Dear Jeff, The error indicates that the vector you entered as covariate does not have the right length (two times the total number of subjects). Depending on what you actually want to do with the ANCOVA, you might want to use a two-sample t-test after contrasting the within-subject factor at the first level ([1 1] or [1 -1]) and including the covariate there (potentially interacting with group). Best regards, Guillaume. On 16/06/2019 20:39, Jeff Browndyke wrote: > Is it possible to run a 2x2 ANCOVA in SPM12? Right now I have subject, group and condition factors in my model. > > I keep on getting the following error when trying to add a covariate to my model: > > ---------- > > 16-Jun-2019 15:34:52 - Running 'Factorial design specification' > 16-Jun-2019 15:34:54 - Failed 'Factorial design specification' > Index exceeds matrix dimensions. > In file "/Applications/Neuroimaging/functional/spm12/spm_design_within_subject.m" (v4837), function "spm_design_within_subject" at line 82. > In file "/Applications/Neuroimaging/functional/spm12/config/spm_run_factorial_design.m" (v6219), function "spm_run_factorial_design" at line 515. > > No executable modules, but still unresolved dependencies or incomplete module inputs. > > ---------- > > Thanks, > Jeff Browndyke > -- Guillaume Flandin, PhD Wellcome Centre for Human Neuroimaging UCL Queen Square Institute of Neurology London WC1N 3BG