Personally I despise post-hoc power calculations, they're pointless
and do nothing to enlighten the results of your test.
Its like trying to tell someone who's just won the lottery that they
shouldn't buy lottery tickets as the probability of winning is low.
They don't care, they've already got their result (and were fortunate
enough to have won that time)!
A couple of papers on this are...
Hoenig, John M. and Heisey, Dennis M. (2001), The Abuse of Power: The
Pervasive Fallacy of Power Calculations for Data Analysis The American
Statistician, 55, 19-24.
Levine M, Ensom MH (2001) Post hoc power analysis: an idea whose time
has passed? Pharmacotherapy 21.4:405-409
I'm sure there are more references out there.
The only purpose they serve is to inform the researcher about an
appropriate sample size for their next study.
Neil
--
"We should make things as simple as possible, but not simpler." -
Anon (not Albert Einstein)
Email - [log in to unmask]
Website - http://slack.ser.man.ac.uk/
Photos - http://www.flickr.com/photos/slackline/
|