One of the aims of the Open Research movement is to make the research process more transparent, accountable and reproducible.
Writing a Registered Report has been proposed as a way of eliminating publication bias from the scholarly communication process. Usually studies reporting positive results are easier to get published than those that report negative or inconclusive ones. This publication bias can skew research priorities in some subjects, lead to selective reporting of results and possibly the changing of hypotheses in order to fit unexpected results.
What is a Registered Report?
The Registered Report format is suitable for all areas of science where there might be the potential for bias. The author writes the intended methods and proposed analyses that they will use in their study in the Registered Report format. The protocols are then peer reviewed prior to the research being conducted (Stage 1). After the proposed study has passed peer review, the study begins. Once the author has completed the study, they write the results and discussion sections of the article and then resubmit the article for Stage 2 peer review. If the authors can show that they conducted their study according to the initial protocols, the publication of the article should be straight forward. The advantages of the Registered Report format are that the study should be much easier to replicate by others and, as the journal commits to publishing the report regardless of the outcome, there is no possibility of the methods or analyses being changed to alter the final results of the study.
Many journals in several disciplines now include the Registered Report format as an article type. According to the Center for Open Science, there are now over 300 journals that include Registered Reports in their list of acceptable formats. The subject range includes journals on psychology, neuroscience, linguistics, biology, ecology, oncology, veterinary science, computer science, nutrition, sociology, medicine and education. Several publishers also provide guidelines for authors, including Wiley, Royal Society, Taylor & Francis and BioMed Central.
What do researchers think of the Registered Report initiative? I spoke to Dr Jayne Morriss who has used the Registered Report format several times in her work on the neurobiology of anxiety.
Please introduce yourself and tell us a little bit about your research
I’m a senior postdoctoral research fellow based at the School of Psychology and Clinical Language Sciences at the University of Reading. My research focuses on the neurobiology of anxiety.
Why did you choose to structure your work as a registered report?
There are multiple advantages to a registered report. In particular, with a registered report the timeline of the research project is clear, and it forces the researcher to be up front about their hypotheses and analysis plan. Furthermore, the research is published on the basis of sound hypotheses rather than whether the results are ‘positive’ or not.
Was it difficult to write the registered report?
It is a lot easier to write a registered report because there is a clear timeline of events i.e. write the introduction, hypotheses and analysis plan at stage 1 and the results and discussion at stage 2.
What benefits do you think there are to declaring what you are going to do in your study before you actually start collecting the data?
You can receive useful feedback from reviewers before starting the project. Thus, the rationale and design of the study tends to be more detailed and thorough.
Was it difficult to stick to the protocol that you had registered? Did it cause any issues in the conduct of the research?
Not at all. Although, if you notice potential problems with the design while piloting the experiment, you can go back to the editor and recommend further changes to the protocol. This works as long as you haven’t collected/started the real experiment yet.
Did having the registered report make it easier or harder to write up the final paper?
It was much easier to write up the final paper because the introduction and method had already been finalised at stage 1. Moreover, reviewers can only comment on the results/discussion at stage 2.
What was the peer review like on the final paper?
As the method had already been scrutinised in advance, the peer review on the final paper was much quicker and more streamlined.
Would you publish a registered report again?
Yes, I have published three registered reports so far in different journals.
Do you have any advice for others thinking of doing the same?
The process of preparing a registered report is worth learning. If you have a replication study or a ‘risky’ study, I would recommend the preregistered report route. Another thing to remember is that there is actually some flexibility with a registered report. For example, say you realise that there may be an alternative way of analysing the results, you can do these analyses if you state that it was posthoc and not part of the registered analysis plan. Posthoc tests can be added in the results section under the planned analyses or in a supplementary material.
The Registered Reports written by Dr Morriss are listed below.
Morriss, J., Biagi, N. and Dodd, H. (2020) Your guess is as good as mine: a registered report assessing physiological markers of fear and anxiety to the unknown in individuals with varying levels of intolerance of uncertainty. International Journal of Psychophysiology, 156. pp. 93-104. ISSN 0167-8760 doi: https://doi.org/10.1016/j.ijpsycho.2020.07.009
Morriss, J., Wake, S., Lindner, M., McSorley, E. and Dodd, H. (2020) How many times do I need to see to believe? The impact of intolerance of uncertainty and exposure experience on safety-learning and retention in young adults. International Journal of Psychophysiology, 153. pp. 8-17. ISSN 0167-8760 doi: https://doi.org/10.1016/j.ijpsycho.2020.04.012
Currently at stage 2 of review:
Wake, S. J., Dodd, H., & Morriss, J. (2020, March 30). Intolerance of uncertainty and novelty facilitated extinction: The impact of reinforcement schedule. PsyArXiv Preprints. doi: https://doi.org/10.31234/osf.io/7qgrh