How can I perform wilcoxon Test with paired samples and vectors of differente sizes?

Hello everyone, I need to perform a wilcoxon test with dependent samples, but the vectors containing the samples have different sizes.

What function can I use to implement this?

"signrank" predicts vectors of equal size

"ranksum" applies with independent samples

Am I wrong?

Answer 1

[Jeff Miller]

The Wilcoxon test is for data that come in pairs, so the vectors must be the same size. If you are just missing one score or the other for a few pairs, leave those pairs out (they tell you nothing anyway) and run signrank on the remaining pairs. If that is not the situation, you probably need to explain a bit more about your dependent samples to get any advice.

[BENEDETTA]

Thank you so much!