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

Methods for improving regression analysis for skewed continuous or counted responses.

Abdelmonem A Afifi1, Jenny B Kotlerman, Susan L Ettner

  • 1School of Public Health, University of California-Los Angeles, CA 90095-1772, USA. afifi@ucla.edu

Annual Review of Public Health
|November 23, 2006
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

Effect of State Medicaid Reimbursement Policies on Telehealth Visits in Community-Based Health Centers.

Telemedicine journal and e-health : the official journal of the American Telemedicine Association·2026
Same author

Long COVID Disproportionately Reported by Disadvantaged Individuals: A National Survey of U.S. Working-Age Adults.

Journal of general internal medicine·2025
Same author

Protocol: A mixed-methods study to evaluate implementation and outcomes of U.S. state telemental health policy expansion during the COVID-19 pandemic.

PloS one·2024
Same author

Implementation Of New Mexico's 'No Behavioral Health Cost Sharing' Law: A Qualitative Study.

Health affairs (Project Hope)·2024
Same author

How policymakers innovate around behavioral health: adoption of the New Mexico "No Behavioral Health Cost-Sharing" law.

Health affairs scholar·2024
Same author

A Metascore of Multiple Imaging Methods to Measure Long-Term Glaucoma Structural Progression.

Translational vision science & technology·2022

Standard regression analysis often fails in practice. Health researchers can improve statistical inference validity using methods like smear factor re-transformation, multiple imputation, and bootstrap techniques for continuous data, and specialized models for count data.

Area of Science:

  • Biostatistics
  • Health Services Research
  • Statistical Modeling

Background:

  • Standard regression analysis assumptions are frequently violated in real-world data.
  • Ensuring the validity of statistical inference requires specific adjustments.
  • These essential adjustments are not universally adopted by health researchers.

Purpose of the Study:

  • To review and demonstrate practical methods for valid statistical inference in regression analysis.
  • To illustrate the application of these methods in a health services study.
  • To highlight advancements in statistical software for improved prediction and inference.

Main Methods:

  • For continuous outcomes: re-transformation using the smear factor, multiple imputation for missing data, attrition weights, and bootstrap methods.

Related Experiment Videos

  • For count outcomes: zero-inflated Poisson and negative binomial models, and two-part models for excess zeros.
  • Application of these techniques in a health services research context.

    • Main Results:

    • Demonstrated effective adjustments for regression analysis with continuous and count outcomes.
    • Showcased how methods like smear factor, multiple imputation, and specialized count models address common data issues.
    • Highlighted the utility of bootstrap methods for robust inference.

    Conclusions:

    • Implementing advanced regression techniques is crucial for valid statistical inference in health research.
    • User-friendly software now facilitates the application of these complex methods.
    • Improved prediction and inference are achievable through these modern statistical approaches.