Norway: Ethnic (In)equality in a Social-Democratic Welfare State

This chapter offers a systematic review of social science research in Norway on ethnic inequalities in education, from the period 1980–2017. Three broad research traditions are identified: (1) Ethnic inequalities in educational enrolment, achievement, and attainment; (2) Immigrant families and ethnic minority communities as resources for educational careers; and (3) Curriculum, teacher instruction, and student experiences with inclusion and exclusion. Most of the research conducted in Norway focuses on describing and explaining the educational outcomes of children of immigrants from non-European countries, and is characterized by the use of quantitative research methods and access to rich population-wide registry data. Main groups studied in the qualitative traditions are typically children of immigrants from Pakistan, Vietnam, Sri-Lanka, Morocco and Turkey. This rich body of research is written in both Norwegian and English, and has developed in a context characterized by a close collaborative relationship between researchers and the government, in response to a growing concern about social cohesion within the Norwegian social-democratic welfare state.


Introduction
Research on ethnic inequalities in education has a relatively short history in Norway. This largely reflects that there was generally little awareness of ethnic diversity before the start of non-European immigration around 1970, despite the long presence of several smaller ethnic minorities in the Norwegian population, as well as the indigenous Sami (Brochmann and Kjeldstadli 2008). In recent decades, there has been a growing interest in ethnic inequalities in education from both policymakers and academic researchers, reflecting the rapidly increasing population share of immigrants and their Norwegian-born descendants. Yet, there are few comprehensive reviews on the scientific literature from Norway, although some partial reviews exist, primarily in Norwegian (e.g., Hermansen 2016a). The following review surveys the Norwegian literature, based on a systematic sampling of the literature spanning more than 30 years of research.
The chapter starts with a description of the Norwegian national context and a discussion of the methods we used for the review. Then, we present and discuss three key research traditions, identified based on our analysis of the relevant literature, in the main body of the chapter. Finally, the chapter concludes with a critical discussion of the relationship between the different research traditions, and our conclusions regarding the state and future of Norwegian research in the field.

National Context
This section presents a brief overview of the Norwegian educational system, the history of immigration to Norway and current state of ethnic diversity, and various institutional features and social policy models that may-directly or indirectly-affect ethnic inequalities in education.

The Norwegian Educational System
Norway is a Nordic social-democratic welfare society-with a total population of almost 5.3 million people in 2017-where most educational and basic social services are publicly funded (Esping-Andersen 1999). Norwegian welfare-state policies are governed by universalistic ideals, where access to education and related social services are, in principle, available to all residents. This includes immigrants, either as naturalized citizens or denizens, and their native-born children. After the Second World War, the educational system in Norway was expanded with the objective of fostering economic growth and equality of educational opportunity (Telhaug et al. 2006).  .1 provides a schematic presentation of the Norwegian educational system, which distinguishes between preschool (i.e., barnehage), compulsory primary and lower-secondary school, and elective upper-secondary and tertiary education. Starting in the late 1970s, early childcare coverage was gradually expanded throughout Norway (Havnes and Mogstad 2011). Today, local municipalities must guarantee access to subsidized and high-quality preschool childcare services to children from age one until school starts at age six. From 2006, the Ministry of Education has had the official responsibility for preschool childcare centres, as they were redefined as educational institutions.
In 1959, a comprehensive school reform was introduced by the Norwegian parliament, which had three broad goals: (1) increase the minimum level of education by extending compulsory education from seven to nine years, (2) ease the transition into tertiary education, and (3) enhance equality of opportunities along both along socio-economic and geographical dimensions (Lie 1973;Lindbekk 2015). In 1997, compulsory education was extended to 10 years of mandatory schooling, and since then pupils have started school at age six, and typically graduate at age 16. The first seven of these years are spent in primary schools, while the last three years are spent in lower-secondary schools. There is no formal tracking by ability during these years and school attendance is as a general rule based on place of residence. Although it is possible to apply to schools outside the student's local catchment area in some municipalities, this is not very common. Many municipalities publish detailed statistics about the schools' performance on national standardized tests, pupil surveys, share of minority students qualifying for additional Norwegian language training, etc. This allows parents to evaluate their local school and, possibly, move to neighbourhoods with seemingly better performing schools. At the same time, in a comparative perspective, lower-secondary schools in Norway are characterized by modest between-school variation in standardized test scores and socioeconomic stratification (OECD 2016).
Upon finishing compulsory education, most students continue into uppersecondary education, which consists of academic and vocational tracks. Academic upper-secondary tracks last for three years, while vocational uppersecondary tracks last for two years upon which students typically either spend two years in apprenticeship training or one year completing general subjects supplements that provides the pupil with basic entrance requirements for continuation into tertiary education (i.e., generell studiekompetanse). After an educational reform in 1994 (i.e., Reform 94), all pupils gained a legal right to pursue upper-secondary education. The allocation of pupils to different schools is, however, based on their grade point average (i.e., a sum of teacher-assigned grades and grades on centralized exams at the end of lower-secondary school), their own educational preferences, and, in some areas, place of residence. The Norwegian Directorate for Education and Training publishes searchable data on average grade levels, pupil satisfaction surveys and other indicators that may be used as quality indicators of different uppersecondary schools. This information distinguishes between more or less attractive schools based on the composition and achievements of their student bodies. However, availability of relevant study tracks within schools is also an important factor as not all schools offer the same study tracks, and often many of the vocational tracks are available only in selected schools.
After completion of upper secondary education, pupils can enter into various types of tertiary education in universities and university colleges. Most Norwegian universities and university colleges are public and without tuition fees, but there are also a few applied colleges that are privately run. The Norwegian State Educational Loan Fund provides grants and loans to students, to cover living expenses and other direct and indirect costs associated with enrolling in higher education. Grants and loans are awarded independent of parental income levels, effectively reducing the direct influence of parents' financial situation on the decision to enrol. In 2003, a reform (i.e., Kvalitetsreformen) implemented the Bologna convention throughout the entire national system of higher education. This introduced a standard threetier system with three-year Bachelor's degrees, two-year Master's degrees, and three-year doctoral programs leading to a Ph.D. degree. However, there are still some shorter programs as well as one-tier Master's degrees and professional degrees.

