7.2.6.1 Calculations for a simple kappa statistic

This is an archived version of the Handbook. For the current version, please go to training.cochrane.org/handbook/current or search for this chapter here.

Suppose the K studies are distributed according to numbers a to i as in Table 7.2.a. Then

where

is the proportion of studies for which there was agreement, and

is the proportion of studies in which one would expect there to be agreement by chance alone. As an example, from the data in Table 7.2.b,

and so