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

Using binary logistic regression models for ordinal data with non-proportional odds

R Bender1, U Grouven

  • 1Department of Metabolic Diseases and Nutrition, Heinrich-Heine-University of Düsseldorf, Germany.

Journal of Clinical Epidemiology
|October 8, 1998
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

Hyperbaric oxygen therapy for iatrogenic arterial gas embolism after CT-guided lung biopsy : A case report.

Der Anaesthesist·2019
Same author

Response to Letter to the Editor: 'Repigmentation or stimulated skin physiology? Medical needling in combination with non-cultured skin cell transplantation-The way of the melanocyte'.

Burns : journal of the International Society for Burn Injuries·2017
Same author

Tissue specific loss of proliferative capacity of parthenogenetic cells in fetal mouse chimeras.

Roux's archives of developmental biology : the official organ of the EDBO·2017
Same author

Strongly baryon-dominated disk galaxies at the peak of galaxy formation ten billion years ago.

Nature·2017
Same author

Combination of medical needling and non-cultured autologous skin cell transplantation (renovacell) for repigmentation of hypopigmented burn scars in children and young people.

Annals of burns and fire disasters·2017
Same author

[Autologous Skin Cell Transplantation and Medical Needling for Repigmentation of Depigmented Burn Scars on UV-protected and UV-exposed Skin].

Handchirurgie, Mikrochirurgie, plastische Chirurgie : Organ der Deutschsprachigen Arbeitsgemeinschaft fur Handchirurgie : Organ der Deutschsprachigen Arbeitsgemeinschaft fur Mikrochirurgie der Peripheren Nerven und Gefasse : Organ der V...·2016
Same journal

A comparison of five statistical methods used to analyse longitudinal EORTC QLQ-C30 quality of life scores in randomised controlled trials: a simulation study.

Journal of clinical epidemiology·2026
Same journal

Sample Size Determination for Decision-centered Pragmatic Trials.

Journal of clinical epidemiology·2026
Same journal

Many multicenter randomized controlled trials do not account for center effect: a methodological review.

Journal of clinical epidemiology·2026
Same journal

Patient Acceptability of the Modified Zelen Approach to Randomized Trials - A Survey of the CAPS THA Cohort.

Journal of clinical epidemiology·2026
Same journal

Corrigendum to SPICE-GRADE: simultaneous processing of indirect causal evidence in complex pathways using GRADE - an exploratory case study. [Journal of Clinical Epidemiology, 194C (2026) 112219].

Journal of clinical epidemiology·2026
Same journal

Small numbers of clusters in cluster randomised trials: a scoping review of problems and proposed solutions.

Journal of clinical epidemiology·2026
See all related articles

The proportional odds model (POM) can yield invalid results when its core assumption is violated. Analyzing ordinal data with non-proportional odds using separate binary logistic regression models offers a simpler, adequate alternative.

Area of Science:

  • Statistics
  • Biostatistics
  • Epidemiology

Background:

  • The proportional odds model (POM) is widely used for ordinal logistic regression.
  • Violation of the POM's proportional odds assumption can compromise analysis validity.
  • Diabetic retinopathy and smoking data serve as a case study for model assumption violations.

Purpose of the Study:

  • To evaluate the impact of violating the proportional odds assumption in logistic regression.
  • To compare alternative methods for analyzing ordinal data when the POM assumption is not met.
  • To assess the utility of separate binary logistic regression models for ordinal data with non-proportional odds.

Main Methods:

  • Application of the proportional odds model (POM) to ordinal response data.

Related Experiment Videos

  • Dichotomization of the ordinal response variable for separate binary logistic regression analyses.
  • Comparison with polytomous logistic regression and the partial proportional odds model.

    • Main Results:

    • The proportional odds assumption was violated in the study data.
    • Separate binary logistic regression models provided a viable alternative to POM.
    • While slightly less efficient, binary models offer easier model building, goodness-of-fit assessment, and interpretation.

    Conclusions:

    • Careful application of separate binary logistic regression is a simple and adequate tool for ordinal data with non-proportional odds.
    • This approach facilitates analysis when the proportional odds assumption is not fulfilled.
    • The findings highlight the importance of checking model assumptions in ordinal logistic regression.