Ethnic school segregation in Copenhagen: A step in the right direction?

Abstract

Chinese

The Danish school system is based on a general belief in the quality and merits of public schooling. Until 20 years ago, more than 90% of all children attended public school. However, this trend has recently seen a decline because of rising spatial inequalities; nowhere is this more visible than in the major cities, particularly Copenhagen. One visible change has been the rise in the number of children with non-Danish backgrounds in public schools in major cities. Previous studies of Copenhagen showed that, while the level of ethnic residential segregation was moderate, the level of school segregation was remarkably high. The purpose of this paper is to revisit the case of Copenhagen through: (a) quantitatively identifying the level of ethnic school segregation in Copenhagen and the change over the last decade, and (b) qualitatively analysing the considerations regarding the school choice of parents in an ethnically diverse district. The paper identifies decreasing levels of ethnic school segregation in public schools but a markedly higher and increasing level in private schools. The qualitative material points to still-existing concerns regarding specific public schools with high proportions of pupils of non-Danish backgrounds as well as to parents who choose to overlook such concerns and opt for the local public school.

Keywords

ethnic diversity private schools public schools school segregation

Introduction

Education has become increasingly important for individuals and their lives. Well-educated individuals have better chances in life in terms of employment and income as well as social and health conditions. They live longer, have fewer health problems, earn more money and seem to be doing better than less educated individuals. Access to the labour market requires growing competences and skills. Consequently, increasing attention is paid to maximising the educational outcomes of children through school choice (Ball, 2001; Brantlinger et al., 1996; Vincent and Ball, 2007). In Denmark, this has heightened the focus on the quality and performance of the local public school and has increased the number of parents who choose a private school or a public school outside their district (Andersson et al., 2010; Rangvid, 2007, 2010). National politicians have used education, especially primary schools and their postulated declining quality, as a new political battlefield. In particular, non-Western pupils have been the focus of this discussion.¹ It is reflected in the recently passed Strategy Against Parallel Societies, called the ghetto strategy, where one of the six sub-agreements of the strategy addresses challenges of schools in deprived areas (Regeringen, 2018). It involves (a) language tests in the first year of school, followed by intensive support for strengthening language capabilities, and summer schools where needed; (b) an increased focus on the responsibility of parents for their child’s education with the possibility of withdrawing child benefits if children’s non-attendance is too high; and (c) increased possibilities for the state to sanction and even close schools if they continually underperform. The consequences of the reform for school segregation remain to be seen, as the strategy was passed in November 2018.

Especially in districts with high proportions of children from ethnic minorities, concerns regarding the quality of the local public school have increasingly become a reason for parents to choose private schools or a public school in a different district. In Denmark, this is an urban issue, as people from ethnic minorities make up a substantial proportion of the population in urban areas. Previously, most parents did not consider whether or not their child should attend the local public school; this was a default choice (Rangvid, 2007). Private schooling, in contrast, was mostly for children who were sent to boarding school, or whose parents wanted them to attend a school of a specific orientation or religious conviction. In the past, the proportion of pupils in private schools was limited and mostly related to the very upper-class, minor religious groups (e.g. Catholics or Jews), or people with a strong trust in special pedagogic principles, such as those of Rudolf Steiner. Currently, however, new private schools are emerging in bigger cities. Their profile is simply to do better than the local public school by offering lower numbers of pupils per class and a higher quality of schooling (reflected in the level of grades and test scores). Overall, the growing diversification of the urban population in combination with spatial segregation has challenged the tradition of the public school:

Danish parents living in areas with high-concentration schools [ed: schools with a high concentration of people from ethnic minorities] feel that choosing their local public school is no longer a relevant option (because they fear low academic levels and cultural conflicts) and this conflicts with the much-valued concept that the local public school is the natural choice of school for the overwhelming majority of Danes. (Rangvid, 2007: 1331)

Over the last decades, immigration from non-Western countries has increased, which has led to a growing number of children with non-Western backgrounds. In school districts with a high proportion of non-Western children (whether immigrants or their descendants), increasing concern has been voiced regarding what implications this will have for school quality. In Denmark, most children attend childcare institutions for five full days a week from the age of one. Consequently, nurseries, kindergartens and schools play important roles in the daily lives of children and are key arenas for intra-ethnic encounters and integration. Thus, it is imperative to track the development of ethnic school segregation and understand the reasons for such segregation.

In 2007, Rangvid identified interesting patterns of ethnic school segregation in Copenhagen: parental choices result in moderate levels of residential segregation being compatible with high levels of school segregation (Rangvid, 2007). This mirrors the findings of other scholars, of ethnic segregation being higher in schools than in residential areas (Allen, 2007; Burgess et al., 2005; Johnston et al., 2006, 2008). Rangvid called for further research into school segregation in Copenhagen to better understand this. Consequently, the purpose of this paper is two-fold: (1) to quantitatively identify the level of ethnic school segregation in Copenhagen and the change over the last decade, and (2) to qualitatively analyse the considerations regarding the school choice of parents in an ethnically diverse district. Building on the work by Rangvid, the paper updates data on ethnic school segregation in Copenhagen and adds qualitative findings to understand the considerations behind the school choice of parents in Copenhagen.

The situation in Copenhagen is far from unique. While the level of the division of pupils in the Nordic countries is modest, diversity, school segregation and difference in performance are increasing (Arnesen and Lundahl, 2006). Sundell (2014) demonstrated a clear correlation between residential segregation and school segregation in Sweden. In particular, small and medium cities show high levels of ethnic segregation and ethnic separation at the school level. A recent study (Kornhall and Bender, 2018) argued that school segregation is now common all over Sweden. In fact, two parallel school systems have emerged: one for well-educated Swedish middle-class families and one for families with foreign backgrounds. A similar pattern can be seen in Norway. Approximately 30% of pupils in the schools of Oslo are of non-Norwegian background. Birkelund et al. (2010) showed a close relationship between ethnicity, lack of education among parents and negative school performance (lower marks and higher drop-out rates); concluding that socio-economic background was a key factor.

The Nordic welfare states have managed to combine high living standards with relative equality. A key component of the welfare model is the provision of decent housing for all. The strong priority of this objective was achieved during the 1960s and 1970s and produced a huge number of housing units. Around major cities, large social housing estates were constructed. This, in combination with the tradition for district schools, produced socio-economically homogenous schools (Kristensen, 2007). When the ethnic minorities arrived in the 1960s and later, social housing was the only accessible sector for them. As the middle class began to buy rather than rent housing in the same period, the residualisation of social housing began. The middle class and many workers with a majority background moved out of the social housing estates and left an increasingly marginal population behind (Madsen and Hornstrup, 2000). People who were single, ethnic minorities, retired early and inactive in the labour market became a dominant portion of the tenants in many estates. Thus, segregation at the school level took place despite huge welfare expenditures.

Danish education system

The Danish public school is known as Folkeskolen, meaning ‘the school of the people’. Public education in Denmark has been provided by the state since 1814, when elementary education became mandatory. Folkeskolen covers the obligatory ten years of schooling from the year a child turns six, comprising both primary and lower secondary school. Education in public schools is free of charge. Primary schools are district schools, meaning that all children living in a given district have the right to attend that district’s municipal school. Overall, close to 77% of all pupils attend Folkeskolen (2016). However, the proportion attending a private school has increased by 50% over the last 20 years.²

The introduction of parental choice in 2005 has weakened the tradition of the district school. Children are still secured a place in the district school, but parents have the possibility of choosing another school. A central element in the 2005 reform was to provide easily accessible information on school quality to assist parents in choosing a school. Information is available online and allows for comparison between schools (www.uddannelsesstatistik.dk). Information includes grades and the number of teaching hours, staff and pupils in each class. Unfortunately, the published data do not include socio-economic background, and the ‘quality’ of each individual school is simply measured through student marks. While these data attract much attention, the effect on parental choice is moderated by the right to attend the local district school. Children from other parts of the city or other municipalities may attend a school if vacant places are available. Public schools that have good reputations are popular and less frequently opted out of by parents from the district, resulting in fewer empty places. Rangvid (2010) showed that, despite this, half of the children of Copenhagen were affected by school choice, meaning that their parents choose a different school than their local district school. In half of these cases, the choice was a public school outside the district, and the other half chose a private school.

The authorities monitor the quality of private schools but these schools may organise their teaching and disciplines as they wish, providing their pupils’ competences match the competences of those in the public school system. The state partly finances private schools by covering 75% of the average cost per public school pupil. The rest is paid by parents, sometimes supplemented by donations from foundations.

According to the official statistics from the Danish Ministry of Education, the number of public schools has fallen by 24% over the last two decades. This is largely caused by the merging of small rural schools. During the same period, the number of private schools has risen by 20% for several reasons. First, parents fear insufficient educational quality in public schools, which may threaten their children’s future life chances. Second, the rising number of non-Western immigrant children is causing concern. In Denmark, bilingualism is considered a potential challenge to educational quality, particularly if many children in a school have limited Danish language competences because they speak another language at home. Third, the growing diversification of the population has led to an increasing demand for special considerations and more time for the individual child. The latter is difficult because of budget pressure in the public sector, which has resulted in increasing class sizes in public schools. Fourth, local government reform in 2007 forced local governments to reduce their budgets. One method of achieving this has been to merge or close small schools, thus reducing the cost per pupil. A recurrent reaction from local communities has been to transform the village school into a so-called free school, a private school, to provide a local schooling option. The first three reasons have affected Copenhagen, while the latter happens primarily in depopulated areas.

Increasing attention to educational outcome

Over the last decades, the educational system has attracted increasing attention, becoming a key battlefield of modern politics in many countries (Ball, 2001; Brantlinger et al., 1996). Education and competence fit for intensified international competition have become cornerstones in national strategies. This reflects and is reflected in the rising attention among parents who see education as the key factor in determining the future possibilities for their children. As a consequence of the strong focus on educational outcome, the quality of schools has received ever-increasing attention (Pedersen, 2011). Researchers point to the decline of the welfare state in favour of educational consumerism and individualism (Beach and Sernhede, 2011).

Schools are a strong marker of the socio-spatial differentiation of urban space (Gibbons and Telhaj, 2007) with school reputation drastically affecting the recruitment of pupils. In contrast to working-class parents, middle-class parents are active choosers who have managed to maintain their relative privilege in terms of access to high-quality education through parental choice (Butler and Robson, 2003). Consequently, middle-class families either attempt to keep a distance from working-class residential districts (Sibley, 1995), or choose a school outside the catchment area (Butler and Robson, 2003). Thus, while tendencies towards decreasing residential segregation can be identified, this is coupled with an increasing segregation in relation to institutions, consumption and leisure activities (Burgess et al., 2005; Rangvid, 2007).

