The hardware and bandwidth for this mirror is donated by METANET, the Webhosting and Full Service-Cloud Provider.
If you wish to report a bug, or if you are interested in having us mirror your free-software or open-source project, please feel free to contact us at mirror[@]metanet.ch.

GDINA Package for Cognitively Diagnostic Analyses

How to cite the package

Ma, W. & de la Torre, J. (2020). GDINA: An R Package for Cognitive Diagnosis Modeling. Journal of Statistical Software, 93(14), 1-26. https://doi.org/10.18637/jss.v093.i14

Visit the package website https://wenchao-ma.github.io/GDINA/ for examples, tutorials and more information.

Learning resources

• Watch Ma and de la Torre’s (2019) NCME digital module 5 on a gentle introduction to the G-DINA model framework and the use of graphical user interface for CDM analyses

• Check the book chapter by Ma (2019) on an introduction to the GDINA package

• Check de la Torre and Akbay’s (2019) article on how to conduct various CDM analyses using the graphical user interface

• Check Shi, et al (2021) on how to use this package, along with other R packages for CDM analyses

Features of the package

• Estimating G-DINA model and a variety of widely-used models subsumed by the G-DINA model, including the DINA model, DINO model, additive-CDM (A-CDM), linear logistic model (LLM), reduced reparametrized unified model (RRUM), multiple-strategy DINA model for dichotomous responses
• Estimating models within the G-DINA model framework using user-specified design matrix and link functions
• Estimating Bugs-DINA, DINO and G-DINA models for dichotomous responses
• Estimating sequential G-DINA model for ordinal and nominal responses
• Estimating the generalized multiple-strategy cognitive diagnosis models (experimental)
• Estimating the diagnostic tree model (experimental)
• Estimating multiple-choice models
• Modelling independent, saturated, higher-order, loglinear smoothed, and structured joint attribute distribution
• Accommodating multiple-group model analysis
• Imposing monotonic constrained success probabilities
• Accommodating binary and polytomous attributes
• Validating Q-matrix under the general model framework
• Evaluating absolute and relative item and model fit
• Comparing models at the test and item levels
• Detecting differential item functioning using Wald and likelihood ratio test
• Simulating data based on all aforementioned CDMs
• Providing graphical user interface for users less familiar with R

Installation

The stable version of GDINA should be installed from R CRAN at here

To