Does ethnic identification promote integration into the larger society? A study of youth in Oslo

Abstract

This paper uses survey data from Oslo to examine whether strong ethnic identification helps or hinders the integration of adolescents from immigrant families into the larger Norwegian society. Dependent variables are openness to other ethnic groups than one’s own, educational achievement, and positive self-concept. The findings are generally positive for the integration of adolescents with immigrant parents in Oslo. In particular, ethnic identification is positively associated with socio-cultural openness to ethnic others, and with having a positive self-concept. However, ethnic identification does not appear to facilitate educational achievement. Nor does it generally appear to do so when it is combined with strong openness to ethnic others.

Keywords

Immigrant family background integration education self-concept youth ethnic identification openness to ethnic others

Does identification with one’s own ethnic group promote integration of youth from immigrant families into the larger society? Data from a large-scale survey of adolescents aged 14–17 in Oslo will be used to address this question by comparing adolescents of different parental-country origins. To indicate ‘Integration’, the study uses measures under the rubrics of ‘Education’, ‘Health’, and ‘Social Connections’ as proposed for immigrant minorities in the UK by Ager and Strang (2008). ‘Education’ will be indicated by educational effort, achievement, and ambition. Under ‘Health’ a clearly positive self-concept (or self-esteem) is taken to indicate psychological well-being, in keeping with Berry et al. (2006: 329) who argue that high self-esteem better enables minorities to cope with adversity. Drawing on Putnam (2000: 22), Ager and Strang point to ‘bonding social capital’ and ‘bridging social capital’ as ‘Social Connections’ that facilitate integration. ‘Bonding’ will be indicated by socio-cultural identification with one’s ethnic community. ‘Bridging’ will be indicated by openness to ethnic others.

Other analytic points are social-capital arguments about the conduciveness of strong ethnic communities for minority youth’s transition to adulthood, and Sam and Berry’s (2010) theorizing about the context-dependent ‘adaptiveness’ of different acculturation strategy types for minorities. The majority population will be included in the analysis, mainly as a comparison category regarding ethnic identification and openness to other groups.

Ethnic social capital

Coleman argued that cohesive families and strong communities acting in concert with schools are advantageous for adolescents’ education and generally for their transition to adulthood (Coleman, 1988; Coleman and Hoffer, 1987). He emphasized the effect of ‘closed’ intergeneration networks that include adult socialization agents (cohesive families and ethnic communities). Our first focus is to assess whether there are ‘integration benefits’ for minorities from bonding with one’s ethnic group.

Haller et al.’s (2011) ‘segmented assimilation’ theory concerning youth whose immigrant parents have low education and low-status jobs in the US context, likewise sees strong ethnic communities as beneficial for youth’s integration into the larger society. They emphasize risks of ‘downward assimilation’ to an underclass in subordinate manual jobs or with deviant lifestyles if ethnic community ties are weak and when a group suffers racist exclusion.

Qualitative studies emphasizing the benefits of strong families and ethnic communities among minorities include Bankston (2004: 177) who notes that family cohesion among Vietnamese refugees is reinforced by their ethnic community and that the combination of strong families and communities socially regulates youth to value schooling and to stay clear of crime. Other US studies make similar observations: Zhou (1997: 79–80, 2005), and Zhou and Bankston (1994) on East Asians, and Suarez-Orozco (1989) on immigrants from Central America. Modood (2004, 2005) uses similar arguments from social capital theory to explain overrepresentation of youth from South and East Asian parental backgrounds in British higher education. But he notes that such families living remote from co-ethnics also have strong education drives for their offspring.

Studies have noted that immigrant families and communities do not necessarily reproduce home-country patriarchy in education. Sikhs in the US and UK stress education equally for daughters and sons (Gibson and Bhachu, 1991). Basit (1997) shows how educating girls is a family project for UK ethnic Pakistanis and that their daughters strive to excel educationally to repay a ‘gratitude debt’ to parents. In Norway, Fekjær and Stærkebye Leirvik (2011) similarly conclude that gratitude to parents motivates ethnic Vietnamese youth to excel in school. Like Helland (1997), they also found strikingly weak predictive power of parental educational level for adolescents’ academic achievement for this minority. In Norwegian research on Sri Lankan Tamil refugee families, Engebrigtsen and Fuglerud (2007: 3) argue that the second generation’s high social mobility may be due to strong support for education from their ethnic community.

Research has noted that strong social regulation by parents and the ethnic community can be resented by adolescents as an unreasonably strong constraint upon their social life (Engebrigtsen and Fuglerud, 2007: 84; Zhou, 2005), and there is the functionalist argument that peer groups can be valuable learning arenas for autonomy and interacting with equals (Frønes, 1988).

Social regulation by adults and its socialization consequences for adolescents may vary between groups. Ogbu (1987) notes that the parental and community support for upward mobility, which is commonly found among ‘voluntary’ minorities (e.g. immigrants), can be weaker in historically oppressed ‘involuntary minorities’.

The argument about education-conducive ‘ethnic social capital’ has primarily been applied to immigrant minorities. However, quantitative research testing this argument empirically is remarkably scarce. A first task of the present study is therefore to ascertain whether ethnic identification is positively associated with advantageous socialization for integration into the larger society. But it is recognized that such associations do not suffice to demonstrate causality, and that individual variation in ethnic identification and other socialization outcomes could plausibly also result from cognitive capacity to cope with socio-cultural complexity (Brewer and Pierce, 2005; Roccas and Brewer, 2002). Individual variation in a cognitive foundation for attitudes might also apply to moral or civic judgments based on abstract principles rather than context-specific conventions (Lauglo, 2013). It could also extend to identification with larger collectivities than one’s primary and secondary groups.

Acculturation orientations

Berry (2003) posits four acculturation strategies for ethnic minorities. Assimilation refers to adopting the culture of the dominant group but rejecting one’s ethnic background. Integration is socio-cultural retention of one’s ethnic identification combined with openness to the majority. Separation is strong ethnic identification but without such openness. Marginalization would be socio-cultural detachment from one’s ethnic background and from the majority.

Sam and Berry (2010) hypothesize that the ‘adaptiveness’ of different acculturation strategies depends on the majority’s attitudes. A ‘melting-pot society’ with a welcoming common culture could make assimilation adaptive for well-being and functioning in the larger society, but multiculturalist Integration can still be preferred if a minority strongly values its traditional beliefs and lives in ethnic enclaves. They argue that multiculturalist integration is particularly adaptive if the majority genuinely values pluralism. Even separation could be adaptive if diversity is tolerated in social and geographical confinement.

The view that socio-cultural rootedness is a precondition for individual well-being is a legacy of the Romantic Movement in Western countries (Blanning, 2012). In current social psychology a strong group identity is also commonly assumed to affect well-being positively (Tajfel and Turner, 1986). Tellingly, Sam and Berry (2010) perceived no adaptive advantage from marginalization. However, in sociology there is also a ‘positive’ view about detachment from community ties as expressed in Park’s (1928) ‘marginal man’ theory. Notwithstanding his emphasis on the socialization effectiveness of ‘closure’ in intergenerational networks, Coleman (1990: 112) noted that social marginality can be conducive to innovation. Current ‘post-modern’ macro-social diagnoses assert that gender, social-class origin, nationality and ethnicity have come to matter less for individuals’ identity, implying that ethnic marginalization might be less disadvantageous. However, recent findings by Berry et al. (2006) in an international study of more than 5000 youth from immigrant backgrounds in 13 countries (including Norway) support the view that a strong ethnic identity remains advantageous for psychic well-being. Their cluster analysis yielded four groups with distinct acculturation profiles. In accord with Berry (2003), the multicultural ‘Integration’ group (36% of cases) displayed high socio-cultural involvement both with their ethnic group and with the wider society. The next cluster (23%) displayed an ethnic profile: strong involvement with one’s ethnic group but weak involvement in the larger society – akin to Berry’s (2003) separation strategy. A national cluster (19%) displayed positive orientation towards society at large but weak involvement with their ethnic group (resembling Berry’s assimilation strategy). A fourth, diffuse profile (22% of cases), evinced an allegedly ‘confused’ mix of marginalization, separation and assimilation elements. The clusters differed in their ‘psychological adaptation’ as measured by composite scales measuring life satisfaction, self-esteem and absence of diagnosed psychological problems. Those in the integration cluster scored highest, followed by the ethnic cluster, and then by the assimilation-oriented national cluster. The diffuse cluster scored distinctly lowest. Structural equation analysis also supported their conclusion that ‘Ethnic involvement promotes psychological well-being’ (Berry, 2003: 325).

