Posted by Larry Hoover on January 27, 2009, at 7:20:22
In reply to Re: Review Article Trashing Serotonin Hypothesis + Adv » linkadge, posted by Larry Hoover on January 27, 2009, at 7:08:56
"About 90% of studies would show no superiority of either study arm (no significant difference), and 5% would show significant antidepressant or placebo superiority."
Errr, I should have my coffee before I post about this stuff. The numbers should read 95%, 2.5% and 2.5% (two-tailed test of significant difference). If your hypothesis was strictly defined to test only whether placebo was superior to antidepressants (one-tailed test), and the two were actually equivalent, you'd find that placebos were superior 5% of the time (and 1% of the time vastly superior, with p < .01), and 95% of the time there'd be no difference.....in a purely theoretical world.
Just in case anybody's noticing, the data do not show anything like that.
Lar
poster:Larry Hoover
thread:876214
URL: http://www.dr-bob.org/babble/20090104/msgs/876511.html