groups of subjects receive the usual standard of care or a placebo. One experi-
mental group and one control group are measured before and after the inter-
vention, while the other experimental and control groups are measured only
after the intervention (see Table 7-1). Researchers select this four-group design
over the two-group pretest-posttest design to reduce the threat of testing. This
design is also superior to the two-group posttest-only design because selection
bias is minimized. Consider the diabetic teaching intervention example again.
Suppose the researcher is concerned that subjects will learn information by
taking the diabetic knowledge test. By adding two more groups that do not
take the pretest, the researcher can compare posttest scores from subjects who
took the pretest to scores of subjects who did not take the pretest. If testing
is a threat, the researcher can expect that the pretest groups will have higher
posttest scores than those who did not take the pretest.
Although the Solomon four-group design is a strong design, it has limita-
tions. Because there are more groups, sample sizes must be large. This means
that subjects must be available, recruitment of subjects will take longer, and
costs will be increased.
Multiple Experimental Groups Designs
It is possible for researchers to use either the posttest alone or the pretest-
posttest design with multiple experimental groups designs. To conduct this
type of study, researchers would have multiple experimental groups and one
control group. In the example of interventions to reduce pain during invasive
procedures, the researcher might be interested in testing two different interven-
tions. The researcher could randomly assign subjects to experimental group 1,
which would receive music therapy. Other subjects would be randomly as-
signed to experimental group 2, where they would receive therapeutic touch.
Other subjects would be randomly assigned to the control group for the usual
standard of care. The advantage of the multiple-groups design is that it allows
researchers to compare the effect of different interventions on the DV. A major
disadvantage of this design is that a large number of subjects is needed to detect
differences across multiple groups.
Factorial Designs
In the previously discussed designs, only one intervention has been manipu-
lated. Even in multiple-groups experiments, each experimental group receives
only one intervention. Factorial designs allow researchers to manipulate more
than one intervention during the same experiment. Researchers may compare
multiple interventions (e.g., music and therapeutic touch combined) or mul-
tiple levels of interventions (e.g., music with therapeutic touch for 10 minutes,
15 minutes, or 20 minutes). For example, a researcher is interested in increasing
KEY TERMS
multiple
experimental
groups designs:
Experimental
designs using
two or more
experimental
groups with one
control group
factorial designs:
Experimental
designs allowing
researchers to
manipulate
more than one
intervention7.1 Experimental Designs 175