Analyses

Cochran’s Q test

Compare three or more repeated binary measurements.

Edit on GitHub

When to use it

Use Cochran’s Q when the same people are measured three or more times on a yes/no variable. It is the binary cousin of the Friedman test. McNemar is the two-occasion version.

Assumptions

Each column is a repeated binary measure. Values are treated as success or failure. You need at least three columns.

Running it in Tensr

Analyze → Nonparametric Tests → Cochran. In chat: “Cochran’s Q for these binary columns.”
Select at least three measure columns.

Options

Prop

Type

Reading the output

Fifty people, three tasks, each scored pass or fail. The pass rates were built to be nearly the same.

Cochran's Q for 3 related dichotomous variables, n = 50. The primary result is not significant (p = .164). This is not large enough to treat the comparison this page is about as a reliable association. The result is non-significant: the data are still compatible with no effect. Metrics: Q = 3.619; df = 2; p-value = .164.

Reporting (APA 7)

Cochran's Q for 3 related dichotomous variables, n = 50. This result is not significant (p = .164). Report the estimate with that p, and do not describe the pattern as a reliable effect.

Coming from SPSS

Analyze → Nonparametric Tests → Cochran.