In many countries school choice policies have been viewed as a means to desegregate schools by removing the immediate link between home address and school allocation. The central argument has been that a policy of free school choice will diminish the impact of socio-spatial segregation on schools by allowing pupils to select schools in different locations, as well as through encouraging competition between schools. However, numerous studies have demonstrated that free school choice has usually led to increasing social and ethnic segregation between schools,1 as highly educated parents are more equipped to navigate the field of choices. The interconnections between residential and school segregation may even increase through the removal of geographical catchment areas.2

In this chapter, we analyse this interdependency in urban Finland: What is the relationship between school choices, school segregation and residential segregation in the Finnish context? Is the universalist, egalitarian Finnish system able to counteract the international trend of growing segregation between schools, as the academic quality of schools is very high across the board? (See also Kalalahti and Varjo in this book on the change in universalism.) We focus our analysis on the Helsinki metropolitan region, where our research using quantitative and qualitative datasets spans more than two decades. We draw on our earlier findings about vicious circles of segregation3 and add here a new, comprehensive analysis and interpretation of the different patterns interconnecting parental choice with urban segregation and school segregation.

We start by describing the general structure of neighbourhood and school segregation in the metropolitan area of Helsinki, its interconnections to student flows and families’ residential mobility—or choices of neighbourhood—and the way these connect with learning outcomes in schools. These intertwined domains of segregation form the macro-level circle of segregation in education and urban neighbourhoods4 and we mainly draw on statistical and register based studies to look at this macro level. Second, we draw on qualitative research to describe the relations between the families’ individual processes of school choice, the school reputations in the local sphere and their interconnection to school segregation. This micro-level analysis focusses on processes of parental choice, and the importance of rumour and ‘the grapevine’5 as well as the emerging local socio-economic differentiation of classes (soft streaming) within schools.

We conclude by drawing together what is known about school choices and segregation in urban Finland in recent years and also note some blind spots in research thus far. This chapter also links strongly to the chapter by Seppänen, Pasu, and Kosunen in this book which discusses institutional-level policies through which municipalities and schools conduct pupil selection and enrol pupils into schools either through local school allocation or through aptitude-tests to selective classes with a special emphasis. The chapter by Ramos Lobato and Bernelius builds further on the topics addressed in this chapter by considering the resource allocation policies which have been set up as a response to the growing challenges faced in disadvantaged schools and the risk of vicious circles of educational segregation we present.

Macro-level Patterns of Segregation

Compared internationally, Finland has become widely known for its relatively low between-school variation in student composition and educational outcomes. The egalitarian ethos and public provision of education are mirrored in the school network, which is mostly comprised of public schools with high academic quality, and the number and relative share of private schools is low even in the largest cities. In the first PISA assessments in 2000 Finland stood out for its remarkably high attainment not only overall, but particularly in the lowest deciles, and in 2003 the poorest quartile of learners in Finnish schools still outscored the respective groups in other OECD countries by the equivalent of more than 1.5 years of education.6 In the first PISA assessment in 2000, the outcome score difference between the lowest and highest scoring school deciles was also very low even compared to other Nordic countries, and less than half of the OECD average.7

The high level of educational equality appears to have had a strong link to the social and spatial structures in the country. At the time of the introduction of the universal basic education (peruskoulu) in the 1970s, the socially equalising policies of the welfare state also began to reach their peak. Socio-economic gaps were moderate and shrinking throughout the society, and the welfare state policies resulted in particularly small differences in household disposable income. To complement this, several municipalities introduced policies of spatial social mix. For example, Helsinki implemented a strong policy approach whereby all neighbourhoods were set a target level of owner-occupied and social housing. Socio-spatial segregation diminished throughout the following decades, and in the beginning of the 1990s, Helsinki demonstrated the most equal pattern of social mix in its recorded history.8 These developments were reflected in the education system, where both school segregation and the effect of home background on pupil attainment decreased.

As school choice policies were introduced in Finland in the 1990s, the assumption was that increased mobility between schools would encourage the availability of specialised subjects to all pupils to complement the core curriculum.9 Our research suggests that educational equality in Finland has been affected by increasing socio-spatial segregation, and that Finnish school choice is associated with growing school segregation through middle-class choice patterns.

Pupil Flows and Growing School Segregation

The long-time trend towards diminishing segregation in Finnish cities during the latter half of the twentieth century was reversed in the recession of the early 1990s. Unemployment grew rapidly in the neighbourhoods which had initially had lowest levels of highly educated adults, and the growth of socio-spatial segregation was mirrored in growing socio-economic school segregation through the connection between pupils’ residential addresses and school allocation (see Ramos Lobato and Bernelius in this book). At the end of the 1990s differences in educational achievement between schools in the largest urban areas were discovered to be relatively significant as the schools with both poorest and highest educational outcomes were found in the capital region in the first national outcome assessment in 1998.10 While these differences between schools were probably not entirely due to the recession, and reflected some educational gaps between neighbourhoods which the pre-1990s equality policies had not been able to entirely close, the role of growing neighbourhood segregation resulted in further widened gaps between schools in the following decades. On the one hand, this demonstrates the strong relationship between urban socio-spatial segregation and school segregation. On the other, it reflects the additional effects of school choices on school segregation.

