Pretest-Posttest Control Group Design is a quantitative research design method. This research study design is a true experimental design in that there is a degree of randomization, use of a control group, and therefore greater internal validity.
This is one of the earliest true experimental study designs. References to when it was first used were difficult to find. The most general were from http://en.wikipedia.org/wiki/Timeline_of_the_history_of_scientific_method that referenced 1753 when a description of a controlled experiment using two identical populations with only one variable was put forth by James Lind's A Treatise of the Scurvy and then in 1926 when randomized design was established by Ronald Fisher in Statistical Methods for Research Workers and later in The Design of Experiments (1936).
This design is one of the best and most practical to assess the impact of an intervention or treatment on two randomized groups, one control and one treatment,
This design is an improvement on pre-experimental designs in that we can determine whether there is a change in behavior and outcomes after intervention and thus decrease the chances of confounding due to other factors. Thus, there is considerable confidence that any differences between intervention group and control group are due to the intervention. Although not used in this experimental design, other advantages include that the study design can expand to include more than one treatment, it can allow for many comparisons (i.e. between groups, pre- to post-intervention in one group), and that the statistical power can be increased by using the pretest measure as a covariate to statistically equate the groups (reference: http://mccoy.lib.siu.edu/projects/psyc/schmeck/vii2.ppt) .
The main shortcoming in this design, particularly as used in this experiment, is that the design assumes groups are equivalent due to random assignment. In this experiment the small N of the two groups may result in a Type II beta error, i.e. too small of a group that may result in a chance of missing an actual effect. Other disadvantages are that the treatment can become confounded with the pretest (i.e. internal validity), only one group receives treatment and both groups are post-tested (no placebo),
Examples in Informatics
- Guillame VM. Library-Sponsored Instruction Improves Core Informatics Competencies among Allied Health Students: A Research-Based Case Study. J Allied Health, 2005.
- Brown, S.J., Lieberman, D.A., Gemeny, B.A., Fan, Y.C., Wilson, D.M., & Pasta, D.J. Educational video game for juvenile diabetes: Results of a controlled trial. Medical Informatics 1997;22:77-89.
- Yom Y-H, Eun LK. Effects of a CD-ROM Educational Program on Sexual Knowledge and Attitude. Computers, Informatics, Nursing. 2005;23:214-219.
- Lusk SL, Ronis DL, Kazanis A, et al. Effectiveness of a Tailored Intervention to Increase Factory Workers' Use of Hearing Protection. Nursing Res. 2003;52:289-295.
- Goeppinger J, Arthur MW, Baglioni AJ Jr, Brunk SE, Brunner CM. A reexamination of the effectiveness of self-care education for persons with arthritis. Arthritis Rheum. 1989;32:706-16.
- Mohr DL, Clemmer DI. Evaluation of an occupational injury intervention in the petroleum drilling industry. Accid Anal Prev. 1989;21:263-71.