Why Process Matters for Causal Inference

In this paper we provide a formal account of how information about causal processes (i.e., knowledge of the causal chain linking an explanatory variable to an outcome variable) can be used to sharpen causal inferences. All of this is done within a Bayesian potential outcomes causal model. The methods discussed in this paper empower researchers by providing them with a richer palette of causal assumptions than typically employed. At the same time, because the methods are embedded in a rigorous counterfactual causal model, researchers are held to high standards of transparency and logical consistency. We illustrate these methods with an application to the effects of election day registration on African American turnout. This analysis shows that traditional regression or matching estimates for these effects are likely overstated. ∗The authors thank Matt Chingos for his research assistance and Kevin Clarke, the participants of the 2009 meeting of the Midwest Political Science Association and the participants of the 2009 Causal Workshop at the Banff International Research Station for their helpful comments and suggestions on an earlier draft of this paper. †Assistant Professor, Department of Government and The Institute for Quantitative Social Science, Harvard University, 1737 Cambridge Street, Cambridge, MA 02138. [email protected] ‡Professor of Law, UC Berkeley School of Law, 490 Simon #7200, Berkeley, CA 94720-7200. [email protected]