, Volume 58, Issue 1, pp 37–52

The asymptotic posterior normality of the latent trait in an IRT model

  • Hua-Hua Chang
  • William Stout

DOI: 10.1007/BF02294469

Cite this article as:
Chang, HH. & Stout, W. Psychometrika (1993) 58: 37. doi:10.1007/BF02294469


It has long been part of the item response theory (IRT) folklore that under the usual empirical Bayes unidimensional IRT modeling approach, the posterior distribution of examinee ability given test response is approximately normal for a long test. Under very general and nonrestrictive nonparametric assumptions, we make this claim rigorous for a broad class of latent models.

Key words

item response theoryempirical Bayesposterior distributionability estimationconfidence intervalmanifest probablyDutch Identity

Copyright information

© The Psychometric Society 1993

Authors and Affiliations

  • Hua-Hua Chang
    • 1
  • William Stout
    • 2
  1. 1.Educational Testing ServicePrinceton
  2. 2.Department of StatisticsUniversity of IllinoisChampaign