Improving the accuracy of DEA efficiency analysis: a bootstrap application to the health care foodservice industry

Abstract

This article analyses the efficiency of health care foodservice operations and its determinants using a Data Envelopment Analysis (DEA) bootstrapping approach. The purpose of using the bootstrapping approach is two-fold: first, to obtain the bias corrected estimates and the confidence intervals of DEA-efficiency scores and second, to overcome the correlation problem of DEA-efficiency scores and to provide consistent inferences in explaining the determinants of health care foodservice efficiency. The approach was implemented on a sample of 89 health care foodservice operations. The results showed the presence of inefficiency in the sample, with an average efficiency level of 72.6%. Further, the results from analysing the determinants of health care foodservice operations provided policy implication regarding the factors that might improve the efficiency of health care foodservice operations.