Date of Award

Spring 1-1-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Political Science

First Advisor

Sävje, Fredrik

Abstract

This dissertation demonstrates a demand-driven approach to political methodology research. The demand-driven approach entails studying and addressing technical problems while remaining in dialogue with political science as actually practiced. This dissertation, in particular, focuses on conducting credible research in politics and policy amid problems related to causal inference, statistical power, and data limitations. Chapter 2, "How Much Should We Trust Modern Difference-in-Differences Estimates?," takes up the challenge of estimating policy effects in observational settings. In particular, this chapter asks how well a range of difference-in-differences (DID)-style methods work for the empirical settings and sample sizes commonly used in political science, with the U.S. states as the running example. It provides a simulation study of the performance of seven DID methods---including the two-way fixed effects estimator and six heterogeneity-robust modern methods---under either constant or heterogeneous effects, in an N=50 setting that mimics the American federalism natural experiment. I find that many modern methods (1) produce confidence intervals that do not include the true average effect at the specified rate and (2) are underpowered. I show that many cases of coverage problems with modern DID estimators can be addressed using the block bootstrap to estimate standard errors. However, I also show that large average effect sizes---at least 0.5 standard deviations---may be needed for detect effects in this sample. I illustrate this challenge in the case of estimating the effects of strict voter identification laws on turnout. Chapter 3, "Beyond Retraumatization: Toward Trauma-Informed Political Science," responds to two trends in political science research. The first is the widespread inclusion of retraumatization as a concern in research ethics guides for political scientists and the second is the rise in human subjects research on sensitive topics, from political violence to abortion policy. In the chapter, I clarify what is at stake when we talk about research participant distress and develop a new framework for trauma-informed political science research in different contexts, which details best practices abound informed consent, debriefing, and more. Chapter 4, "Inducing Emotions in Online Surveys" (coauthored with Ekin Dursun), addresses the problem of weak treatments in studies of the effect of emotions on policy attitudes. The validity of such experiments depends on the success of the emotional manipulations. If manipulations do not succeed in evoking the target emotion, experiments can produce misleading null results. If they evoke some other emotion, significant results can also be misleading. This chapter first argues that we should view emotions as intermediate outcomes of emotion inductions, then presents the results of two experiments evaluating the effectiveness of vignettes, images, autobiographical emotional memory tasks, and self-statement exercises for inducing cynicism, gratitude, and anger. We find that vignettes are broadly reliable.

Download

Share

COinS