Influence of multi-dimensional social capital on structure of scientific collaboration networks: MDSC@SciCoNet

Thumbnail Image
Files
Date
2025-04-16
Authors
Gayen, Avijit
Chakraborty, Somyajit
Chakraborty, Nilotpal
Jana, Angshuman
Journal Title
Journal ISSN
Volume Title
Publisher
Springer Nature
Published Version
10.1007/s42979-025-03945-y
Research Projects
Organizational Units
Journal Issue
Abstract
Understanding the existence and effects of social capital in various social systems has been a focus for social scientists for many years. Research indicates that social capital is a social attribute influenced by the structural, relational, and cognitive components of an individual’s social network. Social capital can significantly impact career success, and its effects have been studied in various domains, including scientific collaborations. In scientific collaborations, an author’s social capital has been shown to influence the scientific impact of their research. However, few studies have explored how components of social capital contribute to variations in collaboration network structures across different scientific fields. In this paper, we analyze the evolution of scientific collaboration networks across different domains by examining the social capital components of authors within the network. For structural and relational capital measures, we use degree centrality and the count of prolific coauthors, respectively. Given the importance of tie strength in both relational and cognitive capital, we separately investigate the role of tie strength among collaborators in the evolution of these networks across various domains. We specifically examine the attachment probabilities of nodes in the network based on: (a) degree centrality, (b) count of prolific coauthors, and (c) tie strength among collaborators. We then model these attachment probabilities and validate them with empirical data. Our observations reveal that while the attachment of nodes with respect to these parameters is preferential, their impact on network structure varies. This study identifies key differences in collaboration dynamics between CSE and physics. In CSE, new collaborations are driven by network degree and prolific co-authors, with minimal influence from social tie strength. In contrast, physics shows a stronger role for social ties in collaboration formation. Both fields emphasize node tie strength in republishing, with physics favoring smaller, closed groups, while CSE exhibits broader networks. These findings highlight the varying roles of structural and social capital across disciplines.
Description
Keywords
Scientifc collaboration networks , Co-authorship networks , Social capital , Tie strength , Preferential attachment , ANOVA
Citation
Gayen, A., Chakraborty, S., Chakraborty, N. and Jana, A. (2025) 'Influence of multi-dimensional social capital on structure of scientific collaboration networks: MDSC@ SciCoNet', SN Computer Science, 6(5), 394 (23pp). https://doi.org/10.1007/s42979-025-03945-y
Link to publisher’s version
URI
https://hdl.handle.net/10468/17558
Collections
Computer Science - Journal Articles
Copyright
© 2025, the Authors, under exclusive licence to Springer Nature Singapore Pte Ltd. This version of the paper has been accepted for publication, after peer review (when applicable) and is subject to Springer Nature’s AM terms of use, but is not the Version of Record and does not reflect post-acceptance improvements, or any corrections. The Version of Record is available online at: https://doi.org/10.1007/s42979-025-03945-y