Two-stage research performance assessment of Turkish HEI using DEA and beta regression

Abstract

In this paper, we study the research efficiency of the Turkish higher education sector in a two-stage DEA model with variable returns to scale. The aim of this paper is to benchmark or rank universities according to their research performance and to identify exogenous factors that may affect an institution’s efficiency score. DEA scores are a prime example of fractional data - a fact that has been disregarded by many previous DEA models which used popular Tobit regression for censored data in the second stage. Using a sample of 50 private and public universities, the first stage of our model calculates the efficiency scores and determines the efficient reference set for inefficient universities. In the second stage, we use beta regression and bootstrapped hypothesis testing to estimate the effects that external factors (age, size and ownership status) have on efficiency scores. We find that 27 universities in our sample are research efficient. Beta regression summary statistics suggest that extra-large universities tend to be less research efficient than large universities (p=0.1), while both age and ownership status of the university do not have a statistically significant impact on an institution’s efficiency score.