Immigration and Ethnic Minorities in Norway
Immigration to Norway from non-European origin countries started relatively late compared to many other Western European countries (Brochmann and Kjeldstadli 2008;Dustmann and Frattini 2013). While Norway experienced net emigration throughout large periods of the twentieth century, this trend was reversed in the late 1960s and the pace of immigration gradually increased. Thus, Norway has rapidly become a multi-ethnic society and the population share of foreign-born individuals residing in Norway today is broadly comparable to countries such as the Netherlands, Germany, France, and the United Kingdom (OECD 2015b).
Immigration to Norway before 1970 primarily consisted of citizens from the Nordic countries and other Western Europeans who came to seek work or immigrated due to family connections. Non-European immigration began around 1970 and consisted of young, unskilled, male labour migrants from Pakistan, Turkey, and Morocco. In 1975, a moratorium on unskilled labour immigration was introduced. Later adopted as a permanent measure, this moratorium ended unskilled labour immigration from outside Western Europe, but allowed for immigration according to three main principles. First, demand for specific skilled labour. Second, entry of refugees and political asylum seekers granted protection on humanitarian grounds. Third, family-based immigration for kin of immigrants already in Norway (i.e., either through reunification with existing family members or as family formation through entry into marriage with a foreign-born spouse, typically found in the same origin country) (Brochmann and Kjeldstadli 2008).
In the period after 1975, admission to Norway from outside Western Europe was primarily confined to immigration due to humanitarian principles and family-based immigration (i.e., for the kin of both the original migrant workers and humanitarian immigrants). Starting in the late 1970s, the number of refugees and asylum seekers arriving from countries in recent conflict areas, such as Vietnam, Chile, Sri Lanka, and Iran (1980s), the Balkans (early 1990s) and Iraq and Somalia (late 1990s), grew substantially. While post-1975 labour immigration from developing countries was negligible, the original cohorts of migrant workers also continued to grow in this period due to family-based chain migration for the initial migrants and their offspring. After the European Union (EU) enlargements in 2004 and 2007, Norway has experienced a rapid increase in labour immigration flows from new EU member states in Eastern Europe, in particular Poland and the Baltic countries. 1 Figure 20.2 shows how the Norwegian immigrant population has increased since 1970. By 2017, immigrants and their Norwegian-born children constituted about 16%-approximately 885,000 persons-of the total Norwegian population. In this population, about 725,000 persons were born abroad and 160,000 persons were born in Norway to immigrant parents (Statistics Norway 2017b). Immigrants from Asia (including Turkey), Africa, and South America made up the majority of this population since about 1990, but today persons arriving from European origin countries are again in the majority as a reflection of the upsurge in immigration from new EU member states in Eastern Europe since the mid-2000s.
1 Norway is not a member of the EU, but as part of the European Economic Area (EEA), the country is part of the internal market for the free movement of labor, services, goods, and capital in the EU and EEA region. Thus, all EU citizens are entitled to apply for work in Norway, as in other EU and EEA countries.   Table 20.1 shows the national-origin composition of Norwegian-born children of immigrants in 2017. Second-generation immigrants in Norway are still young and almost 80% of this population is less than 20 years of age, and within these birth cohorts, they constitute about one out of ten individuals in the total Norwegian population. The major national-origin groups reflect the immigration inflows after 1970, while their age composition reflects the timing of their arrival. Among second-generation immigrants currently above 20 years, the Norwegian-Pakistani minority is by far the largest and the other large groups-Turkey, Morocco, India, Vietnam, and Chile-reflect the early waves of labour immigration and refugee arrivals. In the birth cohorts currently below 20 years of age, the Pakistani, Somali, Iraqi, Polish, and Vietnamese national-origin groups are the largest ones. So far, Norwegian research is more informative about ethnic inequalities in the educational careers of the children of the early waves of labour immigrants and refugees, who arrived in Norway between 1970 and the mid-1990s.

The Sami and Norwegian National Minorities
Before the onset of large-scale immigration, ethnic diversity in Norway primarily reflected the presence of the Sami indigenous people. Today, Norway also recognizes five national minority groups: Jews, Romani (i.e., tatere), Roma (or Gypsies), Norwegian Finns (i.e., kvener), and Forest Finns (i.e., skogfinner). 2 Because of restrictions on the registration of ethnic minority identity in Norwegian public registries, it is not straightforward to estimate the size of these groups today, although the groups are estimated to be small. Historically, Norwegian authorities have oppressed several of Norway's national minority groups, and for some, schools, in particular, have been associated with exclusion and control. In the first half of the twentieth century, a large number of the Romani were sterilized and children were routinely taken from their parents. During the Second World War, Norwegian Jews were stripped of their belongings and deported to concentration camps by the Norwegian Nazi Government.
Some qualitative research on schooling among the Romani and Roma groups has been conducted (Engen 2010;Moen and Lund 2010;Westrheim and Hagatun 2015). This research indicates that children in these communities tend to have high rates of absenteeism and often leave school at the lower secondary level (Engebrigtsen and Lidén 2010;Lund and Moen 2010). In contrast to the national minority groups, some quantitative data is available about the indigenous Sami population. Historically, the Sami population has been exposed to strict assimilationist policies, where schools did not permit the use of their mother tongue, even outside the classroom (Engen 2010). However, a recent state-of-the-art report on discrimination among national minorities, the Sami population, and immigrants in Norway, confirms that very little research has been done on educational disparities among the Sami (Midtbøen and Lidén 2015, p. 37). In the following, we will include research on educational inequalities among the Sami where available.

Integration and Institutional Setting in the Norwegian Welfare State
Ethnic inequalities in education in Norway are interesting from a comparative perspective due to the presence of strong welfare-state institutions (Esping-Andersen 1999). Immigrants and their native-born children are eligible for high-quality basic services, such as full coverage in healthcare services, access to subsidized early childhood education, and other social security benefits important for child well-being, upon arrival. Norway consistently ranks in the very top of the United Nations Development Program's Human Development Index over the past decades (UNDP 2011), has comparatively low economic inequality (OECD 2015a) and low prevalence of child poverty (UNICEF 2016), and, summarized across a large number of domains, it is one of the most 'child-friendly' countries in Europe (Bradshaw and Richardson 2009).
Moreover, the native majority population in Norway exhibits comparatively high rates of intergenerational mobility in education and adult labourmarket status compared to many Western societies (Björklund and Jäntti 2009;Breen and Jonsson 2005). In particular, the consequences of early-life economic deprivation for adult attainments and intergenerational mobility are less pronounced in Norway compared to countries with higher levels of economic inequality and lower presence of welfare-state institutions (Bratsberg et al. 2007;Duncan et al. 2011). Moreover, comparative research indicate that comprehensive educational systems like the one in Norwaywith limited school tracking and a high level of national standardization in curriculum and school autonomy-are particularly beneficial for students with low socioeconomic origin and immigrant origin (Chmielewski and Reardon 2016;Cobb-Clark et al. 2012; Van de Werfhorst and Mijs 2010; Van de Werfhorst et al. 2014). By contrast, early and rigidly selective educational systems seem to reinforce ethnic inequalities in academic achievement, track placement, and subsequent educational attainment. Taken together, the institutional features in Norway are likely to reduce ethnic inequalities in the standards of living and opportunities for educational progress between children in native and immigrant families compared to more unequal host societies (Hermansen 2017b).
Nevertheless, adult immigrants arriving from non-European low-income countries, regardless of entry criteria, experienced declining employment rates and increasing dependency on social welfare assistance over the life cycle. Prior research suggests that universal access to social welfare assistance created work disincentives that in part contributed to low life-cycle employment rates among low-skilled immigrants with many dependent family members (Birkelund and Mastekaasa 2009;Bratsberg et al. 2010Bratsberg et al. , 2014. Despite generous welfare provisions, children of non-European immigrants faced markedly higher risks of exposure to childhood poverty compared to children of native Norwegians (Galloway et al. 2015). Moreover, the degree of ethnic residential segregation in Norway is moderate and comparable to levels found in other Western European countries (Musterd 2005;Wessel et al. 2016).
On the whole, recent immigration has introduced a new dimension of ethnic stratification into Norwegian society. Thus, a key question is to what extent these ethnic inequalities are reproduced or mitigated among children of immigrants who have grown up in the Norwegian welfare-state society.

