 |
 |
Dear FSL experts, Is there any relationship between the size of the mask that we use in randomize and the power of statistical analysis? For example When we use a mask for the whole brain in randomise to study the difference between the groups that means we are using the maximum number of voxels to study the difference between the groups everywhere in the brain ( white and gray matter). 1. If we replace that mask with a mask for gray matter. The number of voxels is lower so can this affect the power ? 2. If I want to do voxel wise analysis to study the differences between the groups only in the gray matter or only in the white matter. Can I replace the whole brain mask with a mask for the white matter or a mask for the gray matter. 3. Are there any specific flags that I need to add to randomise ? Best, jon