I just wondered how much detail people are going into regarding the stat tests. For example, are you getting your students to just rank the differences for Mann Whitney, for example, or getting them to calculate the whole test? I read somewhere that it is only the sign test which they can ask students to fully calculate...however the quote from the spec below is a little ambiguous...all comments and thought would be appreciated.
For example, calculating a non-parametric test of
differences using data from a given experiment.