Learning From Biased Research Designs

39 Pages Posted: 31 Aug 2018

See all articles by Andrew Little

Andrew Little

University of California, Berkeley

Thomas B. Pepinsky

Cornell University - Department of Government

Date Written: August 22, 2018

Abstract

Most contemporary empirical work in political science aims to learn about causal effects from research designs that may be subject to bias. We provide a Bayesian framework for understanding how researchers should approach the general problem of inferring causal effects from potentially biased research designs. The key to our approach is that both researchers and their audiences have prior beliefs about both causal effects and the degree and direction of bias. Once these priors are specified, what a rational researcher should learn from a potentially biased estimate can be derived from Bayes’ rule. We apply this principle to explore when we should learn more or less from basic difference of means estimates, and then extend our analysis to speak to common modern designs intended to uncover causal effects.

Keywords: Research Design, Causality, Treatment Effects, Bayes, Learning, Identification, Bias

Suggested Citation

Little, Andrew and Pepinsky, Thomas B., Learning From Biased Research Designs (August 22, 2018). Available at SSRN: https://ssrn.com/abstract=3236815 or http://dx.doi.org/10.2139/ssrn.3236815

Andrew Little

University of California, Berkeley ( email )

210 Barrows Hall
Berkeley, CA 94720
United States

Thomas B. Pepinsky (Contact Author)

Cornell University - Department of Government ( email )

Ithaca, NY 14853
United States

Register to save articles to
your library

Register

Paper statistics

Downloads
173
rank
163,140
Abstract Views
679
PlumX Metrics