Randomization: Beyond the Closurization Principle
To celebrate the launch of a new short course on the design and analysis of cluster randomised and stepped wedge trials, the Centre for Evaluation is hosting a seminar from Professor Larry Moulton.
This presentation focuses on the randomization of cluster randomized trials, which typically have small numbers of randomization units. Three desirable properties of a randomization procedure are: (1) unbiasedness; (2) validity; and (3) high probability of achieving near-balance on baseline covariates. Methods for achieving not only a high probability but certainty of near-balance are discussed. Validity, a concept relatively unknown among clinical trialists, is explained.
Professor Larry Moulton is a biostatistician at the Johns Hopkins Bloomberg School of Public Health. His research focuses on studies of infectious diseases in low and middle income countries. He is involved in numerous large cluster randomized trials and works on methods for the design, especially the randomization, of such trials, as well as methods of analysis of the resulting correlated data.
This session will be live-streamed and recorded - accessible to both internal and external audience.