 |
 |
Thank you so much for your fast reply. One more quick question, if we are only interested in the interaction of the two variables, is it necessary to include EV2 and EV3? Thanks again, Lindsay ________________________________ From: FSL - FMRIB's Software Library <[log in to unmask]> on behalf of Anderson M. Winkler <[log in to unmask]> Sent: Monday, May 23, 2016 8:01:36 PM To: [log in to unmask] Subject: Re: [FSL] Within group design, 2 variables of interest, 2 nuisance variables Hi Lindsay, I think this design should do it: EV1: intercept EV2: GDS EV3: CAPS EV4: GDS*CAPS EV5: age EV6: ICV The relevant contrasts are then: C1: 0 0 0 1 0 0 C2: 0 0 0 -1 0 0 All the best, Anderson On 23 May 2016 at 21:22, Lindsay Knight <[log in to unmask]<mailto:[log in to unmask]>> wrote: Dear FSL experts, We are currently working on running structural analyses through FIRST. Our sample is within group (all PTSD patients) and our two continuous variables of interest are measures of depressive severity (GDS) and PTSD symptamatology (CAPS). We also have two nuisance variables: age and ICV. Specifically, we would like to look at the interaction of the two behavioral measures (i.e. multiple regression of GDS x CAPS) while controlling for age and ICV, in one model. Our investigation then, is to see how the interaction of these behavioral measures predicts subcortical volume and/or shape. We are having trouble, however, figuring out how to set up the design matrix, contrast codes and F-tests. Any insights would be greatly appreciated! Thank you, Lindsay Knight