The Gatekeepers of Academia: Investigating Bias in Journal Publication Across Topics, Author Backgrounds, and Institutions

Abstract

Building on evidence of the “gatekeeper” effect in a publish or perish culture whereby faculty authors have a more difficult time publishing some studies compared to others, the present study investigates potential biases based on the content of the study. We explore potential relationships with the journal quality (i.e., do higher quality journals publish more liberal or conservative leaning content?), author backgrounds, and institutional characteristics. Using a dataset of over 20 000 article titles and abstracts on 12 controversial topics (e.g., DEI training, gun control), we applied both human and GPT-supported coding to assess bias. Results suggest a slight liberal bias across topics, such that more liberal articles were published, with notable differences based on the topic. Interestingly, journal impact factors and author demographics had only weak relationships with bias ratings. Our findings challenge the assumption that political bias heavily influences academic publishing, suggesting instead that the focus rightly remains on research quality. However, the study highlights the need for further exploration of factors like author self-censorship and editorial gatekeeping that could influence the publication of controversial research.