Co-contributorship Network and Division of Labor in Individual Scientific Collaborations

Chao Lu, Yingyi Zhang, Yong-Yeol Ahn, Ying Ding, Chenwei Zhang, Dandan Ma

Abstract

Collaborations are pervasive in modern science, yet the internal division of labor within these teams remains largely opaque. This study investigates the patterns of scientific collaboration by analyzing co-contributorship networks derived from 138,787 papers published in PLoS journals. By performing a one-mode projection of author-task bipartite networks, we identify three distinct contributor roles: Specialists, Team-players, and Versatiles. Our findings reveal that while Team-players constitute the majority of contributors—particularly in large-scale projects—Specialists and Versatiles appear more frequently than predicted by random-graph null models. Versatiles typically represent senior authors providing supervision and funding, while Specialists occupy two contrasting niches: senior leadership or highly specific technical support. This research provides a granular view of the social architecture of scientific production.

Introduction

For decades, bibliometric research has utilized co-authorship as the primary proxy for scientific collaboration. However, this approach treats the collaboration process as a "black box," failing to distinguish between the varied contributions of individual authors. As scientific problems grow in complexity and team sizes increase, understanding the internal division of labor becomes critical. This paper addresses the lack of granular data on how teams function by mapping authors to specific tasks, such as data analysis, experimental performance, and manuscript drafting, to reveal the underlying social structure of scientific labor.

Methodology

Co-contributorship Networks

The researchers introduced a novel framework using co-contributorship networks to model labor division. The process involves two primary steps:

  1. Bipartite Mapping: A network is constructed connecting authors to specific tasks (e.g., "Author A" linked to "Data Analysis").
  2. One-Mode Projection: This bipartite network is transformed into an author-only network. In this model, an edge between two authors signifies they collaborated on at least one shared task. A self-loop indicates an author performed a specific task entirely alone.

Role Classification

Based on the network topology, contributors are classified into three categories:

  • Specialists: Authors who only possess self-loops, meaning they contribute to all their assigned tasks individually.
  • Team-players: Authors with no self-loops, meaning every task they performed involved collaboration with others.
  • Versatiles: Authors who exhibit both self-loops and collaborative edges, performing some tasks alone and others in a team context.

To validate these roles, the researchers compared the empirical data against two null models: the Erdős–Rényi Model (ERM) and the Configuration Model (CFM).

Data and Empirical Setting

The study analyzed a comprehensive dataset of 138,787 full-text articles published in PLoS journals between 2006 and 2015. PLoS ONE contributed approximately 90% of the data. The researchers utilized mandatory "author contribution statements" to extract task data. While dozens of tasks were identified, the analysis focused on five common categories: conceiving experiments, performing experiments, analyzing data, contributing reagents/tools, and writing the paper.

Key Insights and Results

The Dominance of Team-Players

As team sizes grow, the "Team-player" role becomes the standard mode of contribution. In teams exceeding 20 members, nearly 95% of contributors are classified as Team-players. These individuals form the backbone of labor-intensive tasks such as "performing experiments" and "data analysis."

The Dual Nature of Specialists

Contrary to the assumption that specialists are marginal contributors, the data suggests they often hold pivotal positions. Specialists are significantly more prevalent than expected by random models and generally fall into two categories:

  1. Senior Leaders: Individuals whose "solo" tasks involve high-level oversight, such as "principal investigator" roles or final manuscript revision.
  2. Niche Experts: Individuals providing highly specific resources, such as unique reagents or specialized tools, which do not require integrated team collaboration.

Versatiles as the "Glue"

Versatiles are predominantly senior authors. They act as the bridge between the execution of research (performing experiments) and the management of the project (securing funding and supervision). In small teams, Versatiles comprise roughly 55% of the population, though this percentage drops significantly as teams scale and roles become more rigid.

Efficiency vs. Coordination

The study highlights a tension between the "Adam Smith effect" (increased productivity through specialization) and coordination costs. While specialization allows for complex projects, it introduces overhead. Scientific teams appear to reach a saturation point where they limit further specialization to manage the costs of coordinating disparate parts.

Limitations and Conclusion

While this study provides a new lens for viewing collaboration, it is primarily focused on the life sciences, where "contributing reagents" is a standard task that may not translate to other fields like theoretical physics or the social sciences. Additionally, the model currently treats all tasks as having equal weight, regardless of the actual time or effort required.

In conclusion, the transition from simple co-authorship lists to co-contributorship networks reveals a sophisticated social architecture within scientific teams. Future research tracking the evolution of these roles over a researcher's career—from a "Team-player" trainee to a "Versatile" or "Specialist" leader—will further illuminate the dynamics of the scientific enterprise.

Summary updated

Abstract

Collaborations are pervasive in current science. Collaborations have been studied and encouraged in many disciplines. However, little is known how a team really functions from the detailed division of labor within. In this research, we investigate the patterns of scientific collaboration and division of labor within individual scholarly articles by analyzing their co-contributorship networks. Co-contributorship networks are constructed by performing the one-mode projection of the author-task bipartite networks obtained from 138,787 papers published in PLoS journals. Given a paper, we define three types of contributors: Specialists, Team-players, and Versatiles. Specialists are those who contribute to all their tasks alone; team-players are those who contribute to every task with other collaborators; and versatiles are those who do both. We find that team-players are the majority and they tend to contribute to the five most common tasks as expected, such as "data analysis" and "performing experiments". The specialists and versatiles are more prevalent than expected by a random-graph null model. Versatiles tend to be senior authors associated with funding and supervisions. Specialists are associated with two contrasting roles: the supervising role as team leaders or marginal and specialized contributions.
BibTeX
@article{lu2020co,
  author = {Chao Lu and Yingyi Zhang and Yong-Yeol Ahn and Ying Ding and Chenwei Zhang and Dandan Ma},
  title = {{Co-contributorship Network and Division of Labor in Individual Scientific Collaborations}},
  journal = {J. Assoc. Inf. Sci. Technol.},
  year = {2020},
  url = {https://openreview.net/forum?id=hxk2dmYQNb},
}
Figure