schliessen

Filtern

 

Bibliotheken

A note on the kappa statistic for clustered dichotomous data

The kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation‐based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician–patients dichotomous data, we investigate its special correlation structure and develop a n... Full description

Journal Title: Statistics in Medicine 30 June 2014, Vol.33(14), pp.2425-2448
Main Author: Zhou, Ming
Other Authors: Yang, Zhao
Format: Electronic Article Electronic Article
Language:
Subjects:
ID: ISSN: 0277-6715 ; E-ISSN: 1097-0258 ; DOI: 10.1002/sim.6098
Zum Text:
SendSend as email Add to Book BagAdd to Book Bag
Staff View
recordid: wj10.1002/sim.6098
title: A note on the kappa statistic for clustered dichotomous data
format: Article
creator:
  • Zhou, Ming
  • Yang, Zhao
subjects:
  • Kappa Statistic
  • Clustered Dichotomous Data
  • Confidence Interval
  • Agreement
  • Coverage Probability
  • Physician–Patients
ispartof: Statistics in Medicine, 30 June 2014, Vol.33(14), pp.2425-2448
description: The kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation‐based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician–patients dichotomous data, we investigate its special correlation structure and develop a new simple and efficient data generation algorithm. For the clustered physician–patients dichotomous data, based on the delta method and its special covariance structure, we propose a semi‐parametric variance estimator for the kappa statistic. An extensive Monte Carlo simulation study is performed to evaluate the performance of the new proposal and five existing methods with respect to the empirical coverage probability, root‐mean‐square error, and average width of the 95% confidence interval for the kappa statistic. The variance estimator ignoring the dependence within a cluster is generally inappropriate, and the variance estimators from the new proposal, bootstrap‐based methods, and the sampling‐based delta method perform reasonably well for at least a moderately large number of clusters (e.g., the number of clusters ). The new proposal and sampling‐based delta method provide convenient tools for efficient computations and non‐simulation‐based alternatives to the existing bootstrap‐based methods. Moreover, the new proposal has acceptable performance even when the number of clusters is as small as  = 25. To illustrate the practical application of all the methods, one psychiatric research data and two simulated clustered physician–patients dichotomous data are analyzed. Copyright © 2014 John Wiley & Sons, Ltd.
language:
source:
identifier: ISSN: 0277-6715 ; E-ISSN: 1097-0258 ; DOI: 10.1002/sim.6098
fulltext: fulltext
issn:
  • 0277-6715
  • 02776715
  • 1097-0258
  • 10970258
url: Link


