city between 2009 and 2012. Schools were selected based on a theoretical sampling procedure, as described by Glaser and Strauss (1967). Theoretical sampling of schools was based on three criteria: track composition of the schools, ethnic composition of the school, and geographical region of the school. This resulted in the selection of the following three schools (pseudonyms): St. Bernardus, Mountain High, and Catherina Atheneum.

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

3

Table 1 Highest completed educational level of father (F) and mother (M) by ethnic descent. No formal or primary education

Secondary education

Higher studies

Missing

F

M

F

M

F

M

F

M

Turkish descent Northern African descent Eastern European descent

10 7 2

11 10 2

20 7 7

16 5 5

1 2 5

1 1 5

2 1 2

2 1 2

Total

19

23

34

26

8

7

5

5

First, we considered the hierarchical tracking structure of the Flemish educational system. The four tracks – academic, technical, artistic, and vocational – are intended to prepare students for different futures. As only the academic, artistic, and technical tracks offer direct access to higher education, track choice is important in this educational system (Boone & Van Houtte, 2013). Students in vocational tracks have access to post-secondary, specialized, one-year training attached to secondary schools that allows them to take part in school-based apprenticeships. After completion of this specialization year, students have the opportunity to access higher education in order to enhance professional skills. St. Bernardus is a large (1159 students) Catholic school that offers the vocational, technical, and academic tracks. Mountain High is a medium-sized (444 students) secular school that only offers the academic track. Catherine Atheneum is a medium-sized (404 students) secular school that offers technical and vocational tracks. Second, we only selected schools that had a critical number of students of immigrant descent. The ethnic school composition varies considerably across schools in Flanders. In St. Bernardus, Mountain High, and Catherine Atheneum, respectively, 41.44%, 13.54%, and 27.48% of the students spoke another language at home than Dutch, which serves as an official indicator for the proportion of students of non-Belgian descent. The variation between the ethnic school composition of our sample is due to the tracks offered in these schools, as fewer ethnic minority students are enrolled in schools that primarily offer academic tracks. Third, we selected schools in the same geographical region (one city), as this facilitated the interpretation of the data and the comparison between schools. The immigrant networks of all students of a similar ethnic descent are more similar across schools in the same geographical area and can therefore be more easily compared. Hence, given the importance of tracks/fields of study within tracks, class groups within schools were selected based on the fixed set of courses (e.g. ‘Mathematics-Sciences’ or ‘Construction’): students have to choose within these tracks. Clustered sampling was used as all students from these class groups were observed and later invited to participate in interviews. Students from nine class groups from the penultimate year of secondary education (comparable with Year 12 in the U.K. or 11th grade in the U.S.) and one post-secondary vocational specialization year were included. As we based our sample on all students enrolled in these class groups, the age range of this sample varied between 16 and 23 years, due to grade retention and delayed entry in secondary school (ideally this would be 16–17 years). In line with prevailing definitions of ‘ethnic background’ in Belgium, students were categorized as being of non-Belgian descent if their maternal grandmother was not born in Belgium, which follows Ogbu’s categorization of immigrants (2008). Thus, based on the birthplace of the maternal grandmother, out of the sample of 129 respondents, there were 33 students of Turkish descent, 18 of Northern African – Morocco (13), Algeria (3), and Tunisia (2) – and 15 of Eastern European descent. Eastern Europe is broadly defined in this study and refers mainly to the division made in the Cold War. Students from Albania (5), Armenia (2), the Serbian Republic (1), Kosovo (2), Macedonia (2), Belarus (1), Bulgaria (1), and Russia (1) were included. As an indicator of ethnic descent, the use of the birthplace of maternal grandmother did not lead to a different categorization of ethnic groups than the use of the birthplace of paternal/maternal grandmother/grandfather. Most Turkish and Northern African parents of secondary school students migrated after finishing school in their country of origin, with the result that all but 3 students were born and raised in Belgium. By contrast, all students of Eastern European descent immigrated to Belgium during their school career. While our sample of voluntary immigrants varies according to timing of migration and settlement and (see later) social network characteristics, the educational level of parents of immigrant descent was generally low, compared to parents of Western European descent. We provide an overview in Table 1 of the educational level of the parents of the students, according to ethnic descent, to give an indication of the socioeconomic background of the students. Table 2 provides an overview of the ethnic distribution of female and male students across tracks. Table 2 Track enrolment by gender and ethnic descent. Academic track M

