Graphical regression models for polytomous variables

Carolyn J. Anderson, Ulf Böckenholt

Research output: Contribution to journalArticle

Abstract

When modeling the relationship between two nominal categorical variables, it is often desirable to include covariates to understand how individuals differ in their response behavior. Typically, however, not all the relevant covariates are available, with the result that the measured variables cannot fully account for the associations between the nominal variables. Under the assumption that the observed and unobserved variables follow a homogeneous conditional Gaussian distribution, this paper proposes RC(M) regression models to decompose the residual associations between the polytomous variables. Based on Goodman's (1979, 1985) RC(M) association model, a distinctive feature of RC(M) regression models is that they facilitate the joint estimation of effects due to manifest and omitted (continuous) variables without requiring numerical integration. The RC(M) regression models are illustrated using data from the High School and Beyond study (Tatsuoka & Lohnes, 1988).

Original languageEnglish (US)
Pages (from-to)497-509
Number of pages13
JournalPsychometrika
Volume65
Issue number4
DOIs
StatePublished - Dec 2000

Keywords

  • Conditional independence
  • Graphical models
  • Latent continuous variables
  • Marginal maximum likelihood estimation
  • RC(M) association model

ASJC Scopus subject areas

  • Mathematics (miscellaneous)
  • Psychology(all)
  • Psychology (miscellaneous)
  • Social Sciences (miscellaneous)

Fingerprint Dive into the research topics of 'Graphical regression models for polytomous variables'. Together they form a unique fingerprint.

  • Cite this