New study reveals editorial trends at top science journals

A new study analyzing over 110,000 manuscript submissions to two of the world’s leading science journals has revealed persistent patterns in editorial decisions that favor specific demographics, institutional affiliations, and research topics. Published in Science Advances, the research highlights how factors such as geographic origin, institutional prestige, group size, and subject matter influence the likelihood of a paper being accepted for publication. The study, led by researchers from the University of Vermont and the University of Colorado Boulder, examined anonymized submission data spanning five years, from 2016 to 2020—for Science and Science Advances, both published by the American Association for the Advancement of Science. These journals serve as major gatekeepers in the global scientific community, influencing which research receives attention, funding, and recognition. The data set, developed over several years, provides an unprecedented look into the inner workings of the peer-review process at elite science journals. Among the key findings, the study found that papers authored by larger teams, those with 10 or more contributors, were three times more likely to be accepted than those with fewer authors. This suggests that collaborative efforts may carry an implicit advantage in the current publishing landscape. Additionally, manuscripts with corresponding authors affiliated with the most prestigious institutions saw an acceptance rate of 11.6%, compared to just 3.4% for those from less renowned institutions. This disparity underscores the growing influence of institutional reputation in determining publication outcomes. Geographic differences also played a significant role. Papers with corresponding authors based in China were less than one-third as likely to be published as those with authors located in the United States or Canada. Even authors with Chinese names but based in the U.S. faced similar disadvantages, indicating systemic challenges beyond mere national origin. The study also noted that research on topics such as immunology and RNA, particularly during the early stages of the pandemic, was far more likely to be accepted than studies focusing on social sciences, including those examining the political and societal impacts of the virus. The editorial process itself appears highly selective. Over 80% of submissions to Science and more than 75% to Science Advances were rejected by in-house editors before reaching the peer-review stage. This means that only a small fraction of all submissions ever undergo the formal evaluation by external experts. According to the researchers, editors play a crucial role in filtering content, often making decisions based on preliminary assessments of relevance, novelty, and impact. The study also touched upon broader implications for the future of scholarly publishing. With the increasing use of artificial intelligence tools to generate scientific content, concerns about the quality and integrity of published research have grown. Researchers argue that maintaining rigorous editorial standards is essential to preserving trust in the scientific process. The availability of this anonymized dataset could help journals refine their selection criteria and address potential biases in the system. Looking ahead, the researchers emphasize the importance of transparency and ongoing scrutiny of editorial practices. As the volume of scientific output continues to grow, understanding the mechanisms that determine which work reaches the public will become increasingly vital. The study serves as a critical step toward demystifying the peer-review process and fostering a more equitable and inclusive scientific publishing environment.