Using data from a nation-wide survey on Spanish workers, this paper examines two issues: the determinants of school choice (distinguishing between public and private schools) and the influence of school type on the individual's educational attainment. For that, bivariate probit models are estimated. The analysis is carried out separately by gender and birth cohort. We find that individuals who have attended a private school come from households with higher socio-cultural level and that having attended a private school does not raise the probability of reaching a higher degree, once potential endogeneity of school choice and educational outcomes are taken into account. However, the latter result depends on the instrumental variable used for observed school type in models of demand for higher education.