Publication bias is the tendency of authors and journals to prioritize for publication positive, statistically significant findings. It is a practice that undermines the ability of science to self-correct and incentivizes the adoption of questionable research practices. The publication process means that traditional workflows for peer review are also a form of post-mortem; it happens when the research project is complete, meaning that the authors have no option but to repeat the research if the project is identified as having been fundamentally flawed, which can lead to a significant waste of resources.
Registered reports are an innovative article format, introduced in 2013 and now offered in some form by nearly two hundred journals. They shift the emphasis from the results of research to the significance of the research question and the rigour of the methods. Peer review happens in two stages: first, before any data collection or analysis has taken place (Stage 1) and then again after the data have been collected or analysed (Stage 2). Acceptance in principle is offered at Stage 1 for methodologically robust protocols that meet the journal’s criteria for significance, whereas peer review at Stage 2 serves only to ascertain whether the authors adhered to their protocol. By offering acceptance in principle before the results are known, registered reports neutralize publication bias and incentivize robust scientific practices. Peer review at Stage 1 is also a much more collaborative process, in which reviewers play a significant role in shaping robust, reproducible research projects.
In this talk, I will discuss our experience with registered reports in the context of a highly selective journal - Nature Human Behaviour. I will highlight the opportunities and challenges that the format presents, as well as how it can be used to redefine what constitutes a significant scientific advance. I therefore argue that registered reports help us to align publishability with the scientific values of trustworthiness, reproducibility, and open science.