The Development of Norwegian Education Policies
The Norwegian centralized school system has played a key role in the establishment and symbolic unity of the Norwegian nation state (Lidén 2001;Telhaug 1994). Seland (2011) describes three main phases of national school policy development within the timeframe of the review in this chapter. Under the banner of "the common school" (i.e., fellesskolen), the first phase, from 1974-1987 was characterized by an increase in diversification and individualization of instruction, within the wider frame of equal opportunity for learning. This included options for opting out of religious (Christian) education, but also adaptive measures for inclusion of disabled students in regular schools. This demanded specialized plans for handling linguistic and cultural diversity, while at the same time ensuring commonality and equality of instruction. The ambition proved to be both costly and difficult to achieve in practice (Høgmo 2005(Høgmo , 1990).
The second phase, starting with the new school curriculum plan of 1987 had explicit strategies for the inclusion of ethnic minority students, through mother tongue instruction aiming at functional bilingualism. The right to mother tongue education was already introduced for Sami students in the spring of 1985 (Seland 2013).
Through the revision of the general part of the school curriculum plan in 1993, the third phase was entered, where this pattern of inclusion through diversity was altered towards more uniform instruction. The concept of the "unitary school" (i.e., enhetsskolen) returned after having been absent for several decades. Originally, the unitary school had a central role in Norwegian nation building in the nineteenth and early twentieth century, espousing assimilation and cultural homogeneity (Brossard Børhaug 2008;Engen 2003). Resurfacing in the early 1990s, the unitary school was argued to reduce inequality between students by giving them common references with regard to knowledge, values and culture. Importantly, the image of a national cultural community was strengthened (Telhaug 1994), alongside the strengthening of Christianity in the curriculum through the return of religious education without any opt-out option for non-Christian minorities from 1997 (Seland 2013). Towards the end of the 1990s, the policy documents no longer discuss mother tongue instruction as a value in itself, but rather as a necessary step toward being able to fully participate in Norwegian language instruction.
A recent Government white paper (i.e., St.meld. 6, 2012St.meld. 6, -2013 addressed some of the challenges associated with the inclusion of ethnic minorities in the Norwegian education system, stating that "recognizing multilingualism and cultural diversity means recognizing people's various competencies". As a result, the Ministry of Education initiated a four-year program in 2013 called Competence for Diversity (i.e., Kompetanse for mangfold), which aimed at strengthening educational institutions' competencies for dealing with the challenges that minority children, adolescents and adults meet in the education system (Westrheim and Hagatun 2015). The initiative emphasized educating staff, managers, teachers and other actors in the education system about multicultural pedagogy and multilingualism and other forms of diversity pedagogy.
Regardless of policy phase or terminology, the Norwegian school policies have aimed to ensure equality and community across differences (Imsen and Volckmar 2014;Seeberg 2003;Smette 2015). Although challenges associated with creating inclusive and diverse learning environments are not unique to the Norwegian context, the explicit rhetoric of the unitary school seems to have inspired research on how the educational system in Norway deals with diversity.

Methods
We have systematically sampled all relevant peer-reviewed articles, books, edited books, PhD dissertations, and official reports on the subject of ethnic or racial inequalities in the Norwegian educational system from 1980 onwards. In some cases, we also included articles from non-peer reviewed journals if they met high scientific standards and significantly contributed to the understanding of the subject matter. Publications on all levels of education were included, from preschool through tertiary education. We included literature covering research on immigrants and Norwegian-born children of immigrants as well as some research on educational inequalities among the Sami, where available.
Following Stevens (2007) and Stevens et al. (2011), our sampling procedure consisted in three main stages. First, we used systematic queries to search the international bibliographical databases Web of Science and ProQuest. For the English-language searches, we included literature on Norway, Scandinavia and the Nordic countries, to make sure we would pick up all relevant international publications covering empirical research on the Norwegian case. Second, we used systematic queries to search for Norwegian or Scandinavian language publications in the databases ORIA, NORART, LIBRIS and DANBIB. We used the same search strings, adapted to English or Norwegian, for both systematic queries. 3 Third, we inspected the bibliographies contained in the publications identified in the two abovementioned stages to identify additional publications for review.
This sampling approach resulted in identifying a large body of research, which we have categorized into three broad research traditions: (1) Ethnic inequalities in educational enrolment, achievement, and attainment; (2) Immigrant families and ethnic minority communities as resources for educational careers; and (3) Curriculum, teacher instruction, and student experiences with inclusion and exclusion.

Ethnic Inequality in Education in Norway: Key Research Traditions
We now summarize the main findings of the three key traditions in Norwegian research on ethnic inequalities in education. These traditions are relatively broad, but each represents a collection of studies that address similar types of research questions and use similar types of methods. First, we present the quantitative research tradition that primarily studies patterns of ethnic inequality in educational enrolment, achievement, and attainment. Second, we present the qualitative tradition that studies how immigrant families and ethnic minority communities function as resources shaping ethnic minority students' educational careers. Third, we present the qualitative tradition that study how institutional processes (e.g., curriculum and teacher instruction) shape ethnic minority students' experiences of inclusion and exclusion in schools.

Ethnic Inequalities in Educational Enrolment, Achievement, and Attainment
We refer to the tradition studying quantitative aspects of ethnic inequalities in education in Norway as the ethnic inequalities in educational enrolment, achievement, and attainment tradition. This literature is dominated by sociologists, economists, and other quantitative social scientists using large-scale datasets. The tradition has contributed with both descriptive and explanatory analyses of ethnic inequalities in education. This tradition is part of a larger national and international research tradition that focuses on whether and how educational systems and broader institutional settings shape social inequalities in schooling related to family background (e.g., Breen et al. 2009;Hansen and Mastekaasa 2010;Hernes 1974;Shavit and Blossfeld 1993). More recently, comparative educational research has focused more directly on institutional determinants of ethnic inequalities in education (e.g., Alba et al. 2011;Heath and Brinbaum 2014).
This tradition draws on population-wide data from various administrative registries made available by Statistics Norway, as well as self-reported information on students' educational careers using several large-scale surveys, such as 'Ungdata' and 'Children of Immigrants Longitudinal Study in Norway (CILS-NOR)'. In this tradition, immigrant and ethnic minority background is usually measured using information about individuals' country of birth, parental country of birth, and, among the foreign-born, information on age at immigration. In registry-based studies, ethnicity-related information is based on direct measures of immigrant ancestry from administrative records while similar information is often self-reported in most survey-based studies. 4 Over the historical period we cover, the numerical growth in the population of children and youth with immigrant origin in Norway has enabled quantitative studies to provide increasingly more nuanced descriptions of variation by ethnic minority background over time. Whereas early contributions to this literature often only distinguished between immigrants and non-immigrants or 'Western' and 'non-Western' origin regions, later studies increasingly make more fine-grained distinctions with respect to generational status and various regions of origin, and often also separately by single countries of origin.
Turning to the Sami national minority, there is no data basis for creating individual-based statistics on people of Sami descent or ethnicity according to Statistics Norway (Slaastad 2016). However, two main sources of information about the Sami population have been established over the past 10 years; "Sami Statistics", produced biannually since 2006 by Statistics Norway and "Samiske tall forteller" (i.e., 'What Sami numbers describe'), a report produced yearly since 2008, by a publicly appointed expert group for the analysis of Sami statistics. Statistic information about the Sami student population is approximated in three main ways. One way is to count all students in elementary school who has one of the Sami languages as part of their curriculum. In 2014, less than 0.5% of pupils in Norwegian elementary schools had one of the Sami languages as their languages of instruction, or were registered as studying Sami as their first or second language in school (Slaastad 2016, p. 53). A second way to approximate the population is by identifying those whose permanent residence is in areas eligible for the Sami government's financial support for business development (STN). The third way is even broader, including the Sami settlement areas north of the Saltfjellet mountain range in the Arctic Circle.

