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

Scaling regression inputs by dividing by two standard deviations.

Andrew Gelman1

  • 1Department of Statistics, Columbia University, New York, NY, USA. gelman@stat.columbia.edu

Statistics in Medicine
|October 26, 2007
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

FDA Draft Guidance for the Use of Bayesian Methods in Clinical Trials.

JAMA·2026
Same author

Pathfinder: Parallel quasi-Newton variational inference.

Journal of machine learning research : JMLR·2026
Same author

Multilevel regression and poststratification interface: an application to track community-level COVID-19 viral transmission.

Population health metrics·2026
Same author

A multilevel Bayesian approach to climate-fueled migration and conflict.

Scientific reports·2025
Same author

Meta-analysis with a single study.

Statistical methods in medical research·2025
Same author

Simulation-Based Calibration Checking for Bayesian Computation: The Choice of Test Quantities Shapes Sensitivity.

Bayesian analysis·2025
Same journal

A Causal Framework for Evaluating the Total Effect of Strategies Aiming to Expand Screening and to Improve Outcomes.

Statistics in medicine·2026
Same journal

Causal Effects on Nonterminal Event Time With Application to Antibiotic Usage and Future Resistance.

Statistics in medicine·2026
Same journal

Subgroup Analysis of Interval-censored Failure Time Data With Application to Alzheimer's Disease.

Statistics in medicine·2026
Same journal

Rejoinder to Commentaries on "A Perspective on the Appropriate Implementation of ICH E9(R1) Addendum Strategies for Handling Intercurrent Events".

Statistics in medicine·2026
Same journal

A Multi-Stage Drop-the-Loser Design With Superiority Boundaries.

Statistics in medicine·2026
Same journal

Interpretable ROI Identification in Brain Image Analysis: Overcoming CNN Black Box Challenges With Kriging-Enhanced Adaptive Sampling.

Statistics in medicine·2026
See all related articles

This study introduces a new method for interpreting regression coefficients by rescaling numeric variables using two times their standard deviation. This approach allows for direct comparison of coefficients, improving statistical practice in data analysis.

Area of Science:

  • Statistics
  • Data Analysis
  • Regression Modeling

Background:

  • Regression coefficient interpretation is challenging due to varying input variable scales.
  • Current methods like dividing by standard deviation offer partial solutions.
  • A standardized approach is needed for robust coefficient comparison.

Purpose of the Study:

  • To propose a novel rescaling method for numeric variables in regression analysis.
  • To enable direct comparison of regression coefficients across different predictors.
  • To enhance the routine practice of statistical modeling.

Main Methods:

  • Rescaling numeric variables by dividing them by two times their standard deviation.
  • Implementing the rescaling procedure as a function in the R statistical programming language.

Related Experiment Videos

  • Applying the method to linear and multilevel logistic regression models.

    • Main Results:

    • The proposed rescaling method yields coefficients directly comparable to untransformed binary predictors.
    • Demonstrated effectiveness in analyzing National Election Study data and NYC rodent prevalence data.
    • Facilitates routine comparison of coefficient magnitudes in applied modeling.

    Conclusions:

    • The proposed rescaling method offers an improvement over standard practices.
    • Recommends adopting this rescaling as a default option for routine statistical analysis.
    • Enhances the interpretability and comparability of regression model results.