The advantage of a strong ethnic identity fits the view that minorities need collective action to improve their status, as exemplified in the US Civil Rights struggle. If ethnic identification promotes mental health and resilience by collective action in the face of adversity, it may apply both to Ogbu’s (1987) ‘involuntary minorities’ and to ‘voluntary minorities’ facing discriminatory adversity. However, if a voluntary immigrant minority’s culture differs little from the dominant group, assimilation could also facilitate their integration, especially if a minority is physically indistinguishable from the majority. A distinction between ‘visible’ and ‘invisible’ minorities is made in Statistics Canada’s monitoring of labour-market integration of immigrants. ‘Invisible’ minorities’ different situation from other minorities is also recognized in Modood’s (2012) integration typology. However, Berry et al.’s (2006) findings could indicate that assimilation might also be psychologically burdensome for ‘invisible minorities’.

If mainstream society itself is culturally diverse with hybridized identities and relatively harmonious group relations, and/or if the majority is socio-culturally open, a minority may relatively easily combine a strong ethnic identity with an openness to ethnic others. If so, the prevalence among minority groups of a multicultural integration orientation, and the correlation between ethnic identification and openness to ethnic others can proxy the extent to which a minority perceives its larger societal setting to be an inclusive one. Such perceived inclusiveness could also be more common among ‘invisible’ than among ‘visible’ immigrant minorities if the former are less likely to face racist exclusion.¹

Both sides in a US debate between segmented-assimilation theorists and the more optimistic neo-assimilationists perceive big cities increasingly to have an ethnically hybridized mainstream (Alba et al., 2011). In research on the second generation in big cities in eight Western European countries, Schneider and Crul (2012) similarly argue that such settings are increasingly ethnically hybridized and that the second generation identifies especially strongly with its local urban context ‘in which the majority group is losing its capacity to impose assimilative pressure on members of other ethnic groups’ and the ‘parents’ grip on the second generation is weak among its successful members’ (2012: 29). Like Ager and Strang (2008), they emphasize schooling and socio-cultural openness to other ethnic groups as preconditions for ‘success’ (2012: 34) and see socio-cultural openness both as a marker of already achieved integration and as a facilitator of further integration.

Research questions

With Schneider and Crul (2012), this study assumes that integration opportunities are context dependent. Adaptations of Sam and Berry’s (2010) acculturation strategies and of Phinney’s (1992) ‘Multigroup Ethnic Identity’ and ‘Other-Group Orientation’ scales will be used to construct acculturation types which are here labelled ‘orientations’ because they need not be deliberate strategies.

Phinney’s items related to openness to ‘other ethnic groups than your own’ leave these other groups unspecified. Since minorities in Oslo live in a country and city strongly dominated by ethnic Norwegians, one could assume that, for minority youth, openness to ethnic others will usually also include the majority. However, in Oslo, Phinney’s other-group items will also measure openness to an increasingly multi-ethnic city context, and strongly so in Eastside Oslo. In either case, it will be assumed, with Ager and Strand (2008), that the extent of such general openness to ethnic others will constitute bridging social capital for integration into the mainstream social context of Oslo as a whole. In the case of respondents from the majority group, the ‘openness to ethnic others’ will be assumed to refer to ethnic minorities.

The first research questions are: (1) How strongly do adolescents identify with their own ethnic group? (2) How strong is their openness to other ethnic groups than their own? (3) What is the relationship between these two traits?

A prevalent strong openness to ethnic others, and positive correlations between such openness and ethnic identification with one’s own group would, for ethnic minorities, indicate that they perceive Oslo to be a relatively accepting setting. Among the majority, a prevalent strong openness to ethnic others, and a positive correlation (or at least no negative correlation) between ethnic identification and such openness would indicate that the ‘natives’ – also when they identify strongly as ethnic Norwegians – are relatively open to minorities.

For the minorities, a negative correlation could suggest a prevalent separation orientation. For the majority, it could indicate that ethnic Norwegian adolescents with strong national sentiments resent minorities as intruders.

Further research questions are: (4) How well do minorities perform in school as compared to the majority? (5) How much effort do they devote to their schooling? And (6) How strong is their educational ambition? Previous research (for this, see the Context section) suggests that children of immigrants will show greater effort than the majority and also high education ambitions. However, research also indicates that minority groups on average have lower academic performance, and that this performance gap relative to the majority will be largely attributable to parental socio-economic status and, for some, to arrival after the entry age for compulsory schooling.

Previous qualitative studies stressing ‘ethnic social capital for education’ point to a further hypothesis that, within minority groupings, acculturation orientations which include a strong ethnic identity will be positively associated with academic performance. The international study by Berry et al. (2006) also suggests another hypothesis: that a strongly positive self-concept will be more common among adolescents whose acculturation orientation includes strong ethnic identification.

The context

Norwegian nation-building rationales have historically stressed a shared ethnic origin and ordinary folk as carriers of national culture. This populist nationalism may have facilitated early development of broadly based electoral franchise, welfare policies and structures of formal schooling purportedly bridging social-class divides (Lauglo, 1995). However, such populist and historically oriented nationalism can be hard on ethnic minorities, exposing them to ethnic discrimination and/or strong (in some countries) assimilationist pressures and ethnic discrimination by the majority. Norway’s main indigenous ethnic minority, the Sami people, historically faced such pressures. Only since the 1970s have they been accorded special rights as a recognized indigenous minority. For recent generations, Norwegian populist nationalism has been associated with skepticism towards membership in the European Union. In two plebiscites (1972, 1994) such membership was rejected. Nonetheless, Norway joined the EU-associated European Economic Area (EEA) with its internally open labour market.

Since 1975 immigration from non-Nordic countries has been tightly regulated. For non-EEA countries, it has largely been confined to persons joining family residing in Norway, and to political refugees. Some 12% of the population in 2012 was foreign-born.² Recent immigration is mainly employment based and from within the EU (e.g. Poland, Sweden, Lithuania, Germany, Denmark). From non-EEA countries, immigration is mainly for purposes of family unification. Official statistics show that only 19% of immigrants during 1990–2013 (from outside the Nordic countries) have been admitted as political refugees.³ If one adds those coming to join a family member previously admitted as a refugee, 29% of Norway’s immigrant population in 2013 had a ‘refugee background’.⁴ The percentage for Oslo is the same.⁵ Policy on asylum-seekers is hotly contentious. The right-populist and immigration-skeptical Progress Party has in recent decades gained considerable electoral support and has influenced the other main parties’ immigration policies. Valenta and Bunar (2010: 469–470) have compared Norwegian and Swedish programmes for settlement and integration of refugees and note a greater emphasis on coercive elements in Norway’s refugee-settlement programme.

