Mean centering, multicollinearity, and moderators in multiple regression: The reconciliation redux

Dawn Iacobucci; Matthew J Schneider; Deidre L Popovich; Georgios A Bakamitsos

doi:10.3758/s13428-016-0827-9

Mean centering, multicollinearity, and moderators in multiple regression: The reconciliation redux

Behav Res Methods. 2017 Feb;49(1):403-404. doi: 10.3758/s13428-016-0827-9.

Authors

Dawn Iacobucci¹, Matthew J Schneider², Deidre L Popovich³, Georgios A Bakamitsos⁴

Affiliations

¹ Vanderbilt University, Nashville, TN, USA. dawn.iacobucci@owen.vanderbilt.edu.
² Decision Sciences, Drexel University, Philadelphia, PA, USA.
³ Marketing, Texas Tech University, Lubbock, TX, USA.
⁴ Marketing, Stetson University, DeLand, FL, USA.

PMID: 27800581
DOI: 10.3758/s13428-016-0827-9

Abstract

In this article, we attempt to clarify our statements regarding the effects of mean centering. In a multiple regression with predictors A, B, and A × B (where A × B serves as an interaction term), mean centering A and B prior to computing the product term can clarify the regression coefficients (which is good) and the overall model fit R² will remain undisturbed (which is also good).

Keywords: Mean center; Moderated regression.

MeSH terms

Effect Modifier, Epidemiologic
Humans
Models, Theoretical
Multivariate Analysis*