Amsterdam Centre for Inequality Studies (AMCIS)

Early selection in secondary education increases differences between social groups

Interdisciplinary research on the Dutch educational system

27 October 2015

Countries that sort pupils into secondary-school tracks at a relatively early age, such as the Netherlands and Germany, show greater differences between social groups in academic achievement than countries where sorting takes place at a later age, such as Finland and Sweden. This is one of the findings of a Dutch report released 15 October entitled 'Education systems compared: Learning, working and citizenship'.

Taking an international perspective, researchers from the University of Amsterdam, Maastricht University and Erasmus University analysed how three basic features of the Dutch educational system – early selection, standardised testing, and a strong vocational education segment – contribute to education’s main functions.

Social background differences

The research found that countries that sort pupils into secondary-school tracks at a relatively early age, such as the Netherlands and Germany, show greater differences between social groups in academic achievement than countries where sorting takes place at a later age, such as Finland and Sweden. This is not only due to differences already observed before children begin school. It is also a result of differences that arise at junctures where a choice or selection is made. In the Netherlands, children with lower-educated parents are more likely to flow into lower-level school types than children who perform similarly in academic terms but have better-educated parents.

An objective test (e.g., the CITO standardised test) administered at the transition from primary to secondary school can reduce social background differences in educational achievement. According to the researchers, an objective standard ensures that students are sorted into schooling tracks based on their previous performance, without non-cognitive factors, such as parents’ social background, playing a role. In countries without central testing, such as Belgium, social inequalities are more pronounced than in the Netherlands. Shifting standardised testing to a later time, thereby increasing the weight of the primary school’s recommendation on pupils’ appropriate secondary track, would therefore not be sensible from a social inequality point of view, according to the researchers.

Strong vocational education

The labour market is better served, according to the researchers, by an educational system with an extensive network of vocational education. In general, unemployment rates are lower among graduates of vocational education than among students completing other educational levels. The research finds that this is a product of the strength of employers’ linkages with educational institutions, as reflected in curricular design. Strong engagement ensures that students acquire job-relevant skills, while also reducing the chance that they are educated for jobs that prove to be unavailable in practice.

About the report

Publication of the report concludes the research programme Educational Systems and the Main Functions of Education by NWO’s Netherlands Initiative for Education Research (NRO). Collaborating on this study were sociologists, education specialists and economists from the University of Amsterdam, Maastricht University and Erasmus University Rotterdam.

Published by  AISSR