Is it possible to "evaluate an unplanned exploratory test as being better than a planned confirmatory test"? It seems to me like they are different types of test, and hard to measures on the same scale (precisely because the extent of the unknown unknowns in the case of exploratory tests)
For me, yes, I think it’s possible to evaluate a test as good or bad based on its scientific arguments rather than the timing of the decision about conducting it (there are unknown unknowns in both confirmatory and exploratory). But I know others think confirmatory and exploratory tests are quite different. So, I get why they feel preregistration is useful to help distinguish them.
Is it possible to "evaluate an unplanned exploratory test as being better than a planned confirmatory test"? It seems to me like they are different types of test, and hard to measures on the same scale (precisely because the extent of the unknown unknowns in the case of exploratory tests)
For me, yes, I think it’s possible to evaluate a test as good or bad based on its scientific arguments rather than the timing of the decision about conducting it (there are unknown unknowns in both confirmatory and exploratory). But I know others think confirmatory and exploratory tests are quite different. So, I get why they feel preregistration is useful to help distinguish them.
interesting - thanks! i think i need to watch the videos to be able to usefully comment further
The panel sessions with Chris and Stephan are also good if you have the time.