Information Technology Reference
In-Depth Information
This Table presents both
the unadjusted p -value
and the Bonferroni
adjustment
Differences of Least Squares Means
Standard
Error
Effect
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks*Vehicle
Drinks
Vehicle
Drinks
Vehicle
Estimate
DF
t Value
Pr
> t
Adjustment
Adj P
0
0
0
0
0
0
0
0
0
1
1
1
1
1
3
1
1
1
1
1
2
2
2
2
1
1
1
2
2
1
0
1
1
3
3
1
1
3
3
1
3
3
3
3
3
2
1
2
1
2
1
2
1
2
2
1
2
1
2
2
0.3333
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
0.5900
10
10
10
10
10
10
10
10
10
10
10
10
10
10
10
0.56
0.5846
0.0005
0.0001
<.0001
<
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
Bonferroni
1.0000
0.0071
0.0022
<.0001
0.0007
0.0166
0.0047
<
3.0000
−3.5000
5.08
−5.93
6.5000
11.02
4.0000
6.78
.0001
0.0011
0.0003
<
2.6667
4.52
3.1667
−6.1667
5.37
−10.45
.0001
<
.0001
0.0015
1.0000
0.0022
1.0000
0.0071
1.0000
0.0259
3.6667
6.21
.0001
0.4166
0.0001
0.1210
0.0005
0.4166
0.0017
0.5000
0.85
3.5000
5.93
1.0000
1.69
3.0000
5.08
0.5000
2.5000
0.85
4.24
Figure 11.33
All pairwise comparisons.
=
=
one coded as drinks
1 (car) whose mean (see Fig-
ure 11.30) is 1.000. The second condition is the one codes as drinks
0and vehicle
=
0 and vehicle
=
2 (SUV) whose mean is 1.3333. Given the expression
Condition 1
0.3333
in the Estimate row. The t value is 0.56 and the Bonferroni corrected
probability (labeled Adj P ) associated with that t value for 10 df is shown
as 1.0000 (SAS codes any computed Bonferroni adjusted probabilities
exceeding 1 as 1.0000); we thus conclude that the mean difference is not
statistically significant. Once again, the numerical values for the corrected
probabilities differ from those obtained from SPSS (see Section 10.19).
Condition 2
=
1.000
1.3333
=−
0.3333, we see
11.15 PERFORMING THE POST HOC ANALYSIS IN SAS
To perform the simple effects analysis in SAS Enterprise Guide , configure
the analysis as in Section 11.14. Then in the Least Squares Post Hoc Tests
screen shown in Figure 11.34, in the Effects to use frame, set drinks to
True and all others to False . Then in the Comparisons frame, set Show
p-values for differences to All pairwise differences and set Adjustment
methodfor comparison to Bonferroni .Click Run to perform the analysis.
11.16 SAS OUTPUT FROM THE POST HOC ANALYSIS
The pairwise mean comparisons are shown in Figure 11.35. As can be
seen, all conditions differ significantly from all others.
Search WWH ::




Custom Search