School segregation in urban context: Socio-spatial dynamics and educational inequalities


  • School segregation is both a manifestation and cause of educational inequality and thus one of the key challenges in education.
  • School and residential segregation are tightly interlinked whereby the former is usually higher than the latter.
  • The relationship between residential and school segregation is highly contextual. It strongly depends on the specific educational landscape consisting of both local residential patterns and the educational system. 
  • However, one common mechanism across the globe is that free parental school choice is directly associated with higher levels of segregation of pupils from different socioeconomic and ethnic backgrounds between schools. 


In many cities, richer and poorer people live increasingly less often in the same neighbourhoods (Musterd et al. 2017; Tammaru 2017). This development – referred to as segregation – has severe effects on schools. Since the specific demographics of cities are key ingredients in the mix of school populations, the growing separation on the neighbourhood level is tightly linked to a growing polarisation between schools (Bernelius & Vilkama 2019; Boterman 2019; Butler & Robson 2003). 

School segregation is both a manifestation and cause of educational inequality and thus one of the key challenges in education (Musset 2012). This is related to two main dimensions: First, schools are places for learning, in which the specific composition of school classes can have an effect on pupils’ performances that exceeds the effects of pupils’ individual characteristics (van Ewijk & Sleegers 2010). Second, schools are places of social interaction, where children can learn to deal with social and ethnic diversity (Wilson, H. 2013). 

Schools can thus promote class-crossing networks and social capital (Hanhörster & Weck 2020; Small 2009) or contribute to an increase in segregation along social and ethnic lines for both children and parents (Ramos Lobato 2019). A disconnection of children’s social realities at this early age can thus be a risk for pupil’s educational outcomes and for the social integration and cohesion in cities in general. Therefore, the question arises how cities can react to the growing inequalities in both neighbourhoods and schools. 

Since the relationship between residential and school segregation is crucial to understanding intergenerational social mobility and inequality (Boterman et al. 2019), the topic is of high educational and socio-political relevance. Nevertheless, in-depth knowledge on this relationship is still limited.

School segregation and its effects 

The term “school segregation” refers to the unequal distribution of children of different social and ethnic backgrounds across schools. The topic is of high relevance since the increasingly polarised composition of schools can affect pupils’ performances once individual and family effects have been controlled for (Alegre & Ferrer 2010; Sykes & Kuyper 2013). This can be ascribed to teachers’ adjustment of expectations, classroom dynamics or parents’ resources invested to improve the schools’ conditions (Lupton 2004; Nast 2020). 

Moreover, in some cases, the socially most disadvantaged schools are characterised by the most unfavourable conditions (e.g. in terms of a shortage of teachers or the proportion of cancelled lessons) (Helbig & Nikolai 2019). 
Consequently, school segregation can produce different conditions of learning that reproduce unequal educational outcomes, but does not necessarily do so.

Searching for the ‘right’ school: Parents’ preferences and their effects on school segregation

The expansion of education has led to growing competition for access to universities and prestigious jobs during the last few decades. Education has thus become a sensitive topic and parents are increasingly seen as key players in guaranteeing their children’s educational success (Krüger et al. 2020; Wilson, D. & Bridge 2019), which starts by enabling access to the ‘right’ kindergarten and primary school (Ramos Lobato 2019). 

Since a school’s composition is often directly associated with its educational quality, the definition of the ‘right’ school is increasingly dependent on its social and ethnic composition, which often feeds into growing school polarisation (Butler & Hamnett 2007; Karsten et al. 2003; Wilson, D. & Bridge 2019). Many parents are concerned about their children’s exposure to lower standards of education, to children with inadequate language skills or to the ‘wrong’ types of socialisation (Boterman 2013; Vowden 2012). 

Figure 1_School choice according to parents’ educational attainment

Figure 1. illustrates the selectivity of parents’ school choices after the abolition of primary school catchment areas in the city of Mülheim an der Ruhr, Germany. Only one-third of the higher-educated middle-class parents enrolled their children in the nearest primary school when it had a high proportion of children of benefit recipients / with a migrant background. In contrast, more than two-thirds did when these proportions were low. Thus, middle-class parents still prefer the nearest primary school, but only if it has the ‘right’ composition (Ramos Lobato & Groos 2019). 




