Dear Friends:
I'm using TBSS to assess correlations (both positive and negative)
between FA and age, years of cocaine use, amount of cocaine use per
day, and years of alcohol use of 16 cocaine abusers. So, I have 4
variables (EVs). I will demean every variable first.
I'm wondering if following design and contrast are correct.
1. First Procedure:
Design:
EV1 EV2 EV3 EV4
age coca-y coca-d alco-y
..........
Contrast
1 0 0 0
-1 0 0 0
0 1 0 0
0 -1 0 0
0 0 1 0
0 0 -1 0
0 0 0 1
0 0 0 -1
2. Second Procedure
Assume I want to assess the correlation between FA and age only, I
also want to eliminate the confounds of other three variables such as
years of cocaine use. I will use following design:
First matrix:
EV1
age
...
Contrast
1
-1
Second matrix (after -x)
EV1 EV2 EV3
coca-y coca-d alco-y
3. Assume I want to assess the correlations between FA and each of
the four variables, and also eliminate the confounds of other three
variables for each assessed variable. Do I have to repeat the above
procedure 2 four times? Are there any better methods than repeat
procedure 2 four times?
Best
Jiansong Xu, M.D., Ph.D.
Associate Research Scientist
Dept. of Psychiatry
Yale School of Medicine
Connecticut Mental Health Center, Room S-100
34 Park St.
New Haven, CT 06519
|