Probability of coverage,rate of rejection,and average location relative efficiency for trimmed and partial variance procedures

Abstract

Recent concerns about the normality of the distribution of scores arose from a study that was done by Micceri (as cited in [20]). In addition, the simultaneous occurrence of variance heterogeneity and the treatment effect influenced the decision to execute this study [9]. Heterogeneity of score variances and distributional asymmetry influence one's decision on the most appropriate procedure. Through a Monte Carlo simulation design, I assessed the probability of coverage, average location relative efficiency, and rate of rejection for the selected statistical procedures. When the purpose of the study is to estimate the populations' mean difference, the partial variance procedure provides a better approximation of the confidence level and higher location relative efficiency than that observed for Yuen's procedure.