The Choice of Estimation Method and Its Effect on Efficiency Measurement in Public Education: Stochastic Frontier Regression vs. Data Envelopment Analysis

Abstract

The purpose of this paper is to extend the empirical literature on efficiencymeasurement in public education. Two estimation methods often used in determiningefficiency in the production of public education are used to determine if the results fromthe methods are similar. The Oklahoma public schools are of interest because Oklahomahas a large number of districts with very different characteristics. The methods ofestimation are Stochastic Frontier Regression (SFR) and Data Envelopment Analysis(DEA). SFR estimates the inefficiency model simultaneously with the production or costfunction. In the DEA model, the first stage estimates the efficiency scores and the secondstage uses a Tobit regression model to determine causes of inefficiency. In this study, theempirical results of the SFR and DEA efficiency scores for the majority of Oklahomaschool districts are not identical, suggesting that the method of estimation affects theefficiency scores. In general, SFR generated a more favorable score than that of DEA. Theresults from the two estimation methods in the inefficiency model are also different.However, both methods suggest that the most important determinants of inefficiency aresocioeconomic factors associated with each district. (I21, C13)