{"title":"Using a Satisficing Model of Experimenter Decision-Making to Guide Finite-Sample Inference for Compromised Experiments.","authors":"James J Heckman, Ganesh Karapakula","doi":"10.1093/ectj/utab009","DOIUrl":null,"url":null,"abstract":"<p><p>This paper presents a simple decision-theoretic economic approach for analyzing social experiments with compromised random assignment protocols that are only partially documented. We model administratively constrained experimenters who satisfice in seeking covariate balance. We develop design-based small-sample hypothesis tests that use worst-case (least favorable) randomization null distributions. Our approach accommodates a variety of compromised experiments, including imperfectly documented re-randomization designs. To make our analysis concrete, we focus much of our discussion on the influential Perry Preschool Project. We reexamine previous estimates of program effectiveness using our methods. The choice of how to model reassignment vitally affects inference.</p>","PeriodicalId":50555,"journal":{"name":"Econometrics Journal","volume":null,"pages":null},"PeriodicalIF":2.9000,"publicationDate":"2021-05-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC8478285/pdf/nihms-1690015.pdf","citationCount":"3","resultStr":null,"platform":"Semanticscholar","paperid":null,"PeriodicalName":"Econometrics Journal","FirstCategoryId":"96","ListUrlMain":"https://doi.org/10.1093/ectj/utab009","RegionNum":4,"RegionCategory":"经济学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"2021/6/29 0:00:00","PubModel":"Epub","JCR":"Q1","JCRName":"ECONOMICS","Score":null,"Total":0}
引用次数: 3
Abstract
This paper presents a simple decision-theoretic economic approach for analyzing social experiments with compromised random assignment protocols that are only partially documented. We model administratively constrained experimenters who satisfice in seeking covariate balance. We develop design-based small-sample hypothesis tests that use worst-case (least favorable) randomization null distributions. Our approach accommodates a variety of compromised experiments, including imperfectly documented re-randomization designs. To make our analysis concrete, we focus much of our discussion on the influential Perry Preschool Project. We reexamine previous estimates of program effectiveness using our methods. The choice of how to model reassignment vitally affects inference.
期刊介绍:
The Econometrics Journal was established in 1998 by the Royal Economic Society with the aim of creating a top international field journal for the publication of econometric research with a standard of intellectual rigour and academic standing similar to those of the pre-existing top field journals in econometrics. The Econometrics Journal is committed to publishing first-class papers in macro-, micro- and financial econometrics. It is a general journal for econometric research open to all areas of econometrics, whether applied, computational, methodological or theoretical contributions.