@attributes
ID98413983
RANK0.07
NO1
SEARCH_ENGINEprimo_central_multiple_fe
SEARCH_ENGINE_TYPEPrimo Central Search Engine
LOCALfalse
PrimoNMBib
record
control
sourcerecordid10.1002/sim.6098
sourceidwj
recordidTN_wj10.1002/sim.6098
sourcesystemPC
pqid1532478729
galeid369938030
display
typearticle
titleA note on the kappa statistic for clustered dichotomous data
creatorZhou, Ming ; Yang, Zhao
ispartofStatistics in Medicine, 30 June 2014, Vol.33(14), pp.2425-2448
identifier
subjectKappa Statistic ; Clustered Dichotomous Data ; Confidence Interval ; Agreement ; Coverage Probability ; Physician–Patients
descriptionThe kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation‐based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician–patients dichotomous data, we investigate its special correlation structure and develop a new simple and efficient data generation algorithm. For the clustered physician–patients dichotomous data, based on the delta method and its special covariance structure, we propose a semi‐parametric variance estimator for the kappa statistic. An extensive Monte Carlo simulation study is performed to evaluate the performance of the new proposal and five existing methods with respect to the empirical coverage probability, root‐mean‐square error, and average width of the 95% confidence interval for the kappa statistic. The variance estimator ignoring the dependence within a cluster is generally inappropriate, and the variance estimators from the new proposal, bootstrap‐based methods, and the sampling‐based delta method perform reasonably well for at least a moderately large number of clusters (e.g., the number of clusters ). The new proposal and sampling‐based delta method provide convenient tools for efficient computations and non‐simulation‐based alternatives to the existing bootstrap‐based methods. Moreover, the new proposal has acceptable performance even when the number of clusters is as small as  = 25. To illustrate the practical application of all the methods, one psychiatric research data and two simulated clustered physician–patients dichotomous data are analyzed. Copyright © 2014 John Wiley & Sons, Ltd.
source
version7
lds50peer_reviewed
links
openurl$$Topenurl_article
openurlfulltext$$Topenurlfull_article
search
creatorcontrib
0Zhou, Ming
1Yang, Zhao
titleA note on the kappa statistic for clustered dichotomous data
descriptionThe kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation‐based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician–patients dichotomous data, we investigate its special correlation structure and develop a new simple and efficient data generation algorithm. For the clustered physician–patients dichotomous data, based on the delta method and its special covariance structure, we propose a semi‐parametric variance estimator for the kappa statistic. An extensive Monte Carlo simulation study is performed to evaluate the performance of the new proposal and five existing methods with respect to the empirical coverage probability, root‐mean‐square error, and average width of the 95% confidence interval for the kappa statistic. The variance estimator ignoring the dependence within a cluster is generally inappropriate, and the variance estimators from the new proposal, bootstrap‐based methods, and the sampling‐based delta method perform reasonably well for at least a moderately large number of clusters (e.g., the number of clusters ). The new proposal and sampling‐based delta method provide convenient tools for efficient computations and non‐simulation‐based alternatives to the existing bootstrap‐based methods. Moreover, the new proposal has acceptable performance even when the number of clusters is as small as  = 25. To illustrate the practical application of all the methods, one psychiatric research data and two simulated clustered physician–patients dichotomous data are analyzed. Copyright © 2014 John Wiley & Sons, Ltd.
subject
0Kappa Statistic
1Clustered Dichotomous Data
2Confidence Interval
3Agreement
4Coverage Probability
5Physician–Patients
general
010.1002/sim.6098
1Wiley Online Library
sourceidwj
recordidwj10.1002/sim.6098
issn
00277-6715
102776715
21097-0258
310970258
rsrctypearticle
creationdate2014
addtitle
0Statistics in Medicine
1Statist. Med.
searchscope
0wj
1wiley
scope
0wj
1wiley
lsr30VSR-Enriched:[pages, pqid, galeid]
sort
titleA note on the kappa statistic for clustered dichotomous data
authorZhou, Ming ; Yang, Zhao
creationdate20140630
facets
frbrgroupid6196797001143955337
frbrtype5
creationdate2014
topic
0Kappa Statistic
1Clustered Dichotomous Data
2Confidence Interval
3Agreement
4Coverage Probability
5Physician–Patients
collectionWiley Online Library
prefilterarticles
rsrctypearticles
creatorcontrib
0Zhou, Ming
1Yang, Zhao
jtitleStatistics in Medicine
toplevelpeer_reviewed
delivery
delcategoryRemote Search Resource
fulltextfulltext
addata
aulast
0Zhou
1Yang
aufirst
0Ming
1Zhao
au
0Zhou, Ming
1Yang, Zhao
atitleA note on the kappa statistic for clustered dichotomous data
jtitleStatistics in Medicine
risdate20140630
volume33
issue14
spage2425
epage2448
issn0277-6715
eissn1097-0258
genrearticle
ristypeJOUR
abstractThe kappa statistic is widely used to assess the agreement between two raters. Motivated by a simulation‐based cluster bootstrap method to calculate the variance of the kappa statistic for clustered physician–patients dichotomous data, we investigate its special correlation structure and develop a new simple and efficient data generation algorithm. For the clustered physician–patients dichotomous data, based on the delta method and its special covariance structure, we propose a semi‐parametric variance estimator for the kappa statistic. An extensive Monte Carlo simulation study is performed to evaluate the performance of the new proposal and five existing methods with respect to the empirical coverage probability, root‐mean‐square error, and average width of the 95% confidence interval for the kappa statistic. The variance estimator ignoring the dependence within a cluster is generally inappropriate, and the variance estimators from the new proposal, bootstrap‐based methods, and the sampling‐based delta method perform reasonably well for at least a moderately large number of clusters (e.g., the number of clusters ). The new proposal and sampling‐based delta method provide convenient tools for efficient computations and non‐simulation‐based alternatives to the existing bootstrap‐based methods. Moreover, the new proposal has acceptable performance even when the number of clusters is as small as  = 25. To illustrate the practical application of all the methods, one psychiatric research data and two simulated clustered physician–patients dichotomous data are analyzed. Copyright © 2014 John Wiley & Sons, Ltd.
doi10.1002/sim.6098
pages2425-2448
date2014-06-30