Swedish compulsory schools have become more segregated during the past two decades, with increasing between-school heterogeneity regarding both academic achievement and social background composition of school in-takes (Gustafsson & Yang Hansen, 2009). Educational researchers and theorists argue that the explanations lie in the independent school reform and launch of free school choice policy in Sweden in the early 1990s, since it broke the public school monopoly and introduced market mechanism and competition into the school world.

To estimate the school choice effects on educational segregation is, however, extremely complicated due to methodological difficulties. Even though an intention of free school choice is stated to be to weaken the effects of social, ethnic and residential segregations, the admission to Swedish compulsory schools is still largely residence based. Families also choose their residential area by the existence of quality schools. Recent research indicates that housing segregation has been increasing steadily in Sweden. The school segregation may thus be due to other factors confounded with school choice, such as residential segregation and peer effects. Björklund et al. (2005) pointed out the difficulty of studying the phenomena of school choice, because its effect can be hidden by choice of residential area.

The empirical evidence concerning the effect of school choice on school segregation in Sweden is controversial. It has been shown that school choice is taken advantage of mainly by well-informed high SES parents, using it as a strategy to avoid immigrant-dominated schools, which in turn strengthens the school segregation (Bunar, 2010; SCB, 2007; Szulkin & Jonsson, 2007). Others argue that the increasing residential segregation lies behind the school segregation.

Applying a counterfactual approach, Almgren & Lindbom (2007) studied the extent to which the recent segregation in Swedish primary schools is due to school choice and/or to the residential segregation. They compared the estimated school segregation and residential segregation measures for both actual schools and fictitious schools and claimed that residential segregation caused the educational segregation. However, their assignment of students into fictitious schools neglects the different characteristics of geographic units and the schools, which may cause misclassification of students. In turn the segregation estimates in the fictitious schools may be biased (Gustafsson, 2007). Östh, Andersson & Malmberg (2010) modified the Almgren & Lindbom' s approach, and applied a multilevel model to separate the individual and school level variations in school achievement, controlling for the social background differences at the both level.  The multilevel models were estimated for both actual and fictitious school data. They concluded that school choice is the driving force of the increased school segregation.

However, causes of school segregation may differ between school types, cities and municipalities of different characteristics and contexts. To reach a better understanding of the mechanisms behind school segregation, it is important to control for these variations. The aim of the study thus is to examine the changes in school segregation with respect to achievement, SES and ethnicity 1994 – 2008. The causes of school segregation will be investigated in a longitudinal and multilevel perspective (see below).

Almgren, E., & Lindbom, A. (2007). Valfrihetens effekter på skolornas elevsammansättning: Skolsegregationen i Sverige [the effects of free school choice on school composition: School segregation in Sweden, in Swedish]. In A. Lindbom (Ed.), Friskolorna och framtiden - segregation, kostnader och effektivitet [Independent schools and the future - Segregation, costs and effectiveness]. (pp. 89- 118). Stockholm: Institutet för framtidsstudier. Björklund, A., Clark, M., Edin, P.-E., Fredriksson, P., & Krueger, A. B. (2005). The Market Comes to Education in Sweden. An Evaluation of Sweden’s Surprising School Reforms: Russel Sage Foundation. Bunar, N. (2010). Choosing for Quality or Inequality: Current Perspectives on Implementation of School Choice Policy in Sweden. Journal of Educational Policy, 25, pp. 1-18. Gustafsson, J.-E. (2007). Kommentar till valfrihetens effekter på skolornas elevsammansättning [Comments to effects of free school choice on school composition, in Swedish]. In A. Lindbom (Ed.), Friskolorna och framtiden - segregation, kostnader och effektivitet [Independent schools and the future - Segregation, costs and effectiveness]. (pp. 119-126). Stockholm: Institutet för framtidsstudier. Gustafsson, J.-E., & Yang Hansen, K. (2009). Resultatförändringar i svensk grundskola [Changes in Outcomes in Swedish Compulsory Schools; in Swedish]. In L. M. Olsson (Ed.), Vad påverkar resultaten i grundskolan? (pp. 40-84). Stockholm: Skolverket. SCB. (2007). Barn, boendesegregation och skolresultat [Children, segregated housing and school results]. Stockholm: Statistics Sweden. Szulkin, R., & Jonsson, J. O. (2007). Ethnic Segregation and Educational Outcomes in Swedish Comprehensive Schools. Unpublished manuscript, Stockholm. Östh, J., Andersson, E., & Malmberg, B. (2010). School Choice and Increasing Performance Difference: A Conterfactual Approach. Stockholm: Department of Sociology, Demography Unit.