About

The provided page text does not contain a professional biography or detailed information about Professor Chad Hazlett's research focus, background, or key contributions. Therefore, the JSON output is an empty string for the biography.

Research topics

• Computer Science
• Political Science
• Artificial Intelligence
• Engineering
• Econometrics
• Geography
• Natural resource economics
• Medicine
• Environmental health
• Astrophysics
• Economics
• Physics
• Ecology
• Internal medicine
• Emergency medicine
• Astronomy
• Mathematics
• Statistics

Selected publications

• iv.sensemakr: Sensitivity Analysis Tools for Instrumental Variable Estimates

  2026-02-11

  datasetOpen accessSenior author

  Implements a suite of sensitivity analysis tools for instrumental variable estimates as described in Cinelli and Hazlett (2025) &lt;<a href="https://doi.org/10.1093%2Fbiomet%2Fasaf004" target="_top">doi:10.1093/biomet/asaf004</a>&gt;.

  PublisherOA PDFDOI

• A data-driven approach to identifying determinants of depression in rural Latine adolescents.

  Cultural Diversity & Ethnic Minority Psychology · 2026-01-08

  article

  OBJECTIVES: There is a paucity of research focused on risk and protective factors for depression in rural Latine adolescents. The present study first identified variables commonly described in conceptual models of depression etiology and maintenance in Latine adolescents and rural populations, including demographic (i.e., age, sex), cultural (i.e., acculturation), familial (i.e., family conflict, familism), and contextual factors (i.e., socioeconomic strain, parental education level, discrimination-related stress). A machine learning approach was then used to understand the relative contributions of these variables to depression in rural Latine adolescents. METHOD: = 15.74; 53% female) were Latine adolescents in grades 9-12 recruited from a high school in a low-income rural area, who completed a battery of self-report measures. A data-driven recursive partitioning method was used to examine the joint contribution of these variables to depression severity. RESULTS: Using a conditional inference framework, adolescents with low depression scores were characterized by high familism and low discrimination-related stress, whereas adolescents with high depression scores endorsed low familism. Female adolescents had higher depression severity than their male counterparts. CONCLUSIONS: These findings are consistent with both conceptual models of depression in Latine youth and previous empirical studies, particularly those showing that familism and discrimination-related stress play a significant role as protective and risk factors, respectively. The identification of crucial variables using data-driven approaches could help improve screening for and treatment of depression in rural Latine youth who experience significant mental health inequities. (PsycInfo Database Record (c) 2026 APA, all rights reserved).

  PublisherDOI

• gpss: Gaussian Processes for Social Science

  2026-04-08

  datasetOpen accessSenior author

  Provides Gaussian process (GP) regression tools for social science inference problems. GPs combine flexible nonparametric regression with principled uncertainty quantification: rather than committing to a single model fit, the posterior reflects lesser knowledge at the edge of or beyond the observed data, where other approaches become highly model-dependent. The package reduces user-chosen hyperparameters from three to zero and supplies convenience functions for regression discontinuity (gp_rdd()), interrupted time-series (gp_its()), and general GP fitting (gpss(), gp_train(), gp_predict()). Methods are described in Cho, Kim, and Hazlett (2026) &lt;<a href="https://doi.org/10.1017%2Fpan.2026.10032" target="_top">doi:10.1017/pan.2026.10032</a>&gt;.

  PublisherOA PDFDOI

• Replication Data for: Post-treatment problems: What can we say about the effect of a treatment among sub-groups who (would) respond in some way?

  Harvard Dataverse · 2026-04-09

  datasetOpen access1st authorCorresponding

  Investigators are often interested in how a treatment affects an outcome for units responding to treatment in a certain way. We may wish to know the effect among units that, for example, meaningfully implemented an intervention, passed an attention check, or demonstrated some important mechanistic response. Simply conditioning on the observed value of the post-treatment variable introduces problematic biases. Further, the identification assumptions required of several existing strategies are often indefensible. We propose the Treatment Reactive Average Causal Effect (TRACE), which we define as the total effect of treatment in the group that, if treated, would realize a particular value of the relevant post-treatment variable. By reasoning about the effect among the “non-reactive” group, we can identify and estimate the range of plausible values for the TRACE. We demonstrate the use of this approach with three examples: (i) learning the effect of police-perceived race on police violence during traffic stops, a case where point identification may be possible; (ii) estimating effects of a community policing intervention in Liberia, in communities that meaningfully implemented it, and (iii) studying how in-person canvassing affects support for transgender rights, among participants for whom the intervention would result in more positive feelings towards transgender people.

