Trends in Educational Disadvantage in Primary Education

• In Event: The Cohort Study on Educational Careers in Primary and Secondary Education (COOL5-18)

Abstract

Introduction
Since as far back as the 1970s, the Dutch government has pursued an educational compensatory policy in which schools attended by large numbers of children from disadvantaged backgrounds receive extra financial resources. This policy was implemented due to major differences in the levels of language and numeracy skills in students from different social environments and ethnic backgrounds upon entering primary school. If no extra measures are taken, those differences become more pronounced as the child’s education progresses. The so-called Matthew effect is often used to describe the mechanism behind increasing educational inequality (Stanovich, 2008). This paper examines whether inequality between students from various social and ethnic backgrounds in the period between 1994/95 and 2007/08 has decreased as a result of the educational compensatory policy.

Data and Method
This study used national cohort data collected in the Netherlands between 1994/95 and 2007/08 as part of the COOL5-18 study and its predecessor, the PRIMA study. The data included marks from nationally standardized language and mathematics tests taken by tens of thousands of students at approximately 550 primary schools. The larger part of those schools formed a representative sample of all schools in the Netherlands and the remaining schools formed a supplementary sample comprising schools with an overrepresentation of disadvantaged students.

The development of educational performance was examined using cross-sectional and longitudinal analyses. In the cross-sectional analyses, students’ position in a particular year group (e.g. Kindergarten) from 1994/95 were compared with those of other students in the same year group from other years in the period from 1994/95 to 2007/08. For every year, the average marks for each of the separate socio-ethnic categories of student were compared with those of a reference group comprising children of native Dutch parents with an average level of education ( senior secondary vocational education). Effect sizes (ES; Thompson, 1998) were used to provide an indication of the relevance of the difference.

In the longitudinal analyses, the same pupils were followed from Kindergarten through grade 6. The development of performance levels in language and arithmetic were analyzed using a latent growth curve analysis. Consideration was given to the fact that students had different starting performance levels and also that individual students differed in the degree to which their performance levels increased. These analyses were performed using the Mplus program (Muthén & Muthén, 2004).

Results
The conclusion shows that the development of all the non-western immigrant students was relatively favorable: during their primary school career, they did partly catch up in relation to their initial underperformance level. However, compared with the reference group (children of parents with senior secondary vocational education) they were still significantly educationally disadvantaged with regard to skills (language skills in particular) upon finishing primary school.
Development in educationally disadvantaged native-Dutch students was less favorable. They were also disadvantaged when they began primary education, but their level of underperformance increased during their primary school career.

Authors

• Jaap Roeleveld, University of Amsterdam

• Geert Driessen, Radboud University Nijmegen

• Ineke van der Veen, University of Amsterdam

• Guuske Ledoux, Universityersiteit van Amsterdam