Technical track F

Vocational track

Total

M

F

M

Turkish descent Northern African descent Eastern European descent

5 2 2

9 3 4

1 1 5

3 1 0

6 7 1

F 9 4 3

33 18 15

Total

9

16

7

4

14

16

66

4

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

Class groups were observed in all their school activities for two or three weeks by an ‘observer as a participant’ (Gold, 1958), after which qualitative, semi-structured, face-to-face interviews were conducted with 129 students. In total, the researcher was part of the school environment for about 20 weeks in each school. The periods of observation in each class allowed the researcher to build rapport with respondents and make detailed observations of classroom, playground, and staff room interactions. These observations improved our understanding of the particular school context and the construction of the interview questions. For instance, during the observations, students frequently made remarks about their comparison groups and referred to particular reference groups. Consequently, in subsequent interviews, questions that are more detailed were included about these frames of reference and comparison groups, and how they relate to school-related processes (e.g. choices and motivation). Interviews started after conducting a period of ethnographic observations. In interviews, students were asked about their achievements in school; support from family, peers, and teachers; future aspirations; sense of school belonging; and migration background. The interviews were carried out in Dutch (the instruction language in all three schools and the dominant language in the Flemish region) and afterward translated into English. The average duration of the interviews was slightly less than 50 min, which corresponds to one teaching period. The longest interview lasted for nearly 100 min. The interviews were recorded first and transcribed afterward. These transcriptions were later analyzed, making use of computer software (NVivo9 and NVivo10). Data was collected using an exploratory approach, following a cyclical process of data collection, processing, and analysis – as is customary with a grounded theory (Glaser & Strauss, 1967) approach – using the constant comparison method of analyses. Initial analyses of data (interviews and observations) were guided by the use of particular codes, which can range from general to very specific, that may relate to particular theoretical concepts described in the literature. Ethnographic observations in the first class groups indicated that teachers and students made distinctions between students based on geographical region. Simultaneously, during the first interviews, students referred to distinct frames of reference and comparison groups related to their migration history, settlement pattern, and relationships with the dominant society. This caught the interest of the researcher, who decided to examine this in depth. Subsequently, extra attention was paid to this issue during ethnographic observations, and questions that are more explicit were asked during the next group of interviews. In the analyses, codes, such as ‘migration history’, ‘reference/comparison groups’, and ‘immigrant community’ quickly emerged from the data as meaningful. Due to similarities with Ogbu’s theory, and based on ethnographic observations, analyses that are more extensive were carried out that focused on the frames of reference and comparison groups students use with respect to education. Geographical region itself did not turn out not to be an explaining factor as such, but rather explained the relation and cohesion between other explanatory variables (e.g. social networks, discrimination: see analyses below). As a result, geographical region of origin maintains an important analytical power. The anonymity of the respondents was preserved by replacing their real names with pseudonyms chosen by them. 3. General frames of reference and daily comparison groups Due to the exploratory nature of the present study, we interpreted students’ general frames of reference as the ways in which students refer to specific groups when talking about their educational opportunities. However, it seemed that this overall evaluation of educational opportunities did not necessarily guide students when making daily educational choices, such as the amount of time spent on homework. Consequently, we made a distinction between ‘general frames of reference’, referring to educational opportunities, and ‘daily comparison groups’ that helped students shape their educational behaviour and make educational decisions. General frames of reference appeared to refer to more ideal typical and general motivations, while daily comparison groups were used when referring to students’ personal lives and specific situations. We discussed both separately in the following paragraphs. 3.1. General frames of reference As is common in grounded theory (Glaser & Strauss, 1967), we did not initially categorize students in groups based on geographical region from the start of our analyses. Rather, three distinctive groups seemed to evolve gradually as variations within the voluntary immigrant group (as described by Ogbu, 1987, 2008) emerged from our data. With respect to the general frames of reference, immigrants’ migration history and generational status seemed to be reflected in the general frames of reference students refer to when discussing the educational opportunities they have. Turkish, Northern African, and Southern European immigrants arrived in Belgium after the Second World War to take jobs as guest workers made available by shortages in specific sectors of the labour market, as well as to seek better economic opportunities for themselves. Although these immigrants were expected to return to their countries of origin, they settled permanently (Verhaeghe et al., 2012). Consequently, the second generation of immigrants coming from these regions is currently enrolled in school. This was also reflected in our sample, as all but three students were born and raised in Belgium. Students such as Lola (Moroccan descent) made statements like ‘Morocco is like a holiday destination for me.’ Therefore, it is not surprising that they did not refer to their country of origin but rather to other immigrant network members or their parents when evaluating the situation upon which their future aspirations and educational choices are based. Powergirl (Turkish descent): I wouldn’t want to become like my mother. For example, I know how hard it is for her to stand the whole day and clean offices. . . Actually, I just want to continue and be enrolled in an academic track and

