Based on the projects funded by the National Natural Science Foundation of 31 provinces, autonomous regions and municipalities in China from 2009 to 2018, the improved gravity model was adopted to construct the relevance of basic research among provinces, autonomous regions and municipalities in China. On this basis, the social network analysis method was used to study the spatial correlation network structure and influencing factors of basic research in China. The main results are as follows. Spatial correlation of overall network gradually enhances and the network efficiency has a gentle decline. The eastern economically developed regions do not have a significant impact on basic research in other regions. Spatial association network for basic research can be divided into four functional blocks, including bidirectional spillover block, broker block, net spillover block and net beneficial block. The connections within the block are greater than the connections between the blocks. Spatial correlation structure of basic scientific research is significantly affected by geographical location, research application agencies, scientific researchers and the number of professor and associate professor titles.