Unscripted E2: Moving From Multiple Regression to the Multilevel Model

Dan and Patrick provide a brief review of the traditional multiple regression model and highlight the important assumption that the residuals are independent. This assumption implies that no two residuals are related to one another, a condition that is almost always violated when data are hierarchically nested (e.g., children are nested within schools) or consist of repeated measures (e.g., time is nested within individual). Ignoring residual dependence can bias test statistics and unnecessarily limit the types of research questions that can be empirically assessed. However, the multilevel (or mixed effects) model explicitly allows for nested data structures and the inclusion of predictors at each level of the data (e.g., child-level and school-level predictors). In this episode, Dan and Patrick discuss the basics of the multilevel model, the advantages this approach provides, and potential directions for learning more about this analytic framework. Please visit centerstat.org for additional freely-available instructional materials and other training opportunities. You can also sign up for notifications about future Unscripted episodes at centerstat.org/centerstat-unscripted/