Technical Reports : Probability Models in RCTs


The importance of the normality assumption in large public health data sets.

  1. Summary:  A demonstration of the robustness of the distribution-free inference on the mean. 

  2. Ref: Lumley T, Diehr P, Emerson S, Chen L, Annual Review Public Health 23: 151-69 (2002).

Some observations on the Wilcoxon rank sum test.

  1. Summary: A discussion of the properties of the Wilcoxon rank sum test, particularly noting popular misconceptions about its scientific interpretation and its efficiency relative to the t test.

  2. Ref: Emerson SS, UW Biostatistics Working Paper Series, Working Paper 380, BEPress (2011)

Adjusting for Baseline in Randomized Clinical Trials.

  1. Summary: A discussion of the theory that guides the best approaches for analyzing RCT outcomes when a baseline measurement of that same variable is available.