Modeling Heterogeneous Treatment Effects in Survey Experiments with Bayesian Additive Regression Trees
- 1 September 2012
- journal article
- research article
- Published by Oxford University Press (OUP) in Public Opinion Quarterly
- Vol. 76 (3), 491-511
- https://doi.org/10.1093/poq/nfs036
Abstract
Survey experimenters routinely test for systematically varying treatment effects by using interaction terms between the treatment indicator and covariates. Parametric models, such as linear or logistic regression, are currently used to search for systematic treatment effect heterogeneity but suffer from several shortcomings; in particular, the potential for bias due to model misspecification and the large amount of discretion they introduce into the analysis of experimental data. Here, we explicate what we believe to be a better approach. Drawing on the statistical learning literature, we discuss Bayesian Additive Regression Trees (BART), a method for analyzing treatment effect heterogeneity. BART automates the detection of nonlinear relationships and interactions, thereby reducing researchers’ discretion when analyzing experimental data. These features make BART an appealing “off-the-shelf” tool for survey experimenters who want to model systematic treatment effect heterogeneity in a flexible and robust manner. In order to illustrate how BART can be used to detect and model heterogeneous treatment effects, we reanalyze a well-known survey experiment on welfare attitudes from the General Social Survey.Keywords
This publication has 31 references indexed in Scilit:
- Estimation of Heterogeneous Treatment Effects from Randomized Experiments, with Application to the Optimal Planning of the Get-Out-the-Vote CampaignPolitical Analysis, 2011
- Bayesian Nonparametric Modeling for Causal InferenceJournal of Computational and Graphical Statistics, 2011
- Bayesian ensemble methods for survival prediction in gene expression dataBioinformatics, 2010
- Heterogeneous impacts in PROGRESAJournal of Econometrics, 2008
- Simultaneous confidence intervals based on the percentile bootstrap approachComputational Statistics & Data Analysis, 2008
- When Averages Hide Individual Differences in Clinical TrialsAmerican Scientist, 2007
- Human Research and Data Collection via the InternetAnnual Review of Psychology, 2004
- Subgroup analysis, covariate adjustment and baseline comparisons in clinical trial reporting: current practiceand problemsStatistics in Medicine, 2002
- Racial Attitudes and the “New South”The Journal of Politics, 1997
- Statistics and Causal InferenceJournal of the American Statistical Association, 1986