The relationship between school segregation and school choice is a two-way process, where both phenomena feed into the other. First, mobility towards schools outside one’s own residential area is strongly linked to existing school segregation in the Finnish context. An analysis of pupil flows between schools in Helsinki showed that the flows are systematically and selectively directed towards schools with higher socio-economic status (SES) and better educational outcomes than the pupils’ own catchment area school.11 This finding was consistent in a macro-level analysis where all schools are included, without consideration of local choice patterns between neighbouring schools, or the availability of selective classes. In other words, school segregation, or the SES of the school student body, is strongly linked to the pupils’ choice patterns in Finnish cities, as elsewhere.12

Second, school choices affect the level of school segregation. The families who are particularly active in the school market are, on average, somewhat more educated,13 and the pupils achieve higher educational outcomes than those families and pupils who opt for catchment-area-based classes in their own catchment area school,14 even if there is of course local variation in this pattern. The link between family background and pupil outcomes has also strengthened alongside growing societal segregation. In the first PISA assessment, the relationship between SES background and educational outcomes was clearly below the OECD average for Finnish pupils, but during the last two decades, the statistical effect of home background has increased relative to the OECD average.15 The selective profile of ‘active choosers’ means that the pupil flows carry with them higher SES characteristcs and higher educational outcomes to schools the flows are directed to, and away from the schools which are rejected in these patterns of choice.

Combined with the macro-level choice patterns of pupil flows from more disadvantaged schools towards schools with higher SES levels, the selectivity in the SES and outcome profile of the pupils making up these flows has led to the growth of school segregation. The effect is felt at both ends of the scale: both in the schools losing catchment-area pupils to other schools, as well as the schools receiving pupils from other catchment areas. The effect of these choices has also been quantitatively demonstrated by comparing the actual educational outcomes in all lower-secondary schools in Helsinki to a hypothetical scenario where all pupils have been artificially reallocated to their own catchment area school. The analysis clearly demonstrated that when the real-life school choices were introduced to the scenario, educational outcomes fell in the disadvantaged schools losing catchment-area pupils, and rose in the popular schools. These changes were statistically significant.16

Residential Mobility and Schools as Drivers of Neighbourhood Segregation

The macro-level processes of school choice are structured around residential spatial mobility in cities. Internationally, schools are known to be important motivators in families’ choice of neighbourhood across many urban and national contexts. Families with higher SES are especially active in looking for schools which they expect to most benefit their children.17 In the Finnish context, schools are typically also mentioned in housing preference studies, where parents often note that finding a neighbourhood that is good for children is one of the most important considerations when choosing where to live, and schools are seen as an important ‘part of the package’.18 The importance of schools has been picked up by many Finnish real estate agents, who often include “a good school” in real estate descriptions alongside information about other neighbourhood amenities.

The importance of schools as motivators of housing choice can also be observed in the effect that schools can have on housing prices especially in contexts where catchment area boundaries directly mediate access to a certain school. In one UK study the estimated effect of a popular school was as much as 34% of real estate prices in the catchment areas with most popular schools, when other factors were accounted for.19 A similar study in Helsinki showed corresponding effects in the Finnish context, although the price effect was more modest: one standard deviation increase in school outcomes was associated with a 3% increase in housing prices within the catchment area.20

The ability and tendencies of families to navigate school markets are both linked to questions of class or SES across different national contexts, and this is also the case in Finland. Higher status families’ residential choices appear to be strongly connected to perceived socio-economic and ethnic differences in schools and neighbourhoods, especially internationally, leading to choices correlating rather more with pupil composition than any measurable academic qualities of schools.21 School segregation therefore appears to be one of the key drivers of school-related residential decisions.

