logo BDSP

Base documentaire


  1. Behavior and interpretation of the kappa statistic : Resolution of the two paradoxes.

    Article - En anglais

    Two apparent paradoxes have been identified for the kappa (kappa) statistic : (1) high levels of observer agreement with low K values ; (2) lack of predictability of changes in K with changing marginals.

    The first paradox is a function of prevalence of the trait in the sample, while the second is related to symmetry of observations in the disagreement categories.

    While examining the behavior of K as a function of the distribution of responses in a contingency table, it was discovered that for any measured level of observer agreement (Po) there are three characteristic values of kappa : kappamax, kappamin, and kappanor, each of which is a function only of Po.

    The characteristic values allow an observed kappa (kappao) to be placed into perspective.

    By observing symmetry in agreement and disagreement categories, the behavior of K is readily understood and predictable.

    We define symmetry expressions for agreement (SA) and disagreement (SD) in order to represent and quantify these effects.

    K alone has little interpretive value and we recommend that studies reporting kappa also report Po, SD, and P+ (agreement on the presence of the trait).

    Mots-clés Pascal : Analyse statistique, Indice kappa, Interprétation, Méthode calcul, Epidémiologie, Méthodologie

    Mots-clés Pascal anglais : Statistical analysis, Kappa number, Interpretation, Computing method, Epidemiology, Methodology

    Logo du centre Notice produite par :
    Inist-CNRS - Institut de l'Information Scientifique et Technique

    Cote : 96-0420253

    Code Inist : 002B30A01A1. Création : 10/04/1997.