Journal of Human Resources · 2026 · Dalia Ghanem、Sarojini Hirshleifer、Karen Ortiz‐Becerra
中文摘要
我们将带有基线数据的实地实验中的样本流失问题, 作为面板模型中的一个识别问题来处理。对已有文献的系统性综述表明, 学界对如何检验流失偏误尚未形成共识。我们分别针对受访者和研究总体的处理效应确立了识别假设, 并提出了检验其严格含义的程序。随后, 我们将所提出的检验方法与当前的实证研究实践联系起来, 并证明文献中最常用的检验方法通常并非对内部有效性的一般性检验。我们利用若干实证应用说明了本文分析的相关性。
Abstract
<h3>Abstract</h3> We approach attrition in field experiments with baseline data as an identification problem in a panel model. A systematic review of the literature indicates that there is no consensus on how to test for attrition bias. We establish identifying assumptions for treatment effects for both the respondents and the study population, and propose procedures to test their sharp implications. We then relate our proposed tests to current empirical practice, and demonstrate that the most commonly used test in the literature is not a test of internal validity in general. We illustrate the relevance of our analysis using several empirical applications.