In this paper we use the Oaxaca-Blinder decomposition methodology, applied to the pisa 2012 database, in order to identify the causes of the differences in academic results between public and private Costa Rican schools. In order of importance, these are caused by differences in: a) family characteristics, b) school resources, c) student characteristics, and d) working environment in the schools. The differences in the results are not only explained by the differences in characteristics and factors, but also by the differences in the way they are used. So, while students in public schools have collectively worst characteristics, they make better use of them. However, students in private schools obtain more output from the working environment in their schools.