Dear SPM Community,
I see some odd behavior in my data.
If I choose a voxel wise threshold of FWE p <0.1 very small clusters result
These clusters are on the order of 1 - 20 k
While investigating the cluster level p's I noticed something odd:
My cluster uncorrected p value typically was more conservative than the FWE corrected p value. Why would this happen?
I typically analyze my data starting with a voxel wise threshold of uncorrected p<0.01 and cluster wise threshold of FWE p<0.05. As would be expected the cluster wise FWE p is more conservative than the uncorrected p.
Thank you,
Elizabeth Sharer
|