Jove
Visualize
Contact Us
JoVE
x logofacebook logolinkedin logoyoutube logo
ABOUT JoVE
OverviewLeadershipBlogJoVE Help Center
AUTHORS
Publishing ProcessEditorial BoardScope & PoliciesPeer ReviewFAQSubmit
LIBRARIANS
TestimonialsSubscriptionsAccessResourcesLibrary Advisory BoardFAQ
RESEARCH
JoVE JournalMethods CollectionsJoVE Encyclopedia of ExperimentsArchive
EDUCATION
JoVE CoreJoVE BusinessJoVE Science EducationJoVE Lab ManualFaculty Resource CenterFaculty Site
Terms & Conditions of Use
Privacy Policy
Policies

Related Experiment Videos

On Johnson's (2000) Relative Weights Method for Assessing Variable Importance: A Reanalysis.

D Roland Thomas1, Bruno D Zumbo2, Ernest Kwan1

  • 1a Sprott School of Business , Carleton University.

Multivariate Behavioral Research
|January 15, 2016
PubMed
Summary
This summary is machine-generated.

Related Concept Videos

You might also read

Related Articles

Articles linked to this work by shared authors, journal, and citation graph.

Sort by
Same author

Polychoric correlations under the assumption of elliptical latent traits.

The British journal of mathematical and statistical psychology·2026
Same author

Safer at home? Associations between working from home, workplace mistreatment, inclusion, and well-being in a sample of Canadian hybrid workers.

Work (Reading, Mass.)·2026
Same author

From Linear Geometry to Nonlinear and Information-Geometric Settings in Test Theory: Bregman Projections as a Unifying Framework.

Educational and psychological measurement·2025
Same author

Reliability as Projection in Operator-Theoretic Test Theory: Conditional Expectation, Hilbert Space Geometry, and Implications for Psychometric Practice.

Educational and psychological measurement·2025
Same author

Publisher Correction: A systematic review of latent class analysis in psychology: Examining the gap between guidelines and research practice.

Behavior research methods·2025
Same author

Examining the Between- and Within-Person Associations Between Psychological Stress and Physical Activity: A Random Intercept Cross-Lagged Panel Model.

Stress and health : journal of the International Society for the Investigation of Stress·2025

The relative weights method for assessing variable importance in multiple regression is theoretically flawed and should not be used. This analysis shows it can lead to distorted inferences compared to other metrics.

Area of Science:

  • Statistics
  • Econometrics
  • Psychometrics

Background:

  • The "relative weights" method by J. W. Johnson (2000) is a popular approach for determining variable importance in multiple regression analysis.
  • This method is based on the discredited technique proposed by Green, Carroll, and DeSarbo (1978).

Purpose of the Study:

  • To reanalyze the theoretical underpinnings and practical implications of J. W. Johnson's (2000) relative weights method.
  • To evaluate the validity and reliability of the relative weights method in assessing variable importance.

Main Methods:

  • A theoretical reanalysis of the derivation of the relative weights method.
  • Empirical evaluation using two illustrative examples.
  • Comparison with an alternative importance metric, general dominance (Azen & Budescu, 2003; Budescu, 1993).

Related Experiment Videos

Main Results:

  • The theoretical derivation of the relative weights method is found to be flawed.
  • The method's validity is comparable to the discredited Green, Carroll, and DeSarbo (1978) approach.
  • Application of the method can yield materially distorted inferences, especially when contrasted with general dominance metrics.

Conclusions:

  • The J. W. Johnson (2000) relative weights method is not a valid metric for variable importance in multiple linear regression.
  • Researchers should refrain from using the relative weights method due to its theoretical flaws and potential for misleading results.
  • Alternative methods, such as general dominance, should be considered for reliable variable importance assessment.