Seminal Ideas and Controversies in Statistics

Roderick J. A. Little is Richard D. Remington Distinguished University Professor Emeritus at the University of Michigan, where he also holds emeritus appointments in the Department of Statistics and the Institute for Social Research. After secondary school at Glasgow Academy, he received a B.A. in Mathematics from Gonville and Caius College, Cambridge University, and M.Sc. and Ph.D. degrees in Statistics from the Imperial College of Science and Technology, London University. Professor Little is a pioneer and thought leader in the fields of statistical analysis with missing data, Bayesian inference in sample surveys and causal inference. He has received some of the highest honors in statistics and science, including being elected to the U.S. National Academy of Medicine and American Academy of Arts and Sciences.

“Little’s book was a lot of fun. When the ideas are taken together close in time, it broadens one’s perspective, which may not happen if the papers are read discretely. For me, Little’s through-line is that statistical inference is a toolkit of interconnected ideas that work reliably only when tied to design, modelling assumptions and calibration…I feel that the seminal works are not museum pieces but remain living guidance for modern statistical practice. This reaching back for solid ground feels timely for practising statisticians and scientists, with the foundation of epistemology under attack from many directions, including the proponents of artificial intelligence. I would appreciate more books like it.”~ Luke R. Lloyd-Jones, Australian & New Zealand Journal of Statistics“Little's thought-provoking book brings the debates about 15 important statistical concepts to life through reflections on key journal articles…Little based this book on a doctoral-student seminar he developed at the University of Michigan to enhance students' understanding about fundamental concepts of statistics. While the book is ideal for such a class, it is also a tremendous resource for teachers of statistics who want to deepen their knowledge and make the subject come alive for their students…Seminal Ideas and Controversies in Statistics is not an easy read, but it is a rewarding one. In my opinion, every student and teacher of statistics should read this book. It will give you a new perspective on concepts that you may have taken for granted, and a renewed appreciation for the remarkable contributions that statisticians have made.”~Sharon L. Lohr, Arizona State University, USA; published in the Journal of Applied Statistics, Dec 2025“This book truly is essential reading for all statisticians, PhD students in particular. Historical context is generally lacking in statistics PhD programs, and so this book fills an important void. All statistics graduate programs should require a semester long course on topics resembling those in Professor Little’s book – I am already planning to teach a new PhD seminar course largely focusing on its contents! Beyond pedagogy, this book is an excellent reference for statisticians at all stages of careers. Iconic historical examples are used throughout the book as essential centerpieces of the discussions, also making the book a sort of chronological catalogue of seminal ideas and controversies in statistics…I am immensely impressed by the thoroughness of Professor Little’s book. This truly is an exhaustive synopsis of the past century of the statistical literature and the controversies that ensued. The book lives up to its title.”~Jonathan P Williams (13 May 2026): Seminal Ideas and Controversies in Statistics (by Roderick J. A. Little), Journal of the American Statistical Association

• 1. Maximum likelihood. 2. To C or not to C-- that is the question. 3. Frequentist flaps: significance testing, hypothesis testing, or something else?. 4. Fiducial inference and the Behrens-Fisher problem. 5. Do you like the likelihood principle?. 6. A Bayesian/frequentist compromise: Calibrated Bayes. 7. Baseball averages, foreign cars, and shrinkage estimation. 8. Alternatives to least squares in regression. 9. Multiple perspectives on multiple comparisons. 10. Generalized Estimating Equations. 11.The Bootstrap and Bayesian Monte-Carlo methods. 12. Exploratory data analysis and data science. 13. Randomization in survey sampling. 14. Randomized clinical trials and the Neyman/Rubin causal model. 15. Propensity score methods.