Information Technology Reference
In-Depth Information
White Powder
Sea Green Powder
Baby Blue Powder
Granules
Present
Granules
Absent
Granules
Present
Granules
Absent
Granules
Present
Granules
Absent
Aqua
Box
Orange
Box
Aqua
Box
Orange
Box
Aqua
Box
Orange
Box
Aqua
Box
Orange
Box
Aqua
Box
Orange
Box
Aqua
Box
Orange
Box
Subid
1
6.50
4.00
2.70
3.20
2.10
2.00
2.20
2.30
4.50
5.00
3.60
6.70
2
6.50
4.00
2.70
3.20
2.10
2.00
2.20
2.33
4.50
5.00
3.60
6.70
3
5.50
4.80
3.60
3.80
1.50
2.50
1.60
3.20
3.50
6.00
4.60
7.10
4
7.50
3.20
1.80
2.60
2.70
1.50
2.80
1.50
5.50
4.00
2.60
6.30
5
7.20
4.70
3.70
4.40
1.50
2.40
1.70
3.10
3.40
5.60
4.20
7.50
6
5.80
3.30
1.70
2.00
2.70
1.60
2.70
1.60
5.60
4.40
2.80
5.90
Means
6.50
4.00
2.70
3.20
2.10
2.00
2.20
2.33
4.50
5.00
3.57
6.70
Figure 12.1
Raw data for the numerical example.
of laundry with a designated detergent, they are to rate how satisfied they
were with the outcome. The rating survey is composed of fifteen items
rated on a 9-point summative response scale. For each participant, items
are averaged for each condition; it is these means that are recorded in the
data file and that serve as the dependent variable. Thus, overall satisfaction,
the dependent variable in this study, can range between a low of 1 and a
high of 9 with higher values indicating greater satisfaction.
12.2 PARTITIONING THE VARIANCE INTO ITS SOURCES
The summary table for the ANOVA (shown in Table 12.1) contains both
the between and within partitions of the total variance. The portion of
the summary table labeled as
Between-subjects variance
presents the
between-subjects source of variance. Without a between-subjects variable
in the study, the only source of these differences that we can identify here is
the individual differences among the six participants. With six participants
there are 5
df
.No
F
ratio is ordinarily computed for individual differences.
The portion of the table labeled
Within-subjects variance
summarizes the
within-subjects partitioning of the variance. We highlight several aspects
of the table in our discussion below of the effects of interest.
12.3 EFFECTS OF INTEREST IN THIS DESIGN
With three independent variables in the design there are seven effects of
interest: the main effects of the three independent variables, the three
two-way interactions, and the single three-way interaction. In a within-
subjects design, all of these effects are subsumed in the within-subjects
variance. The hierarchical structure of these effects, as discussed in the
three-way between-subjects design chapter, still holds. Thus, the three-way
