Research On Regional R & D Performance Based On DEA And SNA

As the core of national scientific and technological innovation, R&D activities play a vital role in enhancing international competitive position. At present, China's amount of R&D output has been listed on top in the world. However, the high-level R&D output is few. It is crucial to drive regional scientific and technological development as well as technological diffusion with effective allocation of R&D resources, under the context of lack of energy resources, limitation of R&D input, in view of the unbalance of regional economic and technological development. Meanwhile, logical methods of regional R&D performance evaluation are desired as reference for policy formulation, with the aim to seek effective allocation of R&D resources.In this paper, we focused on the introduction of social network analysis (SNA). With basic theory of SNA, we established network on relative efficiency of regional R&D performance which taken region as node, after regional R&D's data environment analysis. The aim of this idea was to offset the short come of DEA method that could not make further exploration on a relatively effective unit.First of all, after the theoretical review of literature at home and abroad, the author established comprehensive performance evaluation system with the principal component analysis method, based on China Science and Technology Statistic Yearbook of 2008 and 2009. Then the author divided R&D performance into two stage, technology development stage and technology diffusion stage, according to the characters of R&D process.Then, research on regional R&D performance analysis was made through principal components analysis and DEA. We got the results of comprehensive management performance of regional R&D and the effectiveness in two R&D stage separately. According to the limitations of both methods above, social network analysis (SNA) was introduced in discrimination of DEA results.Finally, in view of reference weight in C2GS2 model under different specifications, the network of regional R&D relatively effectiveness was founded. The author ranked regional R&D performance according to eigenvector centrality in the network, and analysis strengths and weakness of different region, as well as sensitive factors. The policy recommendations for regional R&D performance improve were advanced according analysis above.