Matching as Nonparametric Preprocessing for Reducing Model Dependence in Parametric Causal Inference

Kosuke Imai Department of Politics, Princeton University, Princeton, NJ 08544 e-mail: kimai{at}princeton.edu Gary King Department of Government, Harvard University, 1737 Cambridge Street, Cambridge, MA 02138 Elizabeth A. Stuart Departments of Mental Health and Biostatistics, Johns Hopkins Bloomberg School of Public Health, 624 North Broadway, Room 804, Baltimore, MD 21205 e-mail: estuart{at}jhsph.edu e-mail: king{at}harvard.edu (corresponding author) Although published works rarely include causal estimates frommore than a few model specifications, authors usually choosethe presented estimates from numerous trial runs readers neversee. Given the often large variation in estimates across choicesof control variables, functional forms, and other modeling assumptions,how can researchers ensure that the few estimates presentedare accurate or representative? How do readers know that publicationsare not merely demonstrations that it is possible to find aspecification that fits the author's favorite hypothesis? Andhow do we evaluate or even define statistical properties likeunbiasedness or mean squared error when no unique model or estimatoreven exists? Matching methods, which offer the promise of causalinference with fewer assumptions, constitute one possible wayforward, but crucial results in this fast-growing methodologicalliterature are often grossly misinterpreted. We explain howto avoid these misinterpretations and propose a unified approachthat makes it possible for researchers to preprocess data withmatching (such as with the easy-to-use software we offer) andthen to apply the best parametric techniques they would haveused anyway. This procedure makes parametric models producemore accurate and considerably less model-dependent causal inferences. Authors' note: Our thanks to Dan Carpenter and Jeff Koch fordata; Alberto Abadie, Neal Beck, Sam Cook, Alexis Diamond, BenHansen, Guido Imbens, Olivia Lau, Gabe Lenz, Paul Rosenbaum,Don Rubin, and Jas Sekhon for many helpful comments; and theNational Institutes of Aging (P01 AG17625-01), the NationalInstitute of Mental Health (MH066247), the National ScienceFoundation (SES-0318275, IIS-9874747, SES-0550873), and thePrinceton University Committee on Research in the Humanitiesand Social Sciences for research support. Software to implementthe methods in this paper is available at http://GKing.Harvard.edu/matchitand a replication data file is available as Ho et al. (2006).     

Sex determination from os sacrum by postmortem CT

Sex determination in forensic practice is performed mostly on sexually dimorphic bones, including pelvic bones such as the os sacrum. Postmortem CT scan provides an easy and fast method for depicting and measuring bone structures prior to elaborate autopsy preparations. To develop a simple and objective method for sex determination in postmortem CT, metric data were evaluated from CT images of the pelvic-associated os sacrum of 95 corpses (49 men and 46 women) from the Canton of Bern, Switzerland. Discriminant function analysis of the data showed that the best accuracy in determining sex was 76.8% and 78.9% with two different observers. It is concluded that measuring the os sacrumin postmortem CT for sex determination has moderate accuracy and should only be applied in combination with other methods.     

A Minimally Invasive Technique for the Detection and Analysis of Pulmonary Fat Embolism: A Feasibility Study*

We investigated the feasibility of postmortem percutaneous needle biopsy (PNB) for obtaining pulmonary samples adequate for the study of pulmonary fat embolism (PFE). Samples of both lungs were obtained from 26 cadavers via two different methods: (i) PNB and (ii) the double-edged knife technique, the gold standard at our institute. After water storage and Sudan III staining, six forensic pathologists independently examined all samples for the presence and severity of PFE. The results were compared and analyzed in each case regarding the vitality of the PFE and its relationship to the cause of death. The results showed that PFE was almost identically diagnosed and graded on the samples obtained via both methods. The discrepancies between the two techniques did not affect the diagnoses of vitality or cause of death related to PFE. This study demonstrates the feasibility of the PNB sampling method for the diagnosis and interpretation of PFE in the postmortem setting.     

The effect of post-identification feedback, delay, and suspicion on accurate eyewitnesses

We examined whether post-identification feedback and suspicion affect accurate eyewitnesses. Participants viewed a video event and then made a lineup decision from a target-present photo lineup. Regardless of accuracy, the experimenter either, informed participants that they made a correct lineup decision or gave no information regarding their lineup decision. Immediately following the lineup decision or after a 1-week delay, a second experimenter gave some of the participants who received confirming feedback reason to be suspicious of the confirming feedback. Following immediately after the confirming feedback, accurate witnesses did not demonstrate certainty inflation. However, after a delay accurate witnesses did demonstrate certainty inflation typically associated with confirming feedback. The suspicion manipulation only affected participants' certainty when the confirming feedback created certainty inflation. The results lend support to the accessibility interpretation of the post-identification feedback effect and the erasure interpretation of the suspicion effect.     

Pi Alpha Alpha activities and effectiveness: results from a national survey of chapter advisors

Pi Alpha Alpha (PAA) is the national academic honor society for public administration. This study presents the results of a comprehensive survey of PAA Chapter advisors focusing on inductee characteristics, induction requirements, chapter activities, chapter officers, and suggestions for improving PAA chapter effectiveness. NASPAA sponsored the survey and provided relevant logistical support. The results indicate that PAA is meeting its major objectives, but several areas require attention including enhancing the activity mix, improving recruitment practices for chapter officers, and a dedicated means for disseminating information on chapter activities through a newsletter and a chapter directory.