Quasi-experimental Design :: Research methods and statistics

I. Jackson (2012), even-numbered chapter exercises, p 360. 2. The recommended design for this type of study is a non-equivalent control group post-test only design. 4. If a study is confounded, the researcher is not absolutely certain that changes in the dependent variable were caused by the manipulation of the independent variable, or some other uncontrolled variable. In a non-equivalent control group post-test only design, any differences observed between the two classes may be due to the non-equivalence of the groups and not to the injection of quizzes. No pre-test measures were given to establish equivalence. Another confound that may impact the results of this study could be the testing effect. Repeated testing may lead to better or worse performance. Changes in performance on the test may be due to prior experience with the test and not to the independent variable. In addition, repeated testing fatigues the subjects, and their performance declines as a result (Jackson, 2012). Because the professor is interested in determining if the implementation of weekly quizzes would improve test scores, an experimenter and/or an instrumentation effect may also affect results. In a single group post-test only design, possible confounds include the lack of a comparison group and the absence of an equivalent control group. Once again, a testing and an experimenter effect could also contribute to changes in test performance. 6. A single-case design is used when: 1. Only one person is measured. 2. The researcher does not want or need to generalize the results to a population. 3. The researcher believes it is unethical to withhold treatment to one group. 8. A multiple-baseline design differs from a reversal design by attempting to control for confounds through the introduction of treatment at differing time intervals to a few different people, to the same person in different situations, or to the same person across different behaviors. Reversal designs attempt to control for confounds by reversing the baseline and treatment conditions one or more times to assess the impact on behavior (Jackson, 2012). 2. Describe the advantages and disadvantages of quasi-experiments? What is the fundamental weakness of a quasi-experimental design? Why is it a weakness? Does its weakness always matter? Quasi-experimental designs are experimental designs that do not provide for the full control of extraneous variables. Primarily, the absence of control in this design is due to the lack of random assignment to groups. Quasi-experimental research designs are used in the study of cause and effect by manipulating the independent variable.