Embedding Reproducibility and Replicability in a Regression Modelling Assessment

Across different fields, researchers have raised concerns with the reproducibility of results in published articles and how to meaningfully interpret a replication study. Developing core skills in reproducibility and replicability has been identified as a gap in statistics education, particularly before students conduct an independent research project. The assessment I describe in this article uses the concepts of reproducibility and replicability as a medium for students to demonstrate their ability to apply regression modelling techniques. Students analyse open data from an original study to reproduce their results, analyse data from a direct replication of the original study, and argue whether they consider the results to replicate. After outlining the graduate-level statistics and research design course I use the assessment in, I provide an example assessment aimed at psychology and neuroscience students. I reflect on my experiences using the assessment before suggesting how educators can adapt the approach for their students.