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

Normal theory based test statistics in structural equation modelling

K H Yuan1, P M Bentler

  • 1Department of Psychology, University of California at Los Angeles 90095-1563, USA.

The British Journal of Mathematical and Statistical Psychology
|December 17, 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

['Dou Bi' () and 'Xiao Shi' () -- the thinking and methods in medical cases related terms].

Zhonghua yi shi za zhi (Beijing, China : 1980)·2025
Same author

The Relationship Between Systemic Lupus Erythematosus and Osteoporosis Based on Different Ethnic Groups: a Two-Sample Mendelian Randomization Analysis.

Calcified tissue international·2024
Same author

[Palliative radiotherapy for AIDS related diffuse large B cell lymphoma: a report of six cases with literature review].

Zhonghua xue ye xue za zhi = Zhonghua xueyexue zazhi·2017
Same author

Invariant Standardized Estimated Parameter Change for Model Modification in Covariance Structure Analysis.

Multivariate behavioral research·2016
Same author

A NEW MATRIX FOR THE ASSESSMENT OF FACTOR CONTRIBUTIONS.

Multivariate behavioral research·2016
Same author

Brief Report: An Additional Minimal Transformation To Orthonormality.

Multivariate behavioral research·2016

Psychology research often uses statistics assuming normal data, but real data is rarely normal. This study introduces new asymptotically distribution-free (ADF) test statistics for better analysis of covariance structure models with non-normal data.

Area of Science:

  • Psychometrics
  • Statistical modeling
  • Psychology

Background:

  • Covariance structure models in psychology often rely on the assumption of multivariate normality.
  • This assumption is frequently violated in real-world psychological data sets.
  • Consequently, conclusions drawn from normal theory methods may be unreliable.

Purpose of the Study:

  • To develop and evaluate new test statistics for covariance structure models that do not assume multivariate normality.
  • To address the limitations of traditional normal theory methods when applied to non-normal psychological data.
  • To provide robust statistical tools for analyzing psychological data.

Main Methods:

  • Development of three new asymptotically distribution-free (ADF) test statistics.

Related Experiment Videos

  • Utilizing Monte Carlo simulations to assess the finite sample behavior of the proposed ADF statistics.
  • Analysis of the sensitivity of ADF test statistics to model degrees of freedom and complexity.

    • Main Results:

    • An ADF test statistic demonstrated good performance in finite sample situations, outperforming traditional methods.
    • The study found that ADF test statistics are more sensitive to model degrees of freedom than to model complexity.
    • A novel index was proposed for assessing the robustness of rescaled statistics.

    Conclusions:

    • The proposed ADF test statistics offer a more appropriate and reliable approach for analyzing non-normal data in covariance structure models.
    • Recommendations are provided for the practical application of these new test statistics in psychological research.
    • The findings highlight the importance of considering data distribution in statistical inference for psychological studies.