Learning from the Heavyweights: Predicting the Future of Biostatistics

While not exactly as contentious as Ali vs. Frazier or Sugar Ray vs. Duran, debates occasionally arise about the fundamental nature of evolving academic fields.  In his XL-Files article for the IMS Bulletin, “Statistics vs Data Science: a 3-year-old prediction”, Professor Xiao-Li Meng describes a classic match-up on the direction of Biostatistics vs. Biostatistical science, and its potential relevance for other fields.  

Visionary scholars and pioneers of biostatistics, Marvin Zelen of the Harvard T.H. Chan School of Public Health and Chin Long Chiang of the University of California, Berkeley, long championed different theories about the identity of biostatistics as a discipline.  While Zelen predicted that ‘biostatistical science’ would emerge as a discipline that would place more emphasis on knowledge in substantive fields and training in computing, Chiang argued that biostatistics was already a discipline, “concerned with the development and application of statistical theory and methods for the study of phenomena arising in the life science”. 

While differing in their emphasis on computing and substantive scientific knowledge versus theoretical model-building, as well as the relative importance of breadth versus depth in defining the scope of the field, the predictions of both Zelen and Chiang have both proved true, and not simply for biostatistics. The emerging field of Data Science depends on both the volume and velocity provided by computing, and the variety of more sophisticated stochastic modeling. Ultimately, Meng’s article suggests that the outcome of the match may be a draw in the sense that simultaneously deepening the foundations and expanding the horizons of academic inquiry will prove key to the future of all existing and emerging disciplines.