Methodological study proposes an agreement statistic for multinomial data, providing dual derivations and asymptotic error to improve categorical reliability analysis.
A kappa-like statistic is proposed as a measure of agreement for a set of multinomial random variables arrayed in a two-way layout. This statistic is shown to arise either via the chancecorrection of a certain proportion of agreement or via an analysis of variance for a two-way layout. The asymptotic standard error is derived under the assumption of complete independence among the random variables.
No takes yet. Share an insight, caveat, or question.
Davies et al. (1982) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: