Schools, ability, and the socioeconomic gradient in education choices

This paper provides evidence on the socioeconomic gradient in dropout and progression in upper secondary education in Norway. Using a rich data set covering all students transferring from compulsory education to upper secondary education in 2002, we find that student achievement at the end of compulsory education is the main predictor of dropout and delayed progression. The socioeconomic gradient is sensitive to the inclusion of prior achievement in the model. We find that the gradient is modest and non-linear, and related to dropout behavior and not grade repetition. The results are remarkably robust to controlling for a full set of school fixed effects.

► The paper examines the socioeconomic gradient in progression in upper secondary education using Norwegian register data. ► The estimated gradient is sensitive to whether the empirical model includes prior student achievement or not. ► The effect of both parental education and parental income is highly nonlinear. ► The estimated socioeconomic gradient is remarkably stable to inclusion of school fixed effects. ► The estimated gradient seems not to be biased by endogenous sorting of students or allocation of resources across schools.