Jeffrey E. Cohen is professor of Political Science at Fordham University. He received his Ph.D. from the University of Michigan in 1979. His research focuses primarily on the presidency and other executives. He has published in numerous journals in addition to Presidential Studies Quarterly, including the American Political Science Review, the American Journal of Political Science, and the Journal of Politics. His book, Going Local: Presidential Leadership in the Post-Broadcast Age, (Cambridge University Press, 2010) won both the Neustadt Award of the American Political Science Association and Harvard University’s Shorenstein Center Goldsmith Award.
What initially sparked your interest in exploring experimental political science?
I have been interested in experiments in political science since my graduate school days, 40 years ago, when I read some of early and classic political science experiments, such as Sam Eldersveld, who used his own campaign for the mayoralty in Ann Arbor to conduct an experiment on campaign tactic effects. But as my research turned to the presidency, I put my interest in experiments on the back burner, not really sure how to apply experimental designs in studying the presidency. My interest in experiments became more active in recent years for several reasons: colleagues who use experiments in their own research, some fine experimental research on the presidency, and the ability to conduct important experiments inexpensively with MTURK.
What are the challenges that accompany experimental political science?
One major challenge confronting experiments on the presidency is that the overwhelming number of people know who the current president is and have opinions about that president. Results of experimental treatments may be affected by the attitudes of respondents towards the incumbent president. This can be a curse but also a blessing for experimental studies. The curse arises when we want to learn something about people’s attitudes towards the office as opposed to the incumbent. Many respondents may conflate the two. But there is a blessing too, as a way to overcome this issue is to do essentially the same experiment across other presidents and other executives.
What kind of response do you hope your work elicits from your readers? Ideally, what kind of critical thought do you hope your article inspires?
First, I hope my essay will spark greater interest by political scientists, both inside the presidency subfield and outside, to consider experiments as a way to study the office. I am certain that there are a host of questions, which I have not even begun to consider, for which an experimental design is appropriate and illuminating. At the same time, I caution against a rush to experimentation. Like all other research techniques, designing a useful and strong experiment requires a degree of thoughtfulness and recognition of both the power and pitfalls of using experiments. Hopefully the upcoming generation of presidency and executive politics scholars, especially those in graduate school now, will learn how to design experiments and use experimentation for studying the office.
Professor Jeffrey E. Cohen can be contacted at firstname.lastname@example.org