Beyond “Before versus After”: Matching the Analysis to the Longitudinal Research Question

Pre-post and longitudinal studies are common, but selecting the right analysis is not always straightforward. Should we analyse change scores, adjust for baseline, use a paired t-test, ANCOVA, or a mixed-effects model? This seminar will provide a practical framework for matching the analysis to the study design, research question and estimand. Using applied examples, it will compare common approaches, explain why within-group change is not the same as a treatment effect, and discuss issues such as multiple timepoints, missing data, regression to the mean, and raw versus adjusted results. The seminar will conclude with a decision framework to help researchers choose an analysis that answers the intended scientific question and supports an appropriate interpretation.