区域创新网络演化及其邻近性机理

The Evolution and Proximity Mechanism of Regional Innovation Network

基于我国2002-2018年31个省市的区域创新统计数据,本文借助修正的引力模型和社会网络分析法,分析区域创新空间关联网络的网络密度、中心性、凝聚子群等网络演化特征,并结合QAP回归考察多维邻近性对区域创新空间关联的作用机理。研究发现,我国区域创新空间关联的网络密度稳步提升,“多中心驱动”网络已经形成,创新中心发展遵循着邻近递减与等级扩散的规律。区域创新空间关联网络凝聚子群结构变动较大,各省市扮演的角色也随之不断调整,逐步呈现出“多中心、多层次”的网络创新格局。社会邻近性、认知邻近性、地理邻近性对我国区域创新的空间关联均具有显著促进作用,且促进作用依次减小,地理邻近性与社会邻近性表现为替代效应。据此,提出优化区域创新网络空间布局的对策建议。

Based on the regional innovation statistics of 31 provinces and cities in China from 2002 to 2018,the network evolution characteristics of regional innovation spatial association networks,including network density,centrality,and cohesion subgroups,are analyzed in the paper with means of modified Gravity Model and Social Network Analysis method.QAP Regression is used to investigate the mechanism of multi-dimensional proximity to regional innovation spatial correlation.The study findings are that the density of regional innovation spatial association network in China has steadily increased,the“multi-center driven”network has been formed,and innovation center follows the law of diminishing proximity and hierarchical diffusion.The cohesive subgroup structure of regional innovation spatial correlation network has changed greatly,and the roles played by provinces and cities have been continuously adjusted,gradually presenting a“multi-center,multi-level”network innovation pattern.Social proximity,cognitive proximity,and geographic proximity play a positive role in promoting the spatial correlation of regional innovation in China,and their promotion effects are reduced in turn.Geographic proximity and social proximity appear as substitution effects.Thus,suggestions for optimizing the spatial layout of regional innovation network are proposed.