The Future Strikes Back: Using Future Treatments to Detect and Reduce Hidden Bias
| Autor: | Elwert, Felix, Pfeffer, Fabian |
|---|---|
| Rok vydání: | 2023 |
| Předmět: |
Sociology and Political Science
Computer science DAG SocArXiv|Social and Behavioral Sciences|Sociology|Inequality Poverty and Mobility SocArXiv|Social and Behavioral Sciences|Sociology|Methodology 05 social sciences Confounding 050401 social sciences methods Contrast (statistics) Regression analysis Directed acyclic graph 0506 political science bepress|Social and Behavioral Sciences|Sociology SocArXiv|Social and Behavioral Sciences|Sociology bepress|Social and Behavioral Sciences|Sociology|Quantitative Qualitative Comparative and Historical Methodologies 0504 sociology Causal inference bepress|Social and Behavioral Sciences 050602 political science & public administration Econometrics SocArXiv|Social and Behavioral Sciences causal inference bepress|Social and Behavioral Sciences|Sociology|Inequality and Stratification Advice (complexity) Social Sciences (miscellaneous) |
| Zdroj: | Sociological methodsresearch. 51(3) |
| ISSN: | 0049-1241 |
| Popis: | Conventional advice discourages controlling for postoutcome variables in regression analysis. By contrast, we show that controlling for commonly available postoutcome (i.e., future) values of the treatment variable can help detect, reduce, and even remove omitted variable bias (unobserved confounding). The premise is that the same unobserved confounder that affects treatment also affects the future value of the treatment. Future treatments thus proxy for the unmeasured confounder, and researchers can exploit these proxy measures productively. We establish several new results: Regarding a commonly assumed data-generating process involving future treatments, we (1) introduce a simple new approach and show that it strictly reduces bias, (2) elaborate on existing approaches and show that they can increase bias, (3) assess the relative merits of alternative approaches, and (4) analyze true state dependence and selection as key challenges. (5) Importantly, we also introduce a new nonparametric test that uses future treatments to detect hidden bias even when future-treatment estimation fails to reduce bias. We illustrate these results empirically with an analysis of the effect of parental income on children’s educational attainment. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|