Conditions for causality
A cause must precede its effect, the variables must covary, and rival explanations must be ruled out. Experiments use control groups and random assignment to meet the third condition, a core idea of quantitative research.
Pre-experimental designs
- One-shot case study: X O.
- One-group pretest-posttest: O X O; cannot rule out history or maturation.
- Static-group comparison: posttest-only comparison without random assignment.
True experimental designs
- Pretest-posttest control group: R O X O / R O O.
- Posttest-only control group: R X O / R O, avoiding testing effects.
- Solomon four-group: combines both to separate pretest and treatment effects.
- Factorial designs test several factors and their interactions.
Quasi-experimental designs
Used when randomisation is impractical: nonequivalent control group (O X O / O O), interrupted time series (O O O X O O O) and control-series designs, common in programme evaluation.
Validity
Stronger designs better control threats to internal validity such as history, maturation, testing, instrumentation, regression, selection and attrition, though artificial settings can reduce external validity. Human experiments also need ethical review.
References
- Campbell, D. T., & Stanley, J. C. (1963). Experimental and Quasi-Experimental Designs for Research. Rand McNally.
- Shadish, W. R., Cook, T. D., & Campbell, D. T. (2002). Experimental and Quasi-Experimental Designs for Generalized Causal Inference. Houghton Mifflin.
- Rubin, A., & Babbie, E. R. (2017). Research Methods for Social Work (9th ed.). Cengage.