Chi-square tests for comparing vectors of proportions for several cluster samples

Kenneth J. Koehler, Jeffrey Wilson

Research output: Contribution to journalArticle

24 Citations (Scopus)

Abstract

Test statistics are developed for comparing vectors of proportions obtained from several independent two-stage cluster samples. It is assumed that clusters are selected with probability proportional to size for each sample. Wald's general method of constructing quadratic forms is used to obtain a large sample chi-square test. More easily evaluated chi-square tests are derived from the Dirichlet-multinomial model. Corresponding goodness-of-fit tests for the Dirichlet-multinomial model are also derived.

Original languageEnglish (US)
Pages (from-to)2977-2990
Number of pages14
JournalCommunications in Statistics - Theory and Methods
Volume15
Issue number10
DOIs
StatePublished - Jan 1 1986

Fingerprint

Chi-squared test
Multinomial Model
Proportion
Dirichlet
Goodness of Fit Test
Quadratic form
Test Statistic
Directly proportional

Keywords

  • Dirichlet-multinomial model
  • Pearson chi-square test
  • Wald statistics

ASJC Scopus subject areas

  • Statistics and Probability

Cite this

Chi-square tests for comparing vectors of proportions for several cluster samples. / Koehler, Kenneth J.; Wilson, Jeffrey.

In: Communications in Statistics - Theory and Methods, Vol. 15, No. 10, 01.01.1986, p. 2977-2990.

Research output: Contribution to journalArticle

@article{cd8f562ed5e1460593a7f51dbd34dcab,
title = "Chi-square tests for comparing vectors of proportions for several cluster samples",
abstract = "Test statistics are developed for comparing vectors of proportions obtained from several independent two-stage cluster samples. It is assumed that clusters are selected with probability proportional to size for each sample. Wald's general method of constructing quadratic forms is used to obtain a large sample chi-square test. More easily evaluated chi-square tests are derived from the Dirichlet-multinomial model. Corresponding goodness-of-fit tests for the Dirichlet-multinomial model are also derived.",
keywords = "Dirichlet-multinomial model, Pearson chi-square test, Wald statistics",
author = "Koehler, {Kenneth J.} and Jeffrey Wilson",
year = "1986",
month = "1",
day = "1",
doi = "10.1080/03610928608829290",
language = "English (US)",
volume = "15",
pages = "2977--2990",
journal = "Communications in Statistics - Theory and Methods",
issn = "0361-0926",
publisher = "Taylor and Francis Ltd.",
number = "10",

}

TY - JOUR

T1 - Chi-square tests for comparing vectors of proportions for several cluster samples

AU - Koehler, Kenneth J.

AU - Wilson, Jeffrey

PY - 1986/1/1

Y1 - 1986/1/1

N2 - Test statistics are developed for comparing vectors of proportions obtained from several independent two-stage cluster samples. It is assumed that clusters are selected with probability proportional to size for each sample. Wald's general method of constructing quadratic forms is used to obtain a large sample chi-square test. More easily evaluated chi-square tests are derived from the Dirichlet-multinomial model. Corresponding goodness-of-fit tests for the Dirichlet-multinomial model are also derived.

AB - Test statistics are developed for comparing vectors of proportions obtained from several independent two-stage cluster samples. It is assumed that clusters are selected with probability proportional to size for each sample. Wald's general method of constructing quadratic forms is used to obtain a large sample chi-square test. More easily evaluated chi-square tests are derived from the Dirichlet-multinomial model. Corresponding goodness-of-fit tests for the Dirichlet-multinomial model are also derived.

KW - Dirichlet-multinomial model

KW - Pearson chi-square test

KW - Wald statistics

UR - http://www.scopus.com/inward/record.url?scp=0001303184&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0001303184&partnerID=8YFLogxK

U2 - 10.1080/03610928608829290

DO - 10.1080/03610928608829290

M3 - Article

VL - 15

SP - 2977

EP - 2990

JO - Communications in Statistics - Theory and Methods

JF - Communications in Statistics - Theory and Methods

SN - 0361-0926

IS - 10

ER -