Parameter constraints of the logit form of the reduced RUM

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

The Reduced Reparameterized Unified Model (Reduced RUM) has received considerable attention among educational researchers. Markov chain Monte Carlo (MCMC) or Expectation Maximization (EM) is typically used for estimating the Reduced RUM. Implementations of the EM algorithm are available in the latent class analysis (LCA) routines of commercial software packages (e.g., Latent GOLD, Mplus). Using a commercial LCA routine as a vehicle for fitting the Reduced RUM with the EM algorithm requires that it be reparameterized as a logit model, with complex constraints imposed on the parameters. This article summarizes the general parameterization of the Reduced RUM as a logit model and the associated parameter constraints.

Original languageEnglish (US)
Title of host publicationQuantitative Psychology - 81st Annual Meeting of the Psychometric Society, 2016
EditorsWen-Chung Wang, Marie Wiberg, Steven A. Culpepper, Jeffrey A. Douglas, L. Andries van der Ark
PublisherSpringer
Pages207-213
Number of pages7
ISBN (Print)9783319562933
DOIs
StatePublished - Jan 1 2017
Event81st annual meeting of the Psychometric Society, 2016 - Asheville, United States
Duration: Jul 11 2016Jul 15 2016

Publication series

NameSpringer Proceedings in Mathematics and Statistics
Volume196
ISSN (Print)2194-1009
ISSN (Electronic)2194-1017

Other

Other81st annual meeting of the Psychometric Society, 2016
Country/TerritoryUnited States
CityAsheville
Period7/11/167/15/16

Keywords

  • Cognitive diagnosis
  • EM algorithm
  • Reduced RUM

ASJC Scopus subject areas

  • Mathematics(all)

Fingerprint

Dive into the research topics of 'Parameter constraints of the logit form of the reduced RUM'. Together they form a unique fingerprint.

Cite this