Related Experiment Video
Updated: Mar 13, 2026

Measuring the Functional Abilities of Children Aged 3-6 Years Old with Observational Methods and Computer Tools
Published on: June 20, 2020
Assessing Binary Diagnoses of Bio-behavioral Disorders: The Clinical Relevance of Cohen's Kappa
Domenic V Cicchetti1, Ami Klin, Fred R Volkmar
1*Child Study Center, and Departments of †Biometry and ‡Psychiatry, Yale University School of Medicine, North Branford, CT; §Marcus Autism Center, Atlanta, GA; and ∥Child Study Center, New Haven, CT.
Abstract:
Cohen's (Educ Psychol Meth. 1960;23:37-40) kappa statistic has been criticized recently for providing low chance-corrected coefficients (<0.40) despite high levels of examiner agreement (e.g., ≥85%). The AC1 statistic (Series. 2002;2:1-9; BMC Med Res Methodol. 2013;13:61), the alpha statistic (J Clin Epidemiol. 1996;49:775-782), the B statistic (BMC Med Res Methodol. 2013;13:97), and the delta statistic (Commun Statist Theory Meth. 2008;37:760-772) were developed to "correct" this "problem." Each of these statistics produces inaccurate results because they either derive from levels of chance agreement that are incorrect (Gwet's AC1 statistic) or fail to correct for chance levels of agreement (the alpha, B, and delta approaches). Implications for diagnoses of psychiatric disorders are critically discussed. The importance of kappa is demonstrated in a number of clinical scenarios.
Related Concept Videos
Cochran's Q Test
Diagnostic and Statistical Manual of Mental Disorders (DSM)
Kendall's Coefficient of Concordance
Theory of Attribution II: Kelley's Covariation Theory
Types of Biopharmaceutical Studies: Controlled and Non-Controlled Approaches
Non-controlled studies, commonly employed for initial exploration, lack a control group, rendering them susceptible to biases and external influences. In contrast,...
Biostatistics: Overview
Discrete variables are...