The issue of segregation is of major importance in discussions of school quality. Parents react to the perceived quality of individual schools based on school reputation, which largely depends on the social composition of the school. Consequently, the social and ethnic milieus of schools become central parts of school reputations. For pupils of non-Danish origin, schools are important for their integration, as they are key arenas for encountering mainstream society norms and values (Alwan, 2015; Rangvid, 2007). Among both politicians and parents, this has caused concern regarding schools with a high proportion of ethnic minority pupils (Boterman, 2013; Johnston et al., 2008).

Rangvid (2007, 2010) showed that one outcome of these processes in Copenhagen was a rising separation at the institutional level but without the same drastic rise in residential segregation. Parental choice between public schools and the supply of private schools enables families to remain in mixed neighbourhoods without being forced to send their children to a less attractive school. This illustrates the phrase ‘close together and worlds apart’ (Bourne, 1993). Despite socially and ethnically mixed neighbourhoods, cross-cutting interaction is limited.

Methods and data

The quantitative data originate from official, public statistical databases, primarily the national Statistics Denmark (www.Statistikbanken.dk). Statistics Denmark collects data regularly for the entire country, most of which are produced for administrative purposes. The Ministry of Education has established an online data service with detailed information on individual schools and their pupils, including information on grades, wellbeing and non-attendance (www.uddannelsesstatistik.dk). This is coupled with data from Statistics Copenhagen, covering the City of Copenhagen (www.kk.dk/statistik). Data are not available at a detailed level, nor is it possible to find figures for special social groups without special data deliverables. However, the Danish Registers offer a vast quantity of publicly accessible data, which has been utilised for this paper.

The qualitative material originates from the Danish part of the major European research project DIVERCITIES.³ One part of the project focused on the residents’ perception of and experience with diversity. While school segregation was not a stated focus of the interviews, the interviews clearly point to patterns and challenges with school segregation in the district. For parents in a diverse district, school choice is clearly an imperative concern and a main aspect of living with diversity. Thus, while school segregation was not a predefined theme in the interview guide, it became a central theme in the material. Herein lies the potential of a qualitative approach, allowing for unanticipated themes to emerge. The paper draws on a subsection of the 50 qualitative interviews with residents of Bispebjerg, namely the 35 who were parents. Of these, 18 had children living at home, and these constitute the main basis of the analysis in this paper, supplemented with input from other interviewees. When recruiting interviewees, the goal was to reflect the different population groups in Bispebjerg. Accordingly, prototypical groups mirroring the diversification processes of the district were identified and used to steer recruitment. The goal of the sampling method was not to construct a representative sample but to ensure that the different groups were included. Further information can be found in Skovgaard Nielsen et al. (2016).

School segregation in Copenhagen

The majority of Copenhagen schools have relatively few pupils. A fifth have fewer than 100 pupils, and only three schools have more than 1000 pupils. In Copenhagen Municipality, 70% of children attend one of the 63 public schools at the primary/lower secondary level. The rest attend one of the 57 private schools. Even though the number of pupils in private schools is still less than half of that of public schools, the number of choices available for private schooling is almost as a high as that for public schools.

The proportion of pupils with a non-Danish background⁴ varies substantially between Copenhagen districts. Bispebjerg and Brønshøj-Husum have the highest proportions (43.3% and 35.7%, respectively; see Figure 1). The ethnic composition has challenged the traditional organisation of the public school system, as it has caused many Danish parents to opt out of their local district school. Their solutions have been to choose a private school or a different public school or to move to a less ethnically mixed neighbourhood. The outcome has been a sharp increase in the proportion of pupils with non-Danish backgrounds in some public schools and intensified attempts to find new ways of avoiding ethnically divided schools. In the school year 2016/2017, eight schools in Copenhagen had no pupils with a Danish background, and in 25 schools less than 40% of the pupils had a Danish background. At the other extreme, about 20 schools had no pupils with a non-Danish background, and 68% had less than 20% of pupils with a non-Danish background. However, if we delve into this development over the last decade, we find signs of change towards less ethnic school segregation.

Figure 1.

Proportion of pupils with a non-Danish background in municipal schools in Copenhagen, district level, 2016 (Frederiksberg Municipality excluded).

Table 1 reveals an important change in the pupil composition. The relative proportion of pupils with a non-Danish background peaked in the 2000s and has slowly declined since then. The reason for this is the growth in the number of pupils with a Danish background, which increased by 31% during the period 2010–2016, and a decrease in the number of pupils with a non-Danish background (−6.5%), despite an overall increase in the size of the non-Danish population of Copenhagen (from 17% in 2000 to 24% in 2017). This in itself has reduced ethnic segregation. The main cause of this change is the population growth of younger Danish households caused by the improved housing conditions in combination with a growth in the housing stock of larger dwellings suitable for families (Andersen, 2017). Additionally, serious attempts made by the authorities to reduce school segregation have further diminished segregation at the school level. The index of dissimilarity has fallen from 0.5 to 0.46 for all schools within the City of Copenhagen (2010–2016). However, the ethnic segregation level between municipal schools was lower than that between private schools, increasing moderately since 2010 in private schools and decreasing in public schools.

Table 1.

Number and proportion of pupils with non-Danish background 2010–2017 in schools in Copenhagen Municipality.

	2010/2011	2011/2012	2012/2013	2013/2014	2014/2015	2015/2016	2016/2017	2017/2018
Danish pupils	31,867	32,659	34,091	35,379	37,017	38,566	40,010	41,635
Non-Danish pupils	14,249	14,114	13,998	13,805	13,770	13,559	13,004	12,905
Unknown	205	209	219	297	532	518	572	604
Total	46,321	46,982	48,308	49,481	51,319	52,643	53,586	55,144
Non-Danish/unknown (%)	30.9	30.2	29.1	28.1	27.1	26.0	24.5	23.7

Source: Uddannelsesstatistik.dk.

An ethnic division between schools in Copenhagen has existed for some years. Table 2 shows a modest decline in segregation between pupils with Danish and non-Danish backgrounds from 2010/2011 to 2016/2017 from 0.49 to 0.46. However, it also demonstrates a separation between groups. The index of isolation at the school level was 0.6 in 2010/2011 for Danish pupils. This index shows the probability that a Danish student will randomly meet another Danish student. Non-Western pupils were less isolated with an index of 0.5. In Bispebjerg, however, the index of isolation was 0.8 for Danish pupils. These figures suggest that schools are mixed across ethnic background only to a limited degree. The number of pupils with non-Danish backgrounds has decreased both at district level and within the private school sector, while the number of Danish pupils has increased during the same period. Thus, the declining level of segregation is reflected in municipal schools, while the slight increase in segregation in private schools reveals that ethnic minorities utilise their right to form private schools. With a few exceptions, the private schools in Bispebjerg are faith-based schools.

Table 2.

Index of dissimilarity and isolation for schools in Copenhagen Municipality and Bispebjerg district.

Copenhagen	2010/2011	2016/2017
Index of dissimilarity, all schools	0.49	0.46
Index of dissimilarity, public schools	0.40	0.35
Index of dissimilarity, private schools	0.67	0.69
Index of isolation (DK)	0.66	0.63
Index of isolation (ND)	0.52	0.44
Bispebjerg	2010/2011	2016/2017
Index of dissimilarity, all schools	0.39	0.31
Index of dissimilarity, public schools	0.24	0.16
Index of dissimilarity, private schools	0.35	0.26
Index of isolation (DK)	0.79	0.81
Index of isolation (ND)	0.52	0.40

Notes: DK, Danish; ND, pupils with non-Danish backgrounds.

Source: www.uddannelsesstatistik.dk.

Overall, ethnic school segregation has dropped, even if only by a little. The proportion of pupils with non-Danish backgrounds peaked a decade ago and is now at the level of the 1990s. The number of Danish pupils has increased over the same period. In combination, this has reduced both the proportion of non-Danish pupils as well as the segregation level in general. The establishment of ethnic private schools has, however, produced a high level of segregation in private schools.

Evidence from a Copenhagen district

To understand the processes behind ethnic school segregation and the considerations of parents involved in school choice in Copenhagen, the following section of the paper analyses qualitative evidence from the district of Bispebjerg. Bispebjerg is interesting, as it has the highest proportion of pupils with non-Danish backgrounds and offers more private than public schools.

Diversity and school segregation of Bispebjerg

Copenhagen consists of ten districts. The district of Bispebjerg lies in the northern part of the city and houses about 55,000 inhabitants (2017). It has a highly diverse residential composition, both ethnically and socio-economically (Skovgaard Nielsen et al., 2016). A substantial proportion of the area’s housing is social housing, which remains the main tenure for ethnic minorities in Denmark (Skovgaard Nielsen, 2017). The different waves of immigration to Denmark have also affected the area, leading to a high level of ethnic diversity (Table 3).

Table 3.

Ethnic diversity in Denmark, Copenhagen and Bispebjerg, 2017.

	Denmark	Copenhagen	Bispebjerg
Danish origin	87.1%	75.6%	66.0%
Immigrants, Western (incl. EU-27)	4.1%	8.4%	8.3%
Descendants, Western	0.5%	0.9%	1.0%
Immigrants, non-Western	5.8%	10.2%	16.5%
Descendants, non-Western	2.5%	4.9%	8.1%
Children aged 6–15, Danish origin	88.1%	76.8%	55.5%
Children aged 6–15, Western origin	2.1%	3.5%	4.1%
Children aged 6–15, non-Western origin	9.7%	19.7%	40.4%

Source: Københavns Statistikbank [Copenhagen Statistical Bank] (2017) and Statistikbanken [Statistical Bank Denmark] (2017).

A substantial proportion of Bispebjerg is categorised as deprived by Copenhagen Municipality. The municipality’s definition is based on a combination of the following indicators: small housing units (<60 m²), housing units with installation deficiencies, residents with non-Western origins, residents outside of the workforce, residents with modest or no education and residents with a low income (Copenhagen Municipality, 2011). Growing segmentation (Lindberg and Linden, 1989) has produced marked social differentiation within the district. Although segregation in Copenhagen is still at a modest level, the differences between various forms of ownership are significant. An increasing challenge in Denmark over the last decades has been the concentration of disadvantaged and ethnic minorities in the social housing sector, a tendency that is also seen in Copenhagen. While the Danish social housing sector is aimed, in principle, at the general mass of the population, it has, in reality, increasingly become social housing (Skovgaard Nielsen and Haagerup, 2017). The district of Bispebjerg is a diverse but divided district. Two key reasons are that ethnic minorities are generally over-represented in social housing (Skovgaard Nielsen, 2017) and that, on average, the socio-economic situation of ethnic minorities is worse than that of the Danish majority. Consequently, socio-economic spatial segregation leads to ethnic spatial segregation. The spatial segregation is reflected in public schools, as some schools cater primarily to affluent residents and others to the less affluent, causing the percentages of ethnic minority children to vary substantially between schools.