  PublisherDOI

• Inference at the Data’s Edge: Gaussian Processes for Estimation and Inference in the Face of Extrapolation Uncertainty

  Political Analysis · 2026-03-10 · 1 citations

  articleOpen accessSenior authorCorresponding

  Abstract Many inferential tasks involve fitting models to observed data and predicting outcomes at new covariate values, requiring interpolation or extrapolation. Conventional methods select a single best-fitting model, discarding fits that were similarly plausible in-sample but would yield sharply different predictions out-of-sample. Gaussian processes (GPs) offer a principled alternative. Rather than committing to one conditional expectation function, GPs deliver a posterior distribution over outcomes at any covariate value. This posterior effectively retains the range of models consistent with the data, widening uncertainty intervals where extrapolation magnifies divergence. In this way, the GP’s uncertainty estimates reflect the implications of extrapolation on our predictions, helping to tame the “dangers of extreme counterfactuals” (King and Zeng, 2006). The approach requires (i) specifying a covariance function linking outcome similarity to covariate similarity and (ii) assuming Gaussian noise around the conditional expectation. We provide an accessible introduction to GPs with emphasis on this property, along with a simple, automated procedure for hyperparameter selection implemented in the R package gpss . We illustrate the value of GPs for capturing counterfactual uncertainty in three settings: (i) treatment effect estimation with poor overlap, (ii) interrupted time series requiring extrapolation beyond pre-intervention data, and (iii) regression discontinuity designs where estimates hinge on boundary behavior.

  PublisherOA PDFDOI

• Seeing Like a District: Understanding What Close-Election Designs for Leader Characteristics Can and Cannot Tell Us

  Political Analysis · 2025-03-27 · 1 citations

  articleOpen accessSenior authorCorresponding

  Abstract Many influential political science articles use close elections to study how important outcomes vary after a certain type of candidate wins, such as a Democrat or a Republican. This politician characteristic regression discontinuity (PCRD) design offers opportunities for inferential leverage but also the potential for confusion. In this article, we clarify what causal claims the PCRD licenses, offering a rigorous causal analysis that points to three principal lessons. First, PCRDs do nothing to isolate the effect of the politician characteristic of interest as apart from other politician characteristics. Second, selection processes (regarding both “who runs” and “which elections are close”) can generate and exacerbate such confounding, as noted in Marshall (2024). Third and more fortunately, this approach does make it possible to estimate the average effect of electing a leader of type “A” vs. “B” in the context of close elections, treating the units as districts, not leaders. We also suggest a set of tools that can aid in falsifying key assumptions, avoiding unwarranted claims, and surfacing mechanisms of interest. We illustrate these issues and tools through a reanalysis of an influential study about what happens when extremists win primaries (Hall 2015).

  PublisherOA PDFDOI

• Safe inference outside of randomized trials: Application of the stability-controlled quasi-experiment to the effects of three COVID-19 therapies

  Observational Studies · 2025-01-01

  articleOpen access

  When estimating the effects of medical therapies from their use outside of randomized trials, researchers often rely on assumptions that are difficult to justify and typically impossible to verify. The resulting estimates may thus be far from their intended causal targets, potentially making a harmful treatment appear beneficial or vice versa. We review the stability-controlled quasi-experiment (SCQE), a method suited to settings where a treatment's prevalence changes sharply over a short period, and apply it to assess the effects of remdesivir, hydroxychloroquine, and dexamethasone on COVID-19 mortality. Rather than requiring debate about the absence (or limited strength) of unobserved confounding, about "parallel trends'', or other well-known strategies, the SCQE asks users to reason about a "baseline trend'' assumption. In this setting, this asks"How much could COVID-19 mortality have changed over a short period, absent the treatment change in question?'' Any plausible range for this assumption yields a corresponding range of plausible causal effect estimates. Conversely, SCQE clarifies what baseline trends must be defended or refuted in order to defend or refute a given conclusion about a treatment's efficacy or harm. Using data from two hospital systems early in the COVID-19 pandemic, we show that SCQE could have enabled safe yet informative inferences about treatment effects before clinical trial completion, producing conclusions consistent with eventual trial results.