Somewhat counterintuitively, the socio-economic differences as motivators of school choice may perhaps be even more pronounced in Finland, compared to countries with a stronger level of differentiation in the academic quality of schools. In many countries, the landscape of school quality is highly differentiated due to lack of regulation, divisions between private and public education, or strong dependency on local revenues. In the Finnish context, however, the institutional quality of schools, measured for instance by teacher qualifications, academic curriculum or school resources, is very consistent, especially within the same city or municipality. As a result, a large part of the perceived differences and reputations of schools are constructed around the social composition of schools.22 Thus, the comparatively high institutional stability of Finnish schools might even increase the relative importance of the schools’ pupil composition as a factor influencing the way schools are perceived and the choice of schools that parents make.23

Neighbourhood segregation and school segregation have a similarly reciprocal relationship as school segregation and school choices described above; each affects the other. There are several international studies documenting the interconnections, where the socio-economic structure of the catchment areas affect the schools’ student base and educational outcomes, which then have a further effect on the residential choices made by families with children.24 Finnish studies analysing the links between school catchment area segregation and residential mobility have also found that school-related mobility patterns are not only related to the initial levels of segregation in the school catchment areas, but also considerably exacerbate the level of segregation between these areas.25

In the capital region of Finland, school catchment area segregation has grown considerably during the last two decades. The level of ethnic segregation has grown particularly noticeably in the school catchment areas in Helsinki and the neighbourhoods surrounding schools in Espoo. There is also a strong path dependency in these developments, where initial levels of socio-economic disadvantage or share of ethnic minorities are strong statistical predictors of future developments in the area.

When compared to adults or the population as a whole, children and youngsters are even more segregated between school catchment areas. Figure 11.1 depicts segregation indices (index of dissimilarity) in Helsinki for children under 16, and the population at large. The index describes the share of the compared groups which should theoretically relocate in order to achieve a complete mix in all areas. An index value of 0 would mean a completely mixed spatial distribution with no segregation between the compared groups, whereas the value 1 represents a situation of complete segregation, where all of the individuals in one group would need to relocate in order to achieve a mix between the compared groups. This index value is approximately 10% points larger for children than for the overall population. For example, in order to achieve a spatial mix of children living in high- and low-income households, well over 50% of children in either group should move into other neighbourhoods, while this figure is just over 40% for all population. In practice this means that children are living even more separately than adults, and it has been interpreted as a signal of (white middle-class) families with children being particularly selective in their residential decisions, compared to childless households. The differences in segregation levels between children and adults are also statistically highly significant (p < 0.001).26

Fig. 11.1
A scatter plot shows the index of dissimilarity versus years from 1995 to 2015, with Finnish children versus immigrant children, the majority population versus immigrants, highly educated versus low educated, and children and people in the lowest versus highest income declines plotted with an increased trend.

Segregation indices (index of dissimilarity) for children and total population in school catchment areas in Helsinki 1995–201528

Full size image

Deeper socio-spatial segregation of children and youngsters is further reflected in schools, which as a result of this age-related demographic difference become more segregated than neighbourhoods as a whole, when school allocation is based on residential address. It is empirically difficult to pinpoint the schools’ exact role, or magnitude of the school-related effect, in the growing levels of urban segregation. However, analysing residential patterns and real estate prices close to school catchment area borders shows that even when all other neighbourhood characteristics are similar, access to a particular school does affect the socio-economic and ethnic composition and housing prices.27 In this process, residential patterns and urban segregation are structured by schools and their catchment areas.

The macro-level patterns of urban school choices and residential choices together form a vicious circle of educational segregation. First, residential segregation affects the initial levels of school segregation through the address-based school allocation policies. As children live even more separated than adults, the degree of school segregation would exceed neighbourhood segregation for the total population even in if all pupils would attend their own neighbourhood school. Second, the school segregation patterns are linked to the school choices of families with higher SES, leading to further segregation of schools. As the pupil flows are also implicitly selected by pupils’ educational outcomes,29 the process of school segregation through choices differentiates the schools’ educational outcomes when compared to initial differences between schools. The differences are most visible through choices of emphasised teaching (see Seppänen, Pasu and Kosunen in this book), as well as language choices.30 These processes, in turn, appear to motivate the residential decisions of many families, which then further shape the socio-economic and ethnic landscape of school catchment areas and increase the segregation of school-aged children. All of this feeds a vicious circle (Fig. 11.2).

Fig. 11.2
A process diagram depicts school choice, which incorporates school segregation, educational outcomes and reputation, neighborhood selection, and community segregation based on socioeconomic and ethnic factors.

The vicious circle of school and residential segregation31

Full size image

Micro-level Patterns of Segregation

The micro-level processes involved in vicious circles in education are mediated by individual parental choice strategies and local hierarchies between schools and neighbourhoods. Parental choice of schools is a heavily investigated area in Finland. What is generally known is that school choices are conducted successfully by many kinds of families, but primarily by families from higher social classes.32 This is in line with international research33 and with the quantitative aspects of student flows and macro-level choice patterns described above. Parental choices are traditionally constructed in the discussions between members of a family as well as their friends and acquaintances. The earliest studies on parental choice indicated that physical proximity to a school and existing relationships to friends from primary school were key determinants of school choice.34 However in 2010, depending on the municipality in question, some 11–39% of the pupils in lower secondary schools in the largest cities opted out of their local school allocation by applying to selective classes with emphasised teaching.35 In Helsinki, the share of students choosing a school outside their own catchment area has typically been 15–20%.36

