This is a great summary. For a good few years in an impact role, I was worried I hadn't come to my own settled view on how evaluation should be done. But I have gradually come to same view as this last point: "Methods are only as good or bad as their appropriateness for answering particular questions and in relation to how well they are able to understand the primary phenomena under study."
I came up with a top 10 of what's wrong with RCTs (and experiments in general). It's really a note-to-self for the next time I hear the words "gold standard" or "evidence hierarchy." I'll follow this up with a discussion of some recent randomista pivots. #evaluation #learning #rct cc. Thomas Delahais Andi Fugard Alan Hudson Jonathan Said Niki Wood https://lnkd.in/eQ3J49ZW