Efficiency of Finnish Upper Secondary Schools: An Application of Stochastic Frontier Analysis with Panel Data

In this study the efficiency of Finnish upper secondary schools is evaluated with stochastic frontier analysis. Different stochastic frontier models for panel data are used to estimate education production functions. The results in matriculation examination are explained with comprehensive school grade point average, parents? socioeconomic background, resources, length of studies and decentralization of test taking on matriculation examination. Controls for schools with specialized curriculum are also included. The heterogeneity across schools is allowed by estimating both true random and true fixed effects models. The results show that the effect of teaching resources on examination results is even negative when the heterogeneity across schools is taken into account. Length of studies and decentralization of test taking affected negatively on student achievement. The inefficiency and the rankings of schools based on inefficiency score varied quite considerably depending on the type of stochastic frontier model. The lowest estimates for inefficiency were obtained with true random and true fixed effects models that separate time constant random or fixed effects from inefficiency.