A central feature of an educational system with no official school rankings of comprehensive schools (and a minor private sector) is that the reputations of public schools’ flow ‘through the grapevine’ of local discussions amongst parents. The discussions usually concern nearby schools, and thereby the analytically interesting unit are the closest schools, their mutual relations in a social hierarchy as well as their provision of selective classes. While macro-level processes of school choice can be analysed across the city at large, local hierarchies in school choices are constructed on a smaller geographical scale and on relative differences between schools.37 In these local hierarchies even a school or neighbourhood which is qualitatively close to the city average—for example in terms of SES—may become rejected as a choice, if it is compared to more elite concentrations nearby.38 Individual and local processes form micro-level circles of segregation operating locally, and also feed into macro-level processes that are creating segregation in education.

In parental discussions,39 the reputations of schools varied at the level of the city, by selective classes, and by the local catchment area, which in many of the Finnish cases comprises several schools that may be appointed as the ‘local school’. Local discussions usually described reputations and their hierarchical relationships in relation to general classes, even if hierarchies were constructed in relation to selective classes:

Well, these are rumours of course, but, for example, in [this area] and nearby, the schools, more than one of them, are considered good. They have classes with a special emphasis like [with an emphasis on theoretical subjects], one at least in [School 3] and another one here [School 4], and then there are several classes with an emphasis [on art], so I think they are all kind of good at least according to their reputation, and you need high average grades in order to be accepted. Eva, middle class (public sector), son in the general class of the neighbourhood school.

Some local areas have very strong hierarchies of reputation amongst their comprehensive schools. Upper social classes often seem aware of these reputations and actively make school choices away from certain schools and taught classes towards other schools and classes. When discussing the popularity of schools and classes, both push-factors and pull-factors therefore need to be acknowledged.40 Often the logic of action seems to be that unfavourable general classes “push” families into choosing either other schools or selective classes in the local school or some other school. Local concentrations of disadvantage appear to be a particularly important consideration in “push” factors. In the Finnish literature this phenomenon has previously been called the “rejected-school phenomenon”.41 On the other hand, some selective classes with a good reputation—not necessarily the most competitive or demanding—seem to “pull” pupils even from other areas of the city, which also often means longer school journeys and increasing transport costs for the families concerned:

And now I know I’m being selfish, but it is totally clear that I would not have my children in the [general class of the local neighbourhood school], so if only you can, you will go for some other option. Andreas, upper social class (private sector), son in a class with a special emphasis.

This [school]—does not have an especially flattering reputation, and in that situation [being allocated to a general class in the school] we would have seriously considered applying to some other school. Leo, upper social class (private sector), daughter in the general class of the neighbourhood school.

Choices may be towards selective classes, as in the case of Andreas’ family, or to the general classes in particular schools, as in the case of Leo. The higher social classes seem to want and be able to avoid the schools with ‘the worst’ reputations through these choosing strategies, which reflects a well-researched division between skilled, semi-skilled and disconnected choosers.42 Families with more social, cultural and economic capital are more able and often more willing to exercise their choice of schools. On the other hand, families with fewer resources might be as willing but less able to make successful choices with their children.

Reputation-wise, the ‘worst’ classes in schools are considered to have problems with risky behaviour such as using alcohol, cigarettes and drugs, as well as classroom study conditions amongst pupils in terms of peaceful conduct (misbehaviour during lessons, bullying).43 These directly relate the risk of school rejection to questions of socio-spatial segregation and concentrations of disadvantage in neighbourhoods and in schools and school classes. There are also other concerns, such as problems with school buildings and their (microbe-related) quality of air, which has been a noticeable topic of public discussion in Finland in recent years.44 ‘Problem’ schools tend to be avoided, even if the information about them consists mainly of rumour and perception in the neighbourhood:

I don’t know so much about rumours, or I actually don’t even care about them. But of course, if it is generally known that in a certain school there are significant problems with order, or problems with intoxicants or other. And if I’m able to exclude that school from our options, it is crystal clear that I wouldn’t put my own child into that sort of an environment. Tomas, upper social class, son applying for a place in a class with a special emphasis.

Another group of classes to be avoided, according to parents, were those with the most elite reputations:

I said then that I wouldn’t let my child go to [the most prestigious class with a special emphasis focusing on the same theoretical subject], that it was a too competition-oriented environment, so this is a good solution. Lena, middle class (private sector), son in a class with a special emphasis in the neighbourhood school.

