designs, the nonintervention group is referred to as a comparison group rather
than a control group. Researchers are able to measure both groups on the DV
prior to and after the intervention (see Table 7-3). For example, the faculty of
a nursing college is considering changing a nursing research course offered to
junior students. In one semester, a professor administers a pretest measuring
knowledge about nursing research. The professor then teaches the course con-
tent in the usual lecture style. At the end of the semester, the final exam is used
as a posttest to measure students’ knowledge of research. The next semester,
another group of students enter the research course. They are given the same
pretest, but the instruction is online with no class lectures. At the end of the
semester, the same comprehensive final exam is administered. The faculty would
then be able to compare students’ knowledge pre- and postintervention and
make a determination regarding the effect of the different teaching methods
on students’ knowledge about research. Because students were unable to be
randomly assigned to a teaching method, this is not an experimental design.
Without randomization, it is possible that the students are not equivalent on
important factors that may affect how they learned the course materials. This
contributes to selection bias and weakens the design. Threats to internal validity
as a result of testing, maturation, and mortality can exist.One-Group Pretest-Posttest (No Randomization)
Group 1 O 1 X O 2
Nonequivalent Control Group Pretest-Posttest (No Randomization)
Experimental Group O 1 X O 2
Comparison Group O 1 O 2
One-Group Time Series (No Randomization)
Group 1 O 1 O 2 O 3 X O 4 O 5
One-Group Posttest-Only (No Randomization, No Pretest)
Group 1 X O 1
Nonequivalent Control Group Posttest-Only (No Randomization, No Pretest)
Experimental Group X O 1
Comparison Group O 1
Note: O = observation or measurement; X = treatment or intervention.TABLE 7-3 Quasi-Experimental Designs
178 CHAPTER 7 Quantitative Designs: Using Numbers to Provide Evidence