Ambiguity surrounds any importance measure in cases in which predictor variables are correlated. However, a new measure is proposed that has attractive properties, such as providing individual contributions that are both non-negative and sum to R2. The new measure is compared with four other commonly used measures and its advantages over each of them are pointed out.
Get full access to this article
View all access options for this article.
References
1.
DarlingtonRichard B.“Multiple Regression in Psychological Research and Practice,”Psychological Bulletin, 69 (1968), 161–82.
2.
EngelJames F., FiorilloHenry F., and CayleyMurray A., eds. Market Segmentation: Concepts and Applications.New York: Holt, Rinehart and Winston, Inc., 1972.
3.
FrankRonald E., MassyWilliam F., and WindYoram, Market Segmentation.Englewood Cliffs, New Jersey: Prentice-Hall, Inc., 1972.
4.
GibsonW. A.“Orthogonal Predictors: A Possible Resolution of the Hoffman-Ward Controversy,”Psychological Reports, 11 (1962), 32–4.
5.
GreenPaul E., with contributions by, Douglas CarrollJ.Mathematical Tools for Applied Multivariate Analysis.New York: Academic Press, 1976.
6.
GreenPaul E., and WindYoram. Multiattribute Decisions in Marketing: A Measurement Approach.Hinsdale, Illinois: The Dryden Press, 1973.
7.
JohnsonRichard M.“The Minimal Transformation to Orthonormality,”Psychometrika, 31 (March 1966), 61–6.
