Full-Information Item Factor Analysis. Revision.

Abstract

A method of item factor analysis based on Thurstone's multiple factor model and implemented by marginal maximum likelihood estimation and the EM algorithm is described. Statistical significance of successive factors added to the model is treated by the likelihood ratio criterion. Provisions for effects of guessing on multiple choice items, and for omitted and not reached items, are included. Bayes constraints on the factor loadings are found to be necessary to suppress Heywood cases. Numerous applications to simulated and real data are presented to substantiate the accuracy and practical utility of the method. Analysis of the power tests of the Armed Services Vocational Battery shows statistically significant departures from unidimensionality in five of eight tests.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jul 01, 1986
Accession Number
ADA170602

Entities

People

  • Eiji Muraki
  • R. D. Bock
  • Robert V Gibbons

Organizations

  • NORC at the University of Chicago

Tags

Communities of Interest

  • Biomedical
  • Human Systems
  • Materials and Manufacturing Processes

DTIC Thesaurus Topics

  • Algorithms
  • Biological Sciences
  • Cognition
  • Computational Science
  • Computations
  • Education
  • Equations
  • Factor Analysis
  • Information Science
  • Maximum Likelihood Estimation
  • Military Research
  • Personnel Management
  • Probability
  • Psychology
  • Simulations
  • Statistical Algorithms
  • Statistics

Readers

  • Psychometric Testing or Psychological Assessment.
  • Regression Analysis.