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

5

pursue higher education. I would like to become an office clerk and not, for example, a cleaning lady. . . At my desk, getting a higher salary, having less physical problems like most Turks. This extract shows that Powergirl thought about her future job opportunities by referring to her mother’s working conditions, as well as those of other immigrant network members. On different occasions throughout the interview, she expressed a desire not to work in similar employment conditions as her mother. Other students of Moroccan and Turkish descent their desire to put in effort at school, their desire to stay enrolled, and the educational choices they had to make during their school careers reflected this. Additionally, having spent most of their lives in Belgium and with their family and friends who also settled in Belgium, Powergirl and most of the second-generation Moroccan and Turkish students in our sample (40 out of 51), felt differently than their peers who were born and raised in their countries of origin. They may for example exhibit a lower language proficiency in their mother tongue. These findings contrast with Ogbu’s (1987, 2008) predictions, as he argued that voluntary immigrants and their descendants would refer to their country of origin. However, this desire ‘to do better’ than their parents and/or other members of their immigrant group served as a general motivation to do well in school. By contrast, students from Eastern European countries had only recently migrated, and referred mainly to their country of origin when evaluating the educational opportunities they have. Starting in 1990 a new wave of immigrants from Eastern Europe and the former Soviet Republic entered Belgium, accompanied by a wave of undocumented immigrants (Sierens, 2006). Most immigrants arrived in 1990 due to economic, political, social, and cultural push factors in their countries of origin (Paspalanova, 2006). Despite the differences in physical appearance, religion, and cultural habits between students of Eastern European descent included in our sample, they shared very similar conditions and motives for migration that make it theoretically relevant to discuss these immigrants together as one group. Additionally, it appears that, in general, members of the ethnic majority know relatively little about this group (compared to other immigrant groups). Therefore, students who originate from Eastern European countries are often treated as one cultural group and/or confronted with similar ethnic stereotypes. Generally, all students coming from Eastern European regions were motivated to migrate during their school careers in order to have better educational and future employment opportunities. Although they had lived most of their lives in the receiving country (Belgium), 12 (out of 15) students compared their opportunities with those of their friends and family in their countries of origin and considered the hypothesized future opportunities they would have had, had they not migrated: Unikkadon (Albanian descent): Yes, really, I often think of it. Like my friends that are now there, they have nothing. They dropped out of school, work as farmers, or do industrial work and manual labour. I don’t want that. I want to study, have a good job and think about my future. This quote illustrates that students of Eastern European descent like Unikkadon evaluate the possibilities they have in the immigrant country (Belgium) by referring to the rather low educational opportunities of their peers and family members their countries of origin, and that this comparison functions as a source of motivation to do well in school (see Kao & Tienda, 1995). In sum, the data shows that voluntary ethnic minority groups can differ in terms of the reference groups they use for comparison of their own educational situation and opportunities. While second-generation students of voluntary migrant groups from Turkey and Northern Africa compared themselves primarily with the situation of their parents and other immigrant network members in the immigrant country (see Hermans, 2004), first-generation students of voluntary immigrant groups from Eastern European countries compared themselves primarily with peers from their countries of origin. However, while they compare themselves with different groups, the outcome of these comparisons seem to be very similar as it results in the development of positive attitudes towards education and aspirations for relative social mobility. Furthermore, analyses indicated students could not always rely on these general frames of reference for making particular decisions in relationship to their immediate educational careers, such as following a certain high status track or pursuing higher education. For instance, people in the country of origin do not necessarily have to make similar choices (e.g. tracks/fields of study). Therefore, immigrant students searched for comparison groups they could use in everyday situations to make educational decisions and guide their educational behaviour. Thus, in the following sections, we examine how these positive attitudes towards education translate into actual achievement, and the role of daily comparison groups in this process. Such daily comparison groups do not necessarily overlap with the frames of reference students have. Rather, finding such daily comparison groups and the ways they could help students to improve their educational outcomes appeared to depend on structural characteristics, such as immigrant settlement patterns, which varied for the three discussed immigrant groups. The role of these structural characteristics in shaping students’ comparison groups emphasises the need for a more sociological approach in understanding the development and use of students’ frames of reference, and how this relates to their daily comparison groups. 3.2. Daily comparison groups Our analyses suggest that students of immigrant descent seem to search for specific groups with whom they can compare themselves in everyday life. The possibility of having daily comparison groups seems to depend on the social networks – and thus the social capital – students have. However, students do not necessarily find that their situation is comparable to all

6

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

members of their network, and, thus, may not identify with them (see Terry & Hogg, 1996). Comparison groups depend on the context (Merton, 1968) and, hence, on the daily confrontation and meaningfulness of comparison groups (Richer, 1976). The lack of perceived comparability between peers may be related to the distinct difficulties students coming originally from different countries have to face in the receiving country. Dealing with these problems guides their search for comparable peers. Further, this search for comparable peers appears to relate to students’ educational outcomes. In this study, we first focus on the factors that guide students when looking for daily comparison groups and, second, on how these might relate to educational outcomes.

3.2.1. Finding daily comparable groups Our analyses point to two key factors that explain ethnic minority students’ choices for particular daily reference groups. First, immigrant-specific problems, such as their experiences with discrimination or their adaptation to the immigrant country, support the idea that students of immigrant descent could not compare themselves with students of Belgian descent as they struggled with additional problems, especially as the latter group was perceived to some extent as the source of the problem. The majority of the students interviewed appeared to indicate that problems related to their immigrant or ethnic background made them focus more on comparable peers in their immigrant networks than on students of Belgian descent. While most students of Turkish and Northern African descent appeared to encounter problems related to experiences of discrimination, those of Eastern European descent referred more to adaptation problems related to their recent migration. Variation in the extent to which these groups experienced discrimination appeared to depend on the relationship they developed with the dominant society. Students of Turkish and Northern African descent had already lived for an extensive period in Belgium, while those of Eastern European descent were relatively new to the country. The longer stay of students of Turkish and Northern African descent in Belgium seemed to have resulted in the establishment of an ambiguous relationship with the dominant society, characterized by attachment but also by conflict, experiences of discrimination, and hostile attitudes and remarks. For instance, despite the fact that most were born in Belgium, they are still perceived as ‘immigrants’. For example, Giorgio (Turkish descent) stated, ‘This boy came to me and said “you dirty Turk, what are you doing here? Go back to your country”.’ Like many others, Giorgio did not himself migrate, but was still considered an immigrant, allochton (of non-Belgian descent), or, more generally, an outsider. These terms were frequently used in an explicitly derogatory way. In addition to experiences of racism from peers, this study showed that two thirds of the students who had experienced discrimination also referred to teacher discrimination during their school career: Interviewer: How come you lacked motivation in the second grade of secondary school? Chris (Algerian descent): Well, I had a teacher, I really didn’t think he was a good teacher. He put everyone in categories. Some teachers had something against allochtons, and other teachers preferred the ‘others’. . . I thought they wouldn’t accept me and that’s why I didn’t even try to do my best, so I didn’t. Half of the students mentioned encountering derogatory generalizations, discriminatory remarks or attitudes from classmates, or racist remarks in public spaces. Furthermore, irrespective of personal experiences, half of the students seemed to expect labour market discrimination. For instance, many mentioned that people with ‘a foreign name’ or in specific jobs (e.g. manual labourer, money transporter), would be more vulnerable to such experiences. Although 12 students did not experience discrimination themselves, none of these students doubted the existence of discrimination, based on shared experiences within their immigrant networks. Such experiences and perceptions made students feel like they had fallen between ‘two worlds’, never being fully part of the dominant group in a specific territory. This made students refer more to, and compare themselves with, people of their immigrant community who had similar experiences or feelings. By contrast, students of Eastern European descent appeared to encounter substantially fewer experiences of discrimination. Only 6 out of 15 students mentioned experiencing some form of discrimination. These experiences were mostly related to prejudices concerning their religious affiliation (e.g. ‘all Muslims are terrorists’) or adaptation problems related to their recent migration (e.g. language proficiency). Nevertheless, experiences of racism and out-group positioning were less negative and/or considered as such by these students, who more often interpreted them as directed at the individual rather than their ‘in-group’. For instance, Walter (Albanian descent) remarked that teachers were only aware of their immigrant background after a couple of months, when they referred explicitly to their country of origin. Due to the late discovery of these students’ immigrant backgrounds, teachers did not negatively change their behaviour towards them. This lower likelihood of experiencing discrimination is in line with previous research and can be related to the individual migration of these immigrants, their lower visibility in society, the perceived in-group/out-group similarity, and the heterogeneity of this ‘group’ (see for example Piontkowski, Florack, Hoelker, & Obdrzálek, 2000). However, these students more frequently mentioned suffering from adaptation problems, such as language proficiency and knowledge about the Flemish educational system: Yasmina (Russian descent): I had a diploma from my country of origin from the third year [of secondary school]. After studying Dutch for six months, I had sufficient language proficiency to be enrolled in the fourth year. Teachers advised me to redo my third year because I didn’t understand any French and English [obligatory courses]. . . I didn’t understand a lot of these courses, so I learned everything by heart, but failed my year and neighbours advised me to change school.

