The R package 'brms' is only relevant to two recent programs, MBA and BayesianGroupAna.py, not 3dMVM/3dLME.
I suggest that you use 3dMVM in this case because you have a typical ANCOVA. And make sure that you declare the variable 'Morph' as a within-subject variable.
-gltLabel 1 RootFluency_effect -gltCode 1 'Morph : 1*words -1*scrambled Root_fluency :' \
-gltLabel 2 PhonFluency_effect -gltCode 2 'Morph : 1*words -1*scrambled Phon_fluency :' \
These two tests don't make sense with the model you specified:
-model "Morph+Root_fluency+Phon_fluency" \
If you assume that no interaction exists between 'Morph' and 'Root_fluency', then why are you requesting for those two tests? You may consider a model like
-model "Morph*Root_fluency+Morph*Phon_fluency" \
Gang
Edited 1 time(s). Last edit at 03/14/2019 11:16AM by Gang.