Hi all,
I am analysing the group comparison between healthy control(HC) and disease groups (DG), using birth variable covariates such as birth weight (BW) and gestastaional age(GA) as 2nd-level covariates. I defined HC, DG, BW for HC (BW1), BW for DG (BW2) and GA for HC (GA1), GA for DG (GA2) as 2nd-level covariates in the SETUP of conn.
My questions are:
I want to look at the within the disease group, ROI-based approach: correlation between birth variables and mean ß-values of group different clusters. If I want to look at that am I right to define the contrast like: [0 1] for DG and GA2? Actually I am confusing about difference of selecting [0 1]-effect of GA2 or [1 0] effect of DG. Am I looking relation between covariates and connectivity (for example regions has decreased connectivty in DG group with increasing GA) when I select [0 1]-effect of GA2 option? and Am I controlling whether GA has any effect on connectivity in DG when I select [1 0] effect of DG option?
In short, I am curious about how can I make an interpretation on increasing GA cause decreased or increased connectivity between my ROI and related clusters.
Is that way correct to import cluster which found in result of seed to voxel analysis for DG and get the subject spesific beta values of these clusters and check the correlation between these values and GA values in SPSS for checking the effect of covariate? But I am not sure that how to interpret the outcome of SPSS about effect of covariate?
I hope I can express myself clearly
thanks in advance