Political Analysis Advance Access originally published online on August 5, 2009
Political Analysis 2009 17(4):341-357; doi:10.1093/pan/mpp011
| ||||||||||||||||||||||||||||||||||||||||||||||||||
This article appears in the following Political Analysis issue: Special Issue: Natural Experiments in Political Science [View the issue table of contents]
Observing the Counterfactual? The Search for Political Experiments in Nature
Department of Political Science, Binghamton University, State University of New York, PO Box 6000, Binghamton, NY 13902-6000 e-mails: grobinso{at}binghamton.edu (corresponding author), jmcnulty{at}binghamton.edu, jkrasno{at}binghamton.edu
A search of recent political science literature and conference presentations shows substantial fascination with the concept of the natural experiment. However, there seems to be a wide array of definitions and applications employed in research that purports to analyze natural experiments. In this introductory essay to the special issue, we attempt to define natural experiments and discuss related issues of research design. In addition, we briefly explore the basic methodological issues around the appropriate analysis of natural experiments and give an overview of different techniques. The overarching theme of this essay and of this issue is to encourage applied researchers to look for natural experiments in their own work and to think more systematically about research design.
Authors' note: A previous draft of this essay was presented at the 2008 Meeting of the American Political Science Association, Boston, MA. Thanks go to Chris Zorn for turning over the pages of Political Analysis to us and our contributors. The authors thank Ryan Enos, Seth Hill, Amy Lerman, Jeff Lewis, James Lo, Michael McDonald, Nate Monroe, and two anonymous reviewers for their helpful comments. Any remaining errors belong to the authors.