Triangulating outcomes, timeframes, and estimation methods in quasi-experimental research funding evaluation: a case from Poland

Abstract This paper uses an example of the Lider, a prestigious research grant scheme for early-career researchers founded by the Polish National Centre for Research and Development, to show how the results of quasi-experimental outcome evaluation can vary depending on methodological choices. The study compares evaluation results based on (a) 12 outcome variables, (b) six timeframes of analyses, and (c) three quasi-experimental methods: difference-in-differences (DID), propensity score matching (PSM), and regression discontinuity design (RDD). The results show that conclusions based on various combinations of the abovementioned dimensions (a, b, and c) can differ extensively and that only selected effects persist across most variants. This is a warning signal for policymakers, policy analysts, and evaluators because decisions about the scope and methods of the study can impact the evaluation results significantly. Peer Review https://www.webofscience.com/api/gateway/wos/peer-review/10.1162/QSS.a.491