A Nonparametric Method to Assess Treatment Effects for Unbalanced Designs Using SAS/IML
Two-way randomized block designs with non-normally distributed outcomes may be frequently encountered by researchers. The nonparametric test developed by Friedman can be used to compare more than two treatments when the blocks are completely balanced. Skillings and Mack proposed a general method for comparing treatments for incomplete blocks when the observations are missing arbitrarily. A program to calculate the Skillings-Mack test statistic and the p-value for the hypothesis of no treatment effect was developed using SAS/IML in SAS version 9.1. Designed for users with intermediate SAS experience, the program reads the results of the unbalanced two-way design directly into IML for analysis. Use of this program will be demonstrated using quality reporting data from a leading healthcare provider.