Drawing on PISA 2012 data, in this paper we study the impact of school's socio-economic composi-tion on the science test scores for Spanish lower-secondary education students. We adopt a semi-para-metric methodology that enables spillovers to affect all the parameters in the educational productionfunction. We also deal with the issue of endogenous students' sorting into schools with better socio-economic composition. The positive effect of school's socio-economic composition is stronger whencomputed using the semi-parametric approach, suggesting the relevance of slope effects of school com-position. However, the spillovers are substantially reduced when the endogenous sorting of students iscontrolled for.