Misleading heuristics and moderated multiple regression models

Citation
Jr. Irwin et Gh. Mcclelland, Misleading heuristics and moderated multiple regression models, J MARKET C, 38(1), 2001, pp. 100-109
Citations number
18
Categorie Soggetti
Economics
Journal title
JOURNAL OF MARKETING RESEARCH
ISSN journal
00222437 → ACNP
Volume
38
Issue
1
Year of publication
2001
Pages
100 - 109
Database
ISI
SICI code
0022-2437(200102)38:1<100:MHAMMR>2.0.ZU;2-8
Abstract
Moderated multiple regression models allow the simple relationship between the dependent variable and an independent variable to depend on the level o f another independent variable. The moderated relationship, often referred to as the interaction, is modeled by including a product term as an additio nal 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 a nd well understood. The authors argue that researchers have derived from th is simpler type of multiple regression several data analysis heuristics tha t, when inappropriately generalized to moderated multiple regression, can r esult in faulty interpretations of model coefficients and incorrect statist ical analyses, Using theoretical arguments and constructed data sets, the a uthors 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.