About 40% of immigrants and their children live in, or within commuting distance of, Oslo. Oslo is strongly ethnically stratified (Özerk and Kerchner, 2014). In some primary schools, 9 of 10 pupils are from immigrant families. Minorities are in the majority in some secondary schools. In Oslo East, 40% of the respondents in our sample are from immigrant homes, compared to 7% in Oslo West. Daily interaction with ethnic others will thus vary greatly among school catchment areas.⁶

Steinkeller (2013) shows, using Norwegian national cohort-data, that immigrants’ children perform less well on tests than the majority in basic schooling and leave with lower final grades. Using 2006 PISA-data, Dustmann et al. (2011) show that, in many PISA-countries, parental education matters greatly for achievement gaps between children of immigrants and the majority. In Norway, this gap is especially great in Oslo, largely because majority students have an unusually high proportion of parents with higher education (half the majority adolescents in our sample).

There is ethnic discrimination in the labour market. Midtbøen and Rogstad (2012) submitted 1800 fictitious applications to jobs advertised in Oslo or nearby municipalities. Applicants with foreign-sounding names were less likely to be called for an interview than those with identical formal qualifications but Norwegian-sounding names.

Surveys by Statistics Norway show that a substantial minority of the total population is skeptical of immigration and multiculturalism. Although Blom (2013: 5) showed that 72% of the adult population agreed that ‘Most immigrants make an important contribution to Norwegian working life’, this percentage declined from 2012 to 2013 and it was further reduced if the question referred specifically to non-Nordic immigrants. Agreement was prevalent with the assimilationist statement that: ‘Immigrants in Norway should endeavor to become as similar to Norwegians as possible.’

No study has compared Oslo with other European cities in terms of majority group attitudes to minorities, or minorities’ exposure to discrimination. However, Schneider and Crul (2012) found that Stockholm provides relatively conducive conditions for integration of second-generation immigrants compared to a range of other Western European cities. They emphasize the importance of structural features in Sweden’s education system. These closely resemble those in Norway and Oslo. Within Norway, increasingly multi-ethnic Oslo seems to be a relatively ‘accommodating’ setting for immigrant families (Blom, 2013: 5).

According to Schneider and Crul’s (2012) inferences from other Western European big cities, the Norwegian structure and financing mode of education should be relatively favourable for minority youth’s educational opportunities. School textbooks are free. Public schools and higher education institutions charge no tuition fees. Private schools are rare; and students from ethnic minorities (also of non-Western background) are well represented in private, basic-education schools (Lauglo, 2010). Differentiation between academic and non-academic curricula occurs distinctly ‘late’ (after age 16+). In that post-compulsory stage, access is by legal entitlement and students choose their curriculum type. Public higher education incurs no tuition fees and offers grants and soft loans for living expenses.

Data and methods

Data are used from a 2006 questionnaire-survey of youth in Oslo conducted by NOVA – Norwegian Social Research. The sample was designed to be representative of the two top grades (9 and 10) in compulsory education and in the first grade of upper-secondary education in which nearly all youth enter at age 16+. All students in relevant grades were targeted in 75 schools. Of these, 93% voluntarily completed the omnibus-questionnaire (Valset and Øia, 2006). The achieved sample contains 11,440 youth aged 14–17, large enough to have sufficient cases for univariate and bivariate analysis within the most common parental-country backgrounds.

Adolescents with two immigrant parents constitute 27%. Of these minorities, 78% were either Norwegian-born or had arrived before the school-entry age (6). Most minority adolescents have parents from a predominantly Muslim country. The largest country-background group is Pakistani. Of those adolescents, 88% were either Norwegian-born or had ‘arrived early’. Other relatively large groups (n > 100) are those of Turkish parentage (195), Somalis (174), Sri Lankan (predominantly Tamil refugees) (167), Moroccans (166), Vietnamese (148), Iraqis (127), Indians (114) and Iranians (101). To obtain sufficient cases for multivariate analysis, broad aggregates of ‘visible’ and ‘invisible’ minorities are used.

For the minorities, self-perceived ethnicity is not necessarily indicated by country background, because ethnic identification can refer to a minority within the country, e.g. Kurds in Iraq and Turkey, and Tamils in Sri Lanka. Neither do the data enable analysis of each country-background group’s migration circumstances. However, parental-country background will be used to assess the consistency of minority–majority differences across these backgrounds, and to construct and compare the aggregates of ‘visible’ and ‘invisible’ minorities.

Adaptation of Jean Phinney’s scales

Phinney (1992) developed two ‘group-orientation scales’ that have been widely used in the US and internationally (Cokley, 2007). To counter response-sets, in that battery of questions the items on ‘ethnic identification’ are interspersed with those measuring openness to ‘other ethnic groups than your own’ and some items have ‘negative’ contents. The Oslo-survey used a translated sub-set of Phinney’s ‘Multigroup Ethnic Identification’ and ‘Other-Group Orientation’ items. To signal that our scales are based on such sub-sets, our scales will be labelled differently from Phinney’s measures: ‘Ethnic identification’ and ‘Openness to other ethnic groups’.

Findings

Tables 1 and 2 include only the minority groups (respondents with two immigrant parents). Of these, 85% are from a developing-country background (here inferred to be ‘visible minorities’). The others (‘invisible minorities’) are predominantly from a European country background.

Table 1.

Percentage distributions of ‘Ethnic Identification’ items. Adolescents from immigrant families.

Statement	Response options				Total
Statement	Strongly agree	Somewhat agree	Somewhat disagree	Strongly disagree	Total	N
I have tried to find out more about my ethnic group, such as its history, traditions and customs.	27.5	39.4	21.0	12.1	100.0	2854
I participate actively in organizations or voluntary associations whose members are mainly from my own ethnic group.	14.4	28.9	28.3	28.4	100.0	2804
I have a clear sense of my background and what it means for me.	53.1	34.5	9.6	2.8	100.0	2835
I feel strongly that I belong in my ethnic group.	39.8	37.6	15.9	6.8	100.0	2785
I understand quite well what my background means to me, in terms of how I relate to my own group and other groups.	43.4	39.4	12.9	4.2	100.0	2779
I am very proud of my ethnic group.	63.1	25.0	8.2	3.7	100.0	2792
I take part in my ethnic group's traditions (such as those relating to food, music, dance and other customs).	44.5	34.1	14.1	7.3	100.0	2793
I have a strong sense of attachment to my own ethnic group.	39.4	36.9	17.6	6.1	100.0	2738

N including ‘no answer’ = 2979.

Table 2.

Percentage distributions of items measuring ‘Openness to other ethnic groups’. Adolescents from immigrant families.

Statement	Response options				Total
Statement	Strongly agree	Somewhat agree	Somewhat disagree	Strongly disagree	Total	N
I like meeting and getting to know people from ethnic groups other than my own.	59.1	31.2	7.1	2.5	100.0	2846
I often spend time with people from ethnic groups other than my own.	52.6	31.6	12.0	3.8	100.0	2844
I participate in activities with people from other ethnic groups.	40.9	35.0	17.0	7.1	100.0	2804
I like being with people from ethnic groups other than my own.	47.8	33.9	12.8	5.5	100.0	2797
I don't want to be friends with people from other ethnic groups. (Response scoring was reversed for scaling)	6.0	8.0	14.4	71.6	100.0	2806

N including no answer = 2979.

Ethnic identification

Table 1 shows distributions of valid responses on the items used in our Ethnic Identification scale. There is prevalence of ‘agreement’ on all items apart from membership in ethnic organizations. The items are a mix of cultural perceptions or attitudes and indicators of interaction. They are accordingly taken to measure socio-cultural ethnic identification.