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

7

Many first-generation immigrants, like Yasmina, encountered adaptation problems that often hindered their learning in school, and, consequently, complicated the course of their school career considerably (see Kao & Tienda, 1995). These migration-specific problems made them feel different from students of Belgian descent and caused them to search for comparable peers in their disparate immigrant networks. Nevertheless, as these adaptation problems were perceived as ‘transitional’ ones that they could overcome over time, some students of Eastern European descent – particularly those who did not perceive very many adaptation problems – appeared more likely to compare their educational behaviour with students of Belgian descent. In sum, young people of immigrant descent seemed to identify more with other students of immigrant descent than with native Belgians due to their experiences of being frequently positioned as different or as out-group members by the majority group, ethnic discrimination, and/or adaptation problems (Ogbu, 2008; Piontkowski et al., 2000). While these processes explain why certain voluntary immigrant groups are less likely to choose native Belgians as a reference group, the process of finding comparable peers for their daily lives seems to depend more on the immigrant network characteristics, which will be set out in the next paragraph.

3.2.2. Immigrant network characteristics Although most students of immigrant descent felt themselves to be theoretically more comparable to specific immigrant groups, not all students were able to find such daily comparison groups due to variations in availability and density of immigrant networks. Further, depending on the existing resources and shared information within these networks, these daily comparison groups seemed to help students to varying extents and in distinct ways. In Ghent, the city in which this study took place, Turkish immigrants formed a tight community through the processes of family reunification and asylum, leading to a chain migration system. This resulted in the formation of a transplanted community, originating from Emirdag in the Turkish province of Afyon. Consequently, the Turkish students in our sample could rely on an extended, geographically concentrated Turkish community characterized by a considerable amount of social control and local ethnic economy (Verhaeghe et al., 2012). The availability of a relatively large and close Turkish community facilitated the opportunity to identify with and rely on this community when making concrete comparisons, as illustrated by Memoli: Interviewer: How come you want to become a baker? Memoli (Turkish descent): It’s in my family, actually, it’s in ‘our kind’. When you go in this city to a bakery, ninety percent of all bakers are from our city in Turkey. This extract indicates that Memoli’s ideas were mainly shaped within this extensive Turkish immigrant community. In reality, the majority of bakers in Ghent are of Belgian descent. Memoli’s living environment, concentrated within his Turkish community, gave him a biased view of the ethnic proportions of the city, and that view seemed to influence his development of future goals, such as opening a bakery. Other immigrant groups are less dense and homogeneous, resulting in distinctly different daily comparison groups for those immigrant children. Due to the greater variety in regions of origin and educational levels of immigrants, Northern Africans are a more heterogeneous group, and have formed less pronounced transplanted communities. Moroccans, the largest group, mainly come from the rural areas of Souss and Anti-Atlas and from cities, and speak different languages. Moroccan immigrants have extended but heterogeneous networks in Belgium, and their immigrant community is smaller in the selected city (Verhaeghe et al., 2012). Immigrant communities of other Northern African countries appeared to be even more heterogeneous. This heterogeneity was reflected in the references students made to the daily comparison groups; one referred to extended family members, while others mentioned neighbours or people they met during religious activities. For instance, Marjan received much of her information about education from friends she met during a religious course within her community: Interviewer: Do you know many people that can provide you information about higher education? Marjan (Algerian descent): I have some acquaintances that have finished their bachelor and are starting their master. They could help me when preparing their exams and stuff. Interviewer: How do you know them? Marjan: I was enrolled in a course at the Mosque to learn to read and write Arab. I was the youngest of the group. I had the time of my life and we kept contact. During the interview, Marjan made it clear that having friends that pursued higher education made her more familiar with the idea of being enrolled in higher education. In this interview, it appeared that these friends stimulated her over the course of her school career to continue to make an effort in school, and helped her in overcoming problems of discrimination. Her network consisted of a wide variety of students of different Northern African countries, which also appeared to be the case for many students in our sample who come from these regions. Finally, most of the students of Eastern European descent referred to their country of origin when evaluating educational opportunities. However, this did not help them in concrete situations and everyday school life. These students did not mention being part of an extended immigrant network and as result did not experience strong social control or support from within their immigrant group. One example is James:

