Registered Reports: A Q&A with Chief Editor of Nature Human Behaviour

To help us understand more about registered reports and their role in the peer review process, we invited Chief Editor of Nature Human Behaviour, Stavroula Kousta to answer few of our burning questions.

What exactly are Registered Reports?
Registered Reports are an innovative article format that shift the emphasis from the results of research to the significance of the research question and the rigour of the methods. With Registered Reports, peer review happens in two stages: in the first stage, before any data collection or analysis has taken place, reviewers evaluate the significance and robustness of the research protocol, which consists of the introduction, methods, analysis plan and any pilot data. If the editors and reviewers are persuaded that the research protocol is strong and the work suitable for the journal in question, the article is accepted in principle for publication. The authors then collect their data (or analyse their data, if the dataset already exists) and write up the full paper, which is reviewed again. If the authors adhered to their protocol and their conclusions are supported by the data, the paper is formally accepted for publication, regardless of the novelty, direction, or conclusiveness of the results.

When were they originally introduced?
Registered Reports were originally introduced in 2013 at the journals Cortex and Perspectives in Psychological Science. They are now offered in some form by more than 170 journals, including a number of Springer Nature journals, such as Nature Human Behaviour; BMC Biology; BMC Ecology; BMC Medicine; Attention, Perception and Psychophysics; Cognitive Research: Principles & Implications.

What problems do they attempt to solve? Why do we need them?
Publication bias – the tendency of authors or journals to prioritize for publication positive, statistically significant findings – undermines the aims of science and its ability to self-correct.

One study estimated that more than 80% of the research published in the new millenium reports positive results (Fanelli, 2012). But such a high proportion of positive findings is improbable: several studies with null or negative findings go unpublished. As support for them amasses, scientific findings come to be accepted as facts. However, if only positive findings are published, the risk of these ‘facts’ being false is substantial.

Registered Reports were introduced to counteract the damaging effects of publication bias, by shifting the emphasis from the results of research to the questions that guide the research and the methods used to answer them: direction or conclusiveness of the results do not affect the decision to publish. Because of their emphasis on methodological rigour and the requirement that all analyses should be pre-specified, Registered Reports also help to neutralize several questionable research practices, such as p-hacking and HARK-ing (i.e. hypothesizing after the results are known).

What are their benefits?
Science is the quest for truth. However, over the past several decades, what is good for science and what is good for scientists have not been in alignment. As a result, the scientific enterprise is compromised. Publication continues to be the key vehicle for researcher recognition and career advancement. However, if publication decisions are based entirely on the results of research, rather than on the importance of the research question and the quality of the methods, the aims of science and what is required of scientists to succeed are necessarily at odds. Registered Reports help to align what is good for scientists with what is good for science, and promote rigour over statistical significance.

What kind of research are they suitable for?
Registered Reports are suitable for confirmatory research – that is, research that is driven by a priori hypotheses. They are not suitable for exploratory research – that is, research that explores potential relationships between variables a posteriori, without any commitment to pre-existing hypotheses. Exploratory research can be used to generate hypotheses, whereas confirmatory research (dis)confirms these hypotheses. Although Registered Reports are unsuitable for the former, they are invaluable for the latter, especially in fields where publication bias and questionable research practices have a high prevalence.

Stavroula Kousta is Chief Editor of Nature Human Behaviour, where she introduced Registered Reports in 2017. She was previously the Editor of Trends in Cognitive Sciences (2008–2013) and Senior Editor at PLOS Biology (2013-2016). Originally from Greece, Stavroula obtained a PhD in English and Applied Linguistics (psycholinguistics) from the University of Cambridge. She then did four years of post-doctoral research on the psychological and neural underpinnings of language and semantic knowledge at University College London.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.