↓ 跳过正文

Research Design

Making Policy Evidence Traceable: A Modular Preregistration Framework for Policy Evaluation

PSJ 54/4 2026-10-06 Original Making Policy Evidence Traceable: A Modular Preregistration Framework for Policy Evaluation Yiyun Xia, Tianhao Zhao, Yuhao Ba Abstract ABSTRACT Preregistration is increasingly used in policy research, but its value depends on whether preregistered commitments can be clearly linked to reported analyses. Policy studies often involve multiple outcomes, evolving administrative data, consequential design and analytic choices, and adaptive implementation, making it difficult to distinguish prespecified tests from subsequent adjustments. This research note proposes a modular, policy‐oriented companion framework for using existing preregistration infrastructure. The framework combines a shared core that defines the estimand, outcomes, and identification strategy with design‐specific modules for field experiments, administrative‐data quasi‐experiments, and behavioral policy trials. Together, the components serve two purposes: claim traceability, by linking reported results to the original plan, and governed adaptation, by providing a transparent record of changes made during the study. The aim is not to prevent adaptation, but to show which analyses were prespecified, which were revised through documented amendments, and which were developed later. This approach aims to improve the interpretability and transparency of policy research while remaining feasible in applied policy settings.

Heterogeneous Treatment Effects and Causal Mechanisms

APSR — pp. 1-18 2026-04-06 Original Heterogeneous Treatment Effects and Causal Mechanisms JIAWEI FU, TARA SLOUGH Abstract The credibility revolution advances the use of research designs that permit the identification and estimation of causal effects. However, understanding which mechanisms produce measured causal effects remains a challenge. The dominant current approach to the quantitative evaluation of mechanisms relies on the detection of heterogeneous treatment effects (HTEs) with respect to pretreatment covariates. This article develops a framework to understand when the existence of such HTEs can support inferences about the activation of a mechanism. We show first that this design cannot provide evidence of mechanism activation without additional, generally implicit, exclusion assumptions. Further, even when these assumptions are satisfied, the presence of HTEs supports the inference that the mechanism is active but the absence of HTEs is generally uninformative about mechanism activation. We provide novel guidance for interpretation and research design in light of these findings.