“…Although the Type I error rate is widely viewed as being relatively unaffected by non-normality, Bradley (1980) has pointed out conditions in which this is not true. This finding is also evident in the findings of recent studies and published texts (e. g., See Hempel, Ronchetti, & Rousseeuw, 1986;Huber & Ronchetti, 2009;Maronna, Martin, & Yohai, 2006;Micceri, 1989;Schoder, et al, 2006;Staudte & Sheather, 1990;Wilcox, 2012a, b;Wilcox & Keselman, 2003).…”