The Reliability of Dichotomous Judgments: Unequal Numbers of Judges per Subject

Abstract

Consider a reliability study in which different subjects are judged on a dichotomous trait by dif ferent sets of judges, possibly unequal in number. A kappa-like measure of reliability is proposed, its correspondence to an intraclass correlation co efficient is pointed out, and a test for its statistical significance is presented. A numerical example is given.