Information Technology Reference
In-Depth Information
Descriptives
satscore
95% Confidence Interval for
Mean
N
Mean
Std. Deviation
Std. Error
Lower Bound
Upper Bound
Minimum
Maximum
zero months
two months
four months
six months
eight months
Total
7
412.8571
26.90371
10.16865
387.9754
437.7389
370.00
440.00
7
474.2857
47.90864
18.10777
429.9776
518.5938
410.00
530.00
7
552.8571
28.70208
10.84837
526.3121
579.4021
530.00
610.00
7
614.2857
38.23486
14.45142
578.9244
649.6471
560.00
660.00
7
622.8571
36.38419
13.75193
589.2074
656.5069
550.00
660.00
35
535.4286
89.15759
15.07038
504.8019
566.0553
370.00
660.00
Figure 6.5
The first portion of the output from the One-Way ANOVA.
need to rerun the analysis asking for either the
Brown-Forsythe
or the
Welch
significance tests.
In the
Missing Values
area, SPSS uses
Exclude cases analysis by anal-
ysis
as a default. The other choice,
Exclude cases listwise
,isapplicable
when you want to perform many analyses in a single run based on the very
same set of cases. Invoking this choice would remove from all of the anal-
yses
any
case with a missing value on
any
dependent measure. Since we
have only one analysis, the choice is irrelevant and we can leave the default
in place. Click
Continue
to return to the main dialog window. Then click
OK
to run the analysis.
6.7 THE OUTPUT OF THE OMNIBUS ONE-WAY BETWEEN-
SUBJECTS ANOVA IN SPSS
6.7.1 DESCRIPTIVE STATISTICS
The descriptive statistics generated by
One-Way ANOVA
are shown in
Figure 6.5. The
One-Way ANOVA
procedure provides quite a bit of infor-
mation: the mean, standard deviation, standard error with its 95 percent
confidence interval, and the minimum and maximum scores in each
group.
6.7.2 HOMOGENEITY TEST
Figure 6.6 displays the result for Levene's test of homogeneity of vari-
ance. The statistic (symbolized as
W
if you are reporting it) has a value
of 1.459 and, with 4 and 30
df
, has a probability of occurrence of .239 if
the null hypothesis is true. Against an alpha level of .05 this probability
is not statistically significant. We therefore do not reject the null hypoth-
esis that the variances of the five conditions are comparable; in short, it
Test of Homogeneity of Variances
The equal variances assumption is not
violated; the probability (Sig.) is > .05.
satscore
Levene
Statistic
df1
df2
Sig.
1.459
4
30
.239
Figure 6.6
The results of Levene's homogeneity of variance test.
