Chi-square test for R × C contingency tables with clustered data

Sin Ho Jung, Seung Ho Kang, Chul Ahn

Research output: Contribution to journalArticle

10 Citations (Scopus)

Abstract

In ophthalmologic or dental studies, observations are frequently taken from multiple sites (called units), such as eyes or teeth, of each subject. In this case, observations within each subject (called clusters) may be dependent, although those from different subjects are independent. When a categorical observation is made from each site, application of the usual Pearson chi-square tests is invalid since sites within the same subject tend to be dependent. We propose a modified χ2 statistic for testing no treatment effect in these cases. The proposed methods do not require correct specification of the dependence structure within cluster. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to real-life data.

Original languageEnglish
Pages (from-to)241-251
Number of pages11
JournalJournal of Biopharmaceutical Statistics
Volume13
Issue number2
DOIs
Publication statusPublished - 2003 May 10

Fingerprint

Clustered Data
Chi-squared test
Contingency Table
Chi-Square Distribution
Tooth
Dependent
Dependence Structure
Treatment Effects
Categorical
Statistic
Observation
Simulation Study
Tend
Specification
Testing
Unit

All Science Journal Classification (ASJC) codes

  • Pharmacology (medical)
  • Pharmacology, Toxicology and Pharmaceutics(all)

Cite this

@article{3813171937e34b4b8404a2f09dba2368,
title = "Chi-square test for R × C contingency tables with clustered data",
abstract = "In ophthalmologic or dental studies, observations are frequently taken from multiple sites (called units), such as eyes or teeth, of each subject. In this case, observations within each subject (called clusters) may be dependent, although those from different subjects are independent. When a categorical observation is made from each site, application of the usual Pearson chi-square tests is invalid since sites within the same subject tend to be dependent. We propose a modified χ2 statistic for testing no treatment effect in these cases. The proposed methods do not require correct specification of the dependence structure within cluster. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to real-life data.",
author = "Jung, {Sin Ho} and Kang, {Seung Ho} and Chul Ahn",
year = "2003",
month = "5",
day = "10",
doi = "10.1081/BIP-120019269",
language = "English",
volume = "13",
pages = "241--251",
journal = "Journal of Biopharmaceutical Statistics",
issn = "1054-3406",
publisher = "Taylor and Francis Ltd.",
number = "2",

}

Chi-square test for R × C contingency tables with clustered data. / Jung, Sin Ho; Kang, Seung Ho; Ahn, Chul.

In: Journal of Biopharmaceutical Statistics, Vol. 13, No. 2, 10.05.2003, p. 241-251.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Chi-square test for R × C contingency tables with clustered data

AU - Jung, Sin Ho

AU - Kang, Seung Ho

AU - Ahn, Chul

PY - 2003/5/10

Y1 - 2003/5/10

N2 - In ophthalmologic or dental studies, observations are frequently taken from multiple sites (called units), such as eyes or teeth, of each subject. In this case, observations within each subject (called clusters) may be dependent, although those from different subjects are independent. When a categorical observation is made from each site, application of the usual Pearson chi-square tests is invalid since sites within the same subject tend to be dependent. We propose a modified χ2 statistic for testing no treatment effect in these cases. The proposed methods do not require correct specification of the dependence structure within cluster. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to real-life data.

AB - In ophthalmologic or dental studies, observations are frequently taken from multiple sites (called units), such as eyes or teeth, of each subject. In this case, observations within each subject (called clusters) may be dependent, although those from different subjects are independent. When a categorical observation is made from each site, application of the usual Pearson chi-square tests is invalid since sites within the same subject tend to be dependent. We propose a modified χ2 statistic for testing no treatment effect in these cases. The proposed methods do not require correct specification of the dependence structure within cluster. Simulation studies are conducted to show the finite-sample performance of the new methods. The proposed methods are applied to real-life data.

UR - http://www.scopus.com/inward/record.url?scp=0037276122&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=0037276122&partnerID=8YFLogxK

U2 - 10.1081/BIP-120019269

DO - 10.1081/BIP-120019269

M3 - Article

VL - 13

SP - 241

EP - 251

JO - Journal of Biopharmaceutical Statistics

JF - Journal of Biopharmaceutical Statistics

SN - 1054-3406

IS - 2

ER -