In statistics, an index of the degree of association between two variables measured on an ordinal scale or transformed to ranks. Its computation involves examining every pair of items and counting the number of pairs that are similarly ranked (concordant) and the number differently ranked (discordant) relative to each other on the two variables, and the difference between the number of concordant and discordant pairs is then divided by the total number of pairs. See also Kendall coefficient of concordance. [Named after the English statistician Maurice George Kendall (1907–83) who introduced it in 1938]
Subjects: Probability and Statistics — Psychology.