Openness to other ethnic groups

Table 2 shows distributions of valid responses to indicators of ‘Openness to other groups’—again a mix of attitudes and social interaction. Agreement clearly prevails on all statements expressing openness while disagreement prevails on the item expressing social distance from ethnic others. Thus, while these minorities tend to identify with their ethnic origin, on average they also express strong socio-cultural openness to ethnic others.

Scale averages by ‘country background’

Likert-type scales were constructed from the items in Tables 1 and 2. Responses were scored from 1 for ‘Disagree strongly’ to 4 for ‘Agree strongly’, and intermediate responses were assigned values 2 or 3. Scoring was reversed for the one item with ‘negative’ contents in Table 2. Each respondent’s scale-score is the mean item-score. Majority respondents were similarly scored. Both scales showed high internal consistency: α ≥ 0.81 on Ethnic Identification for the whole sample, as well as in separate analyses for the majority and for the aggregate of all the minorities. The Identification scale correlated r ≥ 0.99, with the first factor extracted from its components. Though having only five components, the Openness-scale’s ∝-coefficients were adequately high: 0.76 for total sample, 0.78 for the majority and 0.64 for the combined minorities. It correlated r ≥ 0.98 with its first factor.

Ethnic identification

Phinney (1992) found that minority youth in the US have a clearer concept of their own ethnicity than do the majority. In Oslo too, the mean ‘Ethnic Identification’ score is significantly higher for minorities than for the majority (p(t) < .001). All these means, including for the majority group, show prevalently positive ethnic identification (>2.5). Boys scored significantly higher than girls only in the majority group.

Openness to ethnic others

The mean ‘Openness to other groups’ is consistently strong (≥3), also for adolescents from backgrounds likely to contain mostly refugees from persecution in their country of origin (Kurds in Iraq and Iran, Tamils in Sri Lanka). Excepting the small sample with Iraqi backgrounds, all minorities had significantly higher means than the majority. Girls generally showed stronger openness across these sub-samples, but few gender differences were statistically significant. All in all, Tables 1 –3 show that valuing one’s ethnic group and being socially open to other groups are both strongly prevalent among these adolescents, especially among the minorities.

Table 3.

Mean scores on ‘Ethnic Identification’ and ‘Openness’ for other ethnic groups, and correlations between the scales. By parental-country origin.

Parental-country origin	Mean score on scales		Correlation (r) between the scales	N
Parental-country origin	Ethnic identification	Openness to other groups	Correlation (r) between the scales	N
Norway	2.76^b	3.02	0.04	7140
Pakistan	3.12^a	3.31^a	0.28	727
Turkey	3.18^a	3.28^a	0.27	163
Somalia	3.18^a	3.24^a	0.16	145
Sri Lanka	3.13^a	3.30^a	0.26	148
Morocco	3.22^a	3.48^a	0.42	134
Vietnam	2.92^a	3.31^a	0.27	130
Iraq	2.97^a	3.35	0.26	99
India	3.11^a	3.34^a	0.25	98
Iran	2.99^a	3.51^a	0.43	90
Other ‘visible minorities’ predominantly other Asian or African)	3.05^a	3.37^a	0.29	361
‘Invisible minorities’ (European/OECD countries)	2.94^a	3.34^a	0.29	429
Total	2.85	3.10	0.13	9664

Significantly higher than for the majority at p(t) < .05.

Significantly higher for boys than girls at p(t) < .05.

Correlations between ethnic identification and openness

The relationship between ethnic identification and openness to ethnic others is likely to reflect prevailing inter-group relations in a given context. Using Phinney’s scales, Kosic and Caudek (2005) found in an Italian city (near the Slovenian border) that majority-group adolescents with strong ethnic identification had ‘more negative responses towards ethnic outgroups than did participants with lower ethnic identification’. Among the Slovene minority, however, ethnic identification and outgroup openness were positively related. Similarly, a Dutch study by Masson and Verkuyten (1993) reported that the majority-group adolescents’ prejudicial attitudes to ethnic others were greater among those identifying strongly with their own ethnicity, but they found no such negative association among minority youth.

In Oslo, stronger ethnic identification does not entail reduced openness to ethnic others. Table 3 shows, for all minorities, consistently positive correlations of weak to moderate strength (0.25–0.43) between the scales. All but one are statistically significant. For those from majority backgrounds the Pearson correlation is of totally negligible magnitude (0.04); but it is noteworthy that it is not negative. Thus, in Oslo, stronger national identification as ethnic Norwegian does not tend to entail reduced openness to minorities.

The positive correlations in Table 3 are stronger than in Phinney’s (1992: 167) initial US findings. However, Phinney et al. (1997: 967) later concluded from other US samples that ‘ethnic identification… makes an indirect contribution to [positive] intergroup attitudes’.

In general, ‘openness to ethnic others’ appears to be a strongly dominant attitude among Oslo adolescents. Among minorities it is consistently positively associated with ‘Ethnic identification’. Among the majority, strong identification with Norwegian identity does not tend to occur at the expense of less ‘Openness to ethnic others’, among these adolescents aged 14–17.

Do identity and openness depend on urban ethnic stratification?

Allport’s (1954) classic ‘contact hypothesis’ could imply that openness to others would be enhanced by local opportunities for cross-ethnic interaction. Oslo is ethnically strongly stratified between its Westside and the bulk of the Eastside. Frøyland and Gjerustad (2012) show that this residential ethnic divide does affect the likelihood of other-group inclusion in majority-group adolescents’ friendship networks. However, they did not find such a pattern for the minorities. As previously noted, Schneider and Crul (2012) argue that exposure to assimilationist pressures and ethnic discrimination will tend to ease for second-generation immigrants in European big-city environments as these settings grow more multi-ethnic. Do the mean scores on our two scales and the correlations between them differ between Oslo East and Oslo West?

To examine this possibility, the sample will be geographically dichotomized between Oslo-West and Oslo-East. Two neighbourhoods in Oslo’s eastern hillside (Østensjø and Nordstrand) which have low minority proportions will be grouped with Oslo-West. To achieve large enough numbers of cases for cross-classification, the sample will be further subdivided into three ethnic groupings: (a) ‘Majority population’ (at least one parent born in Norway); (b) ‘Invisible minorities’: both parents immigrated from a European country/country predominantly settled by Europeans; and (c) ‘Visible minorities’ (both parents immigrated from developing countries). Using these aggregated country-background groupings, Table 4 shows that 92% of the sample in Oslo-West are from the majority group, while 40+% are from minorities in Oslo-East.

Table 4.

Distributions by majority/minority groupings in Oslo-West and Oslo-East.

	Oslo-West	Oslo-East	Total
Majority population	91.8	57.5	74.3
‘Invisible Minorities’	2.7	4.3	3.5
‘Visible Minorities’	5.5	38.2	22.2
Total	100.0%	100.0%	100.0%
N	5058	5285	10,343

Table 5 shows, nonetheless, much similarity between the majority-dominated West and the multi-ethnic East in young people’s socio-cultural attitudes. In both divisions, the mean score on both scales is close to 3, indicating quite strong ethnic identification and openness. For the small grouping of ‘invisible minorities’, the mean scores on both scales are higher in the East than in the West. But that is not so for the much larger ‘visible minorities’ whom one might have expected would feel more at home in the strongly multi-ethnic East than in very strongly majority-dominated West.

Table 5.

Scales of ‘Ethnic Identification’ and ‘Openness to Others’. Means and correlation between scales. By ethnic grouping in Oslo-West and Oslo-East (N in parentheses).

