One of the most frequent criticisms leveled against using nonparametric statistics is their alleged lack of power or sensitivity compared to their parametric counterparts. Empirical studies of the power of nonparametric procedures versus their parametric partners have shown that their lack of power, in many situations, is a myth. Where there is evidence that distributions depart appreciably from normality and the variances are not homogeneous, the probability is enhanced that the power advantage can be found in the nonparametric procedures.
Get full access to this article
View all access options for this article.
References
1.
AndrewsF. C.Asymptotic behavior of some rank tests for analysis of variance. Annals of Mathematical Statistics, 1954, 25, 724–736.
2.
AtiqullahM.The robustness of the covariance analysis of a one-way classification. Biometrika, 1964, 91, 365–372.
3.
BlairR. C.HigginsJ. J.On the relative power of the U and t tests. British Journal of Mathematical and Statistical Psychology, 1980, 33, 114–120.
4.
BlairR. C.HigginsJ. J.A note on the A.R.E. of the Wilcoxon sign rank test relative to the independent means #-test under mixtures of two normal distributions. British Journal of Mathematical and Statistical Psychology, 1981, 34, 124–128.
5.
BoneauC. A.The effects of violations of assumptions underlying the f-test. Psychological Bulletin, 1960, 16, 260–261.
BradleyJ. V.Bizarre distribution shapes. American Statistician, 1977, 31, 147–150.
8.
CaponJ.Asymptotic efficiency of certain locally most powerful rank tests. Annals of Mathematical Statistics, 1961, 32, 88–100.
9.
ChernoffH.SavageI. R.Asymptotic normality and efficiency of certain nonparametric statistics. Annals of Mathematical Statistics, 1958, 29, 972–975.
10.
DixonW. J.The statistical sign test. Journal of the American Statistical Association, 1946, 41, 557–566.
11.
DixonW. J.Power under normality of several nonparametric tests. Annals of Mathematical Statistics, 1954, 25, 610–614.
12.
DonaldsonT. S.Robustness of the F test to errors of both kinds and the correlation between the numerator and denominator of the F ratio. Journal of the American Statistical Association, 1968, 63, 660–676.
13.
GlassG. V.PeckhamP. D.SandersJ. R.Consequences of failure to meet assumptions underlying the analysis of variance and covariance. Review of Educational Research, 1972, 42, 237–289.
14.
HodgesJ. L.LehmannE. L.The efficacy of some nonparametric competitors of the t test. Annals of Mathematical Statistics, 1956, 27, 324–335.
15.
HollanderM.WolfeD. A.Nonparametric design and analysis. New York: Wiley, 1973.
16.
KirkR. E.Experimental design: Procedures for the behavioral sciences. (2nd ed.) Belmont, CA: Brooks/Cole, 1982.
17.
LeachC.Introduction to statistics: A nonparametric approach for the behavioral sciences. New York: Wiley, 1979.
18.
LeavertonP.BirchJ. J.Small sample power curves for the two sample location problem. Technometrics, 1969, 11, 299–307.
19.
LeeE. T.DesuM. M.GehanG. A.A Monte Carlo study of the power of some two sample tests. Technometrics, 1975, 62, 425–432.
20.
MarascuiloL. A.McSweeneyM.Nonparametric and distribution-free methods for the social sciences. Monterey, CA: Brooks/Cole, 1977.
21.
NeaveH. R.GrangerW. J.A Monte Carlo study comparing various two sample tests for differences in means. Technometrics, 1968, 10, 509–522.
22.
PrattJ. W.Robustness of some procedures for the two sample location problem. Journal of the American Statistical Association, 1964, 59, 665–680.
23.
PuriM. L.Asymptotic efficiency of a class of c-sample tests. Annals of Mathematical Statistics, 1964, 35, 102–121.
24.
SiegelS.Nonparametric statistics. New York: McGraw-Hill, 1956.
25.
YoungR. K.VeldmanD. J.Heterogeneity and skewness in analysis of variance. Perceptual and Motor Skills, 1963, 16, 588.
