 |
 |
Hi, I have data looking at response to medication, collected at 3 time points (3 different scans - baseline, 1 week, 6 weeks). About half of the people responded to the medication (2 groups - responder and non-responders) and I am trying to set up a 2-way mixed effects ANOVA to test for main effects of time, response group, and time x response group interaction. At first I followed the instructions from: https://fsl.fmrib.ox.ac.uk/fsl/fslwiki/GLM#ANOVA:_2-factors_2-levels_.282-way_between-subjects_ANOVA.29 ...ending up with this first model, which gave me an error that a linear combination of EVs is close to zero, and contrasts will be set to zero. (with EV1 representing Time, and EV2 representing GrpxTime) Subject Group EV1 EV2 EV3 EV4 EV5 EV6 EV7 1 input 1 1 -1 1 1 0 0 0 0 1 input 2 1 1 -1 1 0 0 0 0 2 input 3 1 -1 1 0 1 0 0 0 2 input 4 1 1 -1 0 1 0 0 0 3 input 5 1 -1 1 0 0 1 0 0 3 input 6 1 1 -1 0 0 1 0 0 4 input 7 1 -1 -1 0 0 0 1 0 4 input 8 1 1 1 0 0 0 1 0 5 input 9 1 -1 -1 0 0 0 0 1 5 input 10 1 1 1 0 0 0 0 1 Contrast: C1 Time 1 0 0 0 0 0 0 F1 C2 GrpxTime 0 1 0 0 0 0 0 After looking on forums I found the example suggested from this design (https://dl.dropboxusercontent.com/u/2785709/outbox/mailinglist/design_scott_full.ods) and tried to backtrack from this , but still can't work out the contrasts to get the main effects of group and time, as well as the group x time interaction. (with EV1 representing Grp1, and EV2 representing Grp2) Subject Group EV1 EV2 EV3 EV4 EV5 EV6 EV7 1 input 1 1 1 0 1 0 0 0 0 1 input 2 1 1 0 -1 0 0 0 0 2 input 3 1 1 0 0 1 0 0 0 2 input 4 1 1 0 0 -1 0 0 0 3 input 5 1 1 0 0 0 1 0 0 3 input 6 1 1 0 0 0 -1 0 0 4 input 7 1 0 1 0 0 0 1 0 4 input 8 1 0 1 0 0 0 -1 0 5 input 9 1 0 1 0 0 0 0 1 5 input 10 1 0 1 0 0 0 0 -1 Contrast: C1 Grp1>Grp2 1 -1 0 0 0 0 0 F1 C2 Grp2>Grp1 -1 1 0 0 0 0 0 F1 C3 Grp1xTime 1 0 1 1 1 0 0 F2 C4 Grp2xTime 0 1 1 1 1 0 0 F2 Can anyone help please? Thanks in advance! Elliot.