Open tsalo opened 2 years ago
I think this depends on whether the null distribution should be symmetric and zero-centered or not. For ALE subtraction analysis, any differences in the number of studies in the two datasets will shift the null distribution.
Is that going to be the case with MKDAChi2 as well?
Are there other cases where this should be considered?
@nicholst is this issue something you're familiar with?
One question I have is whether the maximum statistic null distributions should be separated by sign for two-sided tests. For example, for the MKDAChi2, which can have positive and negative chi-squared values, should we retain maximum negative cluster size and maximum positive cluster size distributions separately?
I'm thinking about this primarily because of what I read in Albajes-Eizagirre et al. (2019):
Originally posted by @tsalo in https://github.com/neurostuff/NiMARE/issues/655#issuecomment-1073029709