Ethnic Inequalities in Education by Family Background, Nationality, Gender, and Trends Over Time
Enrolment in preschool childcare is considered important for children of immigrants, as this lays the foundation for later learning through early acqui-sition of Norwegian language and related social competencies. Recently, there has been a steady increase in the preschool childcare enrolment rate among children with immigrant parents in Norway. Among children below five years this figure was at 62% in 2016 compared to about 77% in the population as a whole (Statistics Norway 2017a). A pilot project in Oslo, where access to preschool childcare for children aged 4-5 years were offered without cost in selected city districts, increased the share of immigrant children enrolled in preschool by about 15 percentage points in these areas (Bråten et al. 2014). Importantly, immigrant-origin children in areas where the financial cost of attendance was removed performed better on standardized tests when entering school (Drange and Telle 2015). Thus, increasing preschool enrolment in the Norwegian immigrant population is likely to reduce subsequent ethnic inequalities in the educational system.
In general, children of immigrants born in Norway tend to perform lower on standardized tests, centralized national exams, and teacher-assigned grades both at the end of compulsory lower-secondary education and uppersecondary education (Bakken 2003;Bakken and Elstad 2012;Bratsberg et al. 2012;Haegeland et al. 2004;Krange and Bakken 1998;Lødding 2003b;Opheim and Støren 2001;Raaum and Hamre 1996). However, there is considerable variation between different origin countries (Støren 2006;Sørensen et al. 2016). Moreover, immigrant students seem to improve their academic achievements during the years in lower-secondary education to a higher degree than native students (Wiborg et al. 2011). Further, a robust finding in survey-based studies is that many immigrant-origin students are highly motivated for school, and typically spend more time on homework and report higher ambitions regarding their continuation into higher education relative to comparable native majority peers with similar grade achievement levels or parental education (Bakken 2016;Bakken and Sletten 2000;Friberg 2016;Frøyland and Gjerustad 2012;Hegna 2010;Lauglo 1999Lauglo , 2000Pihl 1998).
Turning to completion of upper-secondary education, a long-term trend towards narrower educational gaps between second-generation immigrants and their native-majority peers has recently been documented . Figure 20.3 shows that about 70% of children in the nativeorigin majority complete upper-secondary education within five years after enrolling throughout the whole period since the early 1990s. By contrast, for second-generation immigrants this level has increased from about 60% early in the period to reach similar levels as the native majority population at the end of the period. This implies that the overall native-immigrant gap has been entirely closed for the latest graduation cohorts we observe. Bratsberg et al. (2012) also show that this catch-up trend is robust to adjustment for changes in the composition of immigrant students by national origin and parental socioeconomic resources over this period, which may suggest that the Norwegian educational system has successfully improved its efforts to meet the need of these students. 5 Despite this overall catch-up trend, there is also considerable variation in upper-secondary completion rate across various second-generation ethnic minorities in Norway (Birkelund and Mastekaasa 2009;Bratsberg et al. 2012;Fekjaer 2006;Hermansen 2016a, b;Lødding 2003a). Table 20.2 provides an overview of upper-secondary completion rates among second-generation immigrants within the major origin countries. For example, children of immigrants from Vietnam, India, Iran and Sri Lanka complete secondary school in 5 For example, a study evaluating an educational reform introduced in the mid-1990s showed that this had a positive impact on the upper-secondary completion rates among second-generation immigrant students (Brinch et al. 2012). There has also been a focus on compensating schools with high shares of students with immigrant background and low-income parents by allocating extra teachers and funding to these schools (Haegeland et al. 2009(Haegeland et al. , 2005. equal to or slightly greater extent than the general population. Descendants from countries such as Morocco, Turkey, Chile and, to a lesser extent, Pakistan have a low completion rate, where between 50-60% of children complete secondary education. Thus, the Norwegian situation seems comparable to other host societies in Western Europe; where many second-generation immigrants from Turkey, the Middle East and North Africa often lag considerably behind the native majority in education, while those of Southeast-Asian origin often outperform their native peers (Alba and Foner 2015;Heath et al. 2008).
In the Sami population, educational attainment among those residing north of Saltfjellet is similar to the distribution for the country as a whole. Among those residing in STN-areas, educational levels are significantly lower, but have also been steadily rising over the past 15 years (Slaastad 2016, p. 44). Nevertheless, only 56% of students residing in the STN-area and 64% of students residing in non-STN areas north of Saltfjellet had completed their education among students who should have completed upper-secondary education within the statutory five-year period in 2014.
Upon completion of upper-secondary education, second-generation immigrants have higher continuation rates into tertiary education compared to . Only a few of the major groups (e.g., the Turkish-Norwegian group) have a lower enrolment rate in higher education than the average for native majority peers of similar age. By contrast, second-generation ethnic minorities originating from India, Sri Lanka and Vietnam have over 40% enrolment, and the Norwegian-Pakistani origin group is also enrolled at higher rates than the native-origin population (Østby and Henriksen 2013). Moreover, there are horizontal ethnic differences in what fields of study are popular among second-generation immigrants, and they are particularly overrepresented as students in the medical professions (e.g., medical doctors, dentists, and pharmacists) which are characterized by comparatively high earnings (Bratsberg et al. 2014;Schou 2009;Østby and Henriksen 2013). Moreover, second-generation immigrants do not seem to face a higher risk of not completing their postsecondary degrees when compared to native students (Helgeland 2009;Reisel and Brekke 2010), but they have slightly lower average grade achievements in their graduation diplomas (Kolby and Østhus 2009). Thus, this literature shows that second-generation immigrants in Norway, as a whole, are currently overrepresented in institutions of higher education, especially in prestigious educational tracks, despite their low socioeconomic origins.
Importantly, the role of parental socioeconomic resources-such as education and labour market attachment-has been a key focus of many studies in this literature Fekjaer 2007;Hermansen 2016b;Støren and Helland 2010). Both with regard to academic achievement, uppersecondary completion, and final educational attainment, a common finding in this literature is that variation in parental socioeconomic resources to a large extent account for educational differences between students in the native majority and different second-generation ethnic minorities Hermansen 2016b). In some cases, second-generation minority groups outperform their native-origin peers after adjustment for socioeconomic origins. While second-generation immigrants tend to complete more education than their native peers in the lower part of the parental distribution of education and economic resources while, the opposite is true among children from more advantaged family backgrounds Fekjaer 2007;Hermansen 2016b). 6 The less steep intergenerational educational gradients among second-generation immigrants indicates that observed parental resources are less important among them, or at least that intergenerational transmission processes operate differently in immigrant-origin and native majority families.
There are also interesting gender differences in education within the Norwegian second-generation population. While women complete more education than men in the native-origin population of most contemporary Western countries, there are still fewer educational opportunities for girls in the origin countries of many immigrants to Norway (Buchmann et al. 2008;Grant and Behrman 2010). Thus, educational investments among girls is an important indicator of the durability of traditional gender values across immigrant generations, and new research shows that immigrants' daughters outperform their male counterparts in many Western countries (Fleischmann et al. 2014). This is also the case in Norway, where second-generation women have a higher tendency to complete upper-secondary education and this female advantage is slightly more pronounced than what we see among natives (Hermansen and Birkelund 2015;Støren and Helland 2010).
However, second-generation boys in upper-secondary vocational education tend choose less gender typical than their counterparts with a majority background, partly because they more often choose the supplementary education that qualifies them for admission to higher education (Reisel 2014). This pattern is sustained at entry into higher education, where second-generation men have sharply increased their enrolment rate in recent years. Currently, it is only second-generation men of Norwegian-Turkish origin among the major groups who have a lower enrolment rate than the average level within the native-origin majority (Østby and Henriksen 2013). There is also some polarization in the male second-generation population. This implies that secondgeneration males are overrepresented among those who drop out of upper-secondary education, but also that this group has higher continuation rates into tertiary education among those who successfully complete, than their counterparts in the native-origin population.
Recent studies document considerable intergenerational progress in education between the parental immigrant generation and their Norwegian-born children (Bratsberg et al. , 2014Hermansen 2016b). For example, only about one-third of immigrant parents from Turkey had completed uppersecondary education while about 60% of their Norwegian-born children had reached this level . A similar pattern is found in most major national-origin groups, and is even more pronounced when observing second-generation immigrants as adults (Bratsberg et al. 2014;Hermansen 2016b). Focusing on the adult second-generation immigrant population, Hermansen (2016b) shows the native-immigrant gaps in completed years of education is reduced by about 75% from the immigrant generation to their children. Moreover, childhood immigrants who arrive in Norway from lowincome origin countries after school-starting age, especially during adolescence, experience lower academic achievement and educational attainment Hermansen 2017a). Together, the considerable improvement across one generation and the variation by age at arrival suggests that early exposure to Norwegian society enables considerable social mobility among children whose parents arrived from countries with limited educational opportunities.
In sum, this tradition is characterized by the use of large datasets, and quantitative research methods. This allows for generalizable findings, and in recent years, the estimation of group differences by parental country of origin among children of immigrants. To the extent that much of the research in this tradition is based on registry data, one major weakness in this tradition is the lack of information about attitudes, expectations and self-identification. The tradition could therefore benefit from integrating subjective measures on immigrant students' ambitions, attitudes, and acculturation-related indicators from survey data with later follow-ups on outcomes from administrative registries. Findings from this tradition show that children of immigrants seem to do relatively well in the educational system and often perform on par with peers in the native-origin population with similar parental socioeconomic resources. Further, they experience substantial upward educational mobility relative to their immigrant parents. There has been a clear trend towards closing of the overall native-immigrant gap in upper-secondary completion over time, and children of immigrants have higher enrolment rates in higher education compared to young adults in the native majority. Nevertheless, noncompletion of upper-secondary education is still a considerable problem in some minority groups, particularly among young men of Turkish and Moroccan ancestry.

