送交者: xj 于 2005-8-20, 00:40:13:
回答: The problem is: 由 cornbug 于 2005-8-19, 23:49:59:
since you may have multiple depedent variables each measured four times, you can stick to GLM model which allows for multivariate format.
Specifically, you organize each depedent variables as:
Dependent Variables
Y1 Y2 Y3 Y4 Y5 Y6 Y7 Y8 Y9 Y10 Y11 Y12
measurements 1 1 1 1 2 2 2 2 3 3 3 3
time 1 2 3 4 1 2 3 4 1 2 3 4
proc glm;
classes group;
model Y1-Y12=group / nouni;
repeated measure 3, time 4;
run;
This is derived from SAS manual under GLM repeated statements. I never used this before.
This way, you practically have three measurements each measured four times.
However, the interpretation of results should be the same as MANOVA.