Abstract

In this paper, we present a comparison, in terms of power, among three classic procedures (Scheffè, Tukey, Bonferroni) and two methods based on the False Discovery Rate (FDR) in the context of multiple comparison. A Monte Carlo simulation is illustrated, in which we performed all the possible pairwise comparisons among k (k = 3, 4, 5, 6) samples with fixed numerosity (20), by manipulating effect size. Three different definitions of power are considered, based on their possible application in psychological research. The results show that FDR-based procedures offer a better performance compared to the classical methods.