Mann-Whitney Test

Hi all,

I've got some data that contains dichotomous independent variables and ordinal dependent variables. As a result, I think the only test that I can run would be a mann-whitney test. My issue is that of the two dichotomized groups (i.e. male/female) the sample sizes are not the same. One's 300 and the other is 109...

This is going to be a problem is it not? What can I do to fix it.

Thanks for your help.



TS Contributor
hi Karabiner,
just wondering if bootstraping would help here. I mean one could take a random sample of 100 from the 300 males compute the effect and the p-value, repeat a lot of times and calculate the effect size with confidence intervals and an average p?



TS Contributor
Should we be suspicious about U-tests on
ordinal data if sample sizes are unequal? If
yes, some simulations would surely be valuable.

It would be nice, by the way, if people who ask
questions here would also report on what they
have already found out by themselves.

With kind regards