	Mean scores on scales 1–4 scales				Correlation (r) of ‘Ethnic Identification’ with ‘Openness to others’
	‘Ethnic Identification’		‘Openness to Others’
Ethnic groupings	Oslo-West	Oslo-East	West	East	West	East
Majority	2.77 (4174)	2.74 (2681)*	2.92 (4373)	3.17 (2827)*	.02 (4069)	.07*^,† (2611)
Invisible minorities	2.72 (120)	3.09 (200)**	3.21 (128)	3.40 (212)**	.31†† (128)	.26^† (195)
Visible minorities	3.08 (245)	3.10 (1690)	3.35 (258)	3.34 (1809)	.37^†† (239)	.26^†† (1628)

Significance East-West difference: *p ≤ .05, **p ≤ .005.

†

Significant r at p ≤ .05, ^††Significant r at p ≤ .005.

In both East and West, the scales are positively correlated for both minority groupings. These correlations are clearly not stronger in the East. Rather, the trend is stronger correlations in the West, but these differences are not statistically significant.

For the majority, however, the East–West difference in mean score is statistically significant on both scales, but the difference is miniscule on ethnic identification. The majority’s openness to ethnic others is on average clearly higher in the East, paralleling Frøyland and Gjerustad’s (2012) findings regarding friendship networks (on other survey data). Similarly, the positive correlation between the scales is also stronger among majority youth on the Eastside, but the magnitude of the ‘r’ is small. However, the findings provide no support at all for any fear of an ‘ethnic backlash’ among majority youth in Oslo.

Overall, it appears that the attitudinal conditions for integration differ very little between those parts of Oslo where cross-ethnic contact is much more likely to occur at school or in the neighbourhoods, and those parts which remain strongly dominated by the majority population. Thus, the inferences for Oslo as a whole, made on the basis of findings in Table 3, are not altered by separate analysis of the strongly majority-dominated neighbourhoods as compared to multi-ethnic neighbourhoods.

The importance of education is stressed in Norwegian policies for integration of immigrant minorities. Schooling matters as a precondition for immigrant children’s future integration though educational achievement is not a sufficient condition for equitable labour market access (Midtbøen and Rogstad, 2012). Though the catchment areas of public schools in Oslo reflect strong residential ethnic stratification, both public and private schools (rare in Oslo) are arenas for some cross-ethnic contact. The medium of instruction and the curriculum of schools transmit a common cultural foundation.

Table 6 shows findings on educational effort, achievement and ambition. Effort is estimated from the amount of time checked by respondents in answers to a question about how much time they spend on average on homework each day. Achievement is the grade-point average (GPA) based on grades (scaled 1-6) in Norwegian, Mathematics and English which students say they received ‘on their most recent report card’. Ambition is the percentage choosing university-level higher education from response options to the question: ‘What education do you plan to obtain?’

Table 6.

Students’ educational effort, achievement (grade-point average) and percent expecting higher education. By parental-country origin.

Parental-country origin	EFFORT. Daily hours spent on homework	ACHIEVEMENT. GPA in Norwegian, Math’s & English	Estimated GPA if mean SES & ‘born in Norway/ arrived before age 6’	% expecting higher education	Smallest N with valid information on any indicator
Norway (majority grp)	0.9^a	3.9^a	3.8	62.2^a	5467
Pakistan	1.3* ^a	3.4* ^a	3.8	64.5 ^a	558
Turkey	1.1* ^a	3.1*	3.5*	48.5* ^a	132
Somalia	1.2*	3.3*	3.8	63.5	115
Sri Lanka	1.7* ^a	3.7*	4.1*	77.9* ^a	131
Morocco	1.3* ^a	3.4*	3.8	59.8^a	107
Vietnam	1.2*	3.8*	4.1*	55.3	103
Iraq	1.1*	3.3*	3.6*	58.8	80
India	1.3* ^a	3.8	4.0*	74.1*	85
Iran	1.1*	3.4*	3.7	77.9*	68
Other African/ Asian country	1.3* ^a	3.5*	3.8	65.2	336
European/ OECD country	1.2*	3.8*	3.9*	67.0	291
Total	1.0^a	3.8^a		62.9^a	7473

Differs significantly from the majority (p(t) < .05).

Significantly higher for girls.

Table 6 shows that all minorities, on average, devote more time to homework than the majority (see column 1). But their GPA is consistently lower (column 2). It also shows that sampled students with Indian, Sri Lankan, Vietnamese or European/OECD-country parentage perform better than other minorities.

Respondents from the majority group in Oslo have highly educated parents: 50% have higher education. Similarly, so do respondents of European/OECD-country parentage (46%). The percentage with such highly educated parents is clearly lower among respondents from the various ‘visible minority’ backgrounds (e.g. 14% if from a Pakistani background, 6% if a Turkish background). About one-fifth of the minority adolescents had arrived in Norway after schooling commences (age 6). Such late arrival tends to affect their Norwegian language development adversely. Both late arrival and lower parental socio-economic status are likely to explain some of the minorities’ lower educational achievement as compared to majority-group adolescents. The question arises whether other home-background characteristics help explain their lower GPA (e.g. home and school language gap or parents being generally less familiar with the school system).

To test this possibility, column 3 shows adjusted GPA-estimates for the minorities, based on regression analysis that counterfactually assumed that all minority students had the same socio-economic family status (SES)⁷ as the whole-sample mean, and that they either were born in Norway or had arrived before age 6. Nearly all negative performance gaps from the majority then disappeared. For some backgrounds the difference even turns positive. It thus appears in these data that the lower mean achievement among minorities can be attributed to lower socio-economic family status or (for some of them) to late arrival.

As shown in column 4, the level of early ‘expectation of higher education’ (those answering ‘not decided yet’ are also part of the percentages’ base) is distinctly high across the parental-country backgrounds, in keeping with earlier research on immigrant children in Oslo (e.g. Lauglo, 2001). Students with Sri Lankan, Indian, or Iranian backgrounds have distinctly high ‘expectations’. Most other minorities do not differ significantly from the majority. Separate analysis for each background group showed either no significant gender effect or that girls, more often than boys, expect higher education. For the majority, and for those with Pakistani, Turkish or Moroccan parental-country backgrounds, logistic regressions also showed a positive effect for being a girl, net of school grades. Though referring only to ‘expectation’ among adolescents (or more cautiously interpreted: ‘ambition’), the findings bode well for prospectively gender-equitable educational attainment among the minority groups in Oslo, even when the parents have immigrated from reputedly strongly patriarchal countries.

Does ethnic identification predict academic performance?

Is a strong ethnic identification an advantage for minority youth’s educational performance? Bivariate linear regression of GPA on ethnic identification was conducted for each country-background category in Table 6. It failed to support the previously noted arguments from qualitative studies of a positive ‘ethnic social capital’ effect upon educational achievement. For the large aggregate sub-sample (n = 2022) of ‘invisible minorities’ (European/OECD-country backgrounds), the bivariate coefficient was even significantly negative but of negligible magnitude (β = −0.11). Nor did control for SES (described in footnote 7) reveal any net association between GPA and Ethnic identification for any of the country backgrounds shown in Table 6.

Berry's (2003) acculturation typology and Sam and Berry’s (2010) theorizing about context-dependent effects of acculturation types suggest the possibility of context-dependent effects of acculturation type. Are there effects of acculturation type in the Oslo-context?

