I'm doing some stats for a piece of Geography work atm (*shudder*) and i don't want to walk all the way down to the library to try to find a book with a table of critical values for the Mann Whitney U test. I'm sure there must be one on the 'net somewhere, but does anyone know where?
Of course i've tried googling, but all i can come up with is several tables which only go up to n=20 (i need n=34), and some kind of automatic calculator which is not what i want at all.
The tables don't go above 20 because when n1 and n2 are large you can use a normal approximation. See http://www.cquest.utoronto.ca/geog/g...y_U_Table.html. You need to calculate Z and compare it to the critical values of the N(0, 1) distribution. If you post your values of n1, n2 and u, I'll do the sums for you.
(Original post by Tom)
Aha. I just realised where i screwed up....my n isn't 34, it's 17. So i can just look it up in the many tables out there that go up to 20. Stupid stats...shouldn't be in geog at all
Thanks anyway though.
You won't have been the first to not count the sample sizes before combining the results - trust me, too easy a mistake to make.
and that will give you a z-value and its associated probability (ie whether to reject your null)
also the critical values up to 30 are here, so you could just use your u value (rather than z value which i realise may be confusing you, but I don't have time to explain right now!) which the above page will calculate: http://www.social-science.co.uk/stats_CV.php