Explaining variation in evidence-based policy making in the American States

Dylan L. Yingling, Daniel J. Mallinson

Research output: Contribution to journalArticlepeer-review

7 Scopus citations


Background: Though evidence-based policy (EBP) has attracted considerable attention from the public, academics, and governments, prior studies have revealed little about how political parties, institutions, and policy context shape the adoption and implementation of these policies in the American states. Aims and objectives: Develop objective criteria for measuring these policies, as well as a hierarchy which describes the features that make some policies more advanced. This paper presents the first comprehensive study on EBP in the American states. Methods: Using assessments by the Pew and MacArthur foundations to measure EBP in the states for four topics: criminal justice, juvenile justice, behavioural health, and child welfare. Assess the relationship between EBP use and state political and institutional factors. Results: Democratic governors, Republican legislatures, state innovativeness are significant predictors of EBP engagement. Discussion and conclusions: This research makes a substantial contribution to the study of EBP and opens new avenues for future research on the political, cultural, and institutional factors that influence EBP adoption and implementation. In an era of extreme partisanship, our study finds that EBP is a policy niche where actors and institutions across political parties use research evidence to inform effective and efficient policies in ways that maximise the electoral incentives that such policies can offer.

Original languageEnglish (US)
Pages (from-to)579-596
Number of pages18
JournalEvidence and Policy
Issue number4
StatePublished - Apr 2020

All Science Journal Classification (ASJC) codes

  • Social Sciences (miscellaneous)


Dive into the research topics of 'Explaining variation in evidence-based policy making in the American States'. Together they form a unique fingerprint.

Cite this