While it is mainly higher-educated parents who find ways to access the ‘right’ schools, working-class parents are often assumed to be less strategic, ambitious, and less able to make the ‘best’ choices for their children (Reay & Ball 1997; Van Zanten 2005). However, working-class parents’ more frequent choice of the local school is not solely the result of lacking information and aspiration. It is also driven by practical and economic constraints and by parents’ concerns about their children or themselves being singled out or excluded at schools with a higher socio-economic composition (Byrne 2006; Reay 2001; van Zanten 2013). Thus, what is often interpreted as apathy and fatalism can also be attributed to structural obstacles and be redefined as active avoidance to select the most ‘elitist’ schools (Ramos Lobato 2019).

Interestingly, even some middle-class parents deliberately opt out of the mainstream. Driven by their appreciation of diversity and a progressive political ideology, these parents enrol their children in socially and ethnically mixed state schools. While most studies show that the compatibility of parents’ social ideals and individual concerns about their own children’s future position often produces tensions and dilemmas (Billingham & McDonough Kimelberg 2013; Bloomfield Cucchiara 2013; Frank & Weck 2018), a comparative analysis from Finland and Germany draws a more differentiated picture. As both countries’ educational systems enable students to achieve the highest levels of education without attending any ‘elite’ institutions, they thus ‘allow’ middle-class parents to consider additional choice criteria, such as wellbeing or spatial proximity (Ramos Lobato et al. 2018).

But how do parents get access to their favourite schools? Numerous studies from different national and local contexts have illustrated the variety of strategies parents use, such as moving to the school’s catchment area, going private, ‘colonising’ local schools or circumventing (mainly illegally) allocation regulations (Butler & Hamnett 2007). Another, more subtle way, is to attend special classes within comprehensive schools (Kosunen 2014; Nast & Blokland 2014). Parents’ choice strategies thus strongly depend on the specific educational landscape they operate in. This landscape comprises two key dimensions: the institutional and spatial contexts, as well as the complex ways both are interrelated, which is discussed in the following sections.

Educational landscapes: How urban structures and educational systems affect parents’ choices and school segregation 

Polarised cities and polarised schools? The role of residential patterns for school segregation 

Research across many countries illustrates that residential patterns are central for understanding school segregation (Bernelius & Vaattovaara 2016; Boterman 2019; Schindler Rangvid 2007). This relationship is mutual. On the one hand, residential patterns affect schools’ compositions. Where children live still largely determines where they go to school. As a consequence, the increasing residential segregation across many European cities is clearly reflected in local school environments. For instance in Helsinki, the Finnish capital, differences between school catchment areas have grown noticeably since 1990, despite generally increasing income and educational levels. The areas’ ranking position (measured by income, education and ethnic background), however, has stayed quite stable (see Figure 2) (Bernelius & Vilkama 2019).

Figure 2: Spatial differences between the school catchment areas in 1995–2011: (a) average income; (b) proportions of highly educated; and (c) proportions of residents speaking languages other than Finnish or Swedish as mother tongue (Bernelius & Vilkama 2019)

Figure 2_Spatial differences between the school catchment areas in 1995–2011


The educational system’s role for school segregation  

School segregation strongly depends on both the extent to which families are allowed to choose schools and the options they have. As regards the former, educational policies across various countries have strengthened parental school choice and open competition among schools during the last two decades (James et al. 2010; Logan et al. 2008; Söderström & Uusitalo 2010). While few countries have a completely open choice system (e.g. the Netherlands), free choice is often conditional on specific requirements (e.g. Greece) or constrained by priority criteria still privileging residential proximity (e.g. Spain, Italy, Germany or Denmark) (Boterman et al. 2019). 

The options available depend on the public–private mix, and the schools’ differentiation in terms of quality, tuition fees, and profiles. In countries with a public school system (e.g. Finland), the differentiation is quite small. In countries with private schools (often managed by religious institutions) that have gained a certain amount of autonomy (e.g. France, Italy and Spain) or schools based on different pedagogical profiles such as Steiner (e.g. the US, the Netherlands or Germany) there are considerably more options (Boterman et al. 2019). If these schools are real options, however, depends on their affordability and thus on the question whether private schools are publicly funded.  

Apart from providing the institutional framework, educational policies can have an additionally powerful effect by shaping parental discourses on norms and values of schooling (Noreisch 2007; Raveaud & Van Zanten 2007). In the federal state of North-Rhine Westphalia/Germany, predominantly middle-class parents seem to interpret the recently introduced free primary school choice as a clear request to choose rather than as a simple option. The reform triggers their concerns about growing quality differences between primary schools. Consequently, even though some parents 
perceive the increasing hype about primary school choice as exaggerated, the risk of making the ‘wrong’ decision by ‘just’ accepting the nearest school is perceived as being too high to avoid that hype (Ramos Lobato & Groos 2019).

Complex links but similar patterns between neighbourhoods and schools

The way residential and school segregation interact depends on both structural and institutional factors as well as the specific urban configurations. According to Boterman et al. (2019), three broader types of urban settings can be identified that show at least some similar patterns as regards the relationship between residential and school segregation. 

  • Southern European cities are characterised by relatively high levels of social polarisation (measured by the Gini index; OECD.Stat, 2018) but for various reasons, they tend to be less segregated in space. Nevertheless, the specific structure, organisation and distribution of school supply leads to quite high levels of school segregation (Bonal et al. 2019; Cordini et al. 2019). 
  • Continental and Northern European urban contexts, in contrast, are less socially polarised but more segregated in residential and even more in terms of schools (Nielsen & Andersen 2019; Oberti & Savina 2019; Ramos Lobato & Groos 2019).
  • In the UK and US, levels of social polarisation, residential and school segregation are distinctively higher than in European cities. While in the US, the combination of free school choice and gentrification have slightly decoupled school and residential segregation patterns within the last few years (Candipan 2019), the opposite seems to have been the case in the UK. At least in London, free choice has led to a situation in which distance to school has become the major criterion as to whether parents achieve their preferred school (Hamnett & Butler 2013).

Thus, although market- and performance-oriented approaches in education have been implemented throughout most countries in Europe and the US, the mechanisms behind the growing levels of school segregation and its relationship with residential polarisation vary. However, one mechanism is common in all countries: free parental school choice has a significant impact on school segregation. Although being advertised for its positive effects on educational equality by breaking the geographical link between pupils and schools, free parental school choice is directly associated with higher levels of segregation of pupils from different socioeconomic and ethnic backgrounds between schools. 

This finding is strikingly consistent (Wilson, D. & Bridge 2019):

  • across all types of choice mechanisms
  • in different countries and cities
  • across choice systems that have been in place for different lengths of time
  • and it holds across a range of sociodemographic characteristics against which segregation is measured (e.g. socioeconomic status, ethnicity and faith). 

Conclusions and needs for further research 

In many countries of the Global North, social inequality is on the rise. This has wide-ranging effects for cities on both neighbourhoods and schools. Socio-spatial differentiations in cities are strongly linked to segmentation in the educational system whereby children with different socio-economic and ethnic backgrounds are usually distinctively more separated at school than in neighbourhoods. The underlying reasons are highly contextual and depend strongly on the relationship between the specific institutional and spatial local contexts. However, in most contexts, school segregation does not only reflect existing socio-spatial inequalities but also contributes to their perpetuation. Their close relationship thus has become a crucial aspect in explaining the reproduction of social 

Therefore, the question arises of how urban education policies can or should react to the increasing levels of inequality and segregation. What does it take to develop effective equality strategies and what elements are needed so that these policies are also supported and viewed as justified by parents? Neither parental school choice nor other strategies, such as busing or the implementation of charter schools in disadvantaged neighbourhoods, turned out to be suitable policy instruments for achieving the integration of pupils across schools by socioeconomic status, ethnicity or faith (Wilson, D. & Bridge 2019). Desegregating catchment areas by carefully redrawing their boundaries might be a valuable anti-segregation policy; however, it has not modelled at a larger scale in Europe yet (Bernelius & Vilkama 2019). 

Besides the previous, unsuccessful attempts to balance the composition of schools, trying to balance the learning conditions between schools has become popular. In some contexts, targeted funding for those schools that face difficulties in coping with their context-related disadvantages, and compensating for them, has therefore been implemented (e.g. in The Netherlands, some federal states in Germany or cities in Finland). First quantitative studies demonstrate that these ‘positive discrimination policies’ can be remarkably effective in improving the educational outcomes of immigrant and low-performing native students (Silliman 2017). However, deeper insights into their qualitative effects inside schools and their embeddedness into wider strategies of the affected schools would help to understand the mechanisms behind in order to improve the targeted funding. Therefore, there still is a need for further investigation.  

Prospects for future research in the Finnish context

But what does that all mean for Finland that has for years been one of the top-ranking countries in both educational outcomes and educational equality? The Finnish welfare state and its educational system still are comparatively equal. However, recent studies have revealed growing differentiations throughout society and urban space (Bernelius & Vaattovaara 2016; Bernelius & Vilkama 2019; Vaattovaara et al. 2018) and pointed to the educational system’s “vulnerability to the segregation operating in neighbourhoods” (Bernelius & Vaattovaara 2016). Albeit still on a different level, the Finnish welfare and educational systems thus seem to face quite similar challenges as most OECD countries. 

