On 10/24/2012 5:31 AM, Stephen P. King wrote:


Woo-woo. Small effect sizes which are *statistically* significant are indicative of bias errors. I'd wager a proper Bayesian analysis of the original data will show they *support* the null hypothesis (c.f. "Testing Precise Hypotheses" Berger & Delampady, Stat Sci 1987 v2 no. 3 317-352 and "Odds Are It's Wrong" Tom Siegfried, Science News 27 Mar 2010). Meta-analyses are notoriously unreliable and should only be considered suggestive at a best.


You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
For more options, visit this group at 

Reply via email to