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

Discriminant analysis and the selection of patients for controlled drinking programs: a methodological note

A C Ogborne, H M Annis, W R Miller

    Journal of Clinical Psychology
    |January 1, 1982
    PubMed
    Summary

    Discriminant function analysis can improve clinical predictions by considering individual membership probabilities, not just overall accuracy. This approach helps minimize diagnostic errors for better patient outcomes in problem drinking treatment.

    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

    Trends in the distribution of breast cancer over time in the southeast of Scotland and review of the literature.

    Ecancermedicalscience·2014
    Same author

    Changes in the microbiological and shelf-life characteristics of beef tongues and livers following transcontinental and transoceanic shipment.

    Meat science·2011
    Same author

    VASOMOTOR REACTIONS IN THE HYPNOTIC STATE.

    Journal of neurology and psychiatry·2011
    Same author

    Growth-control of human ovarian-carcinoma cells by insulin-like growth-factors.

    Oncology reports·2011
    Same author

    Molecular endocrinology and breast cancer--a reed before the wind lives on (a tribute to Professor Mike Reed).

    Steroids·2011
    Same author

    Present trends in psychiatry.

    Journal. Iowa State Medical Society·2010

    Area of Science:

    • Psychology
    • Statistics
    • Clinical Diagnostics

    Background:

    • The standard criterion group classification in discriminant function analysis prioritizes maximizing correctly classified cases.
    • This method may overlook clinical utility by focusing solely on overall accuracy.
    • Minimizing diagnostic errors and enhancing individual predictions are crucial for clinical application.

    Purpose of the Study:

    • To evaluate the clinical utility of discriminant function analysis beyond optimizing correct classification rates.
    • To demonstrate how considering individual membership probabilities can improve diagnostic accuracy and clinical decision-making.
    • To illustrate an enhanced approach for classifying individuals in clinical studies, specifically in problem drinking research.

    Main Methods:

    Related Experiment Videos

  • Utilized discriminant function analysis to classify individuals into distinct groups based on membership probabilities.
  • Employed a classification scheme that considers probabilities for each criterion group of interest, not just the highest probability.
  • Applied the methodology to data from a study on problem drinkers undergoing treatment, categorizing them into abstinent, controlled drinking, or continued heavy drinking groups.

    • Main Results:

    • The study illustrated that focusing solely on the highest membership probability can underestimate the clinical utility of discriminant analysis.
    • Considering probabilities across all relevant groups allows for a more nuanced understanding of individual case classification.
    • The approach demonstrated practical application in identifying characteristics associated with different drinking outcomes post-treatment.

    Conclusions:

    • Clinical utility of discriminant function analysis is enhanced by considering individual membership probabilities for each criterion group.
    • This method aids in minimizing serious diagnostic errors and improving predictive accuracy for individual patients.
    • The findings support a more refined application of discriminant analysis in clinical settings, particularly for treatment outcome prediction.