Search results

Our goal in this paper is to provide a formal explanation for how within-unit causal process information (i.e., data on posttreatment variables and partial information on posttreatment counterfactuals) can help to inform causal inferences relating to total effects—the overall effect of an explanatory variable on an outcome variable. The basic idea is that, in many applications, researchers may be able to make more plausible causal assumptions conditional on the value of a posttreatment variable than they would be able to do unconditionally. As data become available on a posttreatment variable, these conditional causal assumptions become active and information about the effect of interest is gained. This approach is most beneficial in situations where it is implausible to assume that treatment assignment is conditionally ignorable. We illustrate the approach with an example of estimating the effect of election day registration on turnout.

In this paper, we discuss an estimator for average treatment effects (ATEs) known as the augmented inverse propensity weighted (AIPW) estimator. This estimator has attractive theoretical properties and only requires practitioners to do two things they are already comfortable with: (1) specify a binary regression model for the propensity score, and (2) specify a regression model for the outcome variable. Perhaps the most interesting property of this estimator is its so-called "double robustness." Put simply, the estimator remains consistent for the ATE if either the propensity score model or the outcome regression is misspecified but the other is properly specified. After explaining the AIPW estimator, we conduct a Monte Carlo experiment that compares the finite sample performance of the AIPW estimator to three common competitors: a regression estimator, an inverse propensity weighted (IPW) estimator, and a propensity score matching estimator. The Monte Carlo results show that the AIPW estimator has comparable or lower mean square error than the competing estimators when the propensity score and outcome models are both properly specified and, when one of the models is misspecified, the AIPW estimator is superior.

AbstractWe establish the prevalence of partisan schadenfreude—that is, taking "joy in the suffering" of partisan others. Analyzing attitudes on health care, taxation, climate change, and the coronavirus pandemic, we find that a sizable portion of the American mass public engages in partisan schadenfreude and that these attitudes are most expressed by those who are ideologically extreme. Additionally, we find that a sizable portion of the American public is more likely than not to vote for candidates who promise to pass policies that "disproportionately harm" supporters of the opposing political party, and we demonstrate experimental evidence of demand/preference for candidates who promise cruelty among those who exhibit high amounts of schadenfreude. In sum, our results suggest that partisan schadenfreude is widespread and has disturbing implications for American political behavior.

Do the processes states use to select judges for peak courts influence gender diversity? Scholars have debated whether concentrating appointment power in a single individual or diffusing appointment power across many individuals best promotes gender diversification. Others have claimed that the precise structure of the process matters less than fundamental changes in the process. We clarify these theoretical mechanisms, derive testable implications concerning the appointment of the first woman to a state's highest court, and then develop a matched-pair research design within a Rosenbaum permutation approach to observational studies. Using a global sample beginning in 1970, we find that constitutional change to the judicial selection process decreases the time until the appointment of the first woman justice. These results reflect claims that point to institutional disruptions as critical drivers of gender diversity on important political posts.

Abstract
Does providing information about police shootings influence policing reform preferences? We conducted an online survey experiment in 2021 among approximately 2,600 residents of 10 large US cities. It incorporated original data we collected on police shootings of civilians. After respondents estimated the number of police shootings in their cities in 2020, we randomized subjects into three treatment groups and a control group. Treatments included some form of factual information about the police shootings in respondents' cities (e.g., the actual total number). Afterward, respondents were asked their opinions about five policing reform proposals. Police shooting statistics did not move policing reform preferences. Support for policing reforms is primarily associated with partisanship and ideology, coupled with race. Our findings illuminate key sources of policing reform preferences among the public and reveal potential limits of information-driven, numeric-based initiatives to influence policing in the US.