Acculturation types

An acculturation typology was constructed using the scales shown in Tables 1 –3. Those with clearly ‘positive’ scores on both ‘Ethnic identification’ and ‘Openness to others’ (at least 3) were classified as having a ‘Multicultural integration’ acculturation orientation. Those scoring less than 3 on both scales were classified as ethnically ‘Marginalized’ (though this label could exaggerate the meaning of ‘not high’ scores).⁸ Those with high scores on ‘Ethnic identification’ but relatively low scores on ‘Openness to others’ were classified as having an ethnic ‘Separation orientation’. For majority students, who will be included in our analysis to provide a comparison group, the combination of strong ‘Openness to ethnic others’ and ‘Low ethnic identification’ could be taken to signal a ‘Cosmopolitan’ orientation. Among minority respondents, however, relatively low ‘Ethnic identification’ but high ‘Openness to ethnic others’ was assumed to usually indicate an ‘Assimilation’ orientation (though as noted, it could for also for some minority respondents indicate a cosmopolitan orientation).

Table 7 shows percentage distributions of these orientations by type of parental-country origin. To obtain at least 20 valid observations in each cross-classification of acculturation type by parental-country background, all but one country-specific backgrounds were aggregated into the distinction between ‘visible’ and ‘invisible’ minorities, resulting in these groupings: ‘Majority population’; ‘Invisible minorities’ and two groupings of ‘visible minorities’: ‘Pakistani background’ and ‘Other developing countries’. Mostly because of loss of observations, with no response on one or more the 15 items required either for the acculturation typology or for parental-country background, the remaining sample to be analysed (n = 9399) lost 18% of the observations in the original sample of survey respondents.

Table 7.

Distributions of acculturation-orientation types by parental-country origin.

Type of acculturation orientation	Parental-country origin
	Majority population		‘Invisible minorities’ (European/OECD country)		‘Visible minorities’
	Majority population		‘Invisible minorities’ (European/OECD country)		Pakistan		Other developing countries
	%	n	%	n	%	n	%	n
Multicultural integration	24.6	1712	48.3	171	53.8	377	54.9	763
For minorities: ‘Assimilation Integration’. For the majority: ‘Cosmopolitanism’.	34.5	2400	31.1	110	24.3	170	25.1	349
Separation	16.5	1147	6.8	24	11.6	81	9.1	127
Ethnic marginalization	24.4	1695	13.8	49	10.4	73	10.9	151
Total	100.0	6954	100.0	354	100.0	701	100.0	1390

In Table 7, the main contrast in the distribution by acculturation orientation is between the minority groupings and the majority. In keeping with Phinney (1992: 171), the minorities are more likely to have an orientation type involving strong ethnic identification (either ‘Integration’ or ‘Separation’).

The section on context set out the broad contextual characteristics against which these findings may be assessed. It points to Norway’s tradition of a relatively populist variety of nationalism (Lauglo, 1995), little immigration from non-European countries until the most recent generations, as well as the recent rise to political influence of the immigration-skeptical Progress Party. Against these features, the clear prevalence among majority-group adolescents (about 6/10) of orientation types that include openness to minorities (either Multicultural Integration or Cosmopolitanism) might seem surprising. However, most ethnic Norwegian adolescents in Oslo were clearly not minority-skeptical locals in 2006. At least two generations of ethnic Norwegians had, by 2006, grown up without major national events likely to reinforce a defensively Norwegian ethnicity. There has been peace, steadily growing affluence, and a relative absence of ethnic conflicts. The low unemployment facilitated by Norway’s oil wealth may also have induced majority youth to perceive little competition from immigrant minorities. Human rights and internationalism have been much emphasized in recent decades, not only in school curricula, but also in much public discourse.

Immigration skepticism is a minority view among the ethnic majority in Oslo. The Progress Party received 22% of the vote nationally and was the third-largest party in the 2005 parliamentary election (17% in Oslo). In the mock elections concurrently held among upper-secondary students, the party’s nation-wide support was 25%. In Oslo, these mock elections showed 16.2% support for the Progress Party⁹ – nearly identical to the 16.5% of majority students showing strong ‘Ethnic identification’ but relatively low ‘Openness to ethnic others’ in 2006 (Table 7).

Across the four minority groupings, about 50% have a ‘Multicultural Integration’ orientation (strong ‘Ethnic identification’ and ‘Openness to ethnic others’). However, an ‘Assimilation Integration’ orientation is also quite common, and more so for ‘invisible minorities’. Perhaps more surprisingly, it also characterizes about one-fourth of the ‘visible minorities’ for whom ‘Assimilation’ may appear more difficult to achieve. That ‘Multicultural Integration‘ and ‘Assimilation’ together contain nearly 80% of minority youth, bodes well for harmonious inter-group relations in Oslo – though such attitudes can of course be modified later in life (e.g. negatively, by encountered labour-market discrimination).

‘Multicultural Integration’ is both feasible and desirable to the ‘visible minorities’ since nearly half the adolescents from such backgrounds display this orientation type. If it were also generally conducive to a good life, one might expect it to correlate positively with psychic well-being. If it also is conducive for academic achievement (or arising from cognitive capacity for such achievement) one would also expect it to be associated with the GPA measure shown in Table 6. For ‘invisible minorities’, one might expect that ‘Assimilation orientation’ would be equally adaptive for their integration into the mainstream of Norwegian society. If cutting one’s bonds with ethnicity entailed no psychic hardship, one might also expect it to be associated with personal self-esteem. Do the present data support these expectations?

Does acculturation type predict educational achievement?

Table 8 shows the average GPA across three academic subjects by type of ‘Acculturation orientation’ and parental-country origin. Assimilation-oriented ‘invisible minorities’ do obtain better grades than others with that type of parental ethnic background. But significance of this association did not survive control for SES.

Table 8.

Mean grade-point average^a (with n in brackets) by acculturation orientation and parental-country origin.

Parental-country origin	Acculturation orientation				Total	F-ratio	p(F)	ɛ (Eta-coeff.)
Parental-country origin	Marginalization	Separation	Assimilation integration^b	Multicultural integration	Total	F-ratio	p(F)	ɛ (Eta-coeff.)
“Invisible minorities” (Ethnically European)	3.87	3.48	3.98	3.77	3.83	2.78	.04	.16
“Invisible minorities” (Ethnically European)	(n = 47)	(23)	(108)	(164)	(342)	2.78	.04	.16
“Visible minorities” A. Pakistan	3.36	3.39	3.39	3.55	3.47	2.26	.08	.10
“Visible minorities” A. Pakistan	(66)	(75)	(164)	(357)	(662)	2.26	.08	.10
B. Other visible minorities	3.36	3.34	3.62	3.59	3.55	6.06	<.0005	.12
B. Other visible minorities	(138)	(113)	(330)	(715)	(1296)	6.06	<.0005	.12
Total minorities	3.45	3.37	3.62	3.60	3.57	6.87	<.0005	.09
Total minorities	(251)	(211)	(602)	(1236)	(2300)	6.87	<.0005	.09
Majority population	3.90	4.00	4.00	4.00	3.97	6.01	<.0005	.05
Majority population	(1635)	(1120)	(2320)	(1653)	(6728)

Mean GPA for Mathematics, Norwegian and English in most recent self-reported report card from school.

This combination should likely be interpreted as ‘Cosmopolitanism’ for the majority population. In parentheses: n containing valid information. Since each GPA-entry relies on answers to many questions, there is 18% of cumulative loss of observations excluded due to non-response on at least one required item.

Within the relatively large group (n = 662) with Pakistani parentage, those with ‘Multicultural-Integration’ orientation have higher GPA than others – in keeping with multiculturalist education arguments. However, the association with ‘acculturation type’ is of negligible magnitude (as measured by ɛ) and not significant at the 5% level. Among ‘Other developing countries’, adolescents with ‘Assimilation Integration’ or ‘Multicultural Integration’ orientations both had higher mean GPA than others; and these significant differences survived control for parental SES. Both orientations thus appear to be about equally ‘advantageous’ for academic achievement.

