Gender disparities in the dissemination and acquisition of scientific knowledge

Abstract: Recent research has challenged the widespread belief that gender inequities in academia would disappear simply by increasing the number of women. More complex causes might be at play, embodied in the networked structure of scientific collaborations. Here, we aim to understand the structural inequality between male and female scholars in the dissemination of scientific knowledge. We use a large-scale dataset of academic publications from the American Physical Society (APS) to build a time-varying network of collaborations from 1970 to 2020. We model knowledge dissemination as a contagion process in which scientists become informed based on the propagation of knowledge through their collaborators. We quantify the fairness of the system in terms of how women acquire and diffuse knowledge compared to men. Our results indicate that knowledge acquisition and diffusion are slower for women than expected. We find that the main determinant of women's disadvantage is the gap in the cumulative number of collaborators, highlighting how time creates structural disadvantages that contribute to marginalize women in physics. Our work sheds light on how the dynamics of scientific collaborations shape gender disparities in knowledge dissemination and calls for a deeper understanding on how to intervene to improve fairness and diversity in the scientific community.

• The paper shows that women are up to 25% less likely to be early recipients of scientific knowledge compared to men.
• It employs a dynamic temporal coauthorship network of 678,916 APS publications to quantify acquisition and diffusion fairness.
• The study highlights that increasing female participation alone is insufficient without addressing structural inequities in collaborative networks.

Gender Disparities in the Dissemination and Acquisition of Scientific Knowledge

The paper entitled "Gender disparities in the dissemination and acquisition of scientific knowledge" authored by Chiara Zappala, Luca Gallo, Jan Bachmann, Federico Battiston, and Fariba Karimi, investigates the persistent gender inequities in academia, specifically focusing on the dynamics within the American Physical Society (APS) coauthorship network from 1970 to 2020. This study contends that the mere increase in the number of women participating in scientific research is not sufficient to abolish gender disparities rooted in the network structure of scientific collaborations.

Methodology

The authors utilize a rich dataset from the APS, encompassing 678,916 publications, to construct a temporal coauthorship network. Nodes represent authors and links denote coauthored publications. The network evolves annually from 1970 to 2020, dynamically reflecting the changing structure of scientific collaborations. Gender inference is performed on the authors' names, resulting in 34,596 women, 195,490 men, and 200,669 authors with unknown gender.

The essence of the study is to model knowledge dissemination and acquisition as a contagion process, where informed (I) and susceptible (S) states are utilized to simulate the flow of information through the network. The fairness of the dissemination system is quantified via two primary metrics:

1. Acquisition Fairness (AF): Measures the proportion of female authors who acquire knowledge relative to their presence in the network.
2. Diffusion Fairness (DF): Compares the time required for women and men to disseminate knowledge to the same fraction of the network.

Findings

Key Results

1. Knowledge Acquisition:
   • Women are consistently at a disadvantage in acquiring knowledge. Specifically, among the early recipients of knowledge, women are underrepresented by up to 25% compared to their male counterparts.
   • Acquisition fairness shows a U-shaped trend over the decades, suggesting periods of improvement and regression without a clear gradual progression towards parity.
2. Knowledge Diffusion:
   • Contagion processes starting from female authors generally require more time to reach the same fraction of the network compared to those initiated by men. This indicates an inherent disadvantage for women in disseminating knowledge.
3. Role of Network Structure:
   • The gap in the cumulative number of collaborators between men and women plays a significant role in gender disparities. Women having fewer coauthors than men exacerbates their slower rate of knowledge acquisition and diffusion.

Null Model Analysis

To isolate factors contributing to gender disparities, the authors used three null models altering gender assignment while preserving different structural aspects:

1. Balance Model: Equalizes the number of men and women consistently over time and equalizes coauthor distributions. This model shows no unfairness, validating the fairness metrics.
2. Late Arrival Model: Maintains the historical gender proportion while equalizing collaboration patterns. This model, surprisingly, shows no significant unfairness, indicating that the late entry itself is not the sole driver of disparities.
3. Coauthors Model: Preserves both different gender proportions and coauthor distributions over time. This model closely aligns with empirical data, highlighting that the large impact on unfairness comes from differences in network collaboration structure.

Implications and Future Directions

The findings stress the inadequacy of strategies focusing solely on increasing the number of women in STEM fields. Instead, the structural aspects of scientific collaboration need addressing. Policies should aim at:

1. Encouraging Diverse Collaborations: Initiatives should foster wider collaboration networks for female scientists.
2. Preventing Career Dropout: Providing support mechanisms to reduce the attrition of women from scientific careers.
3. Promoting Academic Mobility: Encouraging and facilitating geographic and institutional mobility for women, which could help expand their collaborative networks and mitigate existing structural barriers.

Conclusion

This study significantly contributes to understanding the complex nature of gender disparities within scientific collaboration networks. It underscores that achieving gender parity in academia involves more than just numeric representation; it requires a nuanced approach to dismantle structural inequities embedded in scientific collaboration networks. Future developments in AI and network analysis tools could further refine these insights, offering more tailored interventions to promote fairness and equity in scientific communities.