# Using a post-hoc test after creating an LME model

Is a post-hoc test necessary when using an LME test for repeated measures?

Assume that, in R, a summary of the LME model (lmemodel) would show significant levels of pairwise comparisons.

-
a post-hoc test for..? If you want a definitive answer, more information (e.g. what model, precisely, was fit? what are you trying to conclude with this model?) would be helpful. –  Macro May 2 '12 at 14:54
Maybe this could be of help? stats.stackexchange.com/questions/5250/… –  nico May 2 '12 at 15:31
The result of my model says there are significant differences for the individual effects (group and time) and for the interaction effect. I want to know if this is true between pairs in the groups or between times. –  Luana May 2 '12 at 16:56
Yes, thanks the experiment described there is the same as mine. Would I still need the post-hoc test? –  Luana May 2 '12 at 17:18