Testing the equality of two dependent kappa statistics
Article first published online: 25 JAN 2000
Copyright © 2000 John Wiley & Sons, Ltd.
Statistics in Medicine
Volume 19, Issue 3, pages 373–387, 15 February 2000
How to Cite
Donner, A., Shoukri, M. M., Klar, N. and Bartfay, E. (2000), Testing the equality of two dependent kappa statistics. Statist. Med., 19: 373–387. doi: 10.1002/(SICI)1097-0258(20000215)19:3<373::AID-SIM337>3.0.CO;2-Y
- Issue published online: 25 JAN 2000
- Article first published online: 25 JAN 2000
- Manuscript Accepted: APR 1999
- Manuscript Received: AUG 1998
- Natural Sciences and Engineering Research Council of Canada
- Harvard School of Public Health
- Schering–Plough Corporation
Procedures are developed and compared for testing the equality of two dependent kappa statistics in the case of two raters and a dichotomous outcome variable. Such problems may arise when each of a sample of subjects are rated under two distinct settings, and it is of interest to compare the observed levels of inter-observer and intra-observer agreement. The procedures compared are extensions of previously developed procedures for comparing kappa statistics computed from independent samples. The results of a Monte Carlo simulation show that adjusting for the dependency between samples tends to be worthwhile only if the between-setting correlation is comparable in magnitude to the within-setting correlations. In this case, a goodness-of-fit procedure that takes into account the dependency between samples is recommended. Copyright © 2000 John Wiley & Sons, Ltd.