8

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

James (Albanian descent): You do have Albanians, but the only Albanian boy I know is Walter, who is my classmate. When I was young, I went to a school in which mainly Belgians were enrolled. I never met that much Albanians at school and it stayed that way. Settlement patterns reflect individual immigration, political situations, and ethnic conflicts in the countries of origin (Paspalanova, 2006). Despite the differences in physical appearance, religion, and cultural habits between students of Eastern European descent included in our sample, they shared very similar conditions and motives for migration, which makes it theoretically relevant to discuss these immigrants together as one group. Additionally, it appears that, in general, members of the ethnic majority know relatively little about this group (as compared to other immigrant groups). Therefore, students who originate from Eastern European countries are often treated as one cultural group, and/or are confronted with similar ethnic stereotypes. Consequently, the daily comparison groups of the majority of students of Eastern European descent differed from those of Turkish and Northern African descent. Nine (out of 15) of them mentioned the lack of actual comparison groups to help them deal with daily school life: Walter (Albanian descent): Yes, I compare myself with my friends outside school because I cannot compare myself with the rest, certainly not with those from Albania. I don’t know anyone there, so I cannot compare myself with anyone, it’s a disaster. At school, I cannot compare myself with the situation at home of the other [immigrant] students because I have another situation. In this quote, Walter suggests that he had to make more of an effort and search for a specific daily comparison group because he could not find one in his immediate school environment. Further, Walter’s quote suggests the importance of variations within the voluntary immigrant group. Due to their recent migration and dispersed immigrant community, some students did not encounter a daily comparison group that would help them when making educational choices. As a result, a couple of these students seemed to search for other members of their social networks that could help them. These students mentioned the support of specific ‘informants’, or representatives of Belgian society, such as teachers, neighbours, and colleagues of their parents. These informants could help students mainly with respect to making educational choices. In contrast with daily comparison groups, however, students could not use these informants when evaluating their own educational behaviour. To conclude, the choice of particular daily comparison groups appeared to depend on students’ frames of reference. However, the ability to find such comparable peers seemed to be related to the social networks students have access to, which is often related to the characteristics of their immigrant networks. Students of Turkish descent could rely on an extensive, homogeneous, dense immigrant community with an overall low achievement level, giving them easy access to comparable peers. By contrast, Northern African immigrant communities were extended but more heterogeneous in terms of region of origin, language, and educational background, which was reflected in their daily comparison groups. Finally, students of Eastern European descent – due to individual migration and conflicts in their regions of origin, and variations in language and educational background – had less access to comparable network members than students of other immigrant groups and, therefore, had to make more of an effort to find people that could help them during their school careers. This variety in the structural characteristics of immigrant networks in Ghent seemed to create differences in access to educational resources, support, information, and the development of specific standards of achievement. This will be discussed in detail in the next section. 3.2.3. Daily comparison groups and educational behaviour and choices Based on the interviews, it seemed that finding daily comparison groups matters for students’ educational outcomes, as they support and guide students throughout their educational careers in two primary ways: (1) setting standards of achievement, and (2) providing support. First, considering the achievement levels, the collective (discriminatory) employment experiences of network members and the educational behaviour of peers within these networks appeared to nuance students’ definitions of success and give rise to the development of distinct standards of achievement and occupational aspirations. Many immigrant community members work in poor and difficult conditions or experience labour market discrimination (see also the quote by Powergirl). Consequently, while immigrant community members are frequently used as a reference for what someone should not want to do, this reference does not necessarily lead students to specific future objectives: Interviewer: And what do you want to achieve? Chris (Algerian descent): A normal diploma and a good job. Just normal like they say, like every other person here (. . .). I want to live in better conditions than my father. When he came to Belgium, he had to sleep on a couch. He had nothing. He worked really hard. When I compare my father’s situation with my own situation, I realize that everything went smoothly for me. However, it wasn’t that easy for my father. When my father talks about his past, it occurs to me that I should appreciate the things I have. Second, these daily comparison groups seemed to matter for guidance and support. Students would often turn to people facing similar problems when dealing with their own. Consequently, the extent in which these comparison groups could help students to achieve in school seemed to be associated with the existing knowledge about the educational system, the level of education, and the standards within these comparison groups. This could explain part of the variability in educational behaviour and choices of students of immigrant descent. For students with more educational resources in

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

9

their daily comparison groups, it was easier to translate achievement motivation into actual educational achievement. This was the case for Claudia and Brit, two girls who did not have an extensive immigrant network. While Claudia states that she mainly want to help her siblings, both students mentioned to make their own decisions with respect to educational choices: Claudia (Albanian descent): I always help the younger ones [sisters and brothers], with their homework, and motivate them by saying ‘when you succeed your exercise, you’ll receive a sticker’ – kind of playing their teacher – and they like it. And at the older sisters, I say ‘you should do what you want to do’. Like my younger sister, she always wanted to become a cook, so after her first year of being enrolled in the academic track, she changed to the technical track. She really wanted to do that and I think you should do what you like to do. Interviewer: And you told her that? Claudia: Yes, and my parents said that too. It’s not like they told us ‘you have to do that’ but rather: We are free to choose whatever track we want to enroll in. Interviewer (to Brit): And your parents? Brit: I could choose what I wanted to do. Interviewer: And do they say something afterwards? Brit: They do not know it that well here, like, what it really means. When I say ‘[I study] Economics’, they know what it means but not really what it signifies. They say: ‘You should do whatever you want.’ While Brit and Claudia stay enrolled in the academic track because of their personal interest in this track and the success they have had in their courses, other students, such as Claudia’s sister, opted for a technical or vocational track. The lack of knowledge about the educational system could also imply that parents and students enrol in courses without being aware of the consequences, as is the case for Tarik (Moroccan descent): ‘I didn’t know. I didn’t get any information. My parents didn’t know anything, they just enrolled me in the Electrics course.’ Other students, when talking about changing to less demanding tracks, referred to their friends or family members who were also enrolled in such tracks and who had to do a smaller amount of homework. Other students, like Marjan (see quote above) and Gonca, mentioned being inspired by the educational experiences of their network members, which seemed to help them to achieve more than their parents. Gonca (Turkish descent): Well it is a good thing to study, I don’t want to be like them, I want to study and set an example, so they [Turkish immigrant students] can be motivated to study and want to make an effort to realize that. Interviewer: So, it motivates you to set an example. Gonca: Well for example, there will be a doctor, like a pediatrician, ‘ah it’s [Turkish last name], she is a Turk, a Turk can be a doctor, that could be possible, not only Belgians are able to do that’ and so. When members of these daily comparison groups are less familiar with and have less knowledge about the educational system, or do not have a similar educational background, it is harder to find someone who can help students achieve in school and make educational choices in line with personal capacities and aspirations. Thus, the importance of such daily comparison groups appeared to matter mainly for availability of cultural capital that could be shared and access to different sorts of information. For example, students who compared themselves with peers who had a similar knowledge about the educational system made comparable decisions that resulted in a cyclic process of educational choice making. To conclude, daily comparison groups seemed to matter for students’ educational outcomes in two ways: by setting educational standards and by giving access to supporting and information-rich networks. Depending on the available comparison groups – which are mainly within immigrant networks – and the characteristics of a student’s specific immigrant network, daily comparison groups might have different effects on the educational outcomes of students.

