IMPROVED REPEATED-MEASURES STEPWISE MULTIPLE COMPARISON PROCEDURES

Approximate degrees of freedom omnibus and pairwise test statistics of Johansen (1980) and Keselman, Keselman, and Shaffer (1991), respectiv ely, were used with numerous stepwise multiple comparison procedures ( MCPs) to perform pairwise contrasts on repeated measures means. The MC Ps were compared for their overall familywise rates of Type I error an d for their sensitivity to detect true pairwise differences among mean s when multisample sphericity and multivariate normality assumptions w ere not satisfied. Results indicated that multiple range procedures wh ich were modified according to the method described by Duncan (1957) w ere always robust with respect to Type I errors and were at least as p owerful as the unmodified range procedures, and could result in increa ses in power as large as 22%. Overall, the Welsch (1977a) step-up, Per itz-Duncan (Peritz, 1970), and Ryan-Welsch-Duncan (Ryan, 1960; Welsch, 1977a) multiple range procedures were found to be most powerful.