Data envelopment analysis of the efficiency of academic departments at a public university in Iran

Abstract

This paper examines the performance efficiency of 77 academic departments at a public university in Iran using data envelopment analysis (DEA), which applies a multiple input and output variables approach. To conduct reasonable analyses, four types of academic staff accompanied with salary paid were used as the input measures and two indexed measures of output which include students taught and research performance. Using various DEA models, we obtained different estimates for efficiency scores which show that the type of model used affects the efficiency scores. We also found that average efficiency is relatively high and about one half of all academic departments at the university, based on the results of input-oriented variable return to scale models, perform efficiently. The results of the DEA models employed also illustrate the existence of scale inefficiencies and the relatively large heterogeneity among the departments. However, we did not find sufficient evidence supporting efficiency heterogeneity across different fields of study (i.e., humanities and social sciences against sciences and engineering). The determinants of departmental efficiency were also examined.