Abstract: At present,the study on China's regional economic growth focuses mainly on spatial pattern, but the study on spatial correlation and the cause of formation of growth clustering and disparity distribution is seldom. This paper introduces the spatial correlation index method of Moran I and the computational and test results show that the provincial regional economic growth has an obviously spatial correlation. And the economic growth has an obviously cluster in the geographical space. We also uses the spatial econometric analysis model of spatial-temporal data(Panel Data),and the computational and test results show that the regional economic growth during time and in space takes on a distinct spatial effects. The frequent extension to economic activities produced by international and domestic trade and foreign capital etc. brings spatial correlation,and to a great extent results in the inequalities of 31 provincial economic growth. Spatial cluster makes the spatial cost which is produced by geographical location(distance)decrease during the economic growth process,but the geographical characteristics will deeply affect the relationship between core and periphery of spatial clustering in regional economic growth。At last,the econometric test show that the factors such as international and interregional trade,foreign direct investment,human capital,technological innovation have an extraordinary contribution to China's regional economic growth. But those factors cannot change the regulations of economic geography easily. The non-equilibrium clustering of regional economic growth factors in geographical space results in widely distinct patterns of regional economic growth.