‘Multicultural Integration’ has been promoted by multiculturalist educators (e.g. Banks and Banks, 2010). All in all, the present findings from Oslo do not show any general advantage of this acculturation type as compared to the ‘Assimilation’ type, though there is a weak tendency among adolescents with Pakistani background in support of a multiculturalist stance. Adolescents with a ‘Multiculturalist’ orientation do not, however, perform better than those with an ‘Assimilation’ orientation, neither among the aggregated ‘other visible’ nor among the ‘invisible’ minorities. However, education (and life in general) is not all about academic performance. It could be that ‘Multicultural integration’ stands out positively with regard to young people’s sense of personal well-being.

Is acculturation type associated with self-concept?

A positive self-concept (high self-esteem) is commonly thought to indicate psychic well-being. The data include five questions regarding self-concept which have been adapted and translated into Norwegian from items developed by Susan Harter in the US (Wichstrøm, 1995). In their direct translation back to English the statements were: ‘I am often disappointed in myself’, ‘I don’t like the way I live my life’, ‘I am by and large satisfied with myself’, ‘I like myself the way I am’, and ‘I am very satisfied with how I am’. The response options were: ‘Very much agree’, ‘Mostly agree’, ‘Mostly disagree’ and ‘Very much disagree’. Non-response was rare (3%). A Likert scale of mean score was initially constructed, using 1–4 scores from 1 for ‘Very much disagree’ to 4 for ‘Very much agree’ (reversed for the ‘negative’ items). The scale had high internal consistency (α = .83), but was so strongly skewed towards high scores that it was converted to a dichotomy for ‘Clearly positive self-concept’. Table 9 shows the percentage having such positive self-concept (a mean item-score ≥ 3), by type of acculturation-orientation and parental-country origins.

Table 9.

Percentage having ‘Clearly positive self-concept’. By Acculturation orientation and Parental-country origins.

Parental-country origin	Acculturation orientation				Total	χ ²	p(χ²)
Parental-country origin	Marginalisation	Separation	Assimilation integration	Multicultural integration	Total	χ ²	p(χ²)
“Invisible minorities”	60.4	70.0 ^b	47.1	64.8	59.0	9.25	.026
“Invisible minorities”	(n = 48)	(20)	(102)	(159)	(329)
“Visible minorities” A. Pakistani	52.4	63.8	52.9	69.1	63.1	15.76	.001
“Visible minorities” A. Pakistani	(63)	(69)	(155)	(369)	(656)
B. Other visible minorities	40.1	63.4	55.9	69.8	62.6	51.98	<.0005
B. Other visible minorities	(137)	(112)	(333)	(716)	(1298)
Total minorities	47.2	64.2	53.6	69.0	62.2	67.13	<.0005
Total minorities	(248)	(201)	(590)	(1244)	(2283)
Majority	49.4	59.9	49.0 ^a	61.5	54.0	91.21	<.0005

This combination of low ‘Ethnic identification’ and high ‘Openness to ethnic others’ is for the majority taken to indicate Cosmopolitanism. That could also apply to some cases with a minority background.

This percentage refers to only 14 cases of 20 – too few for interpretation.

A strong self-concept is strongly prevalent in all background groupings, and even more so among minority youth than among the majority. In each grouping of parental-country origin, the differences across acculturation types are statistically significant. In line with Berry et al. (2006), those with ‘Multicultural-Integration’ and those with ‘Separation’ orientations (both of which involve a strong ‘Ethnic Identification’) both have on average stronger self-esteem. Among ‘visible minorities’, it is those with a ‘Multicultural-Integration’ orientation who most often have a strongly positive self-concept. Thus, the orientation-types that include strong ‘Ethnic identification’ are associated with a positive self-concept. For ‘visible minorities’ too, a positive self-concept is more common among persons with a ‘Multicultural-Integration’ orientation than among those with an ‘Assimilation’ orientation (p(χ²) = .003).¹⁰

For the majority too, a positive self-concept is more common among those whose orientation includes a strong ‘Ethnic identification’. Thus, for the majority and minorities alike, high self-esteem is positively associated with acculturation types which include strong ‘Ethnic identification’ – in keeping with Phinney’s (1992: 156) findings regarding minority youth in the US and with the more recent international findings in Berry et al. (2006).

Conclusion

Most adolescents in Oslo strongly identify with their ethnic group. Even more commonly, they express strong ‘Openness to ethnic others’. Minorities stand out as compared to the majority group in these respects. For no group does strong ‘Ethnic identification’ appear to occur at the expense of reduced ‘Openness to ethnic others’. Among the minorities, the consistent trend is for these traits to be positively associated. The types of acculturation orientation (‘Multicultural Integration’ and ‘Assimilation’) that involve strong ‘Openness to ethnic others’, together account for some 80% of adolescents from immigrant families.

Educational achievement will matter for the integration prospects of minority youth. Minority adolescents display high educational ambition and devote more time to homework for school than majority adolescents. Thus, their motivation to make use of educational opportunity is strong. They generally perform less well academically than the majority. But this gap can be explained by their lower parental socio-economic status and by some minority youth having arrived in Norway after schooling has commenced for their age-mates. It seems likely that other possible ‘family disadvantages’ for schooling in Oslo (e.g. a home language other than Norwegian, parents being less familiar with Norwegian education) are outweighed by strong motivation and effort (working harder, having high educational ambition).

Notwithstanding findings in earlier international and Norwegian research about ‘ethnic social capital for education’, a strong ethnic identification does not generally boost academic performance. One might expect that such identification would nonetheless be conducive for educational achievement when it is combined with openness to ethnic others – in accord with multicultural-education arguments. That expectation might, to some extent, fit those with Pakistani parental origin, but it does not, on average, fit other minorities in Oslo. However, an acculturation orientation that includes strong ethnic identification is associated positively with a positive self-concept – in line with findings by Berry et al. (2006). This holds, not only for ‘visible minorities’, but also for ‘invisible’ ones. It also holds for the majority group.

The trends do not support fears about ethnic polarization among adolescents in Oslo. Inter-group relations were predominantly good in 2006 – both in the multi-ethnic parts of the city and in the strongly majority-dominated Oslo-West; and ethnic identity did not impede the integration of minorities into mainstream society. Nor does a strong Norwegian ethnic identity tend to breed xenophobia among majority youth. Exceptions can, of course, occur as shockingly illustrated in July 2011 by the mass killings perpetrated in Oslo environs by an Islamophobic terrorist. But as of 2006 the trends shown in this study indicate broad tolerance and mutual understanding among adolescents in this increasingly multi-ethnic urban context. A strong ethnic identity does not tend to undermine such socially desirable outcomes. It does not appear to promote academic achievement but it is associated with a positive self-concept among the minorities and the majority alike. Among minority adolescents, a stronger ethnic identity is most significantly associated with stronger socio-cultural openness to ethnic others—also in parts of Oslo where minorities are rare. However, the causal mechanism behind these modestly positive associations remains to be unravelled.

Footnotes

Acknowledgements

The author is grateful to NOVA-Norwegian Social Research for access to data, to Fengshu Liu of my own Department for help with the literature search, to Tormod Øia, NOVA and four anonymous referees for helpful critique of a draft and additional literature references, and to Minja Tea Dzamarija of Statistics Norway for a special tabulation on refugee immigrants in Oslo.

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

Notes

References

Ager

Strang

(2008) Understanding integration: A conceptual framework. Journal of Refugee Studies 21(2): 166–191.

Alba

Kasinitz

Waters

(2011) The kids are (mostly) alright: Second-generation assimilation: comments on Haller, Portes and Lynch. Social Forces 89(3): 763–773.

Allport

(1954) The Nature of Prejudice, Cambridge, MA: Perseus.

