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

Are ordinal models useful for classification?

M K Campbell1, A Donner, K M Webster

  • 1Department of Epidemiology & Biostatistics, Lawson Research Institute, University of Western Ontario, London, Canada.

Statistics in Medicine
|March 1, 1991
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

Patient information leaflets for placebo-controlled surgical trials: a review of current practice and recommendations for developers.

Trials·2024
Same author

Optimizing the design of invasive placebo interventions in randomized controlled trials.

The British journal of surgery·2020
Same author

Self-Diffusion in Amorphous Silicon by Local Bond Rearrangements.

Physical review letters·2018
Same author

Inequitable health service use in a Canadian paediatric population: A cross-sectional study of individual- and contextual-level factors.

Child: care, health and development·2017
Same author

Effectiveness of open and arthroscopic rotator cuff repair (UKUFF): a randomised controlled trial.

The bone & joint journal·2017
Same author

Maternal, fetal, and placental conditions associated with medically indicated late preterm and early term delivery: a retrospective study.

BJOG : an international journal of obstetrics and gynaecology·2015
Same journal

Interpretable Bayesian Modeling for Multireader Multicase Studies: Addressing Overdispersion and Limited Sample Size in Diagnostic Enhancement Evaluation.

Statistics in medicine·2026
Same journal

Adaptive Sequential Multiple Hypotheses Testing for Concomitant Vaccine Safety Surveillance.

Statistics in medicine·2026
Same journal

Novel Distance Regression for Repeated Outcomes With Missing Data: Applications to Longitudinal and Crossover Studies of Microbiome Beta-Diversity.

Statistics in medicine·2026
Same journal

Optimal Weighted Tests for Replication Studies and the 'Two-Trials Rule' With Multiple Hypotheses.

Statistics in medicine·2026
Same journal

Identifiable Copula-Double-Cox Models: A Fully Parametric Framework for Dependent Right-Censored Survival Data.

Statistics in medicine·2026
Same journal

Moving From Individualized Risk-Based Prevention to Benefit-Based Prevention: Estimating Individualized Life-Years Gained From Prevention Services as a Basis for Eligibility.

Statistics in medicine·2026
See all related articles

Ordinal models do not improve classification accuracy compared to standard multinomial logistic or normal discriminant analyses. This study suggests avoiding ordinal models when the primary goal is classification.

Area of Science:

  • Statistics
  • Biostatistics
  • Machine Learning

Background:

  • Ordinal response data is common in various fields.
  • Ordinal models are designed for such data, but their application requires careful consideration.
  • Misapplication of ordinal models can lead to inaccurate results.

Purpose of the Study:

  • To compare the classification accuracy of ordinal models against standard multinomial logistic and normal discriminant analyses.
  • To determine if ordinal models offer benefits in classification tasks.
  • To investigate the potential harms of inappropriate ordinal model application.

Main Methods:

  • A simulation study was conducted to evaluate classification performance.
  • Various statistical models, including ordinal, multinomial logistic, and normal discriminant analyses, were compared.

Related Experiment Videos

  • Classification accuracy was the primary metric for comparison.

    • Main Results:

    • Ordinal models demonstrated lower classification accuracy than multinomial logistic and normal discriminant procedures across various scenarios.
    • No significant advantage was found for ordinal models in classification tasks.
    • The study highlighted potential pitfalls in assuming ordinality.

    Conclusions:

    • Ordinal models do not confer an advantage for classification tasks compared to established methods.
    • Further research is needed, but current evidence suggests caution in using ordinal models solely for classification.
    • Multinomial logistic and normal discriminant analyses remain robust choices for classification problems, even with ordinal data.