A modification to testing pairwise comparisons that may provide better control of Type I errors in the presence of non-normality is to use a preliminary test for symmetry which determines whether data should be trimmed symmetrically or asymmetrically. Several pairwise MCPs were investigated, employing a test of symmetry with a number of heteroscedastic test statistics that used trimmed means and Winsorized variances. Results showed improved Type I error control than competing robust statistics.
Kowalchuk, Rhonda K.; Keselman, H. J.; Wilcox, Rand R.; Algina, James; Algina, James; and Algina, James
"Multiple Comparison Procedures, Trimmed Means And Transformed Statistics,"
Journal of Modern Applied Statistical Methods:
1, Article 6.
Available at: http://digitalcommons.wayne.edu/jmasm/vol5/iss1/6