Understanding Gender Gaps in Academic Science through the Lens of Topic Fit: An Interdisciplinary Investigation Using a Big Data Approach

自20世纪60年代以来，妇女参与科学的人数稳步增加。然而，在世界各地的学术科学中，无论是在研究成果的数量上还是在研究成果的影响上，性别差异仍然存在。在2000年之后的十年里，男性研究人员平均比女性研究人员发表的论文多35%，被引用的论文多34%。在美国，这些性别差距甚至比全球平均水平还要大。了解研究中持续存在的性别差距的原因对于改善学术机构的性别平等具有重要意义。目前的项目是一项跨学科的努力，旨在解决这个问题，并促进多个领域的知识进步，包括心理学、社会学、组织行为学、人力资源管理和科学计量学。它将提出一种整合现有观点的新理论，并使用大数据方法对其进行测试。具体地说，该项目调查主题匹配度(即任何新合作与研究人员现有工作之间主题相似或一致的程度)，作为理解研究生产率和影响中的性别差距的关键机制。根据社会角色理论，性别角色规范和期望可能导致女性和男性学者与初级研究人员的合作模式不同，适合女性的主题水平较低，这反过来将解释女性生产力的损失。研究人员正在分析世界各地9000多万研究人员的出版记录，使用新方法计算主题匹配指数，并测试这些指数作为解释性别差距的关键机制，使用个体研究人员层面的路径分析和多水平建模来解释数据的嵌套结构。重要的是，该项目还考察了不同学科之间的差异，并分析了一段时间内的趋势。这一奖项反映了NSF的法定使命，并通过使用基金会的学术价值和更广泛的影响审查标准进行评估，被认为值得支持。

The participation of women in science has steadily increased since the 1960s. However, gender disparities continue to exist in academic science around the world, both in the quantity and in the impact of research output. In the decade after 2000, male researchers, on average, published 35% more papers and received 34% more citations than female researchers. These gender gaps are even wider in the United States than the global average. Understanding causes of persistent gender gaps in research is of grave significance for improving gender equality in academic institutions. The current project is an interdisciplinary effort to address this question and advance knowledge across multiple fields, including psychology, sociology, organizational behavior, human resource management, and scientometrics. It will propose a novel theory that integrates existing perspectives and test it using a big data approach.Specifically, the project investigates topic fit (i.e., the degree of subject topic similarity or congruence between any new collaboration and a researcher’s existing work) as a key mechanism for understanding gender gaps in research productivity and impact. Drawing upon Social Role Theory, it is theorized that gender role norms and expectations may lead female and male scholars to have different collaboration patterns with junior researchers and lower levels of topic fit for women, which, in turn, would explain a productivity penalty for women. The investigators are analyzing publication records of over 90 million researchers around the world, computing topic fit indices using novel methods, and testing these indices as a critical mechanism for explaining gender gaps using both path analysis at the individual researcher level and multilevel modeling to account for the nested structure of the data. Importantly, the project also examines differences across academic disciplines and analyzes trends over time. Findings can inform institutional policies and practices for advancing gender equality in academic science.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.