Search All of the Math Forum:
Views expressed in these public forums are not endorsed by
NCTM or The Math Forum.



Re: post hoc power analysis of ttest
Posted:
Jan 17, 2013 6:32 PM


On Thursday, January 17, 2013 11:42:07 PM UTC+13, Keren wrote:
> I have to groups that differ in their mean values. Using the ttest2 function I performed the t test but now I am having trouble calculating the power of my analysis
Keren,
These articles may help you:
Froman, T., & Shneyderman, A. (2004). Replicability reconsidered: An excessive range of possibilities. Understanding Statistics, 3, 365373.
Yuan, K.H., & Maxwell, S. (2005). On the post hoc power in testing mean differences. Journal of Educational and Behavioral Statistics, 30 (2), 141167.
I'm pretty sure that one will tell you how to compute posthoc power for your situation, but also that both will tell you it's a useless and misleading thing to compute. (sorry)



