Rank-based tests for interactions in a two-way design

✍ Scribed by Rand R. Wilcox

• Publisher: Elsevier Science
• Year: 1999
• Tongue: English
• Weight: 88 KB
• Volume: 29
• Category: Article
• ISSN: 0167-9473
• DOI: 10.1016/s0167-9473(98)00076-0

✦ Synopsis

Among the nonparametric approaches to deÿning and testing the hypothesis of no interactions in a two-way design, the method by Patel and Hoel is appealing because it is a function of the ranks and therefore invariant under monotonic increasing transformations of the data, and it does not require the assumption of no main e ects for the approach to be valid. Marden and Muyot corrected and extended the approach by Patel and Hoel, but published papers strongly suggest that their approach will provide poor probability coverage and poor control over the probability of a type I error when sample sizes are small. This note conÿrms this and considers two other methods for obtaining more satisfactory results, one of which is relatively e ective.