 |
 |
Dear all, I get the warning "Warning: at least one EV is (close to) a linear combination of the others. You probably need to alter your design. (Design matrix is rank deficient - ratio of min:max eigenvalues in SVD of matrix is 1.33432e-16). Contrasts involving these combinations will be set to zero.” Design and Covariance matrix attached But I don’t understand where the problem lies. The design matrix is based on te Feat - user manual page: “Paired Two-Group Difference" I’ve got two lower level copes for each subject. My question is: is the change in activation (session 1 to 2), different for the sleep and wake group. EV1 “SLP” indicates whether the subject had a sleep-interval (1), or a wake-interval (-1). EV2 “SES” indicates whether the input was from session 1 (-1), or from session 2 (1). EV3 “SLPxSES” is the element-wise multiplication of EV1 and EV2: the interaction term. EVs 4 to 55 are there to model each subject’s mean effect, but no contrasts were places on these EVs. Thank you very much in advance for your help, it is much appreciated! Rick Wassing Netherlands Institute for Neuroscience