Solving the “file drawer problem”: How researchers, institutions, publishers and funders can reduce publication bias

Abstract

In 1979, psychologist Robert Rosenthal proposed that “journals are filled with the 5% of the studies that show Type I errors, while the file drawers are filled with the 95% of the studies that show nonsignificant results” (1). Despite being identified 45 years ago, the “file drawer problem” persists today. Failure to publish null and negative results has inflated reported efficacy in social sciences as well as translational and clinical medicine. This, in turn, leads to misleading or exaggerated effect sizes in meta-analyses, as well as a risk that future studies will be underpowered or based on irrelevant or insignificant findings.

Why are researchers relegating their null and negative results to the file drawer? Investigators might be reluctant to submit work they think has a higher chance of being rejected. There is evidence from clinical medicine that reviewers are indeed more critical of manuscripts that report null results. Funding agencies and commercial interests can introduce bias into the types of research questions that are investigated and how these results are framed. Finally, the “publish or perish” culture endemic in academia has also been blamed for driving the preference for reporting positive or confirmatory results. 

Given the risk that publication bias poses to the advancement of knowledge, as well as the entrenched behaviours that drive it, what roles and responsibilities do researchers, institutions, funders and publishers have in addressing this problem? This workshop will bring experts from these fields together with participants to develop actionable recommendations for improving the dissemination of null and negative results and reducing publication bias. 

It aims to raise awareness of the persistent “file drawer problem” and its impact on scientific advancement, while exploring the roles and responsibilities of researchers, institutions, funders and publishers in addressing publication bias. 

Workshop structure:

The workshop will begin with short presentations from various stakeholders—researchers, institutions, funders, and publishers—on their perspectives and initiatives addressing publication bias against null and negative results. Participants will then engage in role-playing exercises, collaborative strategy development, and group presentations, culminating in a synthesis session to identify shared challenges. The workshop will conclude with a focus on actionable next steps that participants can implement in their professional contexts.

Target Audience:

This workshop is designed for anyone committed to improving the dissemination of null and negative results. We hope that attendees from the stakeholder groups mentioned above will participate.