Banks

CAB

(2010) Multicultural Education, Hoboken, NJ: Wiley.

Bankston

(2004) Social capital, cultural values, immigration, and academic achievement: The host country context and contradictory experiences. Sociology of Education 77(2): 176–179.

Basit

(1997) “I Want More Freedom, But Not Too Much”: British Muslim girls and the dynamism of family values. Gender and Education 9: 425–440.

Berry

(2003) Conceptual approaches to acculturation. In: Chun

Organista

Mann

(eds) Acculturation: Advances in Theory, Measurement, and Applied Research, Washington DC: American Psychological Association.

Berry

Phinney

Sam

(2006) Immigrant youth: Acculturation, identity, and adaptation. Applied Psychology 55(3): 303–332.

Blanning

(2012) The Romantic Revolution, New York, NY: Random House.

10.

Blom

(2013) Holdninger til innvandrere og innvandring 2013, Oslo: Statistics Norway.

11.

Brewer

Pierce

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

12.

Cokley

(2007) Critical issues in the measurement of ethnic and racial identity: A referendum on the state of the field. Journal of Counseling Psychology 54(3): 224–234.

13.

Coleman

(1988) Social capital in the creation of human capital. American Journal of Sociology 94(suppl): 95–120.

14.

Coleman

(1990) The Foundations of Social Theory, Cambridge: Harvard University Press.

15.

Coleman

Hoffer

(1987) Public and Private Schools, New York, NY: Basic Books.

16.

Dustmann

Frattini

Lanzara

(2011) Educational achievement of second generation immigrants: An international comparison. Economic Policy 27(69): 143–185.

17.

Engebrigtsen

Fuglerud

(2007) Ungdom i flyktningefamilier. Familie og vennskap – trygghet og frihet?, Oslo: NOVA.

18.

Fekjær

Stærkebye Leirvik

(2011) Silent gratitude: Education among second-generation Vietnamese in Norway. Journal of Ethnic and Migration Studies 37(1): 117–134.

19.

Frønes

(1988) De likeverdige. Om sosialisering og de jevnaldrendes betydning, Oslo: Gyldendal.

20.

Frøyland

Gjerustad

(2012) Vennskap, utdanning og framtidsplaner, Oslo: NOVA Rapport 5/2012.

21.

Gibson

Bhachu

(1991) The dynamics of educational decision making: A comparative study of Sikhs in Britain and the United States. In: Gibson

Ogbu

(eds) Minority Status and Schooling, New York, NY: Garland, pp. 63–95.

22.

Haller

Portes

Lynch

(2011) Dreams fulfilled, dreams shattered: Determinants of segmented assimilation in the second generation. Social Forces 89(3): 733–762.

23.

Heggen

Helland

Lauglo

(2013) Utdanningssosiologi, Oslo: Abstrakt.

24.

Helland

(1997) Etnisitet og skoletilpasning. En undersøkelse av norske, pakistanske og konfutsianske Osloungdommers skoleprestasjoner. Magistergradsavhandling. Department of Sociology, Oslo University.

25.

Kosic

Caudek

(2005) Ethnic identification and outgroup attitudes in minority and majority groups. Review of Psychology 12(2): 103–113.

26.

Lauglo

(1995) Populism and education in Norway. Comparative Education Review 39(3): 255–279.

27.

Lauglo

(2001) Social capital trumping class and cultural capital? Engagement with school among immigrant youth. In: Baron

Field

Schuller

(eds) Social Capital. Critical Perspectives, Oxford: Oxford University Press, pp. 143–167.

28.

Lauglo

(2010) Do Private Schools Increase Social Class Segregation in Basic Education Schools in Norway? LLAKES Research Paper 6, London: Institute of Education, London University.

29.

Lauglo

(2013) Do more knowledgeable adolescents have more rationally based civic attitudes? Analysis of 38 countries. Educational Psychology: An International Journal of Experimental Educational Psychology 33(3): 262–282.

30.

Masson

Verkuyten

(1993) Prejudice, ethnic identity, contact and ethnic group preferences among Dutch young adolescents. Journal of Applied Psychology 23(2): 156–168.

31.

Midtbøen

Rogstad

(2012) Diskrimineringens omfang og årsaker. Etniske minoriteters tilgang til norsk arbeidsliv, Oslo: Institutt for samfunnsforskning.

32.

Modood

(2004) Capitals, ethnic identity and educational qualifications. Cultural Trends 13(50): 87–105.

33.

Modood

(2005) The educational attainment of ethnic minorities in Britain. In: Loury

Modood

Teles

(eds) Ethnicity, Social Mobility and Public Policy, New York, NY: Cambridge University Press, pp. 288–308.

34.

Modood

(2012) Multiculturalism, New York, NY: Wiley.

35.

Ogbu

(1987) Variability in minority group performance: A problem in search of an explanation. Anthropology & Education Quarterly 18: 312–334.

36.

Özerk

Kerchner

(2014) Diversity and educational challenges in Oslo and Los Angeles. International Electronic Journal of Elementary Education 6(3): 441–462.

37.

Park

(1928) Human migration and the marginal man. American Journal of Sociology 33: 881–893.

38.

Phinney

(1992) The multigroup ethnic identity measure: A new scale for use with diverse groups. Journal of Adolescent Research 7(2): 156–176.

39.

Phinney

Ferguson

Tate

(1997) Intergroup attitudes among ethnic minority adolescents. Child Development 68(5): 955–969.

40.

Putnam

(2000) Bowling Alone, New York, NY: Simon & Schuster.

41.

Roccas

Brewer

(2002) Social identity complexity. Personality and Social Psychology Review 6: 88–106.

42.

Sam

Berry

(2010) Acculturation: When individual and groups of different cultural backgrounds meet. Perspectives on Psychological Science 5(4): 472–481.

43.

Schneider

Crul

(2012) Comparative integration context theory. In: Crul

Schneider

Lelie

(eds) The European Second Generation Compared, Amsterdam: Amsterdam University Press, pp. 19–37.

44.

Steinkeller

(2013) Innvandrere og norskfødte med innvandrerforeldre i grunnskolen. En analyse av karakterdata og resultater fra nasjonale prøver i 2012, Oslo-Kongsvinger: Statistics Norway.

45.

Suarez-Orozco

(1989) Central American Refugees and US High Schools: A Psychological Study of Motivation and Achievement, Stanford: Stanford University Press.

46.

Tajfel

Turner

(1986) The social identity theory of intergroup behavior. In: Worchel

Austin

(eds) Psychology of Intergroup Relations, Chicago: Nelson-Hall, pp. 7–24.

47.

Valenta

Bunar

(2010) State assisted integration: Refugee integration policies in Scandinavian Welfare States: The Swedish and Norwegian experience. Journal of Refugee Studies 23(4): 463–483.

48.

Valset

Øia

(2006) Dokumentasjon “Ung I Oslo 2006”, Oslo: NOVA.

49.

Wichstrøm

(1995) Harter’s self-perception profile for adolescents – Reliability, validity and evaluation of question format. Journal of Personality Assessment 65(1): 100–116.

50.

Zhou

(1997) Growing up American. The challenge confronting immigrant children and children of immigrants. American Review of Sociology 23: 63–95.

51.

Zhou

(2005) Ethnicity as social capital: Community–based institutions and embedded networks of social relations. In: Loury

Modood

Teles

(eds) Ethnicity, Social Mobility, and Public Policy, New York, NY: Cambridge University Press, pp. 131–159.

52.

Zhou

Bankston

(1994) Social capital and adaptation of the second generation: The case of Vietnamese youth in New Orleans. International Migration Review 23(4): 821–845.