sdm-help-list > Thresholded p-values are all the same
Showing 1-2 of 2 posts
Display:
Results per page:
Jun 15, 2020  04:06 PM | Anna Kasdan
Thresholded p-values are all the same
Hi all,

When thresholding with either the 'tfce' or 'corrp' voxel maps, regardless of the size of the cluster threshold, all p-values come out to the same value (0.00999). This is true for different analyses, where I get unique SDM-z values, brain regions/MNI coordinates etc., but the p-values are always the same which seems odd. Has anyone else experienced this? What does this mean, and is this a bug? Do the p-values here depend on the number of imputations in the mean analysis? 

Thank you in advance.

Best,
Anna
Jun 16, 2020  05:06 AM | Lydia Fortea - Instituto de Investigaciones Biom├ędicas August Pi i Sunyer (IDIBAPS)
RE: Thresholded p-values are all the same
Dear Anna, 

Is normal to get the same p-values, as you reported. The precision of the obtained p-values depends on the number of permutations you set. So, if you have set 100 permutations, the minimum p-value that can be obtained is 0.01 (depending of the computer would be 0.009999 instead of 0.01).  

Kind regards,
Lydia