The district of Bispebjerg is home to 13 schools, of which eight are private schools. While the majority of the schools are private, the smaller sizes of the private schools cause their proportion of pupils to be just below 30%. The proportion of pupils attending a private school has increased slightly from 2010/2011 to 2016/2017, from 24% to 28%. The number of pupils in Bispebjerg remained stable during the period 2010–2016, and the ethnic balance between pupils of Danish and non-Danish background was relatively constant. Pupils with non-Danish backgrounds make up approximately 44%. While the proportion varies from 30% to 67% in public schools, the variation is between 0% and 85% within private schools. The ethnic composition in private schools is thus strongly polarised and reflects the existence of several Muslim schools and a Jewish school in the district.

From 2010/2011 to 2016/2017, ethnic segregation between schools in the district of Bispebjerg diminished slightly from 0.24 to 0.16 for municipal schools and from 0.35 to 0.26 for private schools (Table 2). Thus, contrary to much debate and ‘conventional wisdom’, ethnic segregation is slowly declining. However, the more surprising fact is that ethnic segregation at the school level is much lower within the ethnically mixed district of Bispebjerg than at the city level. The percentage of bilingual pupils varies immensely between the five public schools of Bispebjerg, ranging from 28% at Holberg School in Emdrup and 33% at Utterslev School to 76% at Tagensbo School (Bakalus, 2014). While the percentage of bilingual pupils in Tagensbo School is much higher than the average percentage of ethnic minorities in Bispebjerg, it corresponds approximately to the proportion of ethnic minorities in the school district of Tagensbo School. Thus, in this case, ethnic school segregation reflects ethnic spatial segregation.

Parental perceptions of public schools

As described in the method section, the recruitment of interviewees was based on an attempt to achieve a broad sample of the different groups in Bispebjerg. The diversity of the interviewees in general is mirrored in the group of parents in the sample. They are of mixed ethnic backgrounds and ages, have different socio-economic positions, and have lived in the district for varying numbers of years. Across these differences, the proportion of ethnic minority pupils is mentioned by the parents as a main concern regarding public schools.

Not to sound racist, but I think that there are a lot of immigrant children out here who don’t know how to behave the way that we want our children to behave. (Female, 25, works in a bakery, ethnic Danish background, social housing)

[Describes how changes to the school district changed the ethnic composition of her son’s school prior to his enrolment]: I thought that was really cool because I would have had a terribly bad conscience if I had had to say that we were going to move our son because we don’t want him to go [to school] with all those immigrants. (Female, 43, export adviser, German-Polish background, owner-occupied housing)

Ethnic pupil composition is also a concern among ethnic minority parents, and concerns such as those above are also expressed by minority parents. They express a wish for their children to interact with children of Danish origin in order for them to get to know Danes and Danish society. This could be a reason for the decreasing isolation index of the quantitative material.

Some of the interviewees expressed ambivalence regarding school choice. On one hand, they support the public school in principle and would ideally want their children to attend the local public school. On the other hand, when the local school is perceived as challenged, they are not prepared to enrol their child in the school or at least are very concerned about doing so. It becomes a conflict between principles and concerns regarding the wellbeing and education of the child:

But again, the paradox is that we all in a way want to save the public school. But when it comes to our own children, then that wasn’t really what we meant [when we said we wanted to save the public school]. It wasn’t them [their own children] that were to save the public school. (Female, 30s, Social Housing+⁵)

In Bispebjerg, school segregation is closely related to the reputation of the specific schools. Three public schools are especially referred to in the material. Interviewees living in the school districts of Holberg and Utterslev feel comfortable with sending their children to the local school, whereas this is not the case for interviewees living in the district of Tagensbo School. Although improvements to the social and educational environments at Tagensbo School are acknowledged by the interviewees, accounts of a fragile social order and a harsh environment are expressed by all parent interviewees living in the district, whether or not their children attend Tagensbo School:

We were convinced that he was going to go to Utterslev School, as this is about 100 metres closer and because it is a great school. So, it was like: ‘Yippee, he will be going to Utterslev School!’ But then we got that letter [from the municipality] that now the child is starting school and then it was Tagensbo. My girlfriend was almost in tears; I think actually she did cry. And then all the parents from our housing association were out on the stairs [discussing the matter] and set up Facebook groups: ‘What are we going to do?’ Because no one had really heard anything good about that school. (Female, 30s, Social Housing+)

Some interviewees have chosen, despite concerns, to enrol their children in Tagensbo School. This was the case for the interviewee quoted above. She is a member of a group of residents in a newly built social housing estate aimed at more-affluent residents. The residents are predominantly young families with children, are well-educated with stable employment, and have many personal and social resources. There is a comprehensive social life within the estate among children and adults alike, and the residents provide mutual support for each other. A large group of these residents made the common decision to send their children to Tagensbo School, despite its poor reputation. This finding mirrors that of Boterman (2013), who identifies a strategy of the ‘mixing’ of ‘black schools’ in Amsterdam, with more highly educated native Dutch parents agreeing to simultaneously send their children to the same neighbourhood school to change the pupil composition of the school. Vowden (2012) described this as a strategy of safety in numbers in a study of middle-class parents of primary school children in London. Social mix is acceptable and even desirable, provided that the majority or a substantial minority were from middle-class white backgrounds, and the overwhelming majority were proficient in English when starting school (Vowden, 2012). This could be another cause of the decreasing segregation identified quantitatively.

