When should the McNemar test be applied?

The McNemar test is specifically designed for use with related samples, particularly when dealing with nominal data. This non-parametric statistical test evaluates the differences in proportions between two related groups. It is typically employed in situations where you have paired observations, such as before-and-after scenarios or matched subjects, and you aim to determine if the frequencies of two categorical outcomes differ significantly.

In this context, the test assesses whether the number of pairs that change from one category to another differs significantly from the number that do not change, thus offering insight into the relationship between two related categorical variables. This test is particularly useful in clinical settings or studies involving repeated measures where each subject serves as their own control, further validating the use of nominal data.

The other options do not align with the specific conditions required for the McNemar test. Analyses involving independent groups, more than two groups, or ordinal data would require different statistical approaches suitable for their design and data types.