4. Discussion This study builds on a large body of research on the importance of students’ frames of reference in making educational choices by focusing on how ethnic minority students’ use of particular reference groups relates to structural characteristics of their social environment and how this, in turn, informs their educational achievement. The present study continues a long line of (mainly U.S.-based) research on Ogbu’s cultural ecological theory, and criticism of this theory, by examining ethnic minority students’ frames of reference in three different voluntary immigrant groups in the context of Flanders. In so doing, this study considers variability between voluntary immigrant groups in terms of their migration history, relationship with the dominant society, and immigrant community characteristics. This is important, as the situation of immigrant groups in the receiving country is not determined by their initial migration motivation, as argued by Ogbu (2008), but seems to depend more on immigrant networks, generational status, minority position in society, and perceptions and experiences of collective problems (Gibson, 1997; Hermans, 2004). These findings underscore criticisms levelled at Ogbu’s theory for neglecting variance between voluntary minority groups in terms of the structural characteristics that inform their educational aspirations (e.g. Foster, 2004; Gibson, 1997; Hermans, 2004). Finally, by focusing on the educational experiences of Eastern European, Turkish, and Northern African groups, this study compares two groups that are extensively studied in Europe (Turkish and Northern African immigrants) with each other, and with a group (Eastern

10

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

European immigrants) that is relatively under-studied in a European context. In sum, this study was innovative as it explored the varying ways students’ frames of reference could affect their educational outcomes, and how this affect seemed to vary across different immigrant groups. Our analysis of student interviews in three Flemish secondary schools shows that a distinction can be made between general frames of reference and daily comparison groups. Analogous to Mickelson’s (1990) distinction between abstract and concrete school attitudes, each student’s actual living situation and related opportunities appeared to matter for his/her educational behaviour. Furthermore, our findings are in line with the conceptual framework of Rosen (1956), in which achievement orientations are studied separately from the conditions in which achievement is realized. Although scholars have debated the value of abstract school attitudes for educational outcomes (e.g. Downey, Ainsworth, & Qian, 2009; Mickelson, 1990), our data suggests that students used their general frames of reference when evaluating educational opportunities, and that these appeared to relate to study motivation. While recently arrived students of Eastern European descent refer to the poorer educational opportunities in the country of origin and their parents’ migration motivations, students of Turkish and Northern African descent framed their opportunities in reference to the conditions in the immigrant country. Despite the distinct countries students referred to, all students seemed to be motivated to achieve in school because of this overall comparison with the living conditions of people within their frames of reference. When translating such general achievement motivation into educational behaviour, students referred to daily comparison groups that helped them to make decisions in everyday situations, such as doing homework, and educational choices, such as track choice. The importance of daily comparison groups seemed especially critical in educational systems like the Flemish one that rely more on individual agency when making educational choices (Boone & Van Houtte, 2013). However, when making such choices, and successfully realizing these choices (e.g. by putting effort into their schoolwork), students seemed to refer to specific comparable peers, mostly within their specific immigrant communities, as those peers had to face similar migration-specific problems, such as adaptation problems and problems related to ethnic discrimination. This is understandable, as the choice for particular reference groups depends on the meaningfulness and salience of these groups (Richer, 1976), and are context specific (Merton, 1968). The specific reason why students of immigrant descent seem to search for comparable peers within their immigrant networks lies in the complex relationship they have developed with the dominant society. In line with a rich tradition of research in the U.K. on minority students’ experiences of racism (Gillborn & Gipps, 1996; Stevens, 2007), immigrant youth that have resided longer in the immigrant country, such as the students of Turkish and Northern African descent in Belgium, have developed a more ambiguous relationship with the dominant society. Building further on previous research in Belgium (e.g. D’hondt et al., 2015; De Witte, 1999; Hermans, 2004; Siongers, 2011, 2013), we found that there was a variation in the ethnic stereotypes immigrant groups are confronted with, which appears to have far-reaching consequences for the organization of students’ daily (school) decision-making processes. Although not all students were confronted with discrimination on a daily basis, perceived (future) discrimination shapes students’ lives profoundly and should be incorporated into future research to understand the choices students make throughout their school careers. Differentiating between the frames of reference that inspire students to be motivated in school and daily comparison groups that help them to guide their educational behaviour seems a useful conceptual distinction for gaining a better understanding of the factors that might contribute to the educational outcomes of ethnic minority students. Given their migration history, students of immigrant descent aspire to upward social mobility. While Merton (1968) argued that upward mobility aspirations could increase the chance that students would opt for reference groups outside their networks, it seems that our respondents search for comparable peers within their social networks. However, although students of immigrant descent may be part of larger social networks, and members of those networks may provide useful information about the educational system, those networks do not necessarily guide students’ educational behaviour. It is exactly in this distinction between social network members and comparable peers that we think we can make a useful contribution to the existing literature. These findings add to previous research suggesting that immigrant networks possess varying degrees of ethnic capital, including both social and cultural capital (from their countries of origin) (e.g. Zhou & Bankston, 1994). Students of immigrant descent would benefit from having access to networks that offer relevant information (see also Granovetter, 1973) or set (high) standards of success for them. Social capital may be helpful in sharing information about the educational system (see also Granovetter, 1973), or setting (high) standards of success, but does not necessarily guide students through everyday life by helping students evaluate their educational success and the effort they put into school (see also Terry & Hogg, 1996). Hence, analysis needs to consider both elements in developing a greater understanding of ethnic minorities’ educational choices and expectations. Nevertheless, while connectedness with one’s immigrant group and finding comparable peers could enhance educational outcomes, the relationship with achievement is not guaranteed (Oyserman et al., 2007). Actual educational behaviour appeared to depend on the prevailing standards of achievement and the available support and educational resources within immigrant networks. For example, our results indicate that being part of a network with ample resources, which was the case for students of Northern African descent with higher-educated family members (‘strong ties’), helped to create a demanding comparison group that guides studying practices. While the dense, homogeneous Turkish networks seemed to help to strengthen students’ ethnic identity and sense of belonging, there are less ‘bridging’ ties to other networks that offer information (Granovetter, 1973). Finally, students of Eastern European descent lacked an extended immigrant network that could help them cope with adaptation problems, compare their educational strategies, and base their educational aspirations on. However, they sometimes benefited from the knowledge and information shared through ‘weak ties’, such as neighbours or teachers. These findings further demonstrate the importance of ethnic (social and cultural) capital, as most students of immigrant descent seemed to refer mainly to immigrant networks, if available (e.g.

