Quality indicators in education: Discriminant analysis of the performances in Prova Brasil

Abstract

The present study identifies the contextual variables that best differentiate the performance obtained by students from the final grades of primary education in state schools of Espírito Santo in the Prova Brasil of 2013 through discriminant analysis, with a sample of 124 schools. The results showed that the age-series distortion, the teacher regularity index and the abandonment rate formed an optimal set of variables to discriminate the schools with "better" and "worse" school performance. The technique used contributes to the work of researchers and managers, as they can appropriate this method with the purpose of tracing the profile (description), differentiation (inference) or classification (prediction) of schools. Moreover, researchers and managers can reorganize their actions and investments based on priority contextual variables, with broad potential to change the school performance of their regions in Prova Brasil.