Psychometrika

, Volume 46, Issue 4, pp 443–459

Marginal maximum likelihood estimation of item parameters: Application of an EM algorithm

Authors

  • R. Darrell Bock
    • Department of Behavioral SciencesThe University of Chicago
  • Murray Aitkin
    • University of Lancaster
Article

DOI: 10.1007/BF02293801

Cite this article as:
Bock, R.D. & Aitkin, M. Psychometrika (1981) 46: 443. doi:10.1007/BF02293801

Abstract

Maximum likelihood estimation of item parameters in the marginal distribution, integrating over the distribution of ability, becomes practical when computing procedures based on an EM algorithm are used. By characterizing the ability distribution empirically, arbitrary assumptions about its form are avoided. The Em procedure is shown to apply to general item-response models lacking simple sufficient statistics for ability. This includes models with more than one latent dimension.

Key words

estimation of item parametersEM algorithmitem analysislatent traitdichotomous factor analysisLaw School Aptitude Test (LSAT)
Download to read the full article text

Copyright information

© The Psychometric Society 1981