Proposals for formalising evidence evaluation fascinate n disturb me. It's notable that a certain type of study strongly predicts policy success. But the key lesson from recent #philsci that evidence is only strong when you can *combine* different types of it into a robust theory of change.
Reposted from Kai Ruggeri
In a study of 100 policies & 250 policymakers, we find that a) there's no such thing as the "best" type of evidence, and b) the scale of evidence used in decisions can be a strong predictor of policy effectiveness.

Freely available to all @policysciences.bsky.social

Comments