Some Common Myths About Centering Predictor Variables in Moderated Multiple Regression and Polynomial Regression

By Dev K. Dalal, Michael J. Zickar. 2012

460 citations recorded by the source corpus.

Subfields: Computational Theory and Mathematics, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/2ffee7d93bc90df6e51d794a85578c6c14ec35c0

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computational Theory and Mathematics

1. Felix D. Schönbrodt
2. Dan S. Chiaburu
3. Corinna S. Martarelli
4. W. Wolff
5. J. Schüler
6. Debra L. Shapiro
7. Ning Li
8. Maik Bieleke

In Statistics and Probability

1. Felix D. Schönbrodt
2. J. Dawson
3. Herman Aguinis
4. Kyle J. Bradley
5. Dan S. Chiaburu
6. J. Edwards
7. Corinna S. Martarelli
8. W. Wolff

Citations in the corpus

• The role of information asymmetry in the choice of entrepreneurial exit routes (2014, unknown): “…the squared and interaction terms, which could entail inflated standard errors and hence insignificance in hypotheses testing (Aguinis, 1995; Hair et al., 2006), we centered all of the variables by subtracting their respective means (Aiken andWest, 1991; Dalal and Zickar, 2012; Li and Tang, 2010).”
• An Introduction to Mixed Models for Experimental Psychology (2019, unknown): “A common way to deal with this is to center the continuous covariate such that the test of the lower order e ff ect is performed at its mean (e.g., Dalal & Zickar, 2012).”
• Testing Fit Patterns with Polynomial Regression Models (2015, background): “From an algebraic point of view, recentering or rescaling predictors in an unconstrained polynomial regression by additive shifts or multiplicative scaling factors is rather meaningless (J. Cohen, 1978; Dalal & Zickar, 2012).”
• Moderation in Management Research: What, Why, When, and How (2014, background): “In truth, with the exception of cases of extreme multicollinearity, the decision does not make any difference to the testing of the interaction term; the p value for the interaction term and the subsequent interaction plot should be identical whichever way it is done (Dalal and Zickar 2012; Kromrey ”
• What residualizing predictors in regression analyses does (and what it does not do) (2014, background): “Jaeger and Snider (2013) did it "since the two predictor variables were correlated" (p.”
• Revisiting How and When Perceived Organizational Support Enhances Taking Charge (2015, methodology): “To increase interpretability (Aguinis, 2004; Dalal & Zickar, 2012), we standardized the predictor variables and then computed the higher order terms (Aiken & West, 1991; Liao, Liu, & Loi, 2010; Marquardt, 1980).”
• High Boredom Proneness and Low Trait Self-Control Impair Adherence to Social Distancing Guidelines during the COVID-19 Pandemic (2020, methodology): “Continuous variables were mean centered to enhance interpretability and remove non-essential multi-collinearity [29].”
• Improving Our Understanding of Moderation and Mediation in Strategic Management Research (2017, background): “…the test for moderation does not involve XZ in its raw form, but rather the partialed XZ product, which is necessarily uncorrelated with X and Z (Cronbach, 1987; Dalal & Zickar, 2012), and the test of the partialed XZ product is unchanged regardless of how X and Z are rescaled (Cohen, 1978).”