Proof: By Property 1 of Wilcoxon Rank Sum Test, R1 + R2 = n(n+1)/2. Thus
Property 2: For n1 and n2 large enough the U statistic is approximately normal N(μ, σ) where
Proof: From Property 2 of Wilcoxon Rank Sum Test, the mean of R1 is . Similarly the mean of R2 is . Since
it follows from Property 3 of Expectation that the mean of U1 is
Similarly the mean of U2 is . Since U = min(U1, U2), it follows that the mean of U is also .
By Property 3 of Expectation, the variance of U1 is the same as the variance of R1, which by Property 2 of Wilcoxon Rank Sum Test is .
Similarly the variance of U2 is the same as the variance of R2, which is again . Thus the variance of U is this same amount.
sir,
i have sample size 86 and 82 a have calculated U statistic 348.5
now how can I find critical value for my large sample size .How can I analyse null hypothesis
my data is as under
SUM OF RANK COUNT U STATISTIC
EXP GROUP 10444.5 86 348.5
CONTROL GROUP 3751.5 82 6703.5
Use the normal approximation as described at
https://real-statistics.com/non-parametric-tests/mann-whitney-test/
Charles
I’d like to be able to construct an Excel spreadsheet that could compute its own Mann Whitney critical values. I can find lots of existing tables online, but I can’t seem to find the formula by which those critical values have been generated. Any help would be greatly appreciated.
Hello Brad,
You can use the Mann-Whitney exact test, as described on this website.
Charles
Thanks for your great work!
The materials on this site are excellent. Thank you for all of the work that has gone into generating them. I was wondering if you could provide a bit more detail on the normal approximation of U. The referenced proof for W invokes the central limit theorem, but I don’t see how that is applicable here. That would seem to reduce to showing that U (or W) is the mean of some distribution. The 1947 Mann Whitney paper presents a fairly complex derivation of the limit of U, without using the central limit theorem. Thanks!
Yes, you are correct. I just changed the referenced webpage to reflect this. Thanks for catching this mistake.
Charles