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

11

Suárez-Orozco, Pimentel, & Martin, 2009; Zhou & Bankston, 1994). They also demonstrate how daily comparison groups affect educational outcomes differently. Some future research recommendations can be made. First, the theoretical sampling was based on school criteria, such as the proportion of students of non-Belgian descent, and not on individual characteristics. Future research that also included students of non-Belgian descent that are enrolled in schools with mainly students of Belgian descent, or a focus on the inclusion of particular ethnic groups, could be rewarding. Second, it could be interesting to test the impact of these comparison groups. In the present study, achievement results could not be included due to the lack of a centralized evaluation system in Belgium, which made it impossible to compare outcomes between tracks and schools. Finally, future research could build on this study by comparing the attitudes of parents and students (Hermans, 2004), by including additional school and societal factors (such as school management and media), and by interacting with other issues related to identity and culture (Ogbu, 2008). Based on these findings, policymakers could actively work on the dispersion of knowledge about the educational system and future opportunities, raise standards of achievement for all students, and strengthen ethnic communities (e.g. through role models). For instance, organizing extracurricular activities and homework guidance for students of immigrant descent in their neighbourhoods might be helpful. Another policy implication of this paper concerns the impact of discrimination and perceived ‘blocked opportunities’ (e.g. Kao & Tienda, 1995; Mickelson, 1990) of students’ comparison groups and, by extension, educational careers. This suggests the need for social policies that provide students coping strategies to deal with discrimination (Brondolo, Brady, Pencille, Beatty, & Contrada, 2009) or that actively (and structurally) challenge discrimination in the labour market, schools, and media.

Acknowledgement Funding for this research was provided by the Research Foundation Flanders (FWO).

