Abstract:In this paper, based on the data of 30 provinces of China from 2012 to 2016, considering the influence of random factors, the Bootstrap-SE-DEA model is used to measure the collaborative innovation efficiency of colleges and universities in China; according to the measurement results, the non-parametric test is used to analyze the trend of efficiency change and regional difference. The results show that during the analysis period, at least 75% of the regions’ collaborative innovation efficiency of colleges and universities are less than 1 every year, and only about 50% of the regions are between 0.5 and 1; the non-parametric test results show that the efficiency of collaborative innovation in Chinese colleges and universities has generally increased in recent years, the collaborative innovation efficiency of colleges and universities in the eight comprehensive economic zones shows different characteristics; the K-W test shows that except the Northeast Comprehensive Economic Zone and the Yangtze River Midstream Comprehensive Economic Zone, the distribution of internal efficiency value in other comprehensive economic zones is significantly different.