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

Multivariate analysis: the need for data, and other problems.

B S Everitt

    The British Journal of Psychiatry : the Journal of Mental Science
    |March 1, 1975
    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

    A Monte Carlo Investigation Of The Likelihood Ratio Test For The Number Of Components In A Mixture Of Normal Distributions.

    Multivariate behavioral research·2016
    Same author

    A Note on Parameter Estimation for Lazarsfeld's Latent Class Model using the EM Algorithm.

    Multivariate behavioral research·2016
    Same author

    A Monte Carlo Investigation of the Likelihood Ratio Test for Number of Classes in Latent Class Analysis.

    Multivariate behavioral research·2016
    Same author

    Analysis of drop-out data in treatment trials.

    The British journal of psychiatry : the journal of mental science·1999
    Same author

    Mixture model mapping of the brain activation in functional magnetic resonance images.

    Human brain mapping·1999
    Same author

    The use of multivariate statistical methods in psychiatry.

    Statistical methods in medical research·1998
    Same journal

    The pressurised leaky funnel: rethinking recruitment, selection and retention in the UK psychiatry workforce.

    The British journal of psychiatry : the journal of mental science·2026
    Same journal

    Cutting through stigma: psychiatry and neurosurgery working together.

    The British journal of psychiatry : the journal of mental science·2026
    Same journal

    A fourth pillar for evidence-based medicine: implications for psychiatry - CORRIGENDUM.

    The British journal of psychiatry : the journal of mental science·2026
    Same journal

    Understanding negative perceptions of psychiatrists on social media: lessons from public discourse and professional self-reflection.

    The British journal of psychiatry : the journal of mental science·2026
    Same journal

    Attachment-informed psychopharmacology in psychiatric care.

    The British journal of psychiatry : the journal of mental science·2026
    Same journal

    Acceptability and accuracy of point-of-care monitoring of lithium levels.

    The British journal of psychiatry : the journal of mental science·2026
    See all related articles

    Multivariate analysis requires careful data screening and critical thinking, not just statistical software. Misuse, particularly in psychology, can lead to misleading results, emphasizing the need for data suitability checks.

    Area of Science:

    • Statistics
    • Data Analysis
    • Psychological Research

    Background:

    • Multivariate analyses are valuable tools but require careful application.
    • The ease of access to statistical software can lead to misuse.
    • Journal editorial policies sometimes encourage inappropriate statistical methods.

    Purpose of the Study:

    • To highlight the importance of critical thinking in multivariate data analysis.
    • To address the potential for misleading results due to improper application of statistical methods.
    • To advocate for better data screening practices before employing complex analyses.

    Main Methods:

    • Discussion of the prerequisites for meaningful results in multivariate analysis, including sample size, variable type, and distribution.
    • Critique of the over-reliance on statistical software and significance levels.

    Related Experiment Videos

  • Emphasis on the utility of simple graphical techniques for initial data screening.

    • Main Results:

    • Improper use of multivariate methods, unlike univariate methods, can yield seriously misleading results.
    • Subjectivity in interpretation can only be overcome through replication.
    • Many researchers bypass expert statisticians due to software accessibility.

    Conclusions:

    • Researchers should prioritize data screening using graphical techniques to ensure suitability for multivariate analysis.
    • Critical thinking and careful consideration of data characteristics are essential for valid statistical outcomes.
    • Journal editors should encourage robust data analysis practices over mere inclusion of complex statistics.