Pre-registering a study

Definition

A pre-registration is a research plan that is time-stamped, immutable (i.e., read-only), created before the study is carried out, and deposited in an open repository where other people can access it freely. It usually includes a study plan (i.e., research question, objectives, hypothesis, data collection procedures, variables considered, etc.) and an analysis plan (i.e., confirmatory and exploratory data analyses that will be performed, inference criteria, etc.).

Platforms

There exist many platforms where you can pre-register a study. Some of them are the Open Science Framework (OSF), Zenodo, AsPredicted, and GitHub. More information on the benefits and limitations of each platform can be found in this paper:

Haroz, S. (2022, February 24). Comparison of preregistration platforms. MetaArXiv. https://doi.org/10.31222/osf.io/zry2u.

Example

Here is an example of a pre-registration done by lab members: https://osf.io/yem9x

A screenshot of an OSF page showing the first paragraphs of a pre-registration from the lab

Why pre-register studies?

• Pre-registration is considered a good research practice that limits the use of questionable research practices (QRP) such as p-hacking and HARKing. This is all the more important in the context of the current replication crisis in psychology.

• Pre-registration allow researchers to separate confirmatory from exploratory data analyses.

Pre-registration Benefits

• Pre-registering a study minimizes natural biases and selective reporting,

• It pushes you to think through everything clearly, including:

  • Theory

  • Hypotheses

  • Data collection plan (target sample size, inclusion criteria)

  • Data analysis plan,

• It is a great resource to start new collaborations,

• Pre-registering a study adds weight to the final claims of the study,

• It benefits the field by transparently showing how the research process unfolded.

When pre-registration might not be the best fit

If you're doing an exploratory data analysis with no firm hypotheses, then a pre-registration might not be appropriate. Pre-registrations require firm predictions, so if a study is more about creating an initial understanding of a messy, unexplored research area, a pre-registration might be premature. For example, looking at the data patterns of a newly-designed questionnaire, using qualtitative measures to understand a new construct, etc.

Suggested articles to expand on pre-registration

Havron, N., Bergmann, C., & Tsuji, S. (2020). Preregistration in infant research—A primer. Infancy, 25(5), 734–754. https://doi.org/10.1111/infa.12353

Nosek, B. A., Ebersole, C. R., DeHaven, A. C., & Mellor, D. T. (2018). The preregistration revolution. Proceedings of the National Academy of Sciences, 115(11), 2600–2606. https://doi.org/10.1073/pnas.1708274114

Reich, J. (2021). Preregistration and registered reports. Educational Psychologist, 56(2), 101–109. https://doi.org/10.1080/00461520.2021.1900851