Dear Experts,
I tried to run two-way repeated measurement ANOVA analysis using flexible factorial (SPM8). I have 42 subjects, each subject has two condition and each condtion has 23 frequeny bands. So ,it should run a within-subject two-way anova analysis using flexible factorial design.I set 'subject' as a factor,so the factors are configured in flexible factorial design/SPM8 as:
Factor 1: subject
Factor 2: condition (A)
Factor 3: frequency_band(B)
The ordering of each subject's images is: A1B1 A1B2 A1B3 ...A1B23 A2B1 A2B2 A2B3...A2B23, therefore, the conditions for each subject is set as: 1 1; 1 2; 1 3;...,1 23; 2 1; 2 2; 2 3;...,2 23; (according to "Contrast weights in flexible factorial design with multiple groups of subjects by Jan Glascher and Darren Gitelman).
My question is how to set the F-contrast for two main effects and interaction effect.For the main effect of condition, we can set the contrast as [1 -1] by giving a comparison of level 1 > level 2, but how to set the contrast of the interaction effect (because there is no hypothesis across the 23 levels,I mean I have no hypothesis of level 1 < level 2 < level 3 ...< leve23).
Any comments are greatly appreciated in advance!
Bin-Ke
