Study On Scientific Research Efficiency Evaluation In Universities Under Direct Administration Of The Ministry Of Education In China Based On DEA

The college’s scientific research is the important component of national innovation system. How to ensure the fulfillment of expected scientific research goals on time, the high quality of scientific research achievements and the high performance of scientific research inputs have been widely concerned by various circles of society, as the significantly increase of funds invested by government. Especially the universities under direct administration of the Ministry of Education in China, which are the backbone of scientific research, own a large amount of advantageous resources. Therefore, their scientific research efficiency becomes a focus concerned by society. As the scientific research in the universities is a complex system of multi-input and multi-output, data envelopment analysis has a unique advantage in the comprehensive evaluation of this system, because it doesn’t need to estimate the weights and assume the specific form of production function, which is not available by the other methods. Therefore this article will use DEA to evaluate the scientific research efficiency of universities under direct administration of the Ministry of Education objectively and fairly through structuring multi-level classification DEA models which can reflect the evaluators’value-oriented.This paper will firstly introduce background and basic concepts, and then introduce the basic idea and model of DEA, summarize the achievement and inadequacies of existing research literature through carding them. Secondly this paper will build the input and output indicators system of evaluation through analyzing the characteristics of university research activities combined with the availability of data. Thirdly it will build multi-level classification DEA models with considering the deficiencies of existing DEA models, and use different DEA models to carry out the empirical analysis combined the relevant research data of universities under direct administration of the Ministry of Education, including dynamic analysis with years for DMUs and static analysis with universities for DMUs. Finally it will give the conclusions and outlook. The innovations of this paper include the following two aspects:Firstly, it divides the index into two categories of natural science and humanity social science when building evaluation index system with considering the large difference of input and output indicators in different disciplines.Secondly, it builds multi-level classification DEA models which are able to calculate the university scientific research efficiency of different disciplines with combining the characteristics of the evaluation index system.