Performance evaluation of Chinese research universities: A parallel interactive network DEA approach with shared and fixed sum inputs

In recent years, Chinese government and universities have been striving to improve the quality of higher education according to the continuously updated university performance evaluation. The university performance evaluation has played an important role in education reform. As an essential function of university, the research activities show some particular features in China. On the one hand, scientific research activities have parallel interactive network structures. On the other hand, university, as centrally governed institution, faces limited financial support, implying a fixed-sum constraint on government grant funding. However, previous studies have not developed an appropriate method to address the above features, which may lead to biased empirical results. In order to fill the gap of previous studies, this paper intends to provide a novel data envelopment analysis (DEA) approach by considering the shared fixed-sum input in the network DEA model. We first divide the scientific research activity into two sub-systems which interact with each other, and the government grant funding as a shared fixed-sum input is consumed for the two sub-systems. Then, considering the effect of shared fixed-sum input on performance, the minimum amount of input adjustments is calculated to identify the common equilibrium efficient frontier, and the total efficiency is obtained by weighting the sum of sub-system efficiency. Finally, our models are empirically applied to assess the performance of Chinese universities. The main findings show that the overall efficiency of Chinese research universities varies widely, and then some meaningful recommendations for central policymakers are proposed.