Proof in Medicine: The Role of Statistical Thinking
Joel B. Greenhouse
Carnegie Mellon University
In 2009, Hal Varian from Google famously predicted that “the sexy job in the next 10 years will be statisticians.” Was he wrong? Have Data Scientists replaced Statistical Scientists? In this talk I not only agree with Varian (duh!) but argue that statisticians today have never been more critical to contributing to the advancement of the biomedical and public health sciences. Although there is a wide range of activities that engage statistical scientists, the one common element central to all of them is statistical thinking. Good statistical thinking, in part, requires a nontrivial understanding of the real-world problem; involves judgments about the relevance of the data and models; and evaluates the viability of alternative explanations for observed results. Using case studies from the history of medicine, I will illustrate how statistical thinking has always been a core element in what is now called the data sciences, and plays a central role in helping the scientific community reach consensus about causal questions of interest.
Joel B. Greenhouse, Ph.D., is Professor of Statistics at Carnegie Mellon University, and Adjunct Professor of Psychiatry and Epidemiology at the University of Pittsburgh. He is an elected Fellow of the American Statistical Association, the American Association for the Advancement of Science, and an elected Member of the International Statistical Institute. Professor Greenhouse is a recipient of Carnegie Mellon University's Robert E. Doherty Prize for Educational Leadership, and the William & Francis Ryan Teaching Award for Effectiveness in Teaching.
Greenhouse has served as President of ENAR and is currently an elected member of the Executive Board of the International Biometrics Society. He has served on several National Academy of Sciences' committees, including the Committee on National Statistics, the Panel on Combining Information, and the Institute of Medicine's Committee on the Assessment of Family Violence Interventions. He has also served on a number of data and safety monitoring boards and scientific advisory committees for the National Institutes of Health, the Veterans Administration, and the US Departments of Education and Transportation.
Greenhouse is one of the Editors-in-Chief of the journal Statistics in Medicine, and is a past editor of the Institute of Mathematical Statistics' Lecture Notes and Monograph Series. His research interests include applications of Bayesian methods in practice, and methodological issues related to the use of research synthesis, especially as it is used to synthesize evidence for making policy and for scientific discovery.