Hi Andrew, 

We are currently performing resting-state functional connectivity analysis with NBS and we wonder how to built our design matrix knowing that we have 5 independent groups and 4 covariates. We have found some tracks online and we hesitate between two designs: 

Design 1 (the first column is the intercept):
1 1 0 0 0 0 0 0 0 0 = group 1
1 0 1 0 0 0 0 0 0 0 = group 2
1 0 0 1 0 0 0 0 0 0 = group 3
1 0 0 0 1 0 0 0 0 0 = group 4
1 0 0 0 0 1 0 0 0 0 = group 5

Design 2 (the first column is the intercept):
1 1 1 1 -1 0 0 0 0 = group 1
1 -1 0 0 0 0 0 0 0 = group 2
1 0 -1 0 0 0 0 0 0 = group 3
1 0 0 -1 0 0 0 0 0 = group 4
1 0 0 0 1 0 0 0 0 = group 5 


We have performed our analyses with these two designs. In most cases, the two matrices gave the same results. However, sometimes they did not. Could you please help us identify the appropriate design matrix? 

Thank you in advance for your help, 
Regards, 

Quentin

Hi Quentin, 

For Design 1, the first column (intercept) is not necessary. Current matrix is rank deficient and may throw a warning. Other than that it looks ok. 

I am a bit confused by Design 2 - not sure what this is testing without knowing the contrast. 

Best,
Andrew


Originally posted by Quentin DEVIGNES:

Thank you very much for your answer! We will use the Design #1. 

Regards, 

Quentin