An application of statistical interference in DEA models : an analysis of public owned university departments' efficiency

This paper uses Data Envelopment Analysis (DEA) model formulations in order to determine the performance levels of 16 departments of the University of Thessaly. Particularly, the constant returns to scale (CRS) and variable returns to scale (VRS) models have been applied alongside with bootstrap techniques in order to determine accurate performance measurements of the 16 departments. The study illustrates how the recent developments in efficiency analysis and statistical inference can be applied when evaluating institutional performance issues. The paper provides the efficient departments and the target values which need to be adopted from the inefficient departments in order to operate in the most productive scale size (MPSS). Moreover it provides bias corrected estimates alongside with their confidence intervals. The analysis indicates that there are strong inefficiencies among the departments, emphasizing the misallocation of resources or/and inefficient application of departments policy developments.