It's not clear to me which test you want to get power for: you're
trying to power two things: the overall omnibus F test, and the two
planned comparisons (or perhaps 3 post hoc tests).
You need to decide which of those two you want. Also, what software do you use?
These two papers tell you one way to do it in SPSS
http://www.ncbi.nlm.nih.gov/pubmed/11816450
http://link.springer.com/content/pdf/10.3758/BF03192787.pdf
Jeremy
On 26 June 2013 13:02, Daniel H <[log in to unmask]> wrote:
> Dear all
>
> I was hoping someone could help me with the following.
>
> In an experiment with 3 groups (2x treatment 1 x control), we are able to recruit 400 people. We are expecting a 0.05 difference in means between the control and treatment groups. We expect the SD to be 0.3. Significance level 0.05. I'm unsure how to plug this into a power calculator to work out how much power we have. Any help would be greatly appreciated!
>
> Daniel
|