Figure 8.20: Partitioning of sums of
squares in a 2×2 split-plot design
The data if collected in the usual way (one line per case with each repeated time
measurement on the same line) is first rearranged using a data step. For an explanation,
see under the heading Computer Analysis for the related ANOVA. PROC GLM is then
used to produce the ANOVA output see the following SAS code:
data a;
input subj sex t1 t2 t3;
cards;
1 1 5 8 6
2 1 4 8 6
3 1 3 5 5
4 1 6 10 9
5 1 3 6 5
6 0 6 5 7
7 0 4 4 4
8 0 5 6 5
9 0 4 5 5
10 0 3 4 5
;
data b; set a;
drop t1−t3;
time=1; score=t1; output;
time=2; score=t2; output;
time=3; score=t3; output;
proc glm data=b;
Statistical analysis for education and psychology researchers 340