Key points are not available for this paper at this time.
The purpose of this special communication is to describe the application of the Kappa coefficient for the estimation of interobserver agreement. Kappa is a preferred statistic for the estimation of the accuracy of nominal and ordinal data in clinical research by physical therapists. A brief introduction to the properties of the Kappa coefficient is given, and a special case of Kappa for multiple ratings per subject is explained. A FORTRAN program, written specifically for the multiple-ratings situation, is described.
Haley et al. (Wed,) studied this question.