Sample-Size Software for Ordered Categorical Data

Ordered categorical variables arise frequently in cancer clinical trials and other biomedicalstudies. The statistical procedures for analyzing such data are well known and software for performingthe analysis is readily available. The basic idea is to condition on the margins of the contingency tablecreated by the categorical data and thereby obtain a distribution free test that automatically corrects forties. Despite the popularity of this conditional approach for analyzing ordered categorical data there hasbeen very little work done on power and sample-size considerations at the design phase. A biomedicalinvestigator about to launch a clinical trial for comparing two treatments with ordered categoricaloutcomes will find it extremely difficult to determine what sample size is needed. Either the investigatormust assume that the data are continuous, or else that the data are binary, since these are the onlycases for which reliable methods and software are available. Both approaches are inappropriate forordered categorical data. We propose to fill the void by providing new exact and Monte Carlo methodsthat provide accurate power and sample-size estimates for conditional tests on ordered categorical data.