Not sure where else to ask, since statistics forum didn't give any exact answers about my type of situation.
I have a group of 20 subjects, and each of them has two connectivity matrices - before and after a treatment.
Is there a way for making statistics for each cell in matrix, so it would be like a mean of group's value in each cell? I mean, if one of matrix cells is connection's value between electrodes Fp1 and Cz - can I get mean value and standard deviation in that cell? Also can I count a significant difference between two matrices, making evaluation for each matrix cell?
I'm not sure, if I could get anything like that in Brainstorm, so I exported data. Maybe I need different app for data analysis? I'm using R for my statistics mostly, but not sure how to apply it for now. Maybe I need to decompose matrix, do statistics for each matrix cell and compose in back?
You can test before/after differences of connectivity values using two-sample t-tests or permutation randomizations (non-parametric approach) for instance.
1 Like
In the Process2 tab, you can select all the connectivity files (FilesA=before, FilesB=after) and then run non-parametric tests.
https://neuroimage.usc.edu/brainstorm/Tutorials/Statistics#Example_2:_Permutation_t-test
1 Like