Performance evaluation of Turkish Universities by an integrated Bayesian BWM-TOPSIS model

Abstract

This study aims to develop a university ranking model with the aid of performance measures in the "University monitoring and evaluation reports-2019" published by the Council of Higher Education Institution in Turkey. In this context, some of the performance criteria stated in these reports are filtered and 34 sub-criteria under five main criteria are weighted using the Bayesian Best-Worst Method (BBWM). Then, 189 listed public and private universities are ranked using the TOPSIS multicriteria decision-making (MCDM) method. We have adopted the MCDM concept because many evaluation criteria/sub-criteria and alternative universities are taken into account. For this study, we apply an integrated MCDM model. First, we use BBWM to accomplish the first goal and adopt the TOPSIS method for the second purpose, using the BBWM results. Our purpose in using BBWM is due to its probabilistic structure that reduces the loss of information when handling group decisions. In this context, the evaluations of 11 academic experts are combined and a solid weighting is made by obtaining the credal rankings of performance criteria. Using TOPSIS is its logic of proximity to ideal and the ability to evaluate many alter-natives. In the context of the study, state-private university-based, nomenclature of territorial units for statistics-2 (NUTS-2)-based and classical geographical regions-based rankings are also discussed. The study seeks to help universities optimize their performance efficiently. The results of the study can be adapted as a reference for other educational institutions and public institutions in their efforts to evaluate, improve their performance and form various policies.