Information Technology Reference
In-Depth Information
SS
Tests of Within-Subjects Effects
A
Measure: MEASURE_1
Type III Sum
of Squares
Partial Eta
Squared
Source
VEHICLE
df
Mean Square
F
Sig.
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
Sphericity Assumed
Greenhouse-Geisser
Huynh-Feldt
Lower-bound
2.778
2.778
2.778
2.778
1
1.000
1.000
1.000
5
5.000
5.000
5.000
2
1.946
2.000
1.000
10
9.729
10.000
5.000
2
1.330
1.630
1.000
10
6.651
8.149
5.000
2.778
2.778
2.778
2.778
1.761
1.761
1.761
1.761
.242
.242
.242
.242
.260
.260
.260
.260
SS
A×S
Error(VEHICLE)
7.889
7.889
7.889
7.889
157.389
157.389
157.389
157.389
14.944
14.944
14.944
14.944
17.056
17.056
17.056
17.056
3.278
3.278
3.278
3.278
1.578
1.578
1.578
1.578
SS
B
78.694
80.888
78.694
157.389
52.658
52.658
52.658
52.658
.000
.000
.000
.001
.913
.913
.913
.913
DRINKS
SS
B×S
Error(DRINKS)
1.494
1.536
1.494
2.989
SS
A×B
VEHICLE * DRINKS
8.528
12.822
10.465
17.065
26.017
26.017
26.017
26.017
.000
.001
.000
.004
.839
.839
.839
.839
SS
A
×
B
×
S
Error(VEHICLE*DRINKS)
.328
.493
.402
.656
Figure 11.14
Within-subjects ANOVA summary table.
within-subjects variable are correlated to the same extent. With only two
levels of
vehicle
, the test is not interpretable and SPSS does not test its
significance. With three or more conditions available for evaluation with
respect to
drinks
and the
vehicle
drinks
interaction, Mauchly's spheric-
ity statistic is computed and tested for significance; neither is statistically
significant.
The summary table for the within-subjects variance is displayed in
Figure 11.14. With a nonsignificant sphericity test, we can focus on the
rows labeled
Sphericity Assumed
. There are three effects of interest: the
two main effects and the two-way interaction. Each is evaluated by its own
error term and, as we noted in Table 11.1, the main effect for
drinks
and
the
vehicle
∗
drinks
interaction are both statistically significant.
Figure 11.15 shows the between-subjects portion of the variance. The
Intercept
concerns the full model that we do not deal with here. What
∗
This is the between
subjects variance (
SS
S
)
representing individual
differences. SPSS does
not compute an
F
ratio
for this source.
Tests of Between-Subjects Effects
Measure: MEASURE_1
Transformed Variable: Average
Type III Sum
of Squares
Partial Eta
Squared
Mean Square
Sig.
Source
Intercept
Error
df
F
1
5
544.444
30.222
544.444
6.044
90.074
.000
.947
Figure 11.15
Between-subjects ANOVA summary table.
