Pairwise Comparisons of Means in Clothing and Textiles Research

Abstract

PurposeThe purpose of this paper is twofold: (a) to examine methods used by clothing and textiles researchers to conduct pairwise mean comparisons, and (b) to review the issue of Type I error control and power of these and other more recently developed PMC procedures. Given this background, the major intent of this paper is to provide researchers with some recommendations for selection of a PMC procedure. Since ease of application is often a consideration in choosing a multiple comparison procedure (Seaman, Levin, & Serlin, 1991), methods for obtaining . numerical results for the recommended procedures will be discussed, and illustrative examples are provided.

Clothing deprivation, clothing satisfaction, fashion leadership and Hong Kong young consumers

“…Scheffe's post hoc comparison was followed when significant difference was found in the ANOVA results. The reason for using Scheffe's procedure was that it was designed for comparing a large number of means instead of two group means only (Lix 1995). In this study, differences between four fashion groups (fashion innovators, fashion opinion leaders, fashion innovative communicators, and fashion followers) were to be investigated at the same time, therefore Scheffe's procedure was the suitable choice.…”