...
I suppose this depends on what your alternate hypothesis is.
If you're expecting a general shift in the distribution, then you could do a power calculation based on a t-test and see what sample size you come up with.
A Mann-Whitney test will have more power than a t-test to detect a difference when the distributions are not Normal. Ergo, if you have enough power for a t-test, you'll have enough power for a Mann-Whitney.
The down side of this is that you could end up with a sample size that is much greater than you actually need, but that depends on what the actual distribution looks like and what difference you are looking for.
My preferred option would be to estimate the power by simulation.
______________________________________________
Kieran McCaul MPH PhD
WA Centre for Health & Ageing (M573)
University of Western Australia
Level 6, Ainslie House
48 Murray St
Perth 6000
Phone: (08) 9224-2701
Fax: (08) 9224 8009
email: [email protected]
http://myprofile.cos.com/mccaul
http://www.researcherid.com/rid/B-8751-2008
______________________________________________
If you live to be one hundred, you've got it made.
Very few people die past that age – George Burns
-----Original Message-----
From: [email protected] [mailto:[email protected]] On Behalf Of Graham Smith
Sent: Saturday, 10 October 2009 4:59 AM
To: [email protected]
Subject: st: Mann-Whitney/Wilcoxon power analysis
Can anyone point me towards any help for doing a power analysis for a
Mann-Whitney test (I'm still using v8 of stata if it matters)
Many thanks,
Graham
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
*
* For searches and help try:
* http://www.stata.com/help.cgi?search
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/