Hi Lisa,

The results may not match, particularly if:
- Spatial statistics are used (e.g., cluster, TFCE).
- The rank of the F-test diverges from 2 (e.g., if you are comparing 3 or more groups). For a reference, see Hayer (JASA, 1986).

All the best,

Anderson


On 30 November 2016 at 10:26, Lisa Kr <[log in to unmask]> wrote:
Hello,

I have some trouble understanding the discrepancy in the results of my ANOVA. While the F-stat image shows the regions with significant differences between my three groups, the individual t-tests thenĀ  show that the differences between the single groups are in regions other than these F-test has shown (see attached, red=f-test, green and blue: group differences). Is there any explanation for this? I was expecting the differences between groups to fall in the general areas indicated by the F-Test.

Thanks!!

Lisa