Ethnic Segregation, School Resources, and Educational Inequalities
Increasing spatial concentration of immigrants in certain areas over the past few decades has led to increasing concern among policymakers for detrimental consequences, which has been accompanied by the development of a strand within the quantitative research tradition, addressing ethnic segregation in Norwegian schools. In particular, there has been an increase in geo-graphic residential concentration of the immigrant population in the capital, Oslo (Høydahl 2015). Today, immigrant students comprise the majority of the student body in two out of five schools in Oslo and a few schools have an ethnic minority student share of 90% (Oslo kommune 2014). Below we briefly review the quantitative literature on studies related to ethnic school segregation in Norway.
The share of immigrant-origin students in schools may be systematically related to resource allocation and teacher recruitment. Studies from the early 2000s documented that higher shares of ethnic minority students in schools both reduced recruitment and increased turnover among certified teachers in these schools (Bonesrønning et al. 2005;Strøm 2003). However, schools receive targeted resource transfers according to need in Norway. This implies that schools with high shares of students from disadvantaged families have lower student-teacher ratios compared to other schools (Haegeland et al. 2005) and schools serving many children from immigrant families have more teaching assistants for special needs students (Haegeland et al. 2009).
Turning to peer effects related to ethnic school segregation, a high share of immigrant-origin students with relatively low educational achievement and low socioeconomic status may take up a lot of the teachers' time and negatively affect the quality of education. However, it is also possible that concentrations of immigrant students with high educational aspirations and school motivation have a positive influence on the educational outcomes of their schoolmates. While studies from Norway show that students attending schools with high immigrant shares have lower levels of academic achievement and rates of upper-secondary completion, this relationship seems to largely reflect between-school differences in students' socioeconomic background (Birkelund et al. 2010;Fekjaer and Birkelund 2007;Hardoy and Schøne 2013;Hardoy et al. 2017;Hermansen and Birkelund 2015;Wiborg et al. 2011). In lower-secondary schools, Hermansen and Birkelund (2015) did not find that increasing shares of immigrant-origin peers across cohorts within the same schools lead to lower probabilities of completing uppersecondary education by their early twenties among native-origin students, while immigrant-origin students experienced a small advantage of attending cohorts with higher shares of immigrant-origin peers. This (weak) positive peer effect may be due to students with an immigrant background having high educational aspirations and that these aspirations to some extent are transmitted between peers in the same cohort. Interestingly, this study found that the positive immigrant peer effects seem to mainly reflect the presence of minority schoolmates from relatively high-performing origin regions (e.g., Southeast Asia) while there was no corresponding negative effects of exposure to minority classmates from low-achieving origin regions.
By contrast, prior studies have reached contradictory conclusions regarding the influence of immigrant student composition in upper-secondary schools on educational outcomes. While Fekjaer and Birkelund (2007) found a weak positive relationship between attending schools with many immigrant students and educational achievement, Hardoy and Schøne (2013) found that increases in the share of immigrant peers had a small negative effect on native majority students' probability of completing upper-secondary education. 7 More recently, Hardoy et al. (2017) found that the negative relationship between immigrant concentration and native students' upper-secondary completion disappeared after adjusting for the sorting of students into schools based on grade achievement in lower-secondary school. When looking at within-school variation in immigrant composition across cohorts, there was also no negative relationship (Hardoy et al. 2017).
To sum up, the emerging consensus seems to be that the adverse consequences of immigrant concentrations in schools for educational outcomes among both immigrant-origin and native majority students are relatively modest once between-school sorting by family background is taken into account. It is possible that this could reflect that targeted measures and resource compensation to schools with high shares of immigrant-origin minority students has helped stem the potentially adverse consequences of ethnic school segregation.

