The Fertility and Demographic Change Miniconference will take place 10/17/25, 9:30 am - 5:00 pm (Schedule TBA), 4240A Public Affairs Building. This event is jointly sponsored by CCPR, and the development, macro, and applied groups in the economics department. We have limited space, so if you plan to come then please sign up here. […]
How to Make Causal Inferences Using Texts
Texts are increasingly used to make causal inferences: either with the document serving as the treatment or the outcome. We introduce a new conceptual framework to understand all text-based causal inferences, demonstrate fundamental problems that arise when using manual or computational approaches applied to text for causal inference, and provide solutions to the problems we raise. We demonstrate that all text-based causal inferences depend upon a latent representation of the text and we provide a framework to learn the latent representation. Estimating this latent representation, however, creates new risks: we may unintentionally create a dependency across observations or create opportunities to fish for large effects. To address these risks, we introduce a train/test split framework and apply it to estimate causal effects from an experiment on immigration attitudes and a study on bureaucratic responsiveness. Our work provides a rigorous foundation for text-based causal inferences, connecting two previously disparate literatures. (Joint Work with Egami, Fong, Grimmer and Roberts)
Co-sponsored with the Center for Social Statistics
Estimating Heterogeneous Treatment Effects and Optimal Treatment Assignment Policies
This talk will review recently developed methods for estimating conditional average treatment effects and optimal treatment assignment policies in experimental and observational studies, including settings with unconfoundedness or instrumental variables. Multi-armed bandits for learning treatment assignment policies will also be considered.
Co-sponsored with the Center for Social Statistics
Population Vulnerability and Spatial Analytics
There has been a transition from population studies that were relatively data poor to the present day where digital data is plentiful on many fronts. The “Smart City” is fed by sources of information coming from all directions, where sensors observe things about the movement of vehicles and people, infrastructure conditions, air quality, weather, etc. The challenge is to make use of this digital data, and this is precisely the value added offered by a range of big data spatial analytics. This paper examines aspects of population vulnerability, focusing on particular types of risks and hazards in urban areas.
Small steps with Big Data: Using Machine Learning in Resource Economics
This talk looks at how recent developments in Big Data and Machine Learning are being used in conjunction with randomized controlled trials and large population level program evaluations to design, implement and measure efforts to change consumer behavior. We will explore the role played by very detailed consumption data (often at 15 minute intervals), as well as recent techniques such as deep learning to help us better understand individual and population behaviors, and which insights from behavioral sciences are effective at changing behaviors in areas such as energy conservation and efficiency.
Matching Methods for Causal Inference with Time-Series Cross-Section Data
Matching methods aim to improve the validity of causal inference in observational studies by reducing model dependence and offering intuitive diagnostics. While they have become a part of standard tool kit for empirical researchers across disciplines, matching methods are rarely used when analyzing time-series cross-section (TSCS) data, which consist of a relatively large number of repeated measurements on the same units.
We develop a methodological framework that enables the application of matching methods to TSCS data. In the proposed approach, we first match each treated observation with control observations from other units in the same time period that have an identical treatment history up to the pre-specified number of lags. We use standard matching and weighting methods to further refine this matched set so that the treated observation has outcome and covariate histories similar to those of its matched control observations. Assessing the quality of matches is done by examining covariate balance. After the refinement, we estimate both short-term and long-term average treatment effects using the difference-in-differences estimator, accounting for a time trend. We also show that the proposed matching estimator can be written as a weighted linear regression estimator with unit and time fixed effects, providing model-based standard errors. We illustrate the proposed methodology by estimating the causal effects of democracy on economic growth, as well as the impact of inter-state war on inheritance tax. The open-source software is available for implementing the proposed matching methods.
Co-sponsored with the Political Science Department, Statistics Department and the Center for Social Statistics
Internal vs. external validity in studies with incomplete populations
Researchers working with administrative data rarely have access to the entire universe of units they need to estimate effects and make statistical inferences. Examples are varied and come from different disciplines. In social program evaluation, it is common to have data on all households who received the program, but only partial information on the universe of households who applied or could have applied for the program. In studies of voter turnout, information on the total number of citizens who voted is usually complete, but data on the total number of voting-eligible citizens is unavailable at low levels of aggregation. In criminology, information on arrests by race is available, but the overall population that could have potentially been arrested is typically unavailable. And in studies of drug overdose deaths, we lack complete information about the full population of drug users.
In all these cases, a reasonable strategy is to study treatment effects and descriptive statistics using the information that is available. This strategy may lack the generality of a full-population study, but may nonetheless yield valuable information for the included units if it has sufficient internal validity. However, the distinction between internal and external validity is complex when the subpopulation of units for which information is available is not defined according to a reproducible criterion and/or when this subpopulation itself is defined by the treatment of interest. When this happens, a useful approach is to consider the full range of conclusions that would be obtained under different possible scenarios regarding the missing information. I discuss a general strategy based on partial identification ideas that may be helpful to assess sensitivity of the partial-population study under weak (non-parametric) assumptions, when information about the outcome variable is known with certainty for a subset of the units. I discuss extensions such as the inclusion of covariates in the estimation model and different strategies for statistical inference.
Co-sponsored with the Political Science Department, Statistics Department and the Center for Social Statistics
Abortion utilization in Ohio’s changing legislative context
Changes in Ohio, most notably legislation and policy changes since 2011, likely have impacted women’s access to abortion. Many abortion clinics in Ohio have closed in the past seven years, and several others are currently engaged in litigation and are at risk of closure. Clinic closures influence the distance that women travel when seeking abortion. Coupled with the impact of an Ohio law that mandates a 24-hour waiting period after a woman’s initial abortion consultation, loss of a nearby clinic may put abortion out of reach for many women. Other legislation limits where abortions can and cannot be performed and to what gestational stage abortions are performed. This presentation will provide preliminary findings about population-level shifts in abortion utilization, with special attention to change over time, geographic variation, and groups of women who may be underserved.
Co-sponsored with The Bixby Center
Immigration Restrictions as Active Labor Market Policy: Evidence from the Mexican Bracero Exclusion
An important class of active labor market policy has received little impact evaluation: immigration barriers intended to raise wages and employment by shrinking labor supply. Theories of endogenous technical advance raise the possibility of limited or even perverse impact. We study a natural policy experiment: the exclusion of almost half a million Mexican bracero farm workers from the United States to improve farm labor market conditions. With novel labor market data we measure state-level exposure to exclusion and model the absent changes in technology or crop mix. We fail to reject zero labor market impact, inconsistent with this model.
Co-sponsored with the Public Policy and Applied Social Sciences Seminar
Social Science in the Age of Genomics
The cost of genetic information has been dropping at a rate faster than that of Moore's law in microcomputing. As a result, the science of genetic prediction has improved by leaps and bounds in recent years, and with it has emerged a novel field: sociogenomics. Sociogenomics seeks to integrate genetic and environmental information to obtain a more robust, complete picture of the causes of human behavior. This talk will highlight some recent examples of sociogenomic research, touching upon issues such as adolescent peer effects, racial discrimination, assortative mating, and fertility patterns. The talk will conclude by discussing the social and policy implications of genetic prediction.
Co-sponsored with the Public Policy and Applied Social Science Seminar Series
