The main goal of this paper is to study the efficiency of Colombian schools with an eye on understanding what drives school productivity and how much is to be gained by increasing efficiency. The paper uses Data Envelopment Analysis (DEA) to study efficiency and a Tobit model to study the determinants of efficiency. Introduction. The main goal of this paper is to study the efficiency of Colombian schools with an eye on understanding what drives school productivity and how much is to be gained by increasing efficiency. The paper uses Data Envelopment Analysis (DEA) to study efficiency and a Tobit model to study the determinants of efficiency. This is not the first paper to look at the efficiency of secondary schools in Colombia. Gaviria and Barrientos (2001) look at the determinants of test scores in the city of Bogotá. They find that public schools produce lower scores than private schools after controlling for socioeconomic characteristics. More importantly, they find that teacher´s education and infrastructure positively affect scores in private schools but not in public schools, which suggests the existence of inefficiencies in publicly provided education. Nunez et al. (2002) look at the differences between private and public secondary schools in both urban and rural areas for the year 1999. They find that public schools produce, on average, lower scores. However, at the bottom of the distribution, public schools appear to be better than private schools. Finally, Barrera (2003) finds that test scores in both public and private schools increased in the last decade, but that private school scores increased more than public school ones. Initially for the investigation, we analyzed the case of Bogotá. However, given the disparity between this city and the rest of the country and a desire to include municipal level variation, we decided to expand the study to the national level. The results, as we will show, change dramatically in the two samples. We will, therefore, present the results using the national sample and analyze the specific case of Bogotá separately. The paper has the following structure. In the second part, we describe the main methodology used in the analysis. The third part presents the results for the whole country. The next section includes a discussion of the potential role of costs of public versus private education. Finally we close the paper with some general conclusions. In the Appendix we present the analysis of Bogotá as a particular case.