Several interviewees who have chosen to enrol their children in Tagensbo School despite their concerns were positively surprised, saying that the school does not live up to its doubtful reputation. For other interviewees, Tagensbo’s reputation and their own concerns about the school have caused them to either choose a private school or make plans to move. To some, the school district was crucial in their choice of housing purchase: ‘I don’t know if we would have bought it if it had belonged to Tagensbo School down there’ (female, 36, management consultant, ethnic Danish background, owner-occupied housing).

One of the challenges arising from the school segregation of Bispebjerg is that it limits the potential in a neighbourhood to foster diverse social relations and social cohesion. As the majority of children in Denmark attend childcare institutions for five full days a week from the age of one, nurseries, kindergartens and subsequently schools play important roles in the daily lives of children. In diverse neighbourhoods, such institutions can function as key arenas for encounters across differences: ‘My schooldays were a good experience, because we were such a mix in my class. We were all so different that we simply just had to accept each other, and we’re still like that today’ (female, 25, in vocational training, Ghanaian background, cooperative housing).

In most cases, the children of the interviewees attend schools and institutions located in their own neighbourhood. Some parents identify a certain grouping among the children along socio-economic and, to some extent, ethnic lines. However, they also see the formation of friendships across differences. The local institutions provide arenas for the children to engage in diversified activities and relations (Skovgaard Nielsen et al., 2016). However, this requires a fairly equal representation in these arenas of the diverse neighbourhood groups:

The thing is, if all the children are taken out of their neighbourhood [school], this will have a very negative effect on the way we relate to one another, as we will not know each other so well. So, it’s actually quite important. (Female, 38, consultant, ethnic Danish background, owner-occupied house)

Initiatives to change the patterns and effects of school segregation

Copenhagen Municipality considers school segregation along socio-economic and ethnic lines to be an increasing problem. The ‘Action Plan for the Inclusion Policy’ by The Children and Youth Administration (2011) states the following as one of its two main goals (the other being that all children regardless of ethnic and social backgrounds achieve the qualifications necessary to obtain upper secondary education and gain access to the labour market):

To create the best possible distribution of the city’s children in childcare institutions and schools within the given frames with an aim to enhance the relations between children and their parents across ethnic and social backgrounds in order to strengthen the social integration and create a city marked by acquaintance, understanding and tolerance between the residents of the city. (The Children and Youth Administration, 2011)

To counter school segregation, the municipality has altered the public school districts more than once in recent years. Different boundaries between districts have been drawn to cut across the segments of the housing market. Since social classes, to some degree, separate spatially as a result of segmentation in the housing market, economic capacity and personal choice are reflected in the socio-economic distribution of pupils among schools. To counter the socio-economically unequal distribution of pupils, the municipality has redrawn school district boundaries to create more mixed schools by ensuring that the districts cover different housing types with different types of residents. The challenge is that parents are not blind consumers of schools. The most resourceful parents especially have the economic and social resources to choose private schools if the districts are changed in a way that causes their children to belong to a school of which they do not approve. While, on the drawing board, new districts might lead to a new ‘recruitment area’ for a given school, an actual change in pupil composition requires that the parents choose their local school. The high proportion of pupils affected by school choice in Copenhagen suggests that they do not automatically choose their local school. The solution thus does not seem to have the desired effect, at least not if it is not coupled with other efforts (e.g. to change the image of specific schools).

The municipality has made physical improvements and management changes in schools with poor reputations to address the actual challenges in these schools and to change their reputation. Based on the evidence from Bispebjerg, it seems that tackling reputation is the most imminent challenge to avoid school segregation. Furthermore, policy initiatives have been taken to create ethnically mixed schools in Copenhagen, or at least reduce the ethnic segregation between schools. The ‘Action Plan for Inclusion Policy’ is aimed at increasing diversity in municipal schools by distributing bilingual children across schools throughout the city (Skovgaard Nielsen et al., 2016). Attempts at changing the distribution of bilingual children were introduced in 2004 under the name of magnet-schools (later ‘The Copenhagen Model’ in 2007). Bilingual children starting school who belong to school districts with a high proportion of bilingual children are offered a place in a school with a low proportion of bilingual children. At the same time, schools with a high proportion of bilingual children are given additional resources, and new interesting profiles for these schools are created to increase the proportion of Danish parents who choose the school. Municipal public relations campaigns have, in various ways, been aimed at changing the reputation of specific schools and convincing parents to choose their local public school.

When this policy was first introduced, the municipal administration relocated all bilingual children, regardless of whether they had linguistic challenges (Skovgaard Nielsen et al., 2016). However, this resulted in substantial criticism from the public, who deemed the programme discriminatory. Thus, the programme now relocates all children with linguistic challenges, regardless of their ethnic backgrounds. Even though many of the children in the programme are from non-Western backgrounds, this has improved the programme’s effectiveness, as only children who actually have linguistic difficulties are included. The focus of the ghetto strategy on public schools in deprived housing areas will increase the focus on language capabilities of school starters. However, parental concerns over concentration schools (see Rangvid, 2007, quoted in the introduction) are also about cultural conflicts and these are not solved through language programmes.

