Navigating the Trade-offs between Independence and Collaboration: A Network Analytic Method and Case Study

Research independence emphasizes the ability to conduct individual work, while collaboration highlights establishing academic connections for team-based activities. Since the two dimensions have become crucial indicators for individual-based research evaluation, how to handle the trade-offs between research independence and collaboration diversity is raised. This paper exploits network analytics to investigate the two indicators via an integrated network that includes an author-term network and a co-authorship network. Specifically, a diffusion-based network analytic model is proposed to evaluate independence by considering the uniqueness of the use of terms, and, in parallel, collaboration diversity is quantified by the strength and breadth of a researcher’s co-authorships. We applied the proposed method to a dataset with 19,612 information science-related articles. The observed results not only well demonstrate the reliability of the proposed method but also empirically uncover a balancing threshold between the two indicators, providing references for researchers and decision-makers in the discipline.