Individual potential of educational achievement, socioeconomic and family background: an empiric application with latent variables for Colombia

Abstract

This article proposes a method to estimate the individual potential of educational achievement (PCE) for Colombia using Structural Equation Model approach (SEM). The latent variable PCE is derived from a multidimensional array of the neighborhood effects and the family background. The main contribution of this study was to advance in Bayesian estimates for the structural relations suggested by the theory and to compare these outcomes with those derived from Maximum Likelihood Statistical techniques (M.L). The results show evidence that the quality of local linkages effects (specifically the neighborhood environment) makes the largest contribution to variation in PCE. This conclusion was consistent when estimated by the two methodologies