Premium
Assessing agreement with repeated measures for random observers
Author(s) -
Chen ChiaCheng,
Barnhart Huiman X.
Publication year - 2011
Publication title -
statistics in medicine
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.996
H-Index - 183
eISSN - 1097-0258
pISSN - 0277-6715
DOI - 10.1002/sim.4353
Subject(s) - closeness , concordance correlation coefficient , statistics , repeated measures design , correlation , random effects model , mathematics , random error , computer science , medicine , mathematical analysis , meta analysis , geometry
Agreement studies are often concerned with assessing whether different observers for measuring responses on the same subject or sample can produce similar results. The concordance correlation coefficient ( CCC ) is a popular index for assessing the closeness among observers for quantitative measurements. Usually, the CCC is used for data without and with replications based on subject and observer effects only. However, we cannot use this methodology if repeated measurements rather than replications are collected. Although there exist some CCC ‐type indices for assessing agreement with repeated measurements, there is no CCC for random observers and random time points. In this paper, we propose a new CCC for repeated measures where both observers and time points are treated as random effects. A simulation study demonstrates our proposed methodology, and we use vertebral body data and image data for illustrations. Copyright © 2011 John Wiley & Sons, Ltd.