Supplementing the municipal initiatives, the parental association ‘Use Your Local School’ is a voluntary initiative aimed at getting Copenhagen parents to choose their local public school. The association has parents who have chosen the municipal school relay their experience to affluent parents of children starting school. One part of this is to counter the bad reputation of specific schools and the criticism of the academic level at public schools. Another is to ensure that parents of preschool children become familiar with the local school and establish contact between local families before their children start school. However, dependence on involvement and commitment from local parents has proven a potentially fragile construction in disadvantaged areas (e.g. Bispebjerg).

At the same time, religious private schools for primarily ethnic minority children are potentially creating a separate school system with a low level of integration of pupils of Danish and non-Danish backgrounds. While the municipal efforts so far have focused on ethnic distribution across public schools and the school choice of more-affluent Danish parents, future efforts might need to focus on getting ethnic minority parents to choose their local public school. It is paradoxical that, while the efforts have focused on Danish parents, the challenges emerging might refer to a greater extent to the choices of ethnic minority parents.

Conclusion

In Denmark, ethnic school segregation has been on the public agenda for a long time because of a rising number of ethnic minority inhabitants. This is predominantly an urban issue, as the concentration of people from ethnic minorities is highest in the cities. In 2007, Rangvid concluded that, while the overall level of ethnic segregation in Copenhagen was relatively moderate, the segregation of pupils due to ethnic background between schools was surprisingly high, approaching the level of US cities.

Since Rangvid’s study, important changes have taken place. First, the population of Copenhagen has grown considerably. This population growth is predominantly due to the increasing number of children of Danish descent. Second, the overall proportion of school-aged children with non-Danish backgrounds in Copenhagen is declining. The general improvement of the city and its dwellings has increased the city’s attractiveness and transformed districts into more middle-class areas. At the same time, active policies have been implemented to reduce segregation, such as convincing parents of near-school-age children to choose their local public school, redrawing school district boundaries, securing a more equal distribution of bilingual children, and changing the bad reputations of specific schools. The qualitative material shows that the image of specific schools, particularly the proportion of ethnic minorities among their pupils, is a key concern of parents. However, the interviews also identify attempts to go against public opinion and choose the local public school, even if the perception of the public regarding the school is negative.

An important shift since the 1990s is the reduced emigration of well-educated middle-class families from Copenhagen. This change, primarily due to the refurbishment of housing, in combination with the transfer of large parts of the private rented sector into cooperatives and owner-occupied housing, has increased the proportion of young middle-class families living in Copenhagen (cf. Andersen, 2017). Gentrification has changed the residential composition of neighbourhoods and, consequently, the composition of the children of corresponding catchment areas. At the same time, while the non-Danish population of Copenhagen has increased during the last two decades, the proportion of pupils with a non-Danish origin has decreased. A more restrictive immigration policy has reduced the number of younger immigrants and hence the number of children. These changes have affected the distribution of ethnic minorities at the school level, leading to diminished levels of school segregation. Available statistics show a minor decline at the city level, while Bispebjerg, which has the highest proportion of pupils with non-Danish backgrounds, has experienced a major reduction in ethnic school segregation in both public and private schools.

Before dismissing the challenge of separate learning, however, the index of isolation should be considered. This index is generally quite high for Danish pupils, especially in the Bispebjerg district despite the moderate dissimilarity index. This reveals a remarkable development over recent years: the majority of pupils in private schools are mostly of non-Danish backgrounds, while Danish pupils dominate municipal schools. Thus, within each sector, the ethnic composition is relatively homogenous. This warrants an increased focus in the coming years on ethnic school segregation within private and public schools with specific attention regarding the reasons for and consequences of parental choice among ethnic minority parents.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work is based on research findings from the project ‘Divercities – Creating social cohesion, social mobility and economic performance in today’s hyper-diversified cities’. This project has received funding from the European Union’s Seventh Framework Programme for research, technological development and demonstration under grant agreement No. 319970. SSH.2012.2.2.2-1; Governance of cohesion and diversity in urban contexts.

Notes

ORCID iD

Rikke Skovgaard Nielsen

References

Allen

(2007) Allocating pupils to their nearest secondary school: The consequences for social and ability stratification. Urban Studies 44(4): 751–770.

Alwan

(2015) Folkeskolen er Danmarks vigtigste integrationsprojekt [The public school is Denmark’s most important integration project]. Information, 30 January 2015. Available at: https://www.information.dk/debat/2015/01/folkeskolen-danmarks-vigtigste-integrationsprojekt (accessed 24 June 2019).

Andersen

(2017) Copenhagen: The social costs of urban renewal. In: Cucca

Ranci

(eds) Unequal Cities. Abingdon: Routledge, pp. 138–154.

Andersson

Ösht

Malmberg

(2010) Ethnic segregation and performance inequality in the Swedish school system: A regional perspective. Environment and Planning A 42: 2674–2686.

Arnesen

Lundahl

(2006) Still social and democratic? Inclusive education policies in the Nordic welfare states. Scandinavian Journal of Educational Research 50(3): 285–300.

