Research design is broadly categorised into two main categories, quantitative and qualitative research designs. Qualitative designs can be further broken down into descriptive, correlational and explanatory research design. The Quantitative research designs are broken down into experimental, semi-experimental and quasi-experimental research designs. According to Kadzur (2003), Cook & Campbell (1979), and Pedhazur & Schmelkin (1991), the most common threats to internal validity are selection bias, repeated testing, history, statistical regression, instrumentation, attrition, imitation of treatment and unusual treatment of subjects. This paper will examine repeated testing as a threat to the internal validity of the research design and propose control measures to counter it.
In 2003 Kazdin stated that administering pre-tests may have adverse effects on subsequent administrations of the same test. According to Pedhazur & Schmelikin (1991) repeated testing on subjects using the same tests, may cause the respondents to be influenced by factors like the memory of the previous tests, practice in taking the tests and realisation of the expectancies of the researcher. An example of a research design where repeated testing affects the results of an experiment is longitudinal research. In longitudinal research, test subjects are tested continuously on the same variables over short or long period. What happens is as test subjects are repeatedly tested, they will, with time learn the test and start to anticipate the researcher’s questions and expected results, therefore, the results of the study are due to subjects learning what the researcher is doing behind the scenes. The results of the research are compromised because they could be due to the pre-test and not due to treatment administered by the researcher.
Delegate your assignment to our experts and they will do the rest.
Control measures are therefore necessary to eliminate repeated testing as an internal validity factor. The control measures that counter testing include:
Obtaining multiple groups and utilising between-subject design instead of within-subject design among the test subjects (Use of Control groups).
Not giving the same tests to the same people numerous times.
Increasing the time gap between initial tests and subsequent tests with the hope that respondents will forget the initial tests
The control measures for some common research designs are shown in the table below:
| Research Design |
Control Measures Against Testing |
||
| Experimental | Post-test Only Control group | Pre-post-test Control group | Solomon Four group design |
| Quasi-experimental | Time series | Non-equivalent Control group | Separate sample pre-test/post-test |
| Pre-experimental designs for Description | Static group comparison | ||
References
Marczyk, G., DeMatteo, D., & Festinger, D. (2005). Essentials of research design and methodology . John Wiley & Sons Inc.
Cook, T. D., & Campbell, D. T. (1979). The design and conduct of true experiments and quasi-experiments in field settings. In Reproduced in part in Research in Organizations: Issues and Controversies . Goodyear Publishing Company.
Kazdin, A. E. (2003). Research design in clinical psychology.
Pedhazur, E. J. (86). Schmelkin. LP (1991). Measurement, design, and analysis: An integrated approach. Hillsdale, N]: Erlbaum .
