Assessing the efficiency of innovation entities in China: evidence from a nonhomogeneous data envelopment analysis and Tobit

Universities, research institutes, and firms are the main entities in the national innovation system. Owing to the heterogeneity of their outputs, prior studies have focused on their independent efficiency evaluation. This study adopts the nonhomogeneous data envelopment analysis model to assess the efficiency of three innovation entities in 30 provinces in China on a common platform. Results show that firms have the highest efficiency, and research institutes have the lowest efficiency. Innovation entities perform poorly due to the inefficiency of their subunits. Additionally, the 30 provinces are divided into three clusters by using the hierarchical clustering method. Moreover, Tobit regressions are used to estimate the impact of five environmental factors on the innovation efficiency of the three entities. The regression results show that the more open the region, the stronger the positive impact on the innovation efficiency of research institutes and firms. The regional economic environment has different degrees of negative impact on the three innovation entities. The direction and intensity of the impact of education input, government support, and information infrastructure on the three entities exhibit a large dispersion. The results provide important information for improving the efficiency of innovation entities.