论文信息 - Power study of anova versus Kruskal-Wallis test

Power study of anova versus Kruskal-Wallis test

Abstract This paper describes the comparison of the anova and the Kruskal-Wallis test by means of the power when violating the assumption about normally distributed populations. The permutation method is used as a simulation method to determine the power of the test. It appears that in the case of asymmetric populations the non-parametric Kruskal-Wallis test performs better than the parametric equivalent anova method.

T. V. Hecke | T. Hecke | T. Van Hecke

[1] W. Kruskal,et al. Use of Ranks in One-Criterion Variance Analysis , 1952 .

[2] G. Glass,et al. Consequences of Failure to Meet Assumptions Underlying the Fixed Effects Analyses of Variance and Covariance , 1972 .

[3] J. J. Higgins,et al. A Comparison of the Power of Wilcoxon's Rank-Sum Statistic to that of Student'st Statistic Under Various Nonnormal Distributions , 1980 .

[4] P. Good,et al. Permutation Tests: A Practical Guide to Resampling Methods for Testing Hypotheses , 1995 .

[5] R. Wilcox. ANOVA: A Paradigm for Low Power and Misleading Measures of Effect Size? , 1995 .

[6] S. Sawilowsky,et al. Analysis of Likert scale data in disability and medical rehabilitation research. , 1998 .

[7] M. Kenward,et al. An Introduction to the Bootstrap , 2007 .

[8] Morten W Fagerland,et al. The Wilcoxon–Mann–Whitney test under scrutiny , 2009, Statistics in medicine.