How do collaborative features affect scientific output? Evidences from wind power field

The aim of this study is to examine how scientific collaborative features influence scientific collaboration networks and then affect scientific output. In order to explore the influence of scientific collaboration, we define three collaborative features: inertia, diversity and strength. The data are collected from Scopus and the Web of Science databases. Using technique for order preference by similarity to an ideal solution method, we firstly combine h-index, impact factor and SCImago journal rank to rank journals in the field of wind power. Then we construct the collaboration network of institutions and use structural equation model-partial least square to examine the relationship among collaborative features, network structure, and scientific output. The results show that collaborative diversity and strength have positive effects on scientific output, while collaborative inertia has a negative effect. Both of centrality and structural holes fully account for (mediate) the relationships between collaborative features and outputs. The findings have some important policy implications to scientific collaboration: (1) research institutions should actively participate in diverse collaborations; (2) rather than only collaborating with previous partners, they should seek more new partners; and (3) collaborative features are important antecedents of scientific networks.