flirt-package: Flexible item response theory modeling with efficient maximum...

Description Details Author(s) References

Description

This package provides a flexible framework for uni- and multi- dimensional explanatory item response theory modeling for binary and polytomous item responses. The flexibility stems from specifying IRT models as generalized linear and nonlinear mixed models (Rijmen, Tuerlinckx, De Boeck, & Kuppens, 2003).

For estimation, the package flirt uses an efficient modified EM algorithm based on the graphical model framework. The modified EM algorithm is much faster than the traditional EM algorithm. For more details on the modified EM algorithm, refer to e.g., Rijmen, F., Vansteelandt, K., & De Boeck, P. (2008).

Currently, uni- and multi- dimensional Rasch models, two-parameter logistic (2PL) IRT models, and bifactor models are available with extensions to multiple groups, item covariates, person covariates, and differential item functioning analyses.

The package flirt is based on the Matlab code BNLflirt (Rijmen and Jeon, 2013) for estimation that employs sub-functions from the Matlab toolbox BNL (Bayesian Networks with Logistic Regression Nodes; Rijmen, 2006).

flirt requires the Matlab Compiler Runtime (MCR) for Matlab 2014a, Windows. Having the correct version of the MCR is critical. If you have a different version of Matlab on your computer, please make sure to remove the MCR that you have and download/install the correct version in the following link:

Having the correct version of the MCR is critical. If you have a different version of Matlab on your computer, please make sure to remove the MCR that you have and download/install the correct version in the following link:

http://www.mathworks.com/products/compiler/mcr/

To cite flirt,

Jeon, M., Rijmen, F. & Rabe-Hesketh, S. (2014). Flexible item response theory modeling with flirt. Applied Psychological Methods, 38, 404-405

Details

Package: flirt
Type: Package
Version: 1.15
Date: 2015-1-18
License: GPL

The R script containing sample analyses using flirt is available by contacting the first author.

Author(s)

Minjeong Jeon, Frank Rijmen, and Sophia Rabe-Hesketh

Maintainer: Minjeong Jeon<jeon.117@osu.edu>

References

Jeon, M. and Rijmen, F. (2014). A modular approach for item response theory modeling with the R package flirt. Under revision.

Jeon, M. and De Boeck, P. (2015). A generalized item response tree model for psychological assessments. Under revision.

Jeon, M., Rijmen, F. & Rabe-Hesketh, S. (2014). Flexible item response theory modeling with flirt. Applied Psychological Methods, 38, 404-405

Jeon, M., Rijmen, F., and Rabe-Hesketh, R. (2013). Modeling differential item functioning using a generalization of the multiple-group bifactor model. Journal of Behavioral and Educational Statistics, 38, 32-60.

Rijmen, F. (2006). BNL: A Matlab toolbox for Bayesian networks with logistic regression. Technical Report. Vrije Universiteit Medical Center, Amsterdam.

Rijmen, F., Tuerlinckx, F., De Boeck, P., & Kuppens, P. (2003). A nonlinear mixed model framework for item response theory. Psychological Methods, 8, 185-205.

Rijmen, F., Vansteelandt, K., & De Boeck, P. (2008). Latent class models for diary method data: Parameter estimation by local computations. Psychometrika, 73, 167-182.

Rijmen, F. and Jeon, M. (2013). BNLflirt: Flexible item response theory modeling with BNL. Matlab file exchange.


seonghobae/flirt.x32 documentation built on May 29, 2019, 6:54 p.m.