School dropout, social origin, and age-grade distortion: evidence on primary and secondary effects

Abstract

This article investigates school dropout among young people aged 15 to 19 in Brazil, focusing on the role of social origin and age-grade distortion. Drawing on the theoretical distinction between primary and secondary effects, we examine the extent to which the association between social origin and school dropout is mediated by age-grade distortion, used as an indirect indicator of delayed and irregular educational trajectories. Using data from the Continuous National Household Sample Survey (PNAD Contínua), we estimate logit models that offer a methodological innovation by incorporating educational delay among young people who had already dropped out of school, thereby overcoming a recurrent limitation of Brazilian national datasets. The results indicate that, although age-grade distortion is strongly associated with school dropout, its explanatory power is limited compared with that of social origin. Even after controlling for educational delay and labor market participation, substantial differences in dropout probabilities persist across socioeconomic strata. More importantly, only a small proportion of the association between social origin and school dropout is mediated by educational delay. The evidence therefore suggests that the mechanisms linking inequalities of origin to school dropout are not limited to differences in academic performance but largely involve factors related to educational choices, expectations, and the conditions that enable students to remain in the education system. By presenting evidence consistent with the relevance of secondary effects-without disregarding the existence of unobserved primary effects-the article contributes to the debate on the processes shaping school dropout in Brazil and highlights the importance of the social determinants of educational decisions.