Implications of Centering in a Three-Level Hierarchical Linear Model

• In Event: Multilevel Modeling Methods and Applications

Abstract

Hierarchical data in education are becoming increasingly complex, often involving more than two levels. This study investigated the implications of centering within context (CWC) and grand mean centering (CGM) in three distinct three-level models. The goals were to (1) determine equivalencies in the means and variances across the centering options, (2) identify the algebraic relationships between the three-level contextual models, and (3) clarify the interpretation of the parameters. Artificial datasets were used for illustration. Centering decisions in multilevel models are closely tied to substantive hypotheses and require researchers to be clear and cautious about their choices. This work is designed to assist the researcher in making centering decisions for analysis of three-level hierarchical data.

Authors

• Ahnalee Marie Brincks, University of Miami

• Daniel Feaster, University of Miami

• Maria Llabre, University of Miami

• Craig K. Enders, Arizona State University