I was reading a paper in which Kappa Index of Agreement (KIA) was calculated. I searched the web for a reference but I couldn't find any decent definition. My first result was instead a Wikipedia article named Cohen's kappa.

Since I am not familiar with either of them and Wikipedia article does not state that Cohen's kappa is actually KIA can someone confirm the relation between those two? And if they are not the same can provide a link to a definition of Kappa Index of Agreement (KIA)?

The original paper is Detection and classification of mesoscale atmospheric phenomena above sea in SAR imagery.

P.S. I am not sure what tag should I use so feel free to edit.

  • 1
    $\begingroup$ In fact we have a tag [kappa] which refers to the statistic you have been confused about. $\endgroup$ – mdewey Jan 29 '18 at 14:02

I cannot access the entire paper but from the context of the abstract it would appear that the kappa the author's reference is indeed Cohen's $\kappa$. The statistic itself is a measure of agreement on a categorical classification task (typically used when researchers have to rate or code some sort of phenomenon into one of a handful of categories). The advantage of using $\kappa$ over a raw agreement percentage in many cases is that $\kappa$ adjusts for the possibility of inter-rater agreement occurring due to chance.


Your Answer

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

Not the answer you're looking for? Browse other questions tagged or ask your own question.