GPCMlasso: Differential Item Functioning in Generalized Partial Credit Models

Provides a function to detect Differential Item Functioning (DIF) in Generalized Partial Credit Models (GPCM) and special cases of the GPCM. A joint model is set up where DIF is explicitly parametrized and penalized likelihood estimation is used for parameter selection. The big advantage of the method called GPCMlasso is that several variables can be treated simultaneously and that both continuous and categorical variables can be used to detect DIF.

Version: 0.1-1
Depends: ltm
Imports: Rcpp (≥ 0.12.4), wordcloud, TeachingDemos, cubature, car, caret, statmod, mvtnorm, mirt, methods
LinkingTo: Rcpp, RcppArmadillo
Published: 2018-02-12
Author: Gunther Schauberger
Maintainer: Gunther Schauberger <gunther at stat.uni-muenchen.de>
License: GPL-2 | GPL-3 [expanded from: GPL (≥ 2)]
NeedsCompilation: yes
SystemRequirements: C++11
CRAN checks: GPCMlasso results

Downloads:

Reference manual: GPCMlasso.pdf
Package source: GPCMlasso_0.1-1.tar.gz
Windows binaries: r-devel: GPCMlasso_0.1-1.zip, r-release: GPCMlasso_0.1-1.zip, r-oldrel: GPCMlasso_0.1-1.zip
OS X binaries: r-release: GPCMlasso_0.1-1.tgz, r-oldrel: GPCMlasso_0.1-1.tgz

Linking:

Please use the canonical form https://CRAN.R-project.org/package=GPCMlasso to link to this page.