How to analyze the first scientific collaboration in our study?

As a result, this study focused on 1,487 core scholars in the field of Chinese library and information science. It examined their research cooperation behavior by analyzing the co-authorship relationships in 55,640 papers they published. The study specifically focused on the phenomenon of new collaborative relationships formed when scholars first achieved cooperation, which amounted to 5,597 pairs of the initial collaborative relationship. Using methods such as the Author-Topic Model (ATM), the Louvain algorithm, Propensity Score Matching (PSM), differential expression analysis, and others, the study analyzed the degree of differences between collaborating parties at the time of their initial cooperation and their subsequent cooperative behavior from four perspectives: the scholars’ academic career age, social networks, research interests, and academic abilities. The goal is to understand which characteristics make it more likely for first-time research collaborations to occur between two scholars, the specific traits involved, how it influences their subsequent cooperation, and their future academic development direction.

What are the regularities of the first scientific collaboration?

The results indicate that scholars who have more common co-authors, similar research areas, and publication volume are more likely to establish first-time research collaborations, even when their academic impact and specific research topics differ significantly. Additionally, the initial collaborations formed when scholars publish their first paper tend to be more stable, frequent, and long-lasting, leading to a high degree of similarity in the future research directions and social network relationships of the collaborating parties. Furthermore, while scholars with significant academic background differences may have a lower likelihood of initial collaboration, once they do collaborate for the first time, the higher novelty they offer to each other increases the likelihood of future collaborations. Moreover, there is a certain correlation between the degree of difference in academic backgrounds at the time of the initial collaboration and the development direction of scholars’ future academic social networks and specific research topics. For example, if scholars who collaborate for the first time have a significant difference in academic age or academic impact, their academic circles may generally not intersect in the future.

What are some possible applications of the regularity of first scientific collaborations?

This research analyzes the driving factors of first scientific collaboration, which can be used to optimize collaboration recommendation services of those existing academic service platforms by facilitating new collaborations with unfamiliar scholars who have similar research productivity, overall research direction, and non-overlapping specific research areas. In addition, these findings reveal that first scientific collaborations are correlated with subsequent collaborations and future academic development. This correlation suggests that first scientific collaborations may influence scholars’ future academic trajectories, including the evolution of academic social networks, shifts in research directions, and changes in academic capabilities. This has implications for predicting scholars’ future academic behavior or performance.

This article is a translation of: Peng, X., & Shi, J. (2024). The drivers, features, and influence of first scientific collaboration among core scholars from Chinese library and information field. The Journal of the Association for Information Science and Technology, 1–20. https://doi.org/10.1002/asi.24888