Answers are grounded in this paper and cite their sources.
Modern program evaluation theory posits a rigorous impact evaluation tollgate. Programs are developed and tested via rigorous impact evaluation (often, but not exclusively, random assignment). Programs that pass this tollgate proceed to broader rollout. Results from implementing this theory are disappointing. Few programs pass the tollgate, raising the question: What now? Refine the program or abandon it? Epstein and Klerman ( ) posit that piloting and verifying the intermediate steps of the logic model can screen out programs unlikely to pass the rigorous impact evaluation tollgate. Again, that proposal raises the question: How should we proceed when a program fails to satisfy its own logic model? Refine the program or abandon it? The chapter presents several perspectives on this question and compares them against the Social Innovation Fund's experience with one of its grantees.
No takes yet. Share an insight, caveat, or question.
Epstein et al. (2016) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: