I have a dataset where I’d like to perform a 2-way repeated measures ANOVA (Condition*TIme). I’ve been using 3dANOVA3 -type 4 to achieve this, but I was wondering if there is a way to include 3dClustim in a similar way to 3dttest (-Clustim)?
I’m not sure if this is the best approach, but my plan is to use 3dANOVA to generate Condition*TIme interaction maps with the F-stat. Then, only look at post-hoc comparisons within significant clusters. Therefore, I would like to be able to use 3dClustim to determine significant interaction clusters at individual voxel p<.05.
I've been using 3dANOVA3 -type 4 to achieve this, but I was wondering if there is a way to include 3dClustim in a similar way to 3dttest (-Clustim)?
In this case you can directly use 3dttest++ on each of the contrasts you're interested.
I'm not sure if this is the best approach, but my plan is to use 3dANOVA to generate Condition*TIme interaction maps with the F-stat. Then, only
look at post-hoc comparisons within significant clusters. Therefore, I would like to be able to use 3dClustim to determine significant interaction
clusters at individual voxel p<.05.
It's difficult to define/set a criterion for "the best approach". The cluster approach is usually arbitrary and likely conservative. See discussion here: Thoughts about modeling and result reporting
How do I use the cluster approach to create the Condition*Time interaction map so that I can then do the post-hoc comparisons using 3dttest within the significant clusters?
The
National Institute of Mental Health (NIMH) is part of the National Institutes of
Health (NIH), a component of the U.S. Department of Health and Human
Services.