Invitation to a free/virtual webinar Sponsored by CHANCE magazine and the ASA History of Statistics Interest Group
Where: Zoom
When: September 24, 2025 at 4:00 PM (US-ET)
Registration: https://bit.ly/Sept-Chance-Webinar. Click or tap if you trust this link." rel="noopener">https://bit.ly/Sept-Chance-Webinar
TITLE: A not-well-known Lionel Penrose - Ronald Fisher collaboration
SPEAKER: James Hanley, Professor emeritus, Department of Epidemiology, Biostatistics and Occupational Health, McGill University
ABSTRACT
Lionel Penrose's two influential articles established that maternal age is a significant factor in the probability that a child will be affected by Down Syndrome, while paternal age (1933) and birth order (1934) are not.
I will describe (1) the data, the innovative analyses, the additional methodological article, and the unrecognized-until-now role of Ronald Fisher (2) the challenges we had and the lessons we learned when we tried to reconstruct his 1933 data and apply modern statistical methods (3) our discovery that his main 1934 analysis is based on conditional logistic regression, a technique that statisticians and epidemiologists generally attribute to Prentice & Breslow (1978) and Breslow et al. (1978), while economists attribute it to Nobel laureate McFadden (1973).
BIO
James Hanley received a BSc and MSc in Mathematics/Statistics from University College Cork (National University of Ireland) and a PhD in Statistics/Biometry from the University of Waterloo in Canada. He spent his first 7 years post PhD as a clinical trials statistician at SUNY/Buffalo (1973-1977) and The Sidney Farber Cancer Institute and the Department of Biostatistics in the Harvard School of Public Health (1977-1980). He joined McGill University in Montreal in 1980. There, in addition to teaching in the department's graduate programs in epidemiology and biostatistics, he collaborated widely in research areas from pediatrics to geriatrics. His own early work focused on summaries of the receiver operating characteristics of diagnostic tests. His recent work focused on measuring the mortality reductions produced by cancer screening. He has a keen interest in teaching, in archival research, and in the history of statistics and public health. He has published several articles on these topics, as well as several expository articles on statistical methods.
For more details, and his full cv, see his website: https://jhanley.biostat.mcgill.ca. For a pictures-only bio, see https://jhanley.biostat.mcgill.ca/120_second_bio.pdf
------------------------------
Wendy Martinez
Pronouns: she, they
Senior Mathematical Statistician for Data Science
US Census Bureau
------------------------------