How to interpret the results from "multcompare" function?

Hi, I used "multcompare" function in Matlab and I am not sure how to interpret the results. The graph of group means comparison gives me sometimes different results than the p value in output "c". For example: the group comparison intervals intersect (it means that groups do not differ) and the p value for this pair is lower than alpha= 0,05 (it means that groups differ). Do you have any experience with that? Thanks.

Answer 1

[Jeff Miller]

This article might help. The key point is this: "A frequently encountered misconception is that if 2 independent 95% CIs overlap each other, as they do in Fig. 1, then a statistical test of the difference will not be statistically significant at the 5% level."

[A N]

Thank you! It helped.