  PublisherDOI

• Post-treatment problems: What can we say about the effect of a treatment among sub-groups who (would) respond in some way?

  ArXiv.org · 2025-05-10

  preprintOpen access1st authorCorresponding

  Investigators are often interested in how a treatment affects an outcome for units responding to treatment in a certain way. We may wish to know the effect among units that, for example, meaningfully implemented an intervention, passed an attention check, or demonstrated some important mechanistic response. Simply conditioning on the observed value of the post-treatment variable introduces problematic biases. Further, the identification assumptions required of several existing strategies are often indefensible. We propose the Treatment Reactive Average Causal Effect (TRACE), which we define as the total effect of treatment in the group that, if treated, would realize a particular value of the relevant post-treatment variable. By reasoning about the effect among the "non-reactive" group, we can identify and estimate the range of plausible values for the TRACE. We demonstrate the use of this approach with three examples: (i) learning the effect of police-perceived race on police violence during traffic stops, a case where point identification may be possible; (ii) estimating effects of a community-policing intervention in Liberia, in communities that meaningfully implemented it, and (iii) studying how in-person canvassing affects support for transgender rights, among participants for whom the intervention would result in more positive feelings towards transgender people.

  PublisherOA PDFDOI

• An omitted variable bias framework for sensitivity analysis of instrumental variables

  Biometrika · 2025-01-01 · 13 citations

  articleSenior author

  Abstract We develop an omitted variable bias framework for sensitivity analysis of instrumental variable estimates that naturally handles multiple side effects (violations of the exclusion restriction assumption) and confounders (violations of the ignorability of the instrument assumption) of the instrument, exploits expert knowledge to bound sensitivity parameters and can be easily implemented with standard software. Specifically, we introduce sensitivity statistics for routine reporting, such as (extreme) robustness values for instrumental variables, describing the minimum strength that omitted variables need to have to change the conclusions of a study. Next, we provide visual displays that fully characterize the sensitivity of point estimates and confidence intervals to violations of the standard instrumental variable assumptions. Finally, we offer formal bounds on the worst possible bias under the assumption that the maximum explanatory power of omitted variables is no stronger than a multiple of the explanatory power of observed variables. Conveniently, many pivotal conclusions regarding the sensitivity of the instrumental variable estimate (e.g., tests against the null hypothesis of a zero causal effect) can be reached simply through separate sensitivity analyses of the effect of the instrument on the treatment (the first stage) and the effect of the instrument on the outcome (the reduced form). We apply our methods in a running example that uses proximity to college as an instrumental variable to estimate the returns to schooling.

  PublisherDOI

• Inference with weights: Residualization produces short, valid intervals for varying estimands and varying resampling processes

  ArXiv.org · 2025-07-25

  preprintOpen access

  Weighting procedures are used in observational causal inference to adjust for covariate imbalance within the sample. Common practice for inference is to estimate robust standard errors from a weighted regression of outcome on treatment. However, it is well known that weighting can inflate variance estimates, sometimes significantly, leading to standard errors and confidence intervals that are overly conservative. We instead examine and recommend the use of robust standard errors from a weighted regression that additionally includes the balancing covariates and their interactions with treatment. We show that these standard errors are more precise and asymptotically correct for weights that achieve exact balance under multiple common resampling frameworks, including design-based and model-based inference, as well as superpopulation sampling with a finite sample correction. Gains to precision can be quite significant when the balancing weights adjust for prognostic covariates. For procedures that balance only approximately or in expectation, such as inverse propensity weighting or approximate balancing weights, our proposed method improves precision by reducing residuals through augmentation with the parametric model. We demonstrate our approach through simulation and re-analysis of multiple empirical studies.

  PublisherOA PDFDOI

Frequent coauthors

• Josh Kertzer

  The Ohio State University

  50 shared

• Davide Panagia

  50 shared

• Dale Smith

  University of Illinois Chicago

  50 shared

• Rohan Kalyan

  Virginia Commonwealth University

  50 shared

• Paul Piccard

  University of California, Los Angeles

  50 shared

• Matthew T. Pietryka

  Florida State University

  50 shared

• Melivin Rogers

  The Ohio State University

  50 shared

• Inken Von Borzyskowski

  University of Oxford

  50 shared

Education

• Ph.D., Political Science

  Massachusetts Institute of Technology

  2014