Key Guidelines and Demonstrations Using Dominance Analysis to Determine Predictors’ Relative Importance in Multilevel Models (Poster 32)

• In Event: AERA Poster Session Thursday 8:00 a.m.
  In Poster Session: Advancements in Educational Data Analysis: Machine Learning, Multilevel Models, and Beyond

Abstract

Various educational studies show that Dominance Analysis (DA) can provide useful information on the relative importance of predictors by comparing the change in model fit (i.e., R2). However, the DA procedure has not been studied extensively with Multilevel Linear Models (MLMs). The current study aimed to evaluate which multilevel R² measures produce the most accurate relative importance results to extend the DA application to diverse multilevel contexts. The study using a Monte Carlo simulation found that the choice of multilevel R² measures is important and provided recommendations on which measures enhance the accuracy of DA results under various multilevel model types and mean-centering methods. This study presented three demonstrations of how to apply the DA procedure using real-world datasets.

Authors

• Soonhwa Paek, University of Wisconsin - Milwaukee

• Razia Azen, University of Wisconsin - Milwaukee