* make data for a*b*c 3 by 2 by 3 design with ; * factor c repeated data test; do a = 1 to 3; do b = 1 to 2; do c = 1 to 3; x = ranuni(-1); output; end; end; end; run; * test bcomp (b1 vs b2) at a2 c3 ; * note order of variables in class statement ; * determines how quickly vars change in contrast ; * statement. a changes slowest, then b, and c changes ; * fastest ; proc mixed data=test; class a b c ; model x = a b c a*b a*c b*c a*b*c ; contrast 'bcomp at a2 c3' a 0 0 0 b 1 -1 c 0 0 0 a*b 0 0 1 -1 0 0 a*c 0 0 0 0 0 0 0 0 0 b*c 0 0 1 0 0 -1 a*b*c 0 0 0 0 0 0 0 0 1 0 0 -1 0 0 0 0 0 0 ; run;