MCMC estimation and some model-fit analysis of multidimensional IRT models

By A. Béguin, Cees A. W. Glas. 2001

368 citations recorded by the source corpus.

View Paper: https://www.semanticscholar.org/paper/612357ffb32fe4a9585a3e02bd4925ddfa56fef3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Statistics and Probability

Citations in the corpus

Application of Multidimensional Item Response Theory Models to Longitudinal Data (2006, methodology): “The first is a Bayesian procedure using a Markov Chain Monte Carlo algorithm (see, for instance, Gelman, Carlin, Stern, & Rubin, 1995), which was suggested by Béguin and Glas (2001).”
Estimation of Compensatory and Noncompensatory Multidimensional Item Response Models Using Markov Chain Monte Carlo (2003, methodology): “…offer much potential for estimation of complex statistical models (Gilks, Richardson, & Spiegelhalter, 1996) and are receiving growing attention in item response theory (Albert, 1992; Baker, 1998; Béguin & Glas, 2001; Kim, 2001; Patz & Junker, 1999a, 1999b; Wollack, Bolt, Cohen & Kim, Lee 2001).”
A hierarchical framework for modeling speed and accuracy on test items (2007, methodology): “For the version of the 3PNO model in (9) without guessing parameter ci and independent priors for the other parameters, Albert (1992) introduced Gibbs sampling with data augmentation.”
Performances of LOO and WAIC as IRT Model Selection Methods (2017, methodology): “With an expanding body of literature on Bayesian IRT (e.g., Beguin & Glas, 2001; Bolt & Lall, 2003; Cao & Stokes, 2008; Fox & Glas, 2003; Jiao & Zhang, 2015), the topic of Bayesian model checks and comparison in the IRT context has also received increasing attention (e.g., Levy, Mislevy, &…”
A Comparison of Alternative Models for Testlets (2006, background): “Béguin and Glas (2001) have formulated a similar Gibbs sampling algorithm for the general multidimensional IRT model, which can also be modified to accommodate the special cases of the multidimensional model considered above.”
A Comparative Study of Test Data Dimensionality Assessment Procedures Under Nonparametric IRT Models (2004, methodology): “Second, due to the larger number of parameters, the estimation of multidimensional IRT models is more complicated than the estimation of unidimensional IRT models (e.g., see Béguin & Glas, 2001, who used Markov chain Monte Carlo techniques for estimating a multidimensional normal ogive model).”
Experiences With Markov Chain Monte Carlo Convergence Assessment in Two Psychometric Examples (2004, methodology): “More sophisticated MCMC algorithms, for example, data augmentation (Beguin & Glas, 2001, and the references therein), the resampling and adaptive switching of the transition kernel (Gelfand & Sahu, 1994), multichain annealing or “simulated tempering” (Geyer & Thompson, 1995), and working parameter…”
Assessing Fit of Unidimensional Item Response Theory Models Using a Bayesian Approach. (2005, methodology): “Beguin and Glas (2001) used a similar plot for detecting violation of a multidimen sional IRT model.”
Metropolis-Hastings Robbins-Monro Algorithm for Confirmatory Item Factor Analysis (2010, methodology): “Due to the flexibility of the MCMC method, current MCMC algorithms for IFA arrive in many flavors (Albert, 1992; Beguin & Glas, 2001; Patz & Junker, 1999; Segall, 1998; Shi & Lee, 1998), with no consensus on the relative algorithmic efficiency.”
Latent Variable Selection for Multidimensional Item Response Theory Models via $$L_{1}$$L1 Regularization (2016, positive): “In the literature, various methods have been introduced for the item parameter estimation including the marginal likelihood method (Bock, Gibbons, &Muraki, 1988), Bayesian estimation via Markov chain Monte Carlo (Béguin & Glas, 2001; Bolt & Lall, 2003), least squares method (McDonald, 1982), etc.”
An Integrated Bayesian Model for DIF Analysis (2009, methodology): “For example, Bayesian estimation methods are used in IRT models in several approaches (e.g. Albert (1992), Patz & Junker (1999a), Patz & Junker (1999b), Fox & Glas (2001), Béguin & Glas (2001)).”