Immigrant Families and Ethnic Minority Communities as Resources for Educational Careers
The educational accomplishments of the children of immigrants have attracted substantial scholarly interest, as the group has been closing the academic gap to peers with native background in Norway. As migrant parents often occupy low status jobs with low wages, old explanations invoking socioeconomic resources have been deemed insufficient, and much of the explanatory discussion has circled around the existence of an 'immigrant drive' (Bakken 2016;Birkelund and Mastekaasa 2009;Friberg 2016;Lauglo 1999). A common story about this drive is that the parents' migration history and sacrifices give them a particular motivation for social mobility and that they transmit this 'immigrant optimism' (Kao and Tienda 1998) to their children through inter-connected networks and kinship ties within immigrant communities (Fekjaer and Leirvik 2011;Modood 2004). The research tradition presented in this section primarily aims to give an understanding of the educational success experienced by children of immigrants using qualitative interview data to study intergenerational processes and educational aspirations, and is dominated by sociologists and anthropologists.
In Norway, high educational aspirations from immigrant parents are well documented in qualitative, interview-based research (Hegna and Smette 2017;Kindt 2017a;Leirvik 2010;Prieur 2004;Vassenden and Bergsgard 2012). However, it has also been suggested that children of immigrants feel that they owe their parents to be successful, and that pursuing higher education is a way of expressing gratitude and paying retribution for the hardship and sacrifice experienced through migration (Leirvik 2010). Moreover, a form of social capital embedded in close-knit ethnic communities, often referred to as 'ethnic capital', has been put forward as important in explaining children of immigrants' educational success (Friberg 2016;Lauglo 1999;Leirvik 2010). For instance, Leirvik (2010) asks why young adults with parents from India and Pakistan find it important to follow their parents' wishes about career choice. Based on in-depth interviews, she argues that children internalize and act in accordance with norms about the importance of education, in conjunction with a pronounced honour and respect for their elders.
Elaborating and supporting this, Vassenden and Bergsgard (2012) find that the larger and more tightknit the community is, the more difficult it is to choose an educational pathway in opposition to the community's norms, such as arts or a subject within the creative industry. In a similar vein, Engebrigtsen et al. (2004) suggest that the differences in educational attainment between ethnic groups can be explained by differences in the ethnic communities' group resources and social capital. In Norway, the Sri Lankan-Tamil community is often portrayed as a 'model minority' with high labour market participation, while the Somali immigrant community is portrayed as difficult to integrate, with high levels of unemployment and high dependence on social benefits (Engebrigtsen et al. 2004;Fangen 2008;Fuglerud and Engebrigtsen 2006). Engebrigtsen et al. (2004) suggest a perspective on these differences through the lense of group community resources. The Tamil community typically provides after-school programs for the children in their community, including mother tongue instruction, help with homework as well as music and sport activities.
There is ongoing academic and public debate about the individual costs of the immigrant drive, particularly as children of immigrants' reported sense of duty, debt and gratitude might also be interpreted as an expression of social control (Rogstad 2016). Prieur (2004) argues that having a strict upbringing might explain good educational outcomes, and that this control aspect is gendered. Since girls with immigrant background often are exposed to stricter rules than their male peers, they also perform better in the educational system (Prieur 2004;Østberg 2003). In a recent article, Leirvik (2016) argues that the costs of ethnic capital have been largely under-communicated in previous research. In addition to interview data from her previous studies with children of Pakistani and Indian parents, Leirvik has interviewed 'minority counsellors' working in upper-secondary schools in Oslo. She challenges the 'ethnicity as resource'-perspective, and argues that within tightknit ethnic communities, parents and other adults exercise authority and power over their children in a potentially harmful way. Engebrigtsen (2007), on the other hand, argues that choosing education in accordance with parental wishes does not necessarily signal a lack of autonomy. When Tamil youths choose educational tracks in accordance with their families' desires, they do so knowing that this will give them independence in the future. However, Leirvik (2016) argues that this reasoning fails to take into consideration what happens if youths decide to act against their families' wishes.
Being subject to parental influence might not be understood as equivocal to negative social pressure. Based on survey data, Hegna and Smette (2017) found that although minority students report a stronger parental influence on their educational choice, they experience their parents as positive and supportive. This self-report is mirrored in Kindt (2017b), which focuses on children of immigrants enrolled in prestigious educational tracks. When talking about their educational choices they stress that "I have always loved it" and "I was never pushed". However, Kindt (2017b) suggests that they may be recounting their educational motives in ways acceptable to the majority population, attempting to avoid the stigma of 'a traditional immigrant' subject to family pressure. Some recent Norwegian and international studies have questioned predominant cultural theories about the immigrant drive or ethnic capital (Feliciano and Lanuza 2017;Kindt 2017a). When immigrants experience social degradation after arriving in a new country, in that they find themselves in a lower relative social position than the one they had in their country of origin, their children's school success can be a way to restore the family's status from their home country (Feliciano 2005;Ichou 2014). Based on interviews, Kindt (2017a) argues that when looking more broadly at children of immigrants' social class origin, focusing more on parental status prior to migration, what is understood as an 'immigrant drive' often resembles what studies of the majority population have conceptualized as a 'middle class drive'. Although hesitant to generalize, Prieur (2004) points to a similar tendency in her data: that the educational level of her informants seems to be connected to their fathers' education prior to migration. Leirvik (2012) also acknowledges the need for incorporating pre-migration status in analyses of children of immigrants' educational attainment. However, rather than emphasizing social class resources, she puts emphasis on the role of caste and whether or not immigrants have migrated from rural or urban areas. In contrast, while not entirely dismissing the relevance of parents' pre-migration status, Friberg (2016) argues that young people with immigrant backgrounds are more focused on family obligations than the majority, and that these obligations are directly related to their educational success. Thus, he concludes that a partly culturally determined 'immigrant drive' is real.
In sum, this tradition is characterized by the use of qualitative research methods to answer questions regarding intergenerational processes and educational aspirations. Two major themes in this research tradition are, first, the degree to which the so-called immigrant drive is a reflection of social control and, second, whether the immigrant drive is an expression of social status prior to migration or a product of cultural values within the immigrant network in Norway. One important contribution from this tradition is the emergence of a transnational perspective, and the significance of a wider social context for understanding educational trajectories of children of immigrants. At the same time, this research tradition tends to focus exclusively on the children of immigrants themselves, and rarely collects data from the perspective of the parental generation. Future research would also benefit from investigating educational aspirations and motivations through more longitudinal designs.

