Awhile ago, I wrote a Broadstreet post on the use of PAPs with HPE. Since then, I’ve thought a bit more about this, as part of a chapter I’m working on for my book on historical data. Preregistration is a foundation of experimental work, but is also being suggested for