testing simple main effects after significant interaction in two way repeated measures ANOVA

[jennifer Thornton]

I ran a two way repeated measures ANOVA (both factors GROUP and TIME were repeated) and found a significant interaction between GROUP and TIME. I would like to test for simple main effects but am having trouble implementing the sme command. If anyone can shed some light on this I would greatly appreciate it. Thank you.

[Dave Airey]

use http://www.stata-press.com/data/r14/t713, clear
keep if noise == 1
drop noise
label var subject ""
anova score subject period / subject#period dial / subject#dial period#dial, repeated(period dial) dropemptycells
contrast r.dial@period

This does produce results but I don't know if they are correct. Be nice to have a mixed model equivalent too. It would also be good to compare to a worked example.

[Dave Airey]

mixed score period##dial || subject: || subject: R.period || subject: R.dial, var reml
contrast period##dial // gives same F if chi2 scaled by df
contrast r.dial@period // not the same as above

This is the same model using mixed.

[jennifer Thornton]

A (very) belated thank you for your assistance.