Parents described how some classes were considered very competitive and stressful for children as young as 13 (which in the Finnish context is thought to be far too early). This was also a central reason why some urban parents, even if aware of rumours and social hierarchies concerning schools across the city, were only considering nearby schools as relevant options for their own children. Fierce competition in comprehensive schools was considered irrelevant by many, including those from the higher social classes.45 At the same time, many parents, and increasingly over time, seem to be willing to make a selective choice. This influences neighbourhood schools and their locally allocated classes, as the children attending selective paths are taken out of the ‘ordinary’ groups. The wish for a ‘good-enough school’ was present in the parental discourse:

This is the interesting thing: the specialisation seems to create an A- and B-class division among the pupils … The problem is how to ensure a good education in these general classes as well. It can’t happen that the most enthusiastic and dedicated teachers teach only the classes with a special emphasis, and then the less eager ones cover the general classes: No. Everybody should have the right to as good an education as possible. Lisa, upper middle class (public sector), son applying for a place in a class with a special emphasis.

Selective choices, even if based on the reputations of the schools in the local area, are noted to be creating social class divisions in addition to the circle of urban segregation:

I checked with [my son’s] friends, who were the ones who applied to [a selective class]. Every single one of them came from engineering families. So, is it really the case that this engineering talent for some reason is concentrated in families where the dad works at an IT-enterprise? [laughs] That makes me laugh. Sebastian, upper-middle class, son applying for a place in a class with a special emphasis.

These notions of using pupil selection as a tool for creating social distinction within a public education system are the patterns that also contribute to increasing between- and within-school segregation. It has been noted that the choice of selective classes, the choice to study languages over many years starting in primary school, and the combination of the two, function as the relevant choosing strategies.46 Families whose children have chosen both a lengthy study of a language other than English or in addition to English (which more than 97% of the age cohort study nationally) and a class with a special emphasis, come far more often from housing blocks with highly educated adults and higher average annual incomes per household than those with local school allocation to general class and having English as their first or only foreign language.

In these ways the micro-level circles of segregation feed themselves by the same mechanisms observed in the macro-level: the school choices are informed by the reputations of schools and their classes, which on their part are influenced by the socio-economic structures—or assumptions about these—associated with particular schools. The socio-economic structures are dependent on school enrolments and pupil selection conducted by the city and schools and applied for by families. Since school choice possibilities are applied for predominantly by families from higher social classes, their micro-level logic of choice feeds into growing segregation within and between schools. The individual choices and local processes of segregation in turn feed into the larger-scale processes and macro-level circles of segregation in Finnish education.

Conclusion: Segregation Reshaping the Universalist Landscape of Education

Although school choices were introduced in the egalitarian Finnish context with the expressed hope of equalising access to different educational opportunities for all pupils, the resulting macro- and micro-level choice patterns have a highly selective element within this universalist vision. Geographical analysis shows that shopping for schools is structured not only by a systematic socio-economic selectivity, but also by a spatially systematic selectivity: families with children are spatially even more segregated than other population groups, and residential choices correlate strongly with school catchment area characteristics and the push or pull factors associated with certain local schools. Proximity to popular schools, or the willingness to avoid others, may therefore strongly influence decisions about where to reside, which then contributes both to neighbourhood and school segregation, and school choices add to this effect by flows of students towards schools in wealthier locations.

Processes operating at multiple scales create risks of self-perpetuating vicious circles of segregation, where segregation in schools and neighbourhoods both feed into each other through a complex network of contributing factors and mechanisms. Besides the macro-level patterns of segregation in the cities and their education systems, local hierarchies between neighbouring schools and between school classes may further segregate schools and their individual catchment areas. The micro-level processes may lead to growing segregation even when initial differences are small, as parents compare and navigate the network of schools close to their residential location, and school reputations mediate choices in local school markets.

Our analysis of long-term research evidence demonstrates that in Finland increasing inequality is strongly tied with socio-spatial segregation, which is associated with socially selective school choice. The growth of school segregation and the resulting differences in educational gaps between schools have had a profound effect on the Finnish education system. Nationally, a very large portion of these most disadvantaged schools, as well as of the best-performing schools, are located in urban areas. This highlights the importance of intertwined processes of neighbourhood and school segregation: local processes have wide-reaching consequences for the whole education system.

Our findings unearth multiple mechanisms of growing divides between schools, demonstrating that not even a relatively egalitarian educational system is shielded from circles of segregation which may lead to decreasing equality and risks of educational exclusion. The resulting inequalities are not driven by the educational system as such, but the system appears to lack protections against the internationally well described social and spatial macro- and micro-level processes, which are also operating in Finnish urban schools and neighbourhoods, fuelled mainly by the underlying socio-economic and ethnic differences in neighbourhoods. Our results also demonstrate how some educational policies have had unintended consequences and have contributed to the growing divides. As shown in this chapter, policies allowing a degree of freedom in school choices appear to have provided a pathway to growing segregation between schools.

