School Composition Effects in Spain

XREAP Working Paper No. 2010-13

42 Pages Posted: 22 Apr 2011

Date Written: December 1, 2010


Drawing on the PISA 2006 data set, this study examines the impact of school’s socioeconomic composition on the science test scores of Spanish students enrolled in compulsory secondary education. We define school’s composition in terms of the average parental human capital of students at that same school. These contextual peer effects are estimated using a semi-parametric methodology, which enables spillovers to affect all the parameters in the educational production function. We also deal with the potential problem of student self-selection into specific schools by using an artificial sorting mechanism, which we believe to be independent of a student’s unobserved abilities. The results indicate that the association between a school’s socio-economic composition and test score results is clearly positive and significantly higher when computed using a semiparametric approach. However, we find that the endogenous sorting of students into schools also plays a fundamental role, given that spillovers are significantly reduced when this selection process is eliminated from our measure of school’s composition effects. Specifically, the estimations suggest that contextual peer effects are moderately positive only in those schools where the socio-economic composition is comparatively high. In addition, we find some evidence of asymmetry as to how the external effects and the sorting process actually operate, apparently affecting males and females and high and low performance students differently.

Keywords: Educational Attainments, Peer Effects, PISA, Spain

JEL Classification: I20, I21, I29

Suggested Citation

Di Paolo, Antonio, School Composition Effects in Spain (December 1, 2010). XREAP Working Paper No. 2010-13, Available at SSRN: or

Antonio Di Paolo (Contact Author)

University of Barcelona ( email )

Gran Via de les Corts Catalanes, 585
Barcelona, 08007

Here is the Coronavirus
related research on SSRN

Paper statistics

Abstract Views
PlumX Metrics