Both politics and research should therefore be highly aware of the ongoing international debates and research gaps, in particular the following ones:

1. Increasing levels of school segregation

One already widely discussed topic is the severe impact of parents’ selective school choice practices on school segregation. Selective school choice is not really associated with the comprehensive and egalitarian Finnish education system where the institutional quality differences as drivers of segregation are rather minimised. However, based on quite persistent socio-economic drivers of school choice, segregation between school catchment areas and the schools’ educational outcomes has been increasing in the Finnish context as well (Bernelius & Vaattovaara 2016). Here, classes with special emphasis have become the main mode of school choice and act as a vehicle for selection and distinction – at least in urban areas. Since active school choice is not only growing but is even getting more and more socially accepted in Finland (Bernelius & Vilkama 2019; Kosunen 2014; Kosunen et al. 2016), it might become a more severe topic in the next years and thus needs to be observed carefully through continuous research. 

2. Ethnic diversity and school segregation

Besides the socio-economic drivers of school segregation, first studies have raised awareness for the tight interlinkage of parents’ social and ethnic background for school choice. They have illustrated how principals’ and teachers’ selective recommendations and allocation strategies but also parents’ experience with stigmatisation and stereotypes can affect their choice strategies (Vincent et al. 2012). In light of the growing levels of migration and the persistent educational inequalities between 
immigrant and non-immigrant children (OECD 2015), this perspective is becoming increasingly pressing – but is still missing in the Finnish context. Although Finland has a comparatively short history of international migration, the growing differentiation and stratification between so-called ‘international’ and ‘multicultural’ schools – at least on a symbolic level – clearly illustrates how relevant this topic has become for Finland as well and thus points to the urgent need of further research.

3. Institutional patterns and causes of segregation

Parents’ school choice strategies are highly contextual and thus strongly adapted to the institutional context they operate in. The ongoing implementation of market and performance-oriented approaches in education in many countries results in a growing pressure on schools to become more effective and successful. First studies illustrate that by giving (selective) recommendations or by advertising or canvassing their schools to attract specific groups of children and parents, principals and teachers pursue different strategies to react to the increasing competition in the educational market (Ball & Maroy 2009; Jennings 2010; Ramos Lobato 2017; van Zanten 2013). 

The increasing diversification of schools and the schools’ autonomy as regards the admission process are thus crucial but still less examined dimensions of schools segregation. Since even in Finland schools select a certain proportion of pupils in order to control the enrolment in selective classes (Seppänen & Kosunen 2015), paying attention to the institutional patterns and causes of segregation is a relevant subject for both national and international research.

4. Schools as settings for local social integration

Finally but importantly, schools are not only places for learning but also for social interaction. 
Thus, the emergence of social divisions and the disconnection of social realities in schools affect both educational outcomes and can endanger urban social cohesion. Since school and neighbourhood contexts often seem to work simultaneously to produce unequal outcomes for children (Maloutas et al. 2019; Oberti & Savina 2019), welfare measures aimed at combatting segregation need to focus on both neighbourhood and school contexts. This is relevant for the Finnish context as well, where despite the universally high quality of institutions, the development trajectories of both schools and neighbourhoods are quite sticky (Bernelius & Vilkama 2019). Although the link between neighbourhood and school context is well-demonstrated, the underlying mechanisms explaining how different neighbourhood contexts shape schools and, the other way around, how schools impact the surrounding neighbourhoods are still unclear. 

First studies demonstrate how school and neighbourhood compositions, symbolic meanings of urban areas, and neighbourhood-based policy interventions structure schools and reproduce urban inequality (Nast 2020). Reputation and stigma might thus be relevant mediators of the interconnectedness between school and neighbourhood segregation. That means that even though the diversification of educational provision in Finland seems to be mainly symbolic, it might still have severe impacts on the growing differentiation between neighbourhoods – and vice versa. Understanding schools as important social settings in a neighbourhood and linking them to other forms of local social support might thus have positive effects on both pupils’ learning outcomes and their (local) social integration. Gaining a better understanding of the link between schools and neighbourhoods is thus urgently needed for both urban and educational policies. 

Isabel Ramos Lobato

Postdoctoral researcher
Urban Geography 
City Collaboration