A Critical Perspective on 50 Years of Evaluation Policy: Imagining a Framework for Better, Cheaper, Faster, Usable Evaluation

ABSTRACT National evaluation policy in the United States has substantially improved over the past half century, yet honest assessment requires acknowledging where fundamental challenges persist. This article provides a policy perspective on structural failures in evaluation policy design, drawing on the author's experience helping lead the U.S. Commission on Evidence‐Based Policymaking, negotiating the Evidence Act's passage, and designing implementation frameworks. Using Lee Cronbach's 95 theses as an organizing framework, the article examines six failures in national evaluation policy: pretending evaluation isn't political, aspiring to evaluate everything instead of prioritizing strategically, obsessing over internal validity at the expense of external validity, building dependency rather than sustainable capacity, failing to build knowledge translation infrastructure, and treating evaluation as judgment rather than learning. These failures share a common thread: evaluation policy was designed for researchers' preferences rather than decision‐makers' needs. The article argues that neither political party deserves a gold star on evaluation policy implementation, and that hyperpoliticization of data and evaluation is a bipartisan and longstanding challenge. Looking ahead, the article proposes five adjustments for a future evaluation policy framework oriented toward better, faster, cheaper, and more usable evaluation — including honest acknowledgment of politics, strategic prioritization, increased focus on core components, internal capacity building, and alignment of cost, timeline, and quality.