Power and Robustness of Nonparametric Covariate Adjustment Methods for Ordinal Data

• Activity Number: 256
• Type: Contributed
• Date/Time: Monday, August 2, 2010 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Wei Tian*+ and Muhtarjan Osman
• Companies: Inspire Pharmaceuticals and North Carolina State University
• Keywords: Power ; Robustness ; Covariates Adjustment ; Non-parametric ; ANCOVA ; Ordinal Data

Abstract

The goal of this work is to evaluate the performance of parametric ANCOVA and non-parametric approaches in terms of power and type I error rate for ordinal data. Three non-parametric approaches studied are Cochran-Mantel-Haenszel (CMH) test controlling for baseline score, the randomization-based approach proposed by Koch et al. (1998), and the non-parametric method proposed by Schacht et al. (2008). In general, the results of the simulation study indicate that the parametric ANCOVA method is robust for ordinal data in the sense that the type I error rate is well controlled around the nominal level. The power of the parametric ANCOVA is higher than the CMH test. Additionally, the power of Koch's method is almost identical to the parametric ANCOVA. The approach by Schacht et al. results in the highest power among all methods considered, but its type I error rate is inflated up to 40%.