Two-Rater Kappa Explorer
Describe agreement between two raters using paired binary classifications.
Your entries stay in this page’s memory and are not submitted to us. Downloads are generated locally. Reloading or leaving the page clears your entries. Do not enter patient information.
Describe agreement between two raters using paired binary classifications. All default values are fictional.
Calculation results
Observed agreement: 85%
Marginal expected agreement: 50%
Kappa: 0.7
Assumptions and limits
Unweighted Cohen kappa = (observed agreement − marginal expected agreement) / (1 − marginal expected agreement). Each item must have one binary rating from each of two raters. Items should be independent for standard inference. Kappa depends on category prevalence and marginals; it is not validity, accuracy against truth, or a universal quality grade. No confidence interval or significance test is provided. If marginal expected agreement is 1, kappa is undefined even when observed agreement is 100%.
scikit-learn: Cohen kappa · Read the practical guide · Download blank worksheet (TXT)