The presented findings also demonstrate the complexity of the processes connected to educational equality in an egalitarian education system, and open multiple questions on how to support educational equality in the context of growing socio-spatial divisions. One of the key questions for future research is whether growing neighbourhood and pupil segregation can have a further effect on the institutional quality of schools. This may happen, for instance, if the school system is not equipped to efficiently support institutional equality through increasing school resources in disadvantaged communities.

A recent report published by the Finnish Ministry of Education and Culture suggested that the growing divides between schools may lead to teachers’ workload and working environment becoming so strained in the most socio-economically deprived schools that it can increase the risk of staff recruitment and retention problems in those schools.47 The resulting teacher segregation—the difficulties of acquiring and keeping qualified staff, and related increase in staff burnout—is recognised in several countries.48 Expert interviews used in the report warned of early signals of teacher segregation becoming more noticeable in Finnish basic education and early childhood education, but there is currently no reliable research evidence on the scale of this phenomenon. If growing challenges in disadvantaged neighbourhoods indeed lead to problems in recruiting, maintaining and supporting qualified teachers in such schools, the universalist ideals of stable institutional quality throughout all schools may be put at risk in Finland, and fuel more segregation through simultaneous and correlated pupil segregation and institutional decline in the same schools.

Other open questions relate to educational policies aimed at alleviating school segregation and supporting disadvantaged students and schools. As the processes leading to school segregation operate at multiple levels and through a complex network of factors including residential segregation, the Finnish education system faces challenges in tackling the growing inequalities. Solutions such as considering housing policies together with educational policies to reduce residential segregation between school catchment areas have been discussed in Finland, but large-scale national or local initiatives and related research evidence are still limited. One of the pressing questions is the policy of school choice, and the way this is locally implemented and interpreted in many municipalities causing, for instance variety in pupil selection by schools (see Seppänen, Pasu and Kosunen in this book). It is possible that if parental choices and pupil selection of schools were forbidden, school segregation might increase through residential segregation, if the middle- and upper-class desire to choose schools would have to find an outlet in the housing market: school choice by mortgage. However, it remains an open question whether this would happen if it was only pupil selection by urban schools that was abolished. On the other hand, international evidence points to completely unregulated school choice probably leading to even higher levels of school segregation than the levels caused by closed catchment areas without choice of other schools.49 As Finnish educational policies attempt to find new solutions and innovative strategies for supporting equality in the face of complex socio-spatial challenges, the challenge is to understand multiple factors and mechanisms affecting educational (in)equality, and the effects of different policy responses aimed at reducing inequality.

Notes

  1. 1.

    See for instance, Boterman, W.R. 2021. Socio‐spatial strategies of school selection in a free parental choice context. Transactions of the Institute of British Geographers 46(4). https://doi.org/10.1111/tran.12454.

  2. 2.

    Wilson, D., and G. Bridge. 2019. School choice and the city: Geographies of allocation and segregation. Urban Studies 56: 3198–3215.

    Boterman, W., S. Musterd, C. Pacchi, and C. Ranci. 2019. School segregation in contemporary cities: Socio-spatial dynamics, institutional context and urban outcomes. Urban Studies 56: 3055–3073.

  3. 3.

    Bernelius, V., and M. Vaattovaara. 2016. Choice and segregation in the ‘most Egalitarian’ schools: Cumulative decline in urban schools and neighbourhoods of Helsinki, Finland. Urban Studies 53(15) 3155–3171.

  4. 4.

    See also van Ham, M., T. Tammaru, and H.J. Janssen. 2018. A multi-level model of vicious circles of socio-economic segregation. In Divided cities: Understanding intra-urban inequalities, eds. OECD, 135–153. Paris, France: OECD Publishing.

  5. 5.

    See Ball, S.J., and C. Vincent. 1998.’I heard it on the grapevine’: ‘Hot’ knowledge and school choice. British Journal of Sociology of Education 19(3): 377–400.

  6. 6.

    Törnroos, J., and P. Kupari. 2018. Kotitausta näkyy matematiikan oppimistuloksissa. In Osaaminen kestävällä pohjalla—PISA 2003 Suomessa. Jyväskylän yliopiston Koulutuksen tutkimuslaitos. Opetusministeriön ja OECD:n julkaisuja, eds. P. Kupari and J. Välijärvi, 115–127. Jyväskylän yliopisto.

    See also Välijärvi, J., and P. Linnakylä. eds. 2002. Tulevaisuuden osaajat. PISA 2000 Suomessa.

    Jyväskylä: University of Jyväskylä, Institute for Educational Research. 116–117.

    Bernelius and Vaattovaara, op. cit.

  7. 7.

    Välijärvi and Linnakylä, op. cit.

  8. 8.

