Proximity and impact of university-industry collaborations. A topic detection analysis of impact reports
基于英国415个大学-产业合作项目，分析合作者之间的地理、制度、认知邻近性与经济、社会、知识影响的关系，发现不同类型邻近性对不同影响有替代或距离效应。
The probability to initiate university-industry collaborations (UICs), their intensity and quality, are influenced by the proximity between the collaboration partners. However, little is known about the relationship between collaborators' proximity and impact of UICs. Building on an original database of 415 UICs in the United Kingdom, we analyse the association between collaborators' proximity and the extent to which UICs generate economic, social and knowledge impact. We find that geographical and institutional proximity are substitutes in relation to economic impact, cognitive and institutional proximity are substitutes in relation to knowledge impact, and social impact is associated with cognitive and institutional distance. • Different types of proximity are associated with different types of impact of university-industry collaborations • Geographical and institutional proximity act as substitutes in relation to economic impact • Cognitive and institutional proximity act as substitutes in relation to knowledge impact • Social impact is associated with geographical and institutional distance between partners • Topic modelling of project reports can be used to capture the impact of university-industry collaborations