Selection Bias

This study asks whether respondents’ Big Five personality traits predict their propensity to complete surveys, and how such nonresponse could bias estimated links between personality and political behavior. Using a new data collection effort and replication data, the author reports that Agreeableness and Neuroticism are associated with systematic differences in survey completion and shows how these patterns can alter observed relationships between the Big Five and political behavior. The brief highlights a methodological caution for researchers who use very brief personality inventories in political science.

This paper asks whether a teacher’s measured value added (VA) depends on the VA of the teacher in the previous grade and how that dynamic affects decisions that rely on VA. Using administrative records for North Carolina classrooms, the authors estimate how last grade’s mean teacher VA relates to a teacher’s own VA and evaluate how standard VA approaches would perform under a dismissal policy that targets the bottom 5% of teachers.

The study asks whether feedback systems in markets with information asymmetries work because trustworthy sellers self-select into markets with feedback, or because sellers change behaviour for reputational reasons. Using a field experiment on taxi rides, the authors present an experimental design intended to disentangle selection from reputation mechanisms and discuss the implications for policy on feedback mechanisms.

Governments in the United States must provide free legal services to low-income people accused of crimes. Many jurisdictions rely on assigned counsel systems, where private attorneys represent indigent defendants on a contract basis. Defendants represented by assigned counsel are more likely to be convicted and incarcerated than defendants with privately retained attorneys.

This study asks whether renewable terms compromise judicial independence by creating incentives for judges to curry favor with actors who control their reappointment. Using a 2010 reform at the European Court of Human Rights (ECtHR) that made judges’ terms nonrenewable with immediate effect, the author estimates how removing reappointment opportunities changed judges’ voting behavior on cases involving their nominating states. The paper’s contribution is a focused causal test of the reappointment mechanism versus selection explanations for pro-state voting.

Not all police–civilian encounters appear in administrative datasets available to researchers. One common workaround is to measure the effect of race only among civilians who have already been detained by police, but this selection is made after the treatment (race) and so can distort causal conclusions.

This paper addresses concerns about a recent critique of the 2003 Peruvian Truth and Reconciliation Commission's (TRC) conflict mortality study. The critique attempts to provide an alternative estimate for deaths caused by the Maoist guerrilla group, Sendero Luminoso.

Time-series cross-sectional (TSCS) data, featuring repeated observations of units over time, is common in political science research. This study addresses a critical limitation: standard methods for estimating lagged treatment effects often produce biased results when selection on observables occurs post-treatment. Using potential outcomes as the foundation, the paper demonstrates how popular models like autoregressive distributed lags can be flawed due to conditioning on post-outcome variables.

Does WTO membership influence unrelated economic outcomes? This article replicates ninety-four specifications from sixteen different studies, revealing that the mere existence of a country in the WTO has statistically significant associations with approximately 34 percent of dependent variables, many lacking theoretical justification. Similarly high false positive rates emerge when examining membership in a low-impact environmental treaty (CITES). The authors argue this highlights a critical issue: without theoretically informed sensitivity analysis, seemingly robust findings can mask selection on unobservables.

High rerunning rates among incumbents and the dominance of two major parties in the United States let U.S.-based studies avoid the selection problem of who chooses to run again. In countries where rerunning is uncommon or parties are fluid, applying U.S. methods can produce sample selection bias and produce misleading estimates of incumbency effects.

International migrants are among the largest disenfranchised groups in modern democracies. This paper examines how giving noncitizens suffrage affected public policy, specifically education and social spending, across Swedish municipalities. The author argues that standard political competition models suggest a selection bias when measuring the impact of enfranchising noncitizens. Using unique data from Sweden's municipal-level introduction of voting rights for noncitizens, they find substantial increases in relevant government spending where noncitizens represented a significant portion of the electorate.

Recent work argues that democracies are generally more vulnerable to economic coercion than authoritarian regimes, yet most sanctions target nondemocratic states. This article reconciles that puzzle by arguing that authoritarians are resilient to sanctions most of the time—but become unusually vulnerable during specific "windows of opportunity" created by domestic instability.

International treaties face the challenge of selection bias in measuring their impact.