Cochran's Q is an extension of McNemar's test and is essentially a Friedman's ANOVA for dichotomous data. Which statement best describes it?

• A. McNemar's test for dichotomous related samples
• B. Friedman's ANOVA for continuous data
• C. ANOVA for independent groups
• D. Chi-square test for independence

Answer: (A)

Cochran's Q is used when you have several related (paired) samples and a binary outcome. This test extends McNemar's test, which compares two related samples, to handle more than two conditions. In other words, if you have k related time points or conditions and a Yes/No response for each subject, Cochran's Q checks whether the proportion of Yes responses is the same across all k related samples. When there are only two related samples, Cochran's Q behaves like McNemar's test. The other options don’t fit because they either assume continuous data (Friedman’s ANOVA for ranks of a continuous variable), require independent groups (ANOVA for independent samples), or test independence in unrelated data (Chi-square test for independence).