Curriculum, Teacher Instruction, and Minority Students Experiences with Inclusion and Exclusion
In the following section, we review the literature that seek to understand how the educational institutions in Norway are equipped to meet the challenges of a diverse student body. One strand focuses on the development and implementation of progressive, multicultural pedagogy in preschool and primary school, while another strand focuses on students' experiences in their everyday lives at school. This research tradition is typically undertaken by sociologists, anthropologists, and education researchers. Studies are often based on document analysis, fieldwork in school or preschool environments, and indepth interviews with teachers or students.
In line with Seland's (2011) description of the development of national school curricula discussed in the introduction, other researchers have identified a growing concern with social and cultural cohesion since the mid-1990s. In his book on religious education in Norwegian schools, Iversen (2012) argues that the meaning of the term 'values' has changed in the curriculum over the period he studies, 1974 to 2005. In the 1974 curriculum, 'values' were understood to concern individual actions, with an emphasis on teaching the students to distinguish between right and wrong. By 2005, 'values' referred more to identity and community, intended to help the students understand 'who they are' (Iversen 2012). At the same time, based on a comprehensive analysis of all school books used to teach history, religion and social science in lower and upper secondary schools, Midtbøen et al. (2014a) find a 'maturation of the multicultural field' in Norway over the past 20 years. They identify three signs of maturation. First, descriptions of minorities and diversity have become more nuanced and focused on disrupting stereotypes rather than reinforcing them. Second, the books more often discuss controversial topics such as extremism, freedom of expression and the relationship between welfare and migration. Finally, the newer books more often operate with a more inclusive "we", taking into account that many of the pupils using these books have ethnic minority background (Midtbøen et al. 2014a, pp. 132-134).
A multicultural pedagogy is supposed to make students of different origins feel included in the educational system (Banks 2008;Vasbø 2014;Øzerk 2008). In order to do this, school must adapt knowledge and experience from its diverse students, so that students are able to recognize their own experiences and thereby better understand the schools' curricula. Research on schools' ability to be inclusive for pupils with ethnic, religious or cultural minority background concludes that teachers tend to lack adequate knowledge and appropriate tools (Midtbøen et al. 2014b). A study of newly graduated teaching students find that they often understand and interpret deviant student behaviour as culturally conditioned (Dyrnes et al. 2015). These findings indicate an imbalance between the capacities required to teach diverse learners, and teachers' abilities to do so (Skrefsrud and Østberg 2015). Several researchers have argued that recognizing diversity while simultaneously creating a sense of social cohesion is difficult within the Norwegian unitary educational system (Green et al. 2006;Hagelund 2007;Seeberg 2003;Øzerk 2008).
Through fieldwork and interviews with teachers and other staff members at a Norwegian primary school with a diverse student population, Hagelund (2007) argues that cultural diversity is mostly presented through visible signs, material objects and standardized practices. The classrooms are decorated with flags to represent each student's origin, teachers teach their students songs and lyrics from different countries and in different languages, national holidays are celebrated and students are asked to bring food from their home countries. The study argues that these practices promote a notion of culture that is essentialist, and does not reflect any real divisions within or across cultures. In a similar way, Øzerk (2008) writes about 'ethnification' or 'festivalization' arguing that this way of teaching students about differences reinforces, rather than eradicates, divisions between 'us' and 'them'. However, Hagelund (2007) also identifies boundaries for when 'being different' is no longer accepted. While the particular school studied portrayed itself as tolerant and multicultural, the teachers were explicit about what type of behaviour they could tolerate. Conflicts would arise around participation in swimming lessons, school-camps, and parents' engagement in their children's schoolwork. In these cases, cultural diversity could become a problemand was no longer celebrated as a resource. Thus, Hagelund (2007) contends that even at a school that promotes itself as multicultural, the only way in which children of immigrant background could be legitimately integrated was to master the system set up by the school, and the welfare system more generally.
Schools' strategies and tools for dealing with diversity are embedded in a larger culture, and ways of talking and thinking about these issues. Based on fieldwork in Norwegian schools, Seeberg (2003) found that students with immigrant backgrounds were understood as having an identity that originated from their ethnic background, while students with a Norwegian background were seen as having multiple forms of belonging. She argues that this discourse makes it difficult to deal with difference. Further, instead of addressing diversity and accepting that students from different origins have multiple ways of living their everyday life, Norwegian schools typically handle difference through homogenization. This, Seeberg (2003) suggests is a consequence of the schools' 'hegemonic discourses' where similarity is seen as a precondition for inclusion.

Experiences with Inclusion and Exclusion
While we have already seen that immigrant-origin youth are generally positively inclined towards school and report high educational aspirations, this does not necessarily mean that they are treated well or feel included in the school system. As we documented in the previous section, research on cur-riculum and teaching in Norwegian schools have found that the Norwegian unitary school system is not particularly well equipped to handle student diversity. The research on students' experiences is analogous to this finding; Norwegian schools are struggling to strike the balance between upholding a cohesive community and accommodating cultural diversity and different abilities. Thus, many experience difficulties being 'different' (Hagelund 2007;Imsen and Volckmar 2014;Seeberg 2003).
As one example, Chinga-Ramirez (2015) finds that even though the teachers see themselves as colour-blind, the youth feel that their skin colour is an important marker, which creates distance and exclusion. The study argues that the principle of equality in the Norwegian educational system creates an understanding of what is 'normal' that is not explicit or articulated. The unconscious and unarticulated understanding of this normality creates boundaries between students at school. This is similar to Solbue (2014), which concludes that when people with individual differences are treated similarly it creates a lack of community and a feeling of exclusion. Relatedly, Østberg (2003) argues that students with immigrant backgrounds tend to withdraw from classroom discussions about religion, alcohol or other things where they might feel that their values are different than those commonly expressed. The problem, according to Østberg (2003), is that their ethnic background is made relevant in a confusing manner. Parts of their ethnic identity are celebrated and parts of it are condemned. Ramadan is one example, where Eid is acknowledged as an important holiday, while the practice of fasting is condemned as something one should not take part in.
The low number of teachers with immigrant background has been noted as one of the reasons Norwegian schools fail to give students with immigrant background the recognition they need (Spernes 2014). Norwegian teachers seem to have a limited understanding of cultural diversity, and students report being subjected to stereotypical understandings of what it means to be 'different' (Spernes 2014;Trøften 2010). We also know that students with immigrant backgrounds experience a drop in wellbeing in the transition from lower-secondary to upper-secondary school, something that can partly be explained by their lack of social network and lack of support from teachers (Frøyland and Gjerustad 2012;Hegna 2013).
Lack of cultural sensitivity is also reflected in how school counsellors guide their students. Norwegian schools provide a mandatory counselling session when students are about to make their first educational choice towards the end of tenth grade (Birkemo 2007;Buland 2011;Lødding and Holen 2012;Prieur 2004;Smette 2015;Spernes 2014). In her doctoral thesis, Smette (2015) shows how counsellors and teachers are concerned about the amount of autonomy students with immigrant parents have when they make educational choices. Counsellors often express a concern about immigrant parents' high and unrealistic educational ambitions for their children. In a report evaluating the mandatory counselling, school counsellors report thinking that students' own interests should guide what educational track they choose (Buland 2011). A good choice is understood as one based on the individual's own experiences, while a bad choice is understood as one based on external factors (Birkemo 2007). While tradition and continuity are thought to be in conflict with independence and authenticity in the Norwegian context, they are also believed to be more important and prevalent in the immigrant community (Lidén 2003).
Some researchers warn that the lack of recognition in school, by teachers, counsellors and in curriculum, could potentially be harmful (Eriksen 2013;Skrefsrud and Østberg 2015). Being subjected to discrimination or lack of recognition could lead to different types of responses. Music and Godø (2011) argue that when teachers do not recognize or accept male minority students' ways of taking part in school, they seek recognition from other sources. Often, these other sources are more violent and rough, and they end up embracing a "street culture". This accelerates into a vicious circle because "tough behaviour" is met with even less recognition from the teachers. Fangen and Lynnebakke (2014) define three typical responses to stigmatization: 'avoiding', 'working harder' and 'confronting'. They argue that the best way of dealing with stigmas is to alter between the different strategies. However, as they also point out -not everyone can participate in all three strategies. It depends on previous experiences and resources students with immigrant backgrounds bring with them into the situation.
In sum, this research tradition is preoccupied with the ability of the Norwegian school system to integrate a diverse student body. Currently, this research tradition is made up of a relatively small number of studies. The studies that exist focus mostly on how the educational system in Norway -which is characterized by an emphasis on equality and cohesion -tends to reinforce rather than eradicate ethnic divisions. One weakness of this tradition in the Norwegian context is the relative absence of studies that focus on visible group traits, such as skin colour, rather than culture. This tradition would likely also benefit from more comparative studies. Studying inclusion and integration of a diverse student body across different types of school systems would help shed light on the significance of the hegemonic discourses of similarity and cohesion in the Norwegian society and school system.

