Hi,

I'm trying to run 3dMVM to see if there is significant difference between groups (G1, 2, 3, 4) in terms of neural activation during anticipation of monetary gain. I choose 3dMVM instead of 3dANOVA because we have unbalanced sample. I was wondering what is the accurate way to set up glfCode and gltCode to run F-test to test for the overall group difference, as well as post-hoc ttests to test the difference between the groups.

Below is what I had in the script:

-num_glf 1 \
-glfLabel 1 group -glfCode 1 'group : 1*G1 & 1*G2 & 1*G3 &1*G4' \
-num_glt 6 \
-gltLabel 1 G2-G1 -gltCode 1 'group : 1*G2 - 1*G1' \
-gltLabel 2 G3-G1 -gltCode 2 'group : 1*G3 - 1*G1' \
-gltLabel 3 G4-G1 -gltCode 3 'group : 1*G4 - 1*G1' \
-gltLabel 4 G3-G2 -gltCode 4 'group : 1*G3 - 1*G2' \
-gltLabel 5 G4-G2 -gltCode 5 'group : 1*G4 - 1*G2' \
-gltLabel 6 G4-G3 -gltCode 6 'group : 1*G4 - 1*G3' \

I have tried running the script, but received an error message saying: "Inappropriate coding in test No.1!"

Best,
Tien