It is our pleasure to announce that the two faculty members selected as this year’s Fellows to the Society for Political Methodology are Luke Keele and Rocio Titiunik. Please join us in congratulating them both. The citations for this honor are presented below.
Professor Luke Keele is currently a Research Associate Professor at the University of Pennsylvania. Prior to this, he was a Professor at Georgetown University, an Associate Professor at Pennsylvania State University, and an Associate and Assistant Professor at Ohio State University. He was a post-doctoral fellow in Quantitative Research Methods at Nuffield College and the Department of Politics at Oxford University. He received his PhD in political science from the University of North Carolina at Chapel Hill in 2003. He was the recipient of the Emerging Scholar Award for a young scholar making exceptional contributions to political methodology who is within ten years of their terminal degree, awarded by the Society of Political Methodology in 2013.
Professor Keele’s research centers on statistical methods for causal inference and program evaluation, health services research, and the social sciences. He is particularly interested in matching methods, instrumental variables, randomization inference, and regression discontinuity designs. Substantively, his work has crossed many fields. He has made lasting contributions to the study of voting behavior and elections. He has published work in journals such as the American Political Science Review, Journal of the American Statistical Association, Annals of Applied Statistics, Psychological Methods, Statistical Science, the Journal of the Royal Statistical Society, Series A, Statistics in Medicine, and American Journal of Political Science.
According to data from Google Scholar, Professor Keele has been cited over 10,000 times and he has an h-index of 32. Professor Keele’s most widely cited research is his work on causal mediation analysis with co-authors Kosuke Imai, Dustin Tingley, and Teppei Yamamoto. Political scientists have long focused on understanding the causal processes and associated causal mechanisms that give rise to various outcomes. Keele’s work, which acknowledges the importance of such an understanding, also makes it very clear just how hard it is to achieve such an understanding. In particular, Keele shows how traditional approaches based on linear structural equations modeling rely on untestable and often implausible assumptions and, more surprisingly, that seemingly obvious types of randomized treatment assignment are not sufficient to identify mediation effects. What is particularly nice about this work is that it does not stop with these negative results. Keele goes on to provide concrete, practical advice for identification and estimation of mediation effects and he even goes one step further and provides a method for assessing the sensitivity of the results to various assumptions.
Professor Keele’s writing on technical subjects is well-suited to the classroom due to its organization and clarity. For instance, his 2008 book, Semiparametric Regression for the Social Sciences has been heralded for its clear, straightforward explanations of the models in the pages of TPM. Nonparametric smoothing techniques allow for the estimation of nonlinear relationships between continuous variables. In conjunction with standard statistical models, these smoothing techniques provide the means to test for, and estimate, nonlinear relationships in a wide variety of analyses. Readers are introduced to the principles of nonparametric smoothing and to a wide variety of smoothing methods. The author also explains how smoothing methods can be incorporated into parametric linear and generalized linear models. The use of smoothers with these standard statistical models allows the estimation of more flexible functional forms whilst retaining the interpretability of parametric models.
He has also made contributions via development of award-winning statistical software for matching and causal inference.
Finally, Professor Keele has made important contributions to the development of political methodology as an important professional and academic discipline. He has been active in the Society for Political Methodology, having served for many years as treasurer and member-at-large as well as being an author for the TPM newsletter. He currently serves as an Associate Editor at both the Journal of Causal Inference and Observational Studies. Professor Keele's website contains additional information: http://lukekeele.com/
Professor Rocio Titiunik is currently Professor of Politics at Princeton University. She is also associated with the Center for the Study of Democratic Politics and the Center for Statistics and Machine Learning at Princeton. Prior to Princeton she was the James Orin Murfin Professor of Political Science at the University of Michigan where she began her professional career as an Assistant and then Associate Professor in 2010. At Michigan she was also associated with the ISR Center for Political Studies and the Michigan Institute for Data Science. She received her PhD in Agricultural and Resource Economics from the University of California, Berkeley in 2009. She was the recipient of the Emerging Scholar Award for a young scholar making exceptional contributions to political methodology who is within ten years of their terminal degree, awarded by the Society of Political Methodology in 2016. Her work has been the recipient of many awards, including the Gosnell Prize and the Statistical Software Award from the Society for Political Methodology and the Robert H. Durr Award from the Midwest Political Science Association.
Professor Titiunik's research has focused on causal inference with particular attention to regression discontinuity design and estimation. One important contribution in a series of papers develops methods for determining the best window within which the "as-if" random assumption is appropriate and in which to observe the causal effects attributable to the discontinuity. A second widely cited contribution develops nonparametric confidence intervals for RD designs. A recent piece develops covariate adjustment RD procedures that improve the estimator's efficiency. She and co-authors have developed sophisticated methods for using geographic boundaries as the source of the discontinuity.
The substantive contributions span a broad set of topics. There are multiple papers examining legislative behavior, such as incumbency advantage and productivity, in a number of electoral contexts, from the U. S. Senate to state politics to Brazilian mayor contests. Another set of papers examines turnout and registration, again in a range of contexts. In many instances the direct topic may appear narrow, such as the incumbency advantage of Brazilian mayors and the effects of term limits. But the results and subsequent discussions address much larger questions, such as the effects of weak parties.
The published work includes several expositional pieces making causal inference and RD designs specifically available and comprehensible to a broad range of scholars. This is obviously important work for what otherwise could be a relatively technical and therefore opaque method for non-methodologists. Most notable is a recent monograph from Cambridge University Press titled a "Practical Introduction to Regression Discontinuity Designs."
Professor Titiunik has also made multiple and important software contributions that support the methodological innovations. Her list of publications includes six articles in prominent software journals. As noted above she received the SPM Statistical Software Award for this work.
The scholarly publications appear in the leading journals in multiple disciplines, from Political Science to Economics to Statistics. The recent Google Scholar citation count exceeds 5,000 cites. Again in a broad range of disciplines.
Professor Titiunik has an excellent reputation as teacher who regularly co-authors with graduate students. All describe her as rigorous, demanding, very supportive and an excellent advisor and mentor.
Her service to the discipline includes a term on the SPM Diversity Committee and as a member-at-large (2017 - 2019). She is currently on both the publications and the long-range planning committees. Professor Titiunik also plays important roles in the public sphere. Most significantly, she was a member of the expert panel on Regression Discontinuity Design standards for the Institute of Education Sciences, U. S. Dept. of Education. This was a panel of national experts to advise the Institute on RD designs.