 |
 |
See inline responses below. Best Regards, Donald McLaren, PhD On Mon, Aug 10, 2015 at 3:48 PM, Tracy Ssali <[log in to unmask]> wrote: > Hello SPM Experts, > > I would like to perform a within subjects repeated measures analysis. To > do this I am performing a flexible factorial analysis with the following > factors: > > 1) Subject (1 group/ 7 subjects) > 2) Condition A (3 measures) > 3) Condition B (2 measures) > > (similar to instructions given here: > https://www.jiscmail.ac.uk/cgi-bin/webadmin?A2=ind0912&L=spm&P=R50884&1=spm&9=A&J=on&d=No+Match%3BMatch%3BMatches&z=4 > ) > > I've attached a sample of my design matrix to this email. > >> Please search the list archives for S1G1C1 for a detailed description of how to create contrasts. It all starts with the null hypothesis. Additionally, you need more than 3 subjects. > > To test for differences in Condition A (Main effects), I used to contrast : > [zeros (1,n), 0 0 0 -0.5 .5 -.5 .5 0 0; > zeros (1,n), 0 0 0 -0.5 .5 0 0 -.5 .5 0 0; > zeros (1,n),0 0 0 -0.5 .5 0 0 0 0 -.5 .5] > >> Main Effect of A: [zeros(1,n) 1 -1 0 0 0 .5 .5 -.5 -.5 0 0; zeros(1,n) 0 1 -1 0 0 0 0 .5 .5 -.5 -.5] > > and To test the for differences in Condition B (Main Effects), I used to > contrast: > [zeros (1,n), -1 0 1 0 0 -.5 -.5 0 0 .5 .5; > zeros (1,n), -1 1 0 0 0 -.5 -.5 .5 .5 0 0; > zeros (1,n), 0 -1 1 0 0 0 0 -.5 -.5 .5 .5] > >> Main Effect of B: [zeros(1,n) 0 0 0 1 -1 1/3 -1/3 1/3 -1/3 1/3 -1/3] > > I am not completely sure about whether these contrasts are correct. I > would also like to test for an interaction between condition A and B, > however, I do not know how to generate that contrast. > >> The validity of these main effects assumes that the residual error of each factor is similar. > > Any feedback/advice would be appreciated. > > Thanks, > Tracy >