Exact and asymptotic tests for homogeneity in several 2 x 2 tables

Isildinha Reis, Karim F. Hirji, Abdelmonem A. Afifi

Research output: Contribution to journalArticle

28 Citations (Scopus)

Abstract

This paper presents the results of a Monte Carlo study comparing the performance, in terms of size and power, of six exact and six asymptotic tests for the homogeneity of odds ratios in several 2 x 2 contingency tables. With a small sample size or sparse data structure, the exact tests performed better than the asymptotic tests because they maintained the nominal size and, in some situations, had slightly higher power. Among the exact tests, we recommend the Zelen, Pearson chi-square and scores tests. Among the asymptotic tests, the Breslow-Day and Pearson chi-square tests were slightly better in some situations than the unconditional and conditional score tests. However, both exact and asymptotic tests had low power for small strata sizes, even with moderate to large heterogeneity of odds ratios. Corroborating previous findings, the asymptotic unconditional likelihood ratio test was too liberal in terms of size.

Original languageEnglish
Pages (from-to)893-906
Number of pages14
JournalStatistics in Medicine
Volume18
Issue number8
DOIs
StatePublished - Apr 30 1999
Externally publishedYes

Fingerprint

Asymptotic Test
Exact Test
Chi-Square Distribution
Homogeneity
Tables
Odds Ratio
Chi-squared test
Score Test
Sample Size
Conditional Score
Conditional Test
Sparse Data
Contingency Table
Small Sample Size
Likelihood Ratio Test
Monte Carlo Study
High Power
Categorical or nominal
Data Structures

ASJC Scopus subject areas

  • Epidemiology

Cite this

Exact and asymptotic tests for homogeneity in several 2 x 2 tables. / Reis, Isildinha; Hirji, Karim F.; Afifi, Abdelmonem A.

In: Statistics in Medicine, Vol. 18, No. 8, 30.04.1999, p. 893-906.

Research output: Contribution to journalArticle

Reis, Isildinha ; Hirji, Karim F. ; Afifi, Abdelmonem A. / Exact and asymptotic tests for homogeneity in several 2 x 2 tables. In: Statistics in Medicine. 1999 ; Vol. 18, No. 8. pp. 893-906.
@article{873942cc90804588a408444a78ee8528,
title = "Exact and asymptotic tests for homogeneity in several 2 x 2 tables",
abstract = "This paper presents the results of a Monte Carlo study comparing the performance, in terms of size and power, of six exact and six asymptotic tests for the homogeneity of odds ratios in several 2 x 2 contingency tables. With a small sample size or sparse data structure, the exact tests performed better than the asymptotic tests because they maintained the nominal size and, in some situations, had slightly higher power. Among the exact tests, we recommend the Zelen, Pearson chi-square and scores tests. Among the asymptotic tests, the Breslow-Day and Pearson chi-square tests were slightly better in some situations than the unconditional and conditional score tests. However, both exact and asymptotic tests had low power for small strata sizes, even with moderate to large heterogeneity of odds ratios. Corroborating previous findings, the asymptotic unconditional likelihood ratio test was too liberal in terms of size.",
author = "Isildinha Reis and Hirji, {Karim F.} and Afifi, {Abdelmonem A.}",
year = "1999",
month = "4",
day = "30",
doi = "10.1002/(SICI)1097-0258(19990430)18:8<893::AID-SIM84>3.0.CO;2-5",
language = "English",
volume = "18",
pages = "893--906",
journal = "Statistics in Medicine",
issn = "0277-6715",
publisher = "John Wiley and Sons Ltd",
number = "8",

}

TY - JOUR

T1 - Exact and asymptotic tests for homogeneity in several 2 x 2 tables

AU - Reis, Isildinha

AU - Hirji, Karim F.

AU - Afifi, Abdelmonem A.

PY - 1999/4/30

Y1 - 1999/4/30

N2 - This paper presents the results of a Monte Carlo study comparing the performance, in terms of size and power, of six exact and six asymptotic tests for the homogeneity of odds ratios in several 2 x 2 contingency tables. With a small sample size or sparse data structure, the exact tests performed better than the asymptotic tests because they maintained the nominal size and, in some situations, had slightly higher power. Among the exact tests, we recommend the Zelen, Pearson chi-square and scores tests. Among the asymptotic tests, the Breslow-Day and Pearson chi-square tests were slightly better in some situations than the unconditional and conditional score tests. However, both exact and asymptotic tests had low power for small strata sizes, even with moderate to large heterogeneity of odds ratios. Corroborating previous findings, the asymptotic unconditional likelihood ratio test was too liberal in terms of size.

AB - This paper presents the results of a Monte Carlo study comparing the performance, in terms of size and power, of six exact and six asymptotic tests for the homogeneity of odds ratios in several 2 x 2 contingency tables. With a small sample size or sparse data structure, the exact tests performed better than the asymptotic tests because they maintained the nominal size and, in some situations, had slightly higher power. Among the exact tests, we recommend the Zelen, Pearson chi-square and scores tests. Among the asymptotic tests, the Breslow-Day and Pearson chi-square tests were slightly better in some situations than the unconditional and conditional score tests. However, both exact and asymptotic tests had low power for small strata sizes, even with moderate to large heterogeneity of odds ratios. Corroborating previous findings, the asymptotic unconditional likelihood ratio test was too liberal in terms of size.

UR - http://www.scopus.com/inward/record.url?scp=0033617330&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0033617330&partnerID=8YFLogxK

U2 - 10.1002/(SICI)1097-0258(19990430)18:8<893::AID-SIM84>3.0.CO;2-5

DO - 10.1002/(SICI)1097-0258(19990430)18:8<893::AID-SIM84>3.0.CO;2-5

M3 - Article

VL - 18

SP - 893

EP - 906

JO - Statistics in Medicine

JF - Statistics in Medicine

SN - 0277-6715

IS - 8

ER -