For the pairwise comparison, the p value is less than 0.001. However, the p value for Mauchly's test of sphericity is also <0.001. I believe that SPSS automatically applies the Greenhouse-Geisser or Huynh-Feldt correction. So does this mean that the p value for the pairwise comparison is still valid?
1 Answer
$\begingroup$
$\endgroup$
I assume you only have one repeated (within-subject) factor and the post-hoc tests you are interested in are for this factor.
SPSS Repeated measures ANOVA should give you both GG and HF corrected omnibus tests by default (under the "sphericity assumed" test in the output). Some kind of rule of thumb is that if Mauchly's test's epsilon is < .75, use the GG corrected test and if it's > .75, use HF corrected test, but usually there are no huge differences. In any case, if the corrected omnibus effects are significant, to the best of my knowledge you can also use the post-hoc comparisons like you usually would.
