Extending the reach of the common cause design using meta-analytic methods: Applications and issues

Most meta-analytic methods examine effects across a collection of primary studies. We introduce an application of meta-analytic techniques to estimate effects and homogeneity within a single, primary study consisting of multiple, pretest-intervention-posttest units. This novel assessment was used to validate the recently created “Common Cause” (CC) design. In each case, we established the CC design by eliminating control groups from randomized studies, thereby deconstructing each experiment. This deconstruction enabled us to compare difference-in-difference results in randomized designs with a control group to pretest-posttest differences in a CC design without a control group. Meta-analysis results of multiple OXO effects from the CC designs were compared to meta-analytic effects of multiple randomized studies. This within-study-comparison logic and associated analyses produced consistent similarity between CC and validating-study results when directions of findings and patterns of statistical significance were considered. We provide plausible explanations for varying CC effect-size estimates, describe strengths and limitations, and address future research directions.