I am working on a dataset where two raters indicated whether or not particular features were present in the same individuals. So each individual was assessed by each of these two raters and received a score of 0 for no feature and 1 for a feature. Most of the features were very minor and as a result, many of the individuals received a 0 from both raters on that feature.

I am trying to calculate a Cohen's Kappa in R for each feature to determine reliability between the two raters, but my the kappa values are very low and I am thinking it has to do with the prevalence of the 0 values for the ratings. I saw there used to be a package in R that did something called the Bryt's Kappa for prevalence correction, but this no longer exists.

Is anyone aware of Kappa calculations in R or SPSS that can be done that account for this prevalence?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.