A Comparison of Two Rank Tests for Repeated Measures Designs
نویسندگان
چکیده
The classic rank-based method for comparing J dependent groups is Friedman’s test. Consider a random sample of n vectors from some J-variate distribution. As is well-known, Friedman’s test assigns ranks to the values within each vector and is based on a compound symmetry assumption under the hypothesis of no treatment effect (e.g., Brunner, Domhof, & Langer, p. 68). That is, the distribution is assumed to be invariant under all permutations, which implies that the variances and covariances are equal. Two attempts at improving upon Friedman’s are based in part by assigning ranks to the pooled data instead (Iman, 1974; Quade, 1979). Subsequently, Agresti and Pendergast (1986) proposed a rank-based test that was found to provide better control over the probability of a Type I error and better power. (For relevant theoretical results, see Kepner & Robinson, 1988.) Two alternative methods are
منابع مشابه
New Nonparametric Rank Tests for Interactions in Factorial Designs with Repeated Measures
New rank tests for interactions in factorial designs are presented and applied to some common factorial designs with repeated measures. The resulting p-values of these tests are compared, along with those obtained by parametric and randomization tests.
متن کاملNonparametric Multiple Comparisons in Repeated Measures Designs for Data with Ties
We consider some multiple comparison problems in repeated measures designs for data with ties, particularly ordinal data; the methods are also applicable to continuous data, with or without ties. A unified asymptotic theory of rank tests of Brunner, Puri and Sen (1995) and Akritas and Brunner (1997) is utilized to derive large sample multiple comparison procedures (MCP’s). First, we consider a ...
متن کاملAligned Rank Tests As Robust Alternatives For Testing Interactions In Multiple Group Repeated Measures Designs With Heterogeneous Covariances
متن کامل
Comparison of aligned Friedman rank and parametric methods for testing interactions in split-plot designs
Parametric methods are commonly used despite evidence that model assumptions are often violated. Various statistical procedures have been suggested for analyzing data from multiple-group repeated measures (i.e., split-plot) designs when parametric model assumptions are violated (e.g., Akritas and Arnold (J. Amer. Statist. Assoc. 89 (1994) 336); Brunner and Langer (Biometrical J. 42 (2000) 663))...
متن کاملComparison of electromyographic activity of shoulder muscles during two different pull-up test methods
Pull-up and modified pull-up tests are frequently used for shoulder muscles endurance evaluation by physical educators. Despite of using these tests, the difference of muscle activation pattern in pull-up and modified pull-up tests has not been documented clearly. However, the role of some important muscles such as anterior deltoid, posterior deltoid, trapezius and latissimus dorsi have no...
متن کامل