Discussion and Conclusions
In this chapter, we have identified three research traditions, each covering related strands of research on ethnic inequalities in education in Norway. We have not been comprehensive in the sense that we have presented all research within the main traditions. Instead, we have presented the central contributions within each research tradition in greater detail. Moreover, we have not included all research on ethnic minorities in the Norwegian educational system, but focused on studies that explicitly address educational inequalities. This means that we considered studies of identity, inclusion in sports, bilingualism and other such topics to be outside the scope of the chapter.
The research tradition on educational enrolment, achievement, and attainment typically consists of quantitative research based on large datasets. Within this tradition, we have identified two closely related strands of research. The main strand within this tradition studies the relationship between immigrant origin, family background and educational outcomes in various ways. A key finding is that ethnic inequalities in educational attainment have declined over time, and that once parental socioeconomic status is taken into account immigrant-origin students perform at par with or better than native-origin youth in Norway. Even without controlling for socioeconomic background, Norwegian second-generation immigrants are overrepresented in higher education compared to native origin students. However, it is not clear what factors can explain this positive trend and whether it will be sustained as future immigrant-origin student cohorts complete their schooling.
Thus, the tradition would benefit from more research on what lies behind this trend as well as a comparative focus on which institutional characteristics are most conducive to educational success among children of immigrants. Moreover, research tends to find that some variation in educational outcomes between second-generation immigrant minorities remain even after taking differences in parental socioeconomic resources and residential segregation into account (e.g., Hermansen 2016b). To better understand this variation, future quantitative studies should explore the role of between-group variation in the selectivity of immigrant parents, as captured by their relative educational positions in the distribution of the origin country (Feliciano 2005;Feliciano and Lanuza 2017;Ichou 2014;Van de Werfhorst et al. 2014), as well as the role of group level co-ethnic resources embedded within local immigrant communities (Bygren and Szulkin 2010;Åslund et al. 2011). Finally, research in this tradition would benefit from integrating subjective measures on immigrant students' ambitions, attitudes, and acculturation-related indicators from survey data with later follow-ups on outcomes from administrative registries.
The other, much more limited, strand of research within this tradition, studies ethnic school segregation and potential consequences of segregation for educational outcomes. This strand of research finds modest effects of high shares of immigrant-origin students in schools. Given that ethnic segregation between schools seems to be increasing in Norway, there is need for new studies addressing how teacher allocation and school-level resource compensation is related to the share of ethnic minority students in schools. While several studies have evaluated the consequences of ethnic segregation in schools, the main weakness of this research area is the relative lack of studies that both trace changes in ethnic school segregation over time, as well as studies that examine the causes of ethnic school segregation. The causes of ethnic school segregation should be studied both with respect to the underlying processes driving observed patterns of student composition, and how these patterns are linked to school-level resources such as teacher allocation and teacher-student ratios.
The second research tradition consists of qualitative research on families' and communities' impact on children of immigrants' educational attainment, with research questions typically influenced by findings from the quantitative literature. One key finding is that the immigrant communities are of importance for inspiring higher educational aspirations. Families and ethnic networks can function as a source of social capital contributing to the students' upward educational trajectories. At the same time, the more critical part of this literature emphasizes that pressure exercised by a tightknit community may be oppressive and harmful if choosing differently is associated with fear of negative sanctions. However, none of the above-cited studies interview parents. To better understand the 'immigrant drive', there is a need for more research on what immigrant parents think, want, and expect from their children, how they are involved in their children's lives, and how they lived prior to migrating to Norway. Within this tradition, future research should investigate educational aspirations and motivations through more longitudinal qualitative designs that would enable the tracking of aspirations, choices and outcomes and related experiences and coping mechanisms over time.
The third research tradition, on curriculum, teacher instruction, and minority students' experiences with inclusion and exclusion, finds that the Norwegian unitary school system is not particularly well equipped to handle student diversity, and that this can make it difficult for students with minority backgrounds to fit in. The tradition seems to identify a contested institutional field, where teachers attempt to handle a diverse student body, without adequate curricular tools to do so. At the same time, research suggests that the schoolbooks have become more inclusive and adjusted to a diverse student body over time. A possible critique of this research tradition is that it lacks a comparative perspective. Thus, the general conclusion might overestimate Norwegian schools' lack of ability to handle diversity. It would for example be fruitful to compare the Norwegian school system with other unitary school systems in Europe, such as the French system of laïcité. Further, some of the findings in this research literature are based on small samples of schools, classrooms, teachers and/or students. As a result, we do not know how widespread the challenges it identifies are, and we have little systematic information about pedagogical tools or teaching techniques that are successfully inclusive for ethnic minority students. Further research should also attempt to study processes of institutional change, and the impact such change may have for the inclusion and equal treatment of students of ethnic minority backgrounds. One avenue for this strand of research could be to explore differences in policy texts over time and across contexts, and how these are implemented by teachers in schools.
It is clear that the availability of registry data with information about country of origin has had consequences for the type of research and the categories used in Norwegian research. Self-reported ethnic identity or religion is considered sensitive information according to Norwegian data protection regulations, and is rarely collected in surveys. Similarly, 'race' is not used in Norwegian research on ethnic inequality in education. Barriers and prejudice based on skin colour and other visible traits are understudied, and tend to be deduced from information about country of origin.
Partly because of lack of statistics on achievement and attainment among Sami students, they are virtually absent in most of the literature. This is even more so the case for the five national minority groups. The lack of knowledge and public awareness about educational inequalities in these minority groups, has also contributed to the lack of qualitative studies on educational aspirations and experiences in these groups. Particularly acute is the question of whether and to what extent national minorities and the Sami population are experiencing cumulative discrimination in today's educational system (cf. Midtbøen and Lidén 2016).
While the quantitative tradition on enrolment, achievement and attainment tends to conclude that children of immigrants fare relatively well in the Norwegian school system, the two qualitative traditions tend to be more critical towards the Norwegian institutions and towards the immigrant communities, in various ways. It is worth noting that the quantitative research tradition is much larger, and has so far presumably been more readily funded by the Norwegian government, whether directly or through the Research Council of Norway, than the other two. A potentially promising development in this field is that the Research Council of Norway recently (in 2017) awarded eight doctoral student fellowships earmarked for research on group-based prejudice in schools. Whether the newly funded projects will address any of the questions we have raised in this chapter remains to be seen.
As the composition of the immigrant-origin population in Norway is changing over time, and more children of immigrants come of age, we may find that the patterns and mechanisms identified thus far do not apply to future generations. In particular, it remains to be seen whether the patterns we have observed among children of earlier waves of labour migrants and refugees are replicated among children of newer groups of refugees. Likewise, future policy changes may influence the ability of the Norwegian socialdemocratic welfare state to absorb children of immigrants and provide them with available trajectories for upward mobility.
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons license, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons license and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.