Bakalus

(2014) Her er skolerne vi vælger fra: Se hvor mange tosprogede elever der er på dit barns skole [These are the schools we avoid: See the proportion of bilingual children at your child’s school]. Available at: https://www.bt.dk/danmark/her-er-skolerne-vi-vaelger-fra-se-hvor-mange-to-sprogede-elever-der-er-paa-dit-bar (accessed 24 June 2019).

Ball

(2001) Class Strategies and the Education Market. The Middle Classes and Social Advantage. London: Routledge.

Beach

Sernhede

(2011) From learning to labour to learning for marginality: School segregation and marginalisation in Swedish suburbs. British Journal of Sociology of Education 32(2): 257–274.

Birkelund

Hermansen

Evensen

(2010) Skolesegregering – et problem? Elevsammensetning, frafall og karakterer i Oslo-skolen. Report. Institutt for sosiologi og samfunnsgeografi, Oslo University.

10.

Boterman

(2013) Dealing with diversity: Middle-class family households and the issue of ‘black’ and ‘white’ schools in Amsterdam. Urban Studies 50(6): 1130–1147.

11.

Bourne

(1993) Close together and worlds apart: An analysis of changes in the ecology of income in Canadian cities. Urban Studies 30(8): 1293–1317.

12.

Brantlinger

Majd-Jabbari

Guskin

(1996) Self-interest and liberal educational discourse: How ideology works for middle-class mothers. American Educational Research Journal 33(3): 571–597.

13.

Burgess

Wilson

Lupton

(2005) Parallel lives? Ethnic segregation in schools and neighbourhoods. Urban Studies 42: 1027–1056.

14.

Butler

Robson

(2003) Plotting the middle classes: Gentrification and circuits of education in London. Housing Studies 18(1): 5–28.

15.

Children and Youth Administration, The, Pædagogisk Faglighed (2011) BUU’s handleplan til Inklusionspolitikken 2011–2014. [BUU’s Action plan for the Inclusion Policy 2011–2014]. Copenhagen Municipality.

16.

Copenhagen Municipality (2011) Policy for Disadvantaged Areas [Politik for udsatte byområder]. The Technical and Environmental Administration.

17.

Gibbons

Telhaj

(2007) Are schools drifting apart? Intake stratification in English secondary schools. Urban Studies 44(7): 1281–1307.

18.

Johnston

Burgess

Wilson

, et al. (2006) School and residential ethnic segregation: An analysis of variation across England’s local education authorities. Regional Studies 40: 973–990.

19.

Johnston

, et al. (2008) ‘Sleep-walking towards segregation?’ The changing ethnic composition of English schools, 1997–2003: An entry cohort analysis. Transactions of the Institute of British Geographers 33(1): 73–90 (accessed 24 June 2019).

20.

Københavns Statistikbank [Copenhagen Statistical Bank] (2017). Available at: https://www.kk.dk/artikel/statistikbanken-0 (accessed 24 June 2019).

21.

Kornhall

Bender

(2018) Ett söndrat land. Skolval och segregation i Sverige [A Divided Country. School Choice and Segregation in Sweden]. Stockholm: Arena Idee.

22.

Kristensen

(2007) Housing in Denmark. Centre for Housing and Welfare – Realdania Research, København. Available at: https://boligforskning.dk/sites/default/files/Housing_130907.pdf (accessed 24 June 2019).

23.

Lindberg

Linden

(1989) Social segmentation på den svenska bostadsmarknaden. Report. Sociologiska Institutionen, Lunds Universitet, Lund.

24.

Madsen

Hornstrup

(2000) Analyser af situationen på boligmarkedet. København: Boligselskabernes Landsorganisation.

25.

Pedersen

(2011) Konkurrencestaten [The Competition State]. København: Hans Reitzel.

26.

Rangvid

(2007) Living and learning separately? Ethnic segregation of school children in Copenhagen. Urban Studies 44(7): 1329–1354.

27.

Rangvid

(2010) School choice, universal vouchers and native flight from local schools. European Sociological Review 26(3): 319–335.

28.

Regeringen (2018) Ét Danmark uden parallelsamfund. Ingen ghettoer i 2030 [One Denmark Without Parallel Societies]. Copenhagen: Regeringen.

29.

Sibley

(1995) Geographies of Exclusion: Society and Difference in the West. London: Routledge.

30.

Skovgaard Nielsen

(2017) The potentials of a strong social housing sector: The case of Turks and Somalis in the Copenhagen housing market. Housing, Theory and Society 34(4): 458–476.

31.

Skovgaard Nielsen

Haagerup

(2017) The Danish social housing sector: Recent changes and future challenges. Critical Housing Analysis 4(1): 142–149.

32.

Skovgaard Nielsen

, et al. (2016) DIVERCITIES: Dealing with Urban Diversity: The Case of Copenhagen. Copenhagen: SBi Forlag.

33.

Statistikbanken [Statistical Bank Denmark] (2017). Available at: http://www.statistikbanken.dk/statbank5a/default.asp?w=1680 (accessed 24 June 2019).

34.

Sundell

(2014) Skolsegregation och boendesegregation. Available at: https://politologerna.wordpress.com/2014/04/15/skolsegregation-och-boendesegregation/ (accessed 8 November 2018).

35.

Vincent

Ball

(2007) ‘Making up’ the middle-class child: Families, activities and class disposition. Sociology 4(6): 1061–1077.

36.

Vowden

(2012) Safety in numbers? Middle-class parents and social mix in London primary schools. Journal of Education Policy 27(6): 731–745.