Misleading Heuristics and Moderated Multiple Regression Models

Moderated multiple regression models allow the simple relationship between the dependent variable and an independent variable to depend on the level of another independent variable. The moderated relationship, often referred to as the interaction, is modeled by including a product term as an additional independent variable. Moderated relationships are central to marketing (e.g., Does the effect of promotion on sales depend on the market segment?). Multiple regression models not including a product term are widely used and well understood. The authors argue that researchers have derived from this simpler type of multiple regression several data analysis heuristics that, when inappropriately generalized to moderated multiple regression, can result in faulty interpretations of model coefficients and incorrect statistical analyses. Using theoretical arguments and constructed data sets, the authors describe these heuristics, discuss how they may easily be misapplied, and suggest some good practices for estimating, testing, and interpreting regression models that include moderated relationships.

[1]  France Leclerc,et al.  Can Advertising Copy Make FSI Coupons More Effective ? , 2001 .

[2]  D. Soman The Illusion of Delayed Incentives: Evaluating Future Effort–Money Transactions , 1998 .

[3]  J. Little,et al.  Can Advertising Copy Make FSI Coupons More Effective? , 1997 .

[4]  Gary H. McClelland,et al.  Optimal design in psychological research. , 1997 .

[5]  Herman Aguinis,et al.  Type II Error Problems in the Use of Moderated Multiple Regression for the Detection of Moderating Effects of Dichotomous Variables , 1994 .

[6]  Larry E. Toothaker,et al.  Multiple Regression: Testing and Interpreting Interactions , 1991 .

[7]  C. Judd,et al.  Statistical difficulties of detecting interactions and moderator effects. , 1993, Psychological bulletin.

[8]  C. Coulton,et al.  Interaction Effects in Multiple Regression , 1993 .

[9]  S. Maxwell,et al.  Bivariate median splits and spurious statistical significance. , 1993 .

[10]  G. Vining,et al.  Data Analysis: A Model-Comparison Approach , 1989 .

[11]  L. Cronbach Statistical tests for moderator variables: flaws in analyses recently proposed , 1987 .

[12]  John G. Lynch Uniqueness Issues in the Decompositional Modeling of Multiattribute Overall Evaluations: An Information Integration Perspective , 1985 .

[13]  Jacob Cohen The Cost of Dichotomization , 1983 .

[14]  Subhash Sharma,et al.  Identification and Analysis of Moderator Variables , 1981 .

[15]  H. V. Henderson,et al.  Building Multiple Regression Models Interactively , 1981 .

[16]  Jacob Cohen Partialed products are interactions; partialed powers are curve components. , 1978 .

[17]  Morris B. Holbrook,et al.  Comparing Multiattribute Attitude Models by Optimal Scaling , 1977 .

[18]  A. Goldberger,et al.  On the Exact Covariance of Products of Random Variables , 1969 .