Dear all,
given that I ask for the overall effect of three regressors in a GLM
that as a group explain structural variability in a region (VBM study),
under which circumstances can I interprete the signature of the contrast
estimates?
So far, I understood that for an isolated question on one regressor in a T
contrast this
regressor should be orthogonal in relation to the other regressors.
Does this prerequisite (orthogonality) also hold for plotting/interpreting
the contrast
estimates from an F contrast?
thank you very much for hints on this,
Philipp
Max Planck Institute of Psychiatry
NMR Research Group
Kraepelinsr. 2-10
80804 Munich
Mail: [log in to unmask]
Phone: 0049-89-30622-413
|