A Markov chain Monte Carlo approach to confirmatory item factor analysis

Research output: Contribution to journalArticle

42 Citations (Scopus)

Abstract

Item factor analysis has a rich tradition in both the structural equation modeling and item response theory frameworks. The goal of this paper is to demonstrate a novel combination of various Markov chain Monte Carlo (MCMC) estimation routines to estimate parameters of a wide variety of confirmatory item factor analysis models. Further, I show that these methods can be implemented in a flexible way which requires minimal technical sophistication on the part of the end user. After providing an overview of item factor analysis and MCMC, results from several examples (simulated and real) will be discussed. The bulk of these examples focus on models that are problematic for current "gold-standard" estimators. The results demonstrate that it is possible to obtain accurate parameter estimates using MCMC in a relatively user-friendly package.

Original languageEnglish (US)
Pages (from-to)474-497
Number of pages24
JournalPsychometrika
Volume75
Issue number3
DOIs
StatePublished - Apr 1 2010
Externally publishedYes

Fingerprint

Markov Chains
Factor analysis
Factor Analysis
Markov Chain Monte Carlo
Markov processes
Statistical Factor Analysis
Structural Equation Modeling
Gold
Estimate
Demonstrate
Estimator
Model

Keywords

  • item factor analysis
  • Markov chain Monte Carlo
  • multidimensional item response theory

ASJC Scopus subject areas

  • Psychology(all)
  • Applied Mathematics

Cite this

A Markov chain Monte Carlo approach to confirmatory item factor analysis. / Edwards, Michael.

In: Psychometrika, Vol. 75, No. 3, 01.04.2010, p. 474-497.

Research output: Contribution to journalArticle

@article{2d025f5e01c1492d9d47b1b83e308c84,
title = "A Markov chain Monte Carlo approach to confirmatory item factor analysis",
abstract = "Item factor analysis has a rich tradition in both the structural equation modeling and item response theory frameworks. The goal of this paper is to demonstrate a novel combination of various Markov chain Monte Carlo (MCMC) estimation routines to estimate parameters of a wide variety of confirmatory item factor analysis models. Further, I show that these methods can be implemented in a flexible way which requires minimal technical sophistication on the part of the end user. After providing an overview of item factor analysis and MCMC, results from several examples (simulated and real) will be discussed. The bulk of these examples focus on models that are problematic for current {"}gold-standard{"} estimators. The results demonstrate that it is possible to obtain accurate parameter estimates using MCMC in a relatively user-friendly package.",
keywords = "item factor analysis, Markov chain Monte Carlo, multidimensional item response theory",
author = "Michael Edwards",
year = "2010",
month = "4",
day = "1",
doi = "10.1007/s11336-010-9161-9",
language = "English (US)",
volume = "75",
pages = "474--497",
journal = "Psychometrika",
issn = "0033-3123",
publisher = "Springer New York",
number = "3",

}

TY - JOUR

T1 - A Markov chain Monte Carlo approach to confirmatory item factor analysis

AU - Edwards, Michael

PY - 2010/4/1

Y1 - 2010/4/1

N2 - Item factor analysis has a rich tradition in both the structural equation modeling and item response theory frameworks. The goal of this paper is to demonstrate a novel combination of various Markov chain Monte Carlo (MCMC) estimation routines to estimate parameters of a wide variety of confirmatory item factor analysis models. Further, I show that these methods can be implemented in a flexible way which requires minimal technical sophistication on the part of the end user. After providing an overview of item factor analysis and MCMC, results from several examples (simulated and real) will be discussed. The bulk of these examples focus on models that are problematic for current "gold-standard" estimators. The results demonstrate that it is possible to obtain accurate parameter estimates using MCMC in a relatively user-friendly package.

AB - Item factor analysis has a rich tradition in both the structural equation modeling and item response theory frameworks. The goal of this paper is to demonstrate a novel combination of various Markov chain Monte Carlo (MCMC) estimation routines to estimate parameters of a wide variety of confirmatory item factor analysis models. Further, I show that these methods can be implemented in a flexible way which requires minimal technical sophistication on the part of the end user. After providing an overview of item factor analysis and MCMC, results from several examples (simulated and real) will be discussed. The bulk of these examples focus on models that are problematic for current "gold-standard" estimators. The results demonstrate that it is possible to obtain accurate parameter estimates using MCMC in a relatively user-friendly package.

KW - item factor analysis

KW - Markov chain Monte Carlo

KW - multidimensional item response theory

UR - http://www.scopus.com/inward/record.url?scp=77956060011&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77956060011&partnerID=8YFLogxK

U2 - 10.1007/s11336-010-9161-9

DO - 10.1007/s11336-010-9161-9

M3 - Article

AN - SCOPUS:77956060011

VL - 75

SP - 474

EP - 497

JO - Psychometrika

JF - Psychometrika

SN - 0033-3123

IS - 3

ER -