Two-Way repeated Measures Anova: Pairwise comparison of multiple factor levels using Multcompar

I came across this example of Matlab Doc and was wondering about following.

https://www.mathworks.com/help/stats/repeatedmeasuresmodel.multcompare.html#bt7sh0m-8

Assuming, I have a repeated measures anova with two independent variables which have 3 factor levels. Now I would like to conduct a posthoc comparing each level against each other like so

T = multcompare(R,'Group','By','Gender')

I can see, that there is a significant difference between A and B. but how can I find out which one of these two facator levels is the one that is better? I mean, in a real life setting, the goal would be to say "treatment A is better than C for females". for example.

Thank you for your help!

Answer 1

[Jeff Miller]

Go back to the original data and compute the means for each group of interest, so you can compare which group mean is higher.