A mutilevel decomposition of school performance using robust nonparametric frontier techniques

Abstract

This article proposes a methodology for evaluating educational performance, from a multilevel perspective. We consider the use of frontier techniques rather than regression equations-the latter of which do not explore variations in students' outcomes within the same school, as this variation is hidden behind an average. Similar to some recent literature contributions, we use partial frontier approaches to mitigate the influence of outliers and the curse of dimensionality, yielding statistically robust results. In contrast to previous studies that use partial frontiers, we consider in our estimation idiosyncratic variables at the school, class, and student levels. Our model is applied to a sample of students in the fourth year of primary school in urban schools in Chile. The results are in line with previous ones that found that less than 30% of the variance in students' educational attainment could be attributed to their schools. Our application also corroborates the assertion that a model that considers only student-level variables would yield high inefficiencies that cannot be attributed to the school management, but rather to inadequate resource-endowment policy. In other words, when one does not consider specific variables concerning the resources allocated to the schools, the performance of those school is undervalued, largely because inefficiencies caused by suboptimal resource endowments or difficulties that arise from the socioeconomic environment are instead attributed to poor school management.