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

An SAS/IML procedure for maximum likelihood factor analysis.

Rusan Chen1

  • 1Center for New Designs in Learning and Scholarship, Georgetown University, Washington, DC 20057, USA. chenrs@georgetown.edu

Behavior Research Methods, Instruments, & Computers : a Journal of the Psychonomic Society, Inc
|July 2, 2003
PubMed
Summary

This study presents a SAS/IML program for Maximum Likelihood Factor Analysis (MLFA), offering a transparent alternative to "black box" statistical software for researchers. It aids in understanding MLFA computations and extends to structural equation modeling.

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

Maternal tobacco use: A third-trimester risk factor for small-for-gestational-age pregnancy outcome.

Preventive medicine reports·2020
Same author

Corrigendum to 'Dysfunctional personality disorder beliefs and lifetime suicide attempts among psychiatrically hospitalized military personnel' comprehensive psychiatry 82 (2018) 108-114.

Comprehensive psychiatry·2019
Same author

Dysfunctional personality disorder beliefs and lifetime suicide attempts among psychiatrically hospitalized military personnel.

Comprehensive psychiatry·2018
Same author

The Role of Authoritative Parenting in Adolescent Type 1 Diabetes Management.

Journal of pediatric psychology·2017
Same author

Small-for-gestational age prevalence risk factors in central Appalachian states with mountain-top mining.

International journal of occupational medicine and environmental health·2017
Same author

Anger, hostility, and hospitalizations in patients with heart failure.

Health psychology : official journal of the Division of Health Psychology, American Psychological Association·2017

Area of Science:

  • Multivariate Statistics
  • Psychometrics
  • Computational Statistics

Background:

  • Maximum Likelihood Factor Analysis (MLFA) has a strong mathematical basis, enabling hypothesis testing under normality assumptions with large datasets.
  • Its popularity grew with Jöreskog's 1967 iterative parameter estimation algorithm.
  • Existing MLFA software often functions as opaque 'black boxes', limiting user understanding of computational processes.

Purpose of the Study:

  • To present a concise program for obtaining MLFA solutions using SAS/IML and the NLPQN optimization subroutine.
  • To provide a pedagogical tool that illustrates the step-by-step computational processes of MLFA.
  • To demonstrate the extensibility of this programming approach to other multivariate methods requiring numerical optimization.

Main Methods:

Related Experiment Videos

  • Development of a SAS/IML program incorporating the NLPQN optimization subroutine.
  • Utilizing matrix language for clear, step-by-step computational process illustration.
  • Demonstrating the program's application and extension to related multivariate techniques.

Main Results:

  • A functional SAS/IML program for MLFA was successfully developed.
  • The program effectively demystifies the computational steps involved in MLFA.
  • The approach was shown to be adaptable for other numerical optimization-based multivariate methods.

Conclusions:

  • The developed SAS/IML program offers a transparent and educational approach to MLFA.
  • This method provides valuable insights into the computational underpinnings of MLFA.
  • Researchers can leverage this program for Monte Carlo simulations and understanding complex multivariate analyses like structural equation modeling.