Information Technology Reference
In-Depth Information
Linear Models
The GLM Procedure
satisfac
Level of
politics
Level of
voter
N
Mean
Std Dev
This is the mean
satisfaction score
for conservatives
who voted in the
last election.
1
1
1
2
2
2
1
2
3
1
2
3
10
10
10
10
10
10
34.5000000
24.5000000
15.0000000
31.5000000
25.0000000
25.0000000
5.4619899
12.2678441
8.1513462
6.9161646
9.3689795
12.1860576
satisfac
Level of
voter
Level of
politics
Level of
kidshome
Mean
Std Dev
N
1
1
1
2
2
2
1
1
1
2
2
2
1
2
3
1
2
3
1
2
3
1
2
3
1
1
1
1
1
1
2
2
2
2
2
2
5
5
5
5
5
5
5
5
5
5
5
5
36.0000000
14.0000000
10.0000000
35.0000000
32.0000000
15.0000000
33.0000000
35.0000000
20.0000000
28.0000000
18.0000000
36.0000000
3.53553391
6.51920241
6.67083203
4.52769257
6.51920241
6.51920241
7.0000000
4.52769257
6.51920241
7.51664819
5.70087713
4.00000000
This is the mean
satisfaction score for
moderates who did
not vote in the last
election and who have
no children in the
home.
Figure 9.24
Descriptive statistics output.
9.11 SAS OUTPUT FOR THE OMNIBUS ANALYSIS
A portion of the descriptive statistics generated by
LinearModels
is shown
in Figure 9.24. The mean, standard deviation, and the number of obser-
vations are displayed for each effect.
The summary table for the omnibus analysis is presented in the top
portion of Figure 9.25. Note that SAS provides the
Corrected Total
sum
of squares, informing us that we are viewing the results of the reduced
model. The sum of squares associated with
Model
is a compilation of
the sums of squares for the seven effects (three main effects, three two-
way interactions, and one three-way interaction) that comprise the model
(each effect is treated as a predictor in the model). These seven effects are
evaluated individually in the bottom table. As can be seen, all of the effects
except for
voter
are statistically significant.
