Hi Philip, the p-value for the cluster is not computed from a t or z. It is computed from a simulation in which random noise is generated on the surface, smoothed, clustered. The maximum cluster size is found. This is repeated 10000 times to give a list of 10000 cluster sizes. The p-value for a given cluster in your analysis equals the number of clusters in the list whose area is greater divided by 10000. dopug
On 5/17/13 9:46 AM, Kelly, Philip wrote:
Dear Freesurfer list members,
Within QDEC I have found a number of clusters that show significantly different cortical thickness and local gyrification between two groups.
The cluster summary exhibits the p value, coordinates and other information regarding the cluster, however I wanted to report the t or z values and was unsure in finding or computing this from QDEC.
The p-value must be computed from one of these scores but I could not find it to be reported in any of the output.
Searching on the mailing list and tutorials, I could not find an easy solution to this.
Thanks for your time and I look forward to hearing back.
All the best,
Philip Kelly
Freesurfer mailing list Freesurfer@nmr.mgh.harvard.edu https://mail.nmr.mgh.harvard.edu/mailman/listinfo/freesurfer