Related Experiment Videos
Analytic methods for questions pertaining to a randomized pretest, posttest, follow-up design
Joseph R Rausch1, Scott E Maxwell, Ken Kelley
1Department of Psychology, University of Notre Dame, Notre Dame, IN 46556, USA. jrausch@nd.edu
Summary
This study explores analyzing group differences in randomized pretest, posttest, follow-up (PPF) designs. It recommends using the pretest as a covariate for accurate analysis, with posttest and follow-up usage depending on research questions.
Area of Science:
- Statistics
- Research Methodology
Background:
- Randomized pretest, posttest, follow-up (PPF) designs are common in research.
- Analyzing group differences within these designs requires careful methodological consideration.
Purpose of the Study:
- To delineate key questions regarding group differences in PPF designs.
- To compare statistical methods for analyzing these differences.
- To provide guidance on optimal use of pretest, posttest, and follow-up data.
Main Methods:
- Comparative analysis of Analysis of Variance (ANOVA) and Analysis of Covariance (ANCOVA).
- Discussion of Hierarchical Linear Modeling (HLM) for PPF designs.
- Methodological examination of 5 common research questions.
Main Results:
- The pretest is most effectively used as a covariate in statistical models.
- The utilization of posttest and follow-up data depends on the specific research question.
- ANOVA and ANCOVA offer different approaches to analyzing group differences.
Conclusions:
- Properly utilizing the pretest as a covariate enhances the analysis of group differences.
- Researchers must align their statistical approach with their specific research questions for PPF designs.
- Methodological choices significantly impact the interpretation of group differences in longitudinal studies.