Using matching estimator techniques developed by Heckman, Ichimura, Smith and Todd (1998) and Heckman, Lalonde and Smith (1999), we explore the determinants of achievement rates in standardized tests of Colombian students attending public and private schools. This methodology allows us to control for the fact that choosing among schools is an endogenous decision. After controlling for individual, household and school characteristics, we find a significant difference in favor of students attending private schools. We attribute this finding to characteristics such as distinct levels of teacher unionization and different schemes for teacher evaluation, remuneration and promotion.