Research in Progress

• "Inference at the data's edge: Gaussian processes for estimation and inference in the face of model uncertainty, poor overlap, and extrapolation." (with Chad Hazlett and Doeun Kim) [under review] [draft] [software]
Abstract
Many inferential tasks require estimating uncertainty around predictions of an outcome at points not in the data. Conventional approaches choose and fit a model on the observed data then compute uncertainty estimates using that fit, producing uncertainty estimates that neglect the impacts of model uncertainty further from observed data. The Gaussian Process (GP) offers an attractive solution by estimating (in closed-form) a posterior distribution for the predicted values at test points of interest, combining information about model fit and how far the points in question are from observations. We first offer an accessible explanation and implementation of GPs. While existing implementations typically require three hyperparameters, our approach (available in the R package gpss) simplifies the arrangement of parameters to avoid a multidimensional grid search, leading to a stable and fully automated estimation procedure. We then illustrate the use of GPs in three settings, where model dependency and extrapolation combine to magnify inferential risks: (i) comparisons in which treated and control groups have poor covariate overlap; (ii) interrupted time series (ITS) designs, where models fitted prior to an event are extrapolated to later time points; and (iii) regression discontinuity, where inferences depend on point and uncertainty estimates at the very edges of their supporting data.
• "Estimating dynamic treatment effects of events: An imputation approach to Interrupted Time Series design."
• "Effects of the COVID-19 pandemic and governmental restrictions on firearms purchasing behavior." (with Jack Kappelman, Tanvi Shinkre, Ryan Baxter-King, Haotian Chen, and Abigail Kappelman)
Abstract
How did the onset of the COVID-19 pandemic in the United States affect gun purchasing behavior? This paper explores the significant increase in firearms purchasing rates triggered by the pandemic and subsequent governmental restrictions to curb the virus's spread. Extant scholarship suggests that distrust in government contributed to record-breaking firearms sales, but this study uses a causal, interrupted time series design with administrative data from the National Instant Criminal Background Check System (NICS) and state-issued public health orders to provide more robust, precisely estimated effects and improved counterfactual projections. We isolate the effect of pandemic-era lockdown orders on gun purchases though our stability-controlled quasi-experimental approach, which accounts for confounding factors, data structure issues, and addresses heterogeneity in the timing of public health orders. Taken together, these results demonstrate that credible causal conclusions can be drawn even amid potential confounding. These findings deepen our understanding of the socio-political dynamics of firearm purchasing in crisis contexts and highlight the impact of government restrictions on consumer behavior, with additional findings speaking to the downstream effects this increase in gun purchasing had on firearms-involved mortality and other health outcomes.
• "Non-existent outcomes in research on inequality: A causal approach." (with Ian Lundberg) [software]
Abstract
When studying inequality, a focal outcome may not exist for some individuals. Those who are not employed have no hourly wage, for example. Scholars of wage inequality routinely drop the non-employed. But the same causal process that shapes wage inequality among the employed also shapes which people are employed at all. Researchers who drop those with non-existent outcomes inadvertently induce selection problems and obscure inequality. We show how to use principal stratification methods to study two quantities: (1) the average effect on whether an outcome exists, and (2) the average effect on that outcome among the latent set of people who would have an outcome under either treatment condition. Our technical contribution is to carry out principal stratification within a parametric regression analysis that adjusts for measured confounders. Our applied contribution is to reveal how standard practices in sociology and economics obscure inequality. We illustrate by showing how past work has understated the causal effect of motherhood on the hourly wages of women who would be employed with or without children.
• "Unraveling climate refugee-specific hostility: Exploring deservingness bias in developed countries." (with Jieun Park and Margaret Peters)
Abstract
Climate-induced migration is a pressing global issue, with millions projected to be displaced. Understanding public attitudes towards climate refugees is crucial but lacks research focus, often treating refugees as a homogeneous group. This study investigates attitudes towards climate and war refugees, focusing on the underlying factors driving climate refugee-specific hostility. Introducing the concept of deservingness bias, we argue that climate refugee-specific hostility may stem from the suspicion that those fleeing from natural disasters and climate change are not genuine asylum seekers but economic migrants in disguise. This suspicion is notably pronounced in developed economies: people who are more suspicious about the motives of climate refugees support less climate-induced refugees than war-induced ones in developed economies, but not in other countries. We posit that the presence of well-established systems for addressing domestically displaced populations fosters skepticism regarding the motives of climate refugees, potentially leading to biased attitudes. Drawing on data from original survey experiments and Ipsos Survey 2022, this research aims to gain insights into factors shaping attitudes towards climate and war refugees, with a particular focus on deservingness bias. Our findings contribute to comprehending the group-specific attitudes hypothesis and public attitudes toward refugees in a broader context. It also informs policies to promote integration and reduce tensions in the face of climate-driven displacement.
• "Drivers of attitudes toward Asian Americans: Disentangling stereotyping from racial identity." (with Jieun Park and Jessica Hyunjeong Lee)
Abstract
Despite substantial research on attitudes toward Asian Americans, we know less about the underlying motivations. Prior studies have demonstrated that individuals make inferences about other characteristics based on someone’s race, hence feelings toward Asian Americans could be a consequence of the correlation between race and other perceived traits. To what extent are attitudes toward Asian Americans driven by dislike for racial outgroup per se, xenophobia, fear of group competition, or other presumed stereotypes? Leveraging original survey experiments that manipulate race signaled by name, immigration status, perceived threats, and stereotyped traits in short vignettes, we aim to disentangle the effects of race and other related traits inferred from it on interpersonal attitudes. Distinguishing among these competing motivations is crucial not only for understanding how perceptions of Asian Americans are shaped but also for understanding the disparate implications of interracial relationships. This study will shed light on what is the principal driver of interpersonal affect regarding race.
• "Modeling and assessing controlled direct effects with the regression-with-residuals method." (with Ian Lundberg)