Kosuke Imai’s research while affiliated with Harvard University and other places

A large body of research documents the barriers faced by first-generation, low-income (FGLI) students as “hidden minorities” on elite college campuses. Although existing studies show brief psychological interventions can help mitigate some of these obstacles, universities are investing in more intensive interventions that try to both shift mindsets and mitigate structural disadvantages in FGLI students’ academic preparation. In collaboration with the administrators at a highly selective university, we conducted a randomized controlled trial of a summer bridge program targeted at FGLI students. During summers between 2017 and 2019, we randomly selected 232 out of 418 first-generation or low-income students and invited them to attend an intensive, six-week-long residential summer program featuring courses for academic credit. Students randomized to the control group either interacted with online content offering no academic credit or had no summer intervention. Our preregistered analysis shows that the program encouraged FGLI students to pursue a more ambitious first-year program, increasing the proportion of nonintroductory courses by 7 percentage points. The program also increased the proportion of courses taken for a grade rather than as pass-fail by 6 percentage points. These improvements were accompanied by no discernible impact on first-year grade point averages (GPAs) and academic withdrawal. The findings show the potential to academically integrate FGLI students into selective university communities.

Dropout often threatens the validity of causal inference in longitudinal studies. While existing studies have focused on the problem of missing outcomes caused by treatment, we study an important but overlooked source of dropout, selective eligibility. For example, patients may become ineligible for subsequent treatments due to severe side effects or complete recovery. Selective eligibility differs from the problem of ``truncation by death'' because dropout occurs after observing the outcome but before receiving the subsequent treatment. This difference makes the standard approach to dropout inapplicable. We propose a general methodological framework for longitudinal causal inference with selective eligibility. By focusing on subgroups of units who would become eligible for treatment given a specific treatment history, we define the time-specific eligible treatment effect (ETE) and expected number of outcome events (EOE) under a treatment sequence of interest. Assuming a generalized version of sequential ignorability, we derive two nonparametric identification formulae, each leveraging different parts of the observed data distribution. We then derive the efficient influence function of each causal estimand, yielding the corresponding doubly robust estimator. Finally, we apply the proposed methodology to an impact evaluation of a pre-trial risk assessment instrument in the criminal justice system, in which selective eligibility arises due to recidivism.

Scholars from diverse fields now use highly disaggregated ("microlevel") data with fine-grained spatial (e.g., locations of villages and individuals) and temporal (days, hours, or even seconds) dimensions to test their theories. Despite the proliferation of these data, however, statistical methods for causal inference with spatio-temporal data remain underdeveloped. We introduce an R package, geocausal, that enables researchers to implement causal inference methods for highly disaggregated spatio-temporal data. The geocausal package implements two necessary steps for spatio-temporal causal inference: (1) preparing the data and (2) estimating causal effects. The geocausal package allows users to effectively use fine-grained spatio-temporal data, test counterfactual scenarios that have spatial and temporal dimensions, and visualize each step efficiently. We illustrate the capabilities of the geocausal package by analyzing the US airstrikes and insurgent attacks in Iraq over various spatial and temporal windows.

A century ago, Neyman showed how to evaluate the efficacy of treatment using a randomized experiment under a minimal set of assumptions. This classical repeated sampling framework serves as a basis of routine experimental analyses conducted by today’s scientists across disciplines. In this article, we demonstrate that Neyman’s methodology can also be used to experimentally evaluate the efficacy of individualized treatment rules (ITRs), which are derived by modern causal machine learning (ML) algorithms. In particular, we show how to account for additional uncertainty resulting from a training process based on cross-fitting. The primary advantage of Neyman’s approach is that it can be applied to any ITR regardless of the properties of ML algorithms that are used to derive the ITR. We also show, somewhat surprisingly, that for certain metrics, it is more efficient to conduct this ex-post experimental evaluation of an ITR than to conduct an ex-ante experimental evaluation that randomly assigns some units to the ITR. Our analysis demonstrates that Neyman’s repeated sampling framework is as relevant for causal inference today as it has been since its inception.

Political actors frequently manipulate redistricting plans to gain electoral advantages, a process commonly known as gerrymandering. To address this problem, several states have implemented institutional reforms including the establishment of map-drawing commissions. It is difficult to assess the impact of such reforms because each state structures bundles of complex rules in different ways. We propose to model redistricting processes as a sequential game. The equilibrium solution to the game summarizes multi-step institutional interactions as a single dimensional score. This score measures the leeway political actors have over the partisan lean of the final plan. Using a differences-in-differences design, we demonstrate that reforms reduce partisan bias and increase competitiveness when they constrain partisan actors. We perform a counterfactual policy analysis to estimate the partisan effects of enacting recent institutional reforms nationwide. We find that instituting redistricting commissions generally reduces the current Republican advantage, but Michigan-style reforms would yield a much greater pro-Democratic effect than types of redistricting commissions adopted in Ohio and New York.

The U.S. Census Bureau faces a difficult trade-off between the accuracy of Census statistics and the protection of individual information. We conduct an independent evaluation of bias and noise induced by the Bureau’s two main disclosure avoidance systems: the TopDown algorithm used for the 2020 Census and the swapping algorithm implemented for the three previous Censuses. Our evaluation leverages the Noisy Measurement File (NMF) as well as two independent runs of the TopDown algorithm applied to the 2010 decennial Census. We find that the NMF contains too much noise to be directly useful without measurement error modeling, especially for Hispanic and multiracial populations. TopDown’s postprocessing reduces the NMF noise and produces data whose accuracy is similar to that of swapping. While the estimated errors for both TopDown and swapping algorithms are generally no greater than other sources of Census error, they can be relatively substantial for geographies with small total populations.

A large body of research documents the barriers faced by first-generation, low-income (FGLI) students as ``hidden minorities'' on elite college campuses. Although existing studies show brief psychological interventions can help mitigate some of these obstacles, universities are investing in more intensive interventions that try to both shift mindsets and mitigate structural disadvantages in FGLI students' academic preparation. In collaboration with the administrators at a highly selective university, we conducted the first randomized controlled trial of a summer bridge program targeted at FGLI students. During summers between 2017 and 2019, we randomly selected 232 out of 418 first-generation or low-income students and invited them to attend an intensive, six-week long residential summer program featuring courses for academic credit. Students randomized to the control group either interacted with online content offering no academic credit or had no summer intervention. Our pre-registered analysis shows that the program encouraged FGLI students to pursue a more ambitious first-year program, increasing the proportion of non-introductory courses by 7 percentage points. The program also increased the proportion of courses taken for a grade rather than as pass-fail by 6 percentage points. These improvements were accompanied by no discernible impact on first-year GPAs and academic withdrawal. The findings show the ability to academically integrate FGLI students into selective university communities.