References Ainsworth-Darnell, J. W., & Downey, D. B. (1998). Assessing the oppositional culture explanation for racial/ethnic differences in school performance. American Sociological Review, 63(4), 536–553. Azjen, I., & Fishbein, M. (1980). Understanding attitudes and predicting social behavior. New Jersey: Prentice Hall. Boone, S., & Van Houtte, M. (2013). Why are teacher recommendations at the transition from primary to secondary education socially biased? A mixedmethods research. British Journal of Sociology of Education, 34(1), 20–38. Brondolo, E., Brady, N., Pencille, M., Beatty, D., & Contrada, R. J. (2009). Coping with racism: A selective review of the literature and a theoretical and methodological critique. Journal of Behavioral Medicine, 32(1), 64–88. De Witte, H. (1999). Alledaags racisme in Belgie Een overzicht van de onderzoeksliteratuur over racistische opvattingen en een interpretatie van de betekenis van het opleidingsniveau. Migrantenstudies, 15(1), 2–27. D’hondt, F., Van Praag, L., Stevens, P. A. J., & Van Houtte, M. (2015). Do attitudes toward school influence the underachievement of Turkish and Moroccan minority students in Flanders? The attitude-achievement paradox revisited. Comparative Education Review, 59(2), 332–354. Downey, D. B., Ainsworth, J. W., & Qian, Z. C. (2009). Rethinking the attitude-achievement paradox among Blacks. Sociology of Education, 82(1), 1–19. Foster, K. M. (2004). Coming to terms: A discussion of John Ogbu’s cultural-ecological theory of minority achievement. Intercultural Education, 15(4), 369–384. Gibson, M. A. (1997). Complicating the immigrant/involuntary minority typology. Anthropology & Education Quarterly, 28(3), 431–454. Gillborn, D., & Gipps, C. (1996). Recent research on the achievements of ethnic minority pupils. London: Institute of Education and Office for Standards in Education. Glaser, B. G., & Strauss, A. (1967). The discovery of grounded theory: Strategies for qualitative research. New York: Aldine de Gruyter. Gold, R. (1958). Roles in sociological field observation. Social Forces, 36(3), 217–223. Granovetter, M. (1973). Strength of weak ties. American Journal of Sociology, 78(6), 1360–1380. Hermans, P. (2004). Applying Ogbu’s theory of minority academic achievement to the situation of Moroccans in the low countries. Intercultural Education, 15(4), 431–439. Kao, G., & Tienda, M. (1995). Optimism and achievement: The educational performance of immigrant youth. Social Science Quarterly, 76(1), 1–19. Marsh, H. W., Trautwein, U., Lüdtke, O., & Köller, O. (2008). Social comparison and big-fish-little-pond effects on self-concept and other self-belief constructs: Role of generalized and specific others. Journal of Educational Psychology, 100(3), 510–524. Merton, R. K. (1968). Social theory and social structure. New York: Free Press. Mickelson, R. A. (1990). The attitude-achievement paradox among Black adolescents. Sociology of Education, 63(1), 44–61. Ogbu, J. U. (1987). Variability in minority school performance: A problem in search of an explanation. Anthropology and Education Quarterly, 18(4), 312–334. Ogbu, J. U. (2008). Minority status, oppositional culture, and schooling. New York: Routledge. Oyserman, D., Brickman, D., & Rhodes, M. (2007). Racial-ethnic identity: Content and consequences for African American, Latino and Latina youth. In A. Fuligni (Ed.), Contesting stereotypes and creating identities: Social categories, social identities and educational participation (pp. 91–114). New York: Russell Sage. Paspalanova, M. (2006). Undocumented and legal Eastern European immigrants in Brussels. Leuven: KU Leuven. Piontkowski, U., Florack, A., Hoelker, P., & Obdrzálek, P. (2000). Predicting acculturation attitudes of dominant and non-dominant groups. International Journal of Intercultural Relations, 24(1), 1–26. Portes, A., & Fernández-Kelly, P. (2008). No margin for error: Educational and occupational achievement among disadvantaged children of immigrants. Annals of the American Academy of Political and Social Science, 620(1), 12–36. Richer, S. (1976). Reference group theory and ability grouping: A convergence of sociological theory and educational research. Sociology of Education, 49(1), 65–71. Rosen, B. C. (1956). The achievement syndrome: A psychocultural dimension of social stratification. American Sociological Review, 21(2), 203–211. Sierens, S. (2006). Immigratiesamenleving, onderwijs en overheid in Vlaanderen: Een gespannen driehoeksverhouding. In S. Sierens, M. Van Houtte, P. Loobuyck, DelrueF K., & K. Pelleriaux (Eds.), Onderwijs Onderweg in de Immigratiesamenleving (pp. 9–32). Gent: Academia Press. Siongers, J. (2011). Brusselse jongeren over andere culturen: Van vijand tot lief. In N. Vettenburg, M. Elchardus, & J. Put (Eds.), Jong in Brussel. Bevindingen uit de JOP-monitor (pp. 217–238). Leuven: Acco (TOR 2011/20).

12

L. Van Praag et al. / International Journal of Intercultural Relations 49 (2015) 1–12

Siongers, J. (2013). Bruggen bouwen in Vlaamse grootsteden: Op zoek naar verklaringen voor sociale afstanden tussen verschillende culturen bij Antwerpse en Gentse jongeren. In N. Vettenburg, M. Elchardus, J. Put, & S. Pleysier (Eds.), Jong in Antwerpen en Gent: Bevindingen uit de JOP-monitor Antwerpen-Gent (pp. 224–246). Leuven: ACCO. Suárez-Orozco, C., Pimentel, A., & Martin, M. (2009). The significance of relationships: Academic engagement and achievement among newcomer immigrant youth. Teachers College Record, 111(3), 712–749. Stevens, P. A. J. (2007). Researching race/ethnicity and educational inequality in English secondary schools: A critical review of the research literature between 1980 and 2005. Review of Educational Research, 77(2), 147–185. Suls, J., Martin, R., & Wheeler, L. (2002). Social comparison: Why, with whom, and with what effect? Current Directions in Psychological Science, 11(5), 159–163. Terry, D. J., & Hogg, M. A. (1996). Group norms and the attitude-behavior relationship: A role for group identification. Personality and Social Psychology Bulletin, 22(8), 776–793. Trautwein, U., Lüdtke, O., Marsh, H. W., & Nagy, G. (2009). Within-school social comparison: How students perceive the standing of their class predicts academic self-concept. Journal of Educational Psychology, 101(4), 853–866. Van Kerckem, K., Van de Putte, B., & Stevens, P. A. J. (2013). On becoming ‘too Belgian’: A comparative study of ethnic conformity pressure through the city as context approach. City & Community, 12(4), 335–360. Van Praag, L., Stevens, P. A. J., & Van Houtte, M. (2014). Belgium. In P. A. J. Stevens, & G. A. Dworkin (Eds.), The Palgrave handbook of race and ethnic inequalities in education (pp. 106–137). London: Palgrave. Verhaeghe, P.-P., Van der Bracht, K., & Van de Putte, B. (2012). Migrant zkt Toekomst. Gent op een Keerpunt Tussen Oude en Nieuwe Migratie. Garant: Antwerpen-Apeldoorn. Zhou, M., & Bankston, C. L. (1994). Social capital and the adaptation of the 2nd generation: The case of Vietnamese youth in New Orleans. International Migration Review, 28(4), 821–845.