    Vaattovaara, M. 1998. Pääkaupunkiseudun sosiaalinen erilaistuminen. City of Helsinki Urban Facts Publications. 7.

  9. 9.

    See Seppänen, P. 2003. Patterns of 'public-school markets’ in the Finnish comprehensive school from a comparative perspective. Journal of Education Policy 18(5): 513–531.

  10. 10.

    Jakku-Sihvonen, R., and J. Kuusela. 2002. Mahdollisuuksien koulutuspolitiikan tasa-arvo. Oppimistulosten arvointeja 7. Helsinki: Finnish National Board of Education.

  11. 11.

    Bernelius, V. 2013. Eriytyvät kaupunkikoulut : Helsingin peruskoulujen oppilaspohjan erot , perheiden kouluvalinnat ja oppimistuloksiin liittyvät aluevaikutukset osana kaupungin eriytymiskehitystä , Tutkimuksia 1/2013. Helsinki: City of Helsinki Urban Facts.

    Bernelius and Vaattovaara, op. cit.

  12. 12.

    See also Wilson and Bridge, op. cit.

  13. 13.

    E.g., Kosunen, S. 2016. Families and the social space of school choice in urban Finland. Unigrafia: Helsinki.

  14. 14.

    E.g. Berisha, A.K., and Seppänen, P. 2017. Pupil selection segments urban comprehensive schooling in Finland: Composition of school classes in pupils’ school performance, gender, and ethnicity. Scandinavian Journal of Educational Research 61(2): 240–254.

    Bernelius, op. cit.

  15. 15.

    Salmela-Aro, K., and A.K. Chmielewski. 2019. Socioeconomic inequality and student outcomes in Finnish Schools. In Mahdollisuuksien koulutuspolitiikan tasa–arvo, eds. R. Jakku-Sihvonen and J. Kuusela, 153–168. Singapore: Springer.

    OECD. 2020, Students’ socio-economic status and performance. In PISA 2018 results (volume II): Where all students can succeed, ed. OECD. Paris, France: OECD Publishing. https://doi.org/10.1787/f7986824-en. Accessed 21 Dec 2021.

    Bernelius, V., H. Huilla, and I. Ramos Lobato. 2021. ‘Notorious schools’ in ‘notorious places’? Exploring the connectedness of urban and educational segregation. Social Inclusion 9: 154–165.

  16. 16.

    Bernelius, op. cit.

    Bernelius and Vaattovaara, op. cit.

  17. 17.

    Ball, S.J. 2003. Class strategies and education market. London: Routledge Falmer.

    Byrne, B. 2009. Not just class: Towards an understanding of the whiteness of middle-class schooling choice. Ethnic and Racial Studies 32: 424–441.

    Raveaud, M., and A. van Zanten. 2007. Choosing the local school: Middle class parents’ values and social and ethnic mix in London and Paris. Journal of Education Policy 22: 107–124.

  18. 18.

    Dhalmann, H., M. Vaattovaara, and K. Vilkama. 2013. Hyvää kasvuympäristöä etsimässä. Asuinalueen ja koulun merkitys lapsiperheiden muuttopäätöksille pääkaupunkiseudulla. Yhdyskuntasuunnittelu 51(4): 11–29.

  19. 19.

    Cheshire, P.C., and S.C. Sheppard. 2004. Capitalising the value of free schools: The impact of supply characteristics and uncertainty. The Economic Journal 114: F397–F424.

  20. 20.

    Harjunen, O., M. Kortelainen, and T. Saarimaa. 2018. Best education money can buy? Capitalization of school quality in Finland. CESifo Economic Studies 64: 150–175.

  21. 21.

    Butler, T., and C. Hamnett. 2007. The geography of education: Introduction. Urban Studies 44: 1161–1174.

    Harjunen et al., op. cit.

    Rowe, E.E., and C. Lubienski. 2017. Shopping for schools or shopping for peers: Public schools and catchment area segregation. Journal of Education Policy 32: 340–356.

  22. 22.

    Kosunen, S. 2014. Reputation and parental logics of action in local school choice space in Finland. Journal of Education Policy 29(4): 443–466.

  23. 23.

    See for instance Kauppinen, T.M., M. van Ham, and V. Bernelius. 2010. Understanding the effects of school catchment areas and households with children in ethnic residential segregation. Housing Studies 1–25. https://doi.org/10.1080/02673037.2020.1857707.

  24. 24.

    Andersson, E., J. Östh, and B. Malmberg. 2010. Ethnic segregation and performance inequality in the Swedish school system: A regional perspective. Environment and Planning A: Economy and Space 42: 2674–2686.

    Benson, M., G. Bridge, and D. Wilson. 2015. School choice in London and Paris—A comparison of middle‐class strategies. Social Policy & Administration 49: 24–43.

    Boterman, W.R. 2013. Dealing with diversity: Middle-class family households and the issue of ‘black’ and ‘white’ schools in Amsterdam. Urban Studies 50: 1130–1147. https://doi.org/10.1177/0042098012461673.

    Boterman et al., op. cit.

    Cheshire and Sheppard, op. cit.

    Nieuwenhuis, J., and P. Hooimeijer. 2016. The association between neighbourhoods and educational achievement, a systematic review and meta-analysis. Journal of Housing and the Built Environment 31: 321–347.

    Oberti, M. 2007. Social and school differentiation in urban space: Inequalities and local configurations. Environment and Planning A: Economy and Space 39: 208–227.

    Owens, A. 2017. Racial residential segregation of school-age children and adults: The role of schooling as a segregating force. RSF: The Russell Sage Foundation. Journal of the Social Sciences 3: 63–80.

    Rowe and Lubienski, op. cit.

  25. 25.

    Bernelius, V., and K. Vilkama. 2019. Pupils on the move: School catchment area segregation and residential mobility of urban families. Urban Studies 56(15): 3095–3116.

    Kauppinen et al. 2021, op. cit.

  26. 26.

    Bernelius and Vilkama, op. cit.

    Kauppinen et al. 2021, op. cit.

  27. 27.

    Kauppinen et al. 2021, op. cit.; Harjunen et al., op. cit.

  28. 28.

    Bernelius and Vilkama, op. cit., p. 3106.

  29. 29.

    See Berisha, A.K., and P. Seppänen. 2017. Pupil selection segments urban comprehensive schooling in Finland: Composition of school classes in pupils’ school performance, gender, and ethnicity. Scandinavian Journal of Educational Research 61(2): 240–254.

  30. 30.

    Kosunen, S., V. Bernelius, P. Seppänen, and M. Porkka. 2020. School choice to lower secondary schools and mechanisms of segregation in urban Finland. Urban Education 55(10): 1461–1488.

  31. 31.

    Bernelius and Vaattovaara, op. cit., p. 3167.

  32. 32.

    E.g., Silvennoinen, H., R. Rinne, S. Kosunen, M. Kalalahti, and P. Seppänen. 2015. Yhteiskuntaluokat ja kouluvalinta. In Lohkoutuva peruskoulu : Perheiden kouluvalinnat, yhteiskunta-luokat ja koulutuspolitiikka, eds. P. Seppänen, M. Kalalahti, R. Rinne, and H. Simola, 325–369. Helsinki: Suomen kasvatustieteellinen seura.

  33. 33.

    See for instance Power, S., Edwards, T., & Wigfall, V. 2003. Education and the middle class. Maidenhead: Open University Press.

  34. 34.

    Seppänen, P. 2006. Kouluvalintapolitiikka perusopetuksessa: Suomalaiskaupunkien koulumarkkinat kansainvälisessä valossa (No. 26). FERA: Turku.

  35. 35.

    Simola, H., P. Seppänen, S. Kosunen, and H. Vartiainen. 2015. Oppilasvalikoinnin paluu? In Lohkoutuva peruskoulu: Perheiden kouluvalinnat, yhteiskunta-luokat ja koulutuspolitiikka, eds. P. Seppänen, M. Kalalahti, R. Rinne, and H. Simola, 87–121. Helsinki: Suomen kasvatustieteellinen seura.

  36. 36.

    Bernelius, op. cit.

  37. 37.

    see Kosunen 2014, op. cit.

  38. 38.

    see Bernelius et al. 2021, op. cit.

  39. 39.

    Kosunen 2016, op. cit.

  40. 40.

    see Kosunen et al., 2020, op. cit.

  41. 41.

    Bernelius, op. cit.; Seppänen 2006, op. cit.

  42. 42.

    Ball, S.J., R. Bowe, and S. Gewirtz. 1996. School Choice, Social Class and Distinction: The Realisation of Social Advantage in Education. Journal of Education Policy 11(1), 89–112.

    Gewirtz, S., S.J. Ball, and R. Bowe, R. 1995. Markets, choice and equity in education. Buckingham: Open University Press.

  43. 43.

    Kosunen 2014, op. cit.

  44. 44.

    See Footnote 43.

  45. 45.

    Ramos Lobato, I., V. Bernelius, and S. Kosunen. 2018. Looking for the ordinary? Parental choice and elite school avoidance in Finland and Germany. Nordic Journal of Studies in Educational Policy 4(3): 156–167.

  46. 46.

    Kosunen et al. 2020, op. cit.

  47. 47.

    Bernelius and Huilla, op. cit.

  48. 48.

    OECD 2020, op. cit.

  49. 49.

    Wilson and Bridges 2019, op. cit.; Bernelius and Huilla, op. cit.