Pre-Registration: It’s a Journey

[Excerpts taken from the preprint “Preregistration Is Hard, And Worthwhile” by Brian Nosek and others, posted at PsyArXiv Preprints]
“Preregistration of studies serves at least three aims for improving the credibility and reproducibility of research findings.”
“First, preregistration of analysis plans makes clear which analyses were planned a priori and which were conducted post hoc.”
“Second, preregistration enables detection of questionable research practices such as selective outcome reporting…or Hypothesizing After the Results are Known…”
“Third, preregistration of studies can reduce the impact of publication bias—particularly the prioritization of publishing positive over negative results—by making all planned studies discoverable whether or not they are ultimately published.”
“However, preregistration is a skill that requires experience to hone…Preregistration requires research planning and it is hard, especially contingency planning. It takes practice to make design and analysis decisions in the abstract, and it takes experience to learn what contingencies are most important to anticipate.”
“This might lead researchers to shy away from preregistration for worries about imperfection. Embrace incrementalism…Having some plans is better than having no plans, and sharing those plans in advance is better than not sharing them. With experience, planning will improve and the benefits will increase for oneself and for consumers of the research.”
“There are opportunities to accelerate that skill building. Study registries such as OSF and SREE, and decision tools such as Declare Design provide structured workflows to help researchers anticipate common decisions and provide guidance for articulating those decisions.”
“Other strategies for developing these skills include: (1) refining analysis plans by simulating data to practice making the decisions; (2) splitting the “real” data into exploratory and confirmatory subsamples and preregistering the analysis after viewing the exploratory subset; (3) drafting the study methods and results section in advance to help anticipate what should be done and how you will report it; and (4) submitting the plan as a Registered Report for peer review to get expert feedback on the plan.”
“Researchers can embrace the common back-and-forth between planned and unplanned (confirmatory and exploratory) research activities…The key role of preregistration is to clarify which is which.”
“Preregistration is a plan, not a prison…When deviations from the plan will improve the quality of the research, deviate from the plan.”
“Reporting deviations from the plan can be challenging…If possible, report what occurs following the original plan alongside what occurs with the deviations, and share the materials, data, and code so that others can evaluate the reported outcomes and what would have occurred with alternative approaches.”
To read more, click here.

Leave a Reply

Please log in using one of these methods to post your comment: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: