A two and more independent factor form of the Wilcoxon-Mann-Witney test, extendable to other permutation-based tests

Chris Holland; Tom Holland

A two and more independent factor form of the Wilcoxon-Mann-Witney test, extendable to other permutation-based tests

Chris Holland, Tom Holland

School of Education, Communication & Society

Triveritas Ltd.

Research output: Contribution to journal › Article › peer-review

2 Citations (Scopus)

143 Downloads (Pure)

Abstract

The Wilcoxon-Mann-Witney test is extended to account for a second independent factor. The new test statistic's probability mass function and normal approximation are derived. Critical-values for balanced, unbalanced, and large sample designs are given. The immediate extension of this method to a wide range of non-parametric tests is explained.

Original language	English
Pages (from-to)	123-140
Number of pages	18
Journal	Journal of Modern Applied Statistical Methods
Volume	15
Issue number	1
Publication status	Published - 2016

Keywords

Nuisance factors
Permutation
Rank, order
Two factor nonparametric test
Wilcoxon-Mann-Witney test

Access to Document

A Two and More Independent_HOLLAND_Accepted 20Nov2014_GOLD VoRFinal published version, 512 KBLicence: CC BY

http://digitalcommons.wayne.edu/jmasm/vol15/iss1/8/

Cite this

@article{24265e7f386c441d815a2ae45d941cf9,

title = "A two and more independent factor form of the Wilcoxon-Mann-Witney test, extendable to other permutation-based tests",

abstract = "The Wilcoxon-Mann-Witney test is extended to account for a second independent factor. The new test statistic's probability mass function and normal approximation are derived. Critical-values for balanced, unbalanced, and large sample designs are given. The immediate extension of this method to a wide range of non-parametric tests is explained.",

keywords = "Nuisance factors, Permutation, Rank, order, Two factor nonparametric test, Wilcoxon-Mann-Witney test",

author = "Chris Holland and Tom Holland",

year = "2016",

language = "English",

volume = "15",

pages = "123--140",

journal = "Journal of Modern Applied Statistical Methods",

issn = "1538-9472",

publisher = "Wayne State University",

number = "1",

}

TY - JOUR

T1 - A two and more independent factor form of the Wilcoxon-Mann-Witney test, extendable to other permutation-based tests

AU - Holland, Chris

AU - Holland, Tom

PY - 2016

Y1 - 2016

N2 - The Wilcoxon-Mann-Witney test is extended to account for a second independent factor. The new test statistic's probability mass function and normal approximation are derived. Critical-values for balanced, unbalanced, and large sample designs are given. The immediate extension of this method to a wide range of non-parametric tests is explained.

AB - The Wilcoxon-Mann-Witney test is extended to account for a second independent factor. The new test statistic's probability mass function and normal approximation are derived. Critical-values for balanced, unbalanced, and large sample designs are given. The immediate extension of this method to a wide range of non-parametric tests is explained.

KW - Nuisance factors

KW - Permutation

KW - Rank, order

KW - Two factor nonparametric test

KW - Wilcoxon-Mann-Witney test

UR - http://www.scopus.com/inward/record.url?scp=84975221717&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84975221717

SN - 1538-9472

VL - 15

SP - 123

EP - 140

JO - Journal of Modern Applied Statistical Methods

JF - Journal of Modern Applied Statistical Methods

IS - 1

ER -

A two and more independent factor form of the Wilcoxon-Mann-Witney test, extendable to other permutation-based tests

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this