Based on the data collected from provincial universities in China from 2004 to 2016, SPSS statistical tools were used to evaluate the scientific and technological innovation capability by principal component analysis. The efficiency of scientific and technological innovation was evaluated by two-stage data envelopment analysis model (DEA). The contribution rate of scientific and technological innovation was evaluated by using the "Cobb-Douglas (C-D) production function" and Solow "growth rate equation". Average number for each of the three indicators was calculated, data higher than the average number was defined as "high", and data below the average number was defined as "low". Provinces were divided by their grades of the three indicators:1) high-high-high:Liaoning, Shaanxi, Shandong; 2) high-high-low:Jiangsu, Guangdong, Zhejiang; 3) high-low-high:Beijing, Shanghai, Hubei, Anhui; 4) low-high-high:Jilin, Guangxi, Yunnan, Inner Mongolia, Guizhou, Hainan; 5) high-low-low:Sichuan, Henan; 6) low-high-low:Jiangxi, Fujian, Xinjiang; 7) low-low-high:Hunan, Tianjin, Chongqing; 8) low-low-low:Heilongjiang, Hebei, Shanxi, Gansu. Found in Liaoning, Shaanxi and Shandong, there is a certain linear correlation between scientific and technological innovation capability, efficiency, economic contribution rate and GDP growth rate. The research conclusion is consistent with the situation of "Double First-Class" construction to a certain extent.