Gender and Ethnic Disparities in Science Production and Dissemination

Abstract

There is constant under-representation of women and racial minorities in the research workforce. Past research has demonstrated considerable gender and ethnic disparities in conventional scientific outcomes including careers, funding, publications, and citations. However, addressing the issue of under-representation requires examining the disparities throughout the scientific pipeline from the process of knowledge production to their dissemination in online media. Furthermore, a sizable gap still remains in our understanding of the mechanisms behind these disparities, which can involve different actors and institutions. For instance, the gap in publication counts at leading journals may be caused by the submission volume or the acceptance rate, which in turn may be caused by meritorious factors or bias in review. Disentangling different mechanisms is thus essential for designing effective policy interventions to restore equality in science. In this dissertation, we utilize large-scale bibliometric data, including publicly available datasets and private peer-review data from leading academic journals, to examine the demographic disparities and explore their potential mechanisms in three key activities in science production and dissemination, including (1) ethnic disparity in acceptance rate at top biology journals, (2) ethnic disparity in author mentions in science news, and (3) gender differences in scholarly self-promotion on social media. We leverage novel computational techniques and statistical models to precisely measure the magnitude of the disparities. The fine-grained nature of these data also enables us to pinpoint the actors (e.g., journal editors or peer reviewers) and processes (e.g., publishing stage or dissemination stage) where the disparity is produced, and dig into the mechanisms leading to it. We demonstrate how this computational social science approach can provide insights into practical policy interventions to reduce disparity in science.