Dear SPM experts,
Sorry for the following stupid statistics question, but it
is very important to me: I have some preliminary data (groups of 4 and 4).
Using spm8, I was able to get the contrasts of each subject, but because
my groups are too small and I couldn't get any group result, how can I address
the "effect size"? After reading a bunch of statistics descriptions,
I still don't have a clear understanding. I understand in SPM8 the
"plot" function can plot a voxel's effect size with an error bar, but
I don't really understand what it means. Also, if I provide the z scores
on activated pixels of a representing subject, does it address the statistical
power somehow? Could someone kindly explain or direct me in a simple way?
Thank you very much,
Wei