I am trying to find out how to calculate sample sizes for studies where the
primary endpoint is the Kappa measure of agreement. I have seen reference to
methods when the Kappa table will be a 2x2 table, but the study I am looking
at has greater than 2 classes for each of the methods being compared.
I would be very appreciative if anyone could provide me with any formulae or
point me in the direction of any publications which would help.
Many thanks in advance,
Michael Roughton MSc
|