Generalized estimating equations with model selection for comparing dependent categorical agreement data

Miao-Yu Tsai, Jung Feng Wang, Jia Ling Wu

Research output: Contribution to journalArticle

2 Citations (Scopus)

Abstract

Many studies in biomedical fields are carried out using diagnoses reported by different raters to evaluate the agreement of multiple ratings. The most popular indices of agreement are kappa measures including Cohen's kappa and weighted kappa for binary and ordinal outcomes, respectively. However, when raters assess the same observation on two or more occasions, these ratings are dependent and so the correlation between kappa estimates must be considered when making inferences. In this paper, we focus on testing the equality of correlated kappa coefficients using the generalized estimating equation (GEE) approach and applying quasi-likelihood under the independence model criterion (QIC) measures for model selection. Simulation studies are conducted to compare the performance between GEE with and without QIC measures, weighted least squares (WLS) and independence approaches for binary and ordinal data. Two applications are illustrated: a comparison of two methods for assessing cervical ectopy, and similarity in myopic status for monozygous twins and dizygous twins. We conclude that when performing the QIC model-selection procedure in GEE models and taking into account the correlation between kappa measures, it leads to nominal type I errors and larger powers.

Original languageEnglish
Pages (from-to)2354-2362
Number of pages9
JournalComputational Statistics and Data Analysis
Volume55
Issue number7
DOIs
Publication statusPublished - 2011 Jul 1

Fingerprint

Generalized Estimating Equations
Model Selection
Categorical
Dependent
Cohen's kappa
Ordinal Data
Quasi-likelihood
Type I error
Binary Data
Weighted Least Squares
Selection Procedures
Categorical or nominal
Equality
Simulation Study
Binary
Testing
Evaluate
Coefficient
Model
Estimate

All Science Journal Classification (ASJC) codes

  • Statistics and Probability
  • Computational Mathematics
  • Computational Theory and Mathematics
  • Applied Mathematics

Cite this

@article{d35a4842541b4f06a78a618daea7b922,
title = "Generalized estimating equations with model selection for comparing dependent categorical agreement data",
abstract = "Many studies in biomedical fields are carried out using diagnoses reported by different raters to evaluate the agreement of multiple ratings. The most popular indices of agreement are kappa measures including Cohen's kappa and weighted kappa for binary and ordinal outcomes, respectively. However, when raters assess the same observation on two or more occasions, these ratings are dependent and so the correlation between kappa estimates must be considered when making inferences. In this paper, we focus on testing the equality of correlated kappa coefficients using the generalized estimating equation (GEE) approach and applying quasi-likelihood under the independence model criterion (QIC) measures for model selection. Simulation studies are conducted to compare the performance between GEE with and without QIC measures, weighted least squares (WLS) and independence approaches for binary and ordinal data. Two applications are illustrated: a comparison of two methods for assessing cervical ectopy, and similarity in myopic status for monozygous twins and dizygous twins. We conclude that when performing the QIC model-selection procedure in GEE models and taking into account the correlation between kappa measures, it leads to nominal type I errors and larger powers.",
author = "Miao-Yu Tsai and Wang, {Jung Feng} and Wu, {Jia Ling}",
year = "2011",
month = "7",
day = "1",
doi = "10.1016/j.csda.2011.02.002",
language = "English",
volume = "55",
pages = "2354--2362",
journal = "Computational Statistics and Data Analysis",
issn = "0167-9473",
publisher = "Elsevier",
number = "7",

}

Generalized estimating equations with model selection for comparing dependent categorical agreement data. / Tsai, Miao-Yu; Wang, Jung Feng; Wu, Jia Ling.

In: Computational Statistics and Data Analysis, Vol. 55, No. 7, 01.07.2011, p. 2354-2362.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Generalized estimating equations with model selection for comparing dependent categorical agreement data

AU - Tsai, Miao-Yu

AU - Wang, Jung Feng

AU - Wu, Jia Ling

PY - 2011/7/1

Y1 - 2011/7/1

N2 - Many studies in biomedical fields are carried out using diagnoses reported by different raters to evaluate the agreement of multiple ratings. The most popular indices of agreement are kappa measures including Cohen's kappa and weighted kappa for binary and ordinal outcomes, respectively. However, when raters assess the same observation on two or more occasions, these ratings are dependent and so the correlation between kappa estimates must be considered when making inferences. In this paper, we focus on testing the equality of correlated kappa coefficients using the generalized estimating equation (GEE) approach and applying quasi-likelihood under the independence model criterion (QIC) measures for model selection. Simulation studies are conducted to compare the performance between GEE with and without QIC measures, weighted least squares (WLS) and independence approaches for binary and ordinal data. Two applications are illustrated: a comparison of two methods for assessing cervical ectopy, and similarity in myopic status for monozygous twins and dizygous twins. We conclude that when performing the QIC model-selection procedure in GEE models and taking into account the correlation between kappa measures, it leads to nominal type I errors and larger powers.

AB - Many studies in biomedical fields are carried out using diagnoses reported by different raters to evaluate the agreement of multiple ratings. The most popular indices of agreement are kappa measures including Cohen's kappa and weighted kappa for binary and ordinal outcomes, respectively. However, when raters assess the same observation on two or more occasions, these ratings are dependent and so the correlation between kappa estimates must be considered when making inferences. In this paper, we focus on testing the equality of correlated kappa coefficients using the generalized estimating equation (GEE) approach and applying quasi-likelihood under the independence model criterion (QIC) measures for model selection. Simulation studies are conducted to compare the performance between GEE with and without QIC measures, weighted least squares (WLS) and independence approaches for binary and ordinal data. Two applications are illustrated: a comparison of two methods for assessing cervical ectopy, and similarity in myopic status for monozygous twins and dizygous twins. We conclude that when performing the QIC model-selection procedure in GEE models and taking into account the correlation between kappa measures, it leads to nominal type I errors and larger powers.

UR - http://www.scopus.com/inward/record.url?scp=79953659747&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79953659747&partnerID=8YFLogxK

U2 - 10.1016/j.csda.2011.02.002

DO - 10.1016/j.csda.2011.02.002

M3 - Article

AN - SCOPUS:79953659747

VL - 55

SP - 2354

EP - 2362

JO - Computational Statistics and Data Analysis

JF - Computational Statistics and Data Analysis

SN - 0167-9473

IS - 7

ER -