The article explores the validity and reliability of an instrument for measuring for the School Environment (SE) among 3610 teachers in Medellı ´n, Colombia during 2011. A multilevel exploratory factorial analysis established through 20 items that the instrument identifies four dimensions of SE on an individual level: communication between school and teachers, between teachers and parents, teacher involvement in school decisions, sense of security (respect-safeness) and academic expectations. At a group level, the same items measure a general dimension of SE. Based upon these analyses, the instrument produced a valid and reliable school environment measure that differentiates among schools.

This study analyzes the economic returns to schooling decisions made by high school graduates in Colombia. We wanted to verify if the economic returns (wages) obtained by new postsecondary education graduates compensate for the economic and psychological investment they made to get that academic degree. To answer that question, we estimated these economic returns for each type of postsecondary degree available in Colombia (technical education, technological education, undergraduate studies, graduate studies) by origin of the institution (public or private). Our methodological strategy includes the generation of a micro-database that contains agents' socio-economic backgrounds and also their individual labor market outcomes. Because agents with very similar characteristics and the same schooling decisions might obtain different economic returnsfrom education, we considered as part of our empirical strategy the inclusion of an approximation of agents' cognitive abilities.

This paper provides new evidences to the literature of assignment in the labor market for the Colombian case. Specifically it focuses on the existing relationship between acquired human capital in higher education and its congruence in the labor market. Differing from previews studies, the misallocation analysis is not only based on the horizontal component and the educational mismatch, but it also includes the vertical mechanism (vertical mechanism is related to skills mismatch and horizontal mechanism is related to professional career mismatch). Another contribution is how we measure the abilities through an exploratory factor analysis. The data are taken from the Survey of Graduates of Higher Education Institutions 2014, provided by OLE. We employ a two-step treatment effect method proposed by Heckman (1974Heckman ( , 1979 and Lee (1978)), we found that generic abilities raise the probability of horizontal mismatch and diminish the probability of vertical mismatch. On the other hand, specific abilities lower the probability of both horizontal and vertical mismatch. In terms of wages, we found evidence that confirmed the results of the assignment models because it exists a wage penalty for the mismatched individuals (Sattinger, 1993).JEL classification: C35; J24; J31