Posted by Gary on March 28, 2001 at 10:55:21:
In Reply to: Re: significance levels (-mask issue) posted by B. Douglas Ward on March 28, 2001 at 10:07:07:
My version has 'last revision July, 1999'. I'll try to update it and also
look at my mask again in case I've done something off the mark there.
I know I said just one more question but...
When I enter the rmm value in AlphaSim it is in
original voxel size (I entered 5.1 for nearest neighbour clustering
with an original voxel of 3.125 x 3.125 x 5).
In 3dclust, the voxels are 1x1x1 so the rmm I've entered is 1.1.
to keep the same nearest neighbour approach as the sim.
Does this change in voxel size mess with the predicted
alpha levels? Or does keeping the nearest neighbour criterion
constant across metrics take care of this? Thanks.
Gary