I have seen reference to the W, V and T statistics all within the Wilcoxon Signed rank Test, and this has me confused. First question: Are these used interchangeably?

My test is the Wilcoxon Signed Rank Test because I'm comparing non-parametric matched scores for a before / after ordinal measurement scenario.

(If it's of interest, the pre / post scenario is basically whether respondents give different answers to the same question when extra information is added)

I gather from this very useful post here Output of one-tailed Wilcoxon sign rank test in R by @Glen_b that the V Statistic is simply the Signed Rank Statistics.

I presume this means I cannot extrapolate too much from the size of it? I.e I could have a P score of less than 0.05 with a large V statistic, or a P value of less than 0.05 with a low V statistic.

Second question: What can the V statistic tell me?

I am trying to work within the CrossValidated guidelines so please bear with me if I have made a mistake with this post.




Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Browse other questions tagged or ask your own question.