Hi Anderson,
As always, very insightful and clear, many thanks!
We were pondering about some extra contrasts to set up with the design as you suggested:
EV1: Schizophrenia (coded as 0 or 1)
EV2: Bipolar (coded as 0 or 1)
EV3: Healthy (coded as 0 or 1)
EV4: EV1*Q1
EV5: EV2*Q1
EV6: EV3*Q1
EV7: EV1*Q2
EV8: EV2*Q2
EV9: EV3*Q2
1) How to set up contrasts that probe linear relations between our covariates and TBSS metrics across the 3 groups?
These are the two options we have in mind and wonder which one you would recommend:
Option 1:
Q1 positive: 0 0 0 1 1 1 0 0 0
Q1 negative: 0 0 0 -1 -1 -1 0 0 0
Q2 positive: 0 0 0 0 0 0 1 1 1
Q3 negative: 0 0 0 0 0 0 -1 -1 -1
Option 2:
Q1 positive: 0 0 0 0.33 0.33 0.33 0 0 0
Q1 negative: 0 0 0 -0.33 -0.33 -0.33 0 0 0
Q2 positive: 0 0 0 0 0 0 0.33 0.33 0.33
Q3 negative: 0 0 0 0 0 0 -0.33 -0.33 -0.33
2) You suggest the contrast below (plus F-tests) for probing interaction affects between either of our continues covariates and TBSS metrics:
C1: 0 0 0 1 -1 0 0 0 0
C2: 0 0 0 1 0 -1 0 0 0
C3: 0 0 0 0 0 0 1 -1 0
C4: 0 0 0 0 0 0 1 0 -1
We were wondering whether we might also need the reverse of these contrasts in the model:
C1: 0 0 0 -1 1 0 0 0 0
C2: 0 0 0 -1 0 1 0 0 0
C3: 0 0 0 0 0 0 -1 1 0
C4: 0 0 0 0 0 0 -1 0 1
3) You also mention the possibility to include 3 additional EVs to code the 3-way interaction? Do you mean the Group x Q1 x Q2 interaction?
Would be great to hear your input on the above.
Many thanks!
Cheers,
Ijom
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|