Introduction to Updating (Registrations)
Topics
Updates And The Registration Process
Contributor Permissions And Updating
Updating: What is it
Updating registrations is the process of transparently reflecting necessary changes to a study design. The changes should be implemented only to reflect events outside your control or include unexpected anomalies.
Why updating is important
Preregistration is one of the most important ways to improve the credibility of research findings. For quantitative studies, it makes a clear distinction between planned and unplanned analyses, thus eliminating data-contingent decisions. For qualitative studies, it communicates hypotheses before reviewing the data (ie. transparency) and transparency disclose any deviations. In both instances, there is a clear association between assumptions, data, and data-based decisions. This reduces the likelihood of potentially unethical behaviors such as p-hacking. However, preregistration is not flexible to account for unanticipated anomalies, such as the COVID-19 pandemic. In other words, “preregistrations should be a plan, not a prison”, and researchers should have the ability to update preregistrations in a concise, transparent manner.
Alex DeHaven. Preregistration: A Plan, Not a Prison. COS: Center for Open Science. Https://Www.Cos.Io/Blog/Preregistration-Plan-Not-Prison
When you should update
You should update your preregistration whenever there is an unexpected change to your study design. Make sure you include a thorough explanation as to why the change was necessary, including:
- Description of the change
- Rationale for the change
- Impact on the study
| Example 1 | Example 2 | Example 3 |
|---|---|---|
The sampling location was changed from a depth of 50m to 70m because the population of interest was not present at the expected site. No expected changes on the study results. |
Incentives changed from $20 gift cards to $50 gift cards as the team was unable to obtain the pre-specified sample size. No expected changes on the study results. |
The analysis changed from an ANOVA to the Kruskal-Wallis test as the collected data did not meet the assumptions of the planned analyses. No expected changes on the study results with this more conservative test. |
Updates and the registration process
Updates should be created after a thoroughly completed preregistration is submitted. The chart below illustrates where updating should occur during the registration process.
Contributor permissions and updating
Actions available during the updating workflows depend on your level of permission as a contributor on the registration. Contributors cannot be added or removed nor can permissions be changed after the original registration was submitted.
| Action | Admin Contributor | Write Contributor | Read Contributor |
|---|---|---|---|
| Starting an update | ✓ | ||
| Editing an update | ✓ | ✓ | |
| Viewing an update | ✓ | ✓ | ✓ |
| Submitting an update for approval from Admin contributors | ✓ | ||
| Approve/Reject an update | ✓ | ||
| Return an update to a draft for more editing | ✓ |
This Article Is Licensed Under CCO For Maximum Reuse.
