McNemar's Test

McNemar's test, also known as a test of correlated proportions, is a nonparametric test used with dichotomous nominal or ordinal data to determine whether two sample proportions based on the same individuals are equal. McNemar's test is used in many fields, including the behavioral and biomedical sciences. In short, it is a test of symmetry between two related samples based on the chi-square distribution with 1 degree of freedom (df).

McNemar's test is unique in that it is the only test that can be used when one or both conditions being studied are measured using the nominal scale. It is often used in before-after studies, in which the same individuals are measured at two times, a pretest–posttest, for example. McNemar's test is also often used in ...