Information Technology Reference
In-Depth Information
Figure 12.3
The data file for our three-way within-subjects example.
How the data are structured for a three-way within-subjects design
(the order of the cells in the data file) should be determined by the way
you will be presenting the results to your readers. This structuring, in
turn, is often suggested by the theoretical framework or model within
which the research was generated. If, as is sometimes the case in social and
behavioral research, the study was more exploratory in nature, then the
variables, while they still should be based on a systematic mapping of the
independentvariables,canbeorderedinanystructurethattheresearchers
choose. As can be seen in Figure 12.3, we split the data most globally by
powder color.
The variables (columns) in the data file shown in Figure 12.3 are
ordered in a systematic manner so that they will match up with the way in
which we displayed our independent variables in Figure 12.1. The names
of the variables capture the levels of the three variables:
Powder color is given in the first three letters:
WHT
for white,
GRE
for green, and
BLU
for blue.
Presence of granules is given in the next two or three letters:
YES
for
present and
NO
for absent.
Box color is given in the final two letters:
AQ
for aqua and
OR
for
orange.
In the data file, the variables are incrementing in the same manner as
they are in Figures 12.1 and 12.2. Here are the specifics:
Powder color increments most slowly - the first four columns fol-
lowing
subid
represent the boxes with the white powder, the next
four columns represent the green powder, and the next four columns
represent the blue powder.
Within each of the four columns for each powder,
granules
incre-
ments next most slowly - the first two columns represent the presence
of granules and the last two represent the absence of granules.