Print

Print
Samantha,

Just to be clear, your model has a column for HC, a column for Patients, and a column with the covariate.

If this is the case, then the model is correct and the data are the problem.
(1) How many subjects do you have in each group? Too few subjects would lead to small effects.
(2) Have you checked to make sure the 1st level subject masks overlap? If they don't overlap, then then are no significant voxels.
(3) Have you checked the data for each subject is different?

Best Regards, 
Donald McLaren, PhD


On Fri, Oct 30, 2015 at 6:23 AM, Samantha Brooks <[log in to unmask]> wrote:
Hi SPM Experts,

My colleagues and I tried to run a between group ANCOVA , testing main effect of group (HC versus patients) and main effect of working memory accuracy (high versus low), in the  con1 files only with Total Matter Volume as covariate of no interest, and we get this output error:

Failed  'Model estimation'
Error using spm_est_non_sphericity (line 196)
Please check your data: There are no significant voxels.
In file "C:\Program Files\MATLAB\R2012b\toolbox\spm12\spm_est_non_sphericity.m" (v6015), function "spm_est_non_sphericity" at line 196.
In file "C:\Program Files\MATLAB\R2012b\toolbox\spm12\spm_spm.m" (v6015), function "spm_spm" at line 418.
In file "C:\Program Files\MATLAB\R2012b\toolbox\spm12\config\spm_run_fmri_est.m" (v5809), function "spm_run_fmri_est" at line 33.

The following modules did not run:
Failed: Model estimation

What are we doing incorrectly, can anybody out there help us?

Thanks!

Samantha
_______________
Dr Samantha Brooks, Ph.D (Download Website) 
Dept. of Psychiatry,
J2 Building, Groote Schuur Hospital
Anzio Road
Observatory
Cape Town