Application of DEA to measure the productivity of the academic staff

Abstract

Subject. Increasing the productivity of each academic staff member is an important condition for the achievement of high indicators by the university. The modern system of the effective contract in Russian higher education institutions provides great growth opportunities which need to be identified by analysts. Appropriate analysis methods are needed to secure a rational use of these opportunities. Objectives. Our goal was to find the best option for using the non-parametric method of Data Envelopment Analysis to solve the task related to the assessment of the productivity of each academic staff member at a middle-size university. Methodology. We used single-stage DEA models with constant and variable returns to scale, including non-discretionary outputs. The choice of a specific model was based on both quantitative and content interpretation of the obtained results. Results. It was established that R.D. Banker and R.C. Morey’s model can properly measure the productivity of academic staff and provide the university management with valuable information which can be used to efficiently manage the productivity of the employees. We identified the most probable difficulties regarding the application of similar models to a large sample of employees under real conditions. We obtained data on the productivity distribution and the dependence of productivity on occupation, which require further analysis. Conclusions. The basic single-stage DEA models are sufficient to measure the productivity of employees if non-discretionary outputs are carefully introduced and the solution's sustainability is controlled. However, the features characteristic of all DEA models cannot be ignored: they set achievable goals to improve the existing results, however they do not search new opportunities; they inform that the goals are achieved but not how significant the achieved results are for the university.