| Literature DB >> 2716662 |
Abstract
Formal evaluation of the ability of clinicians and researchers to agree, for example, on the clinical assessment of patients, increasingly is becoming important. Two measures of agreement, kappa and the intraclass correlation coefficient, are described and illustrated. The calculation of confidence intervals that correspond to these statistics by means of the "bootstrap" method also is discussed.Entities:
Mesh:
Year: 1989 PMID: 2716662
Source DB: PubMed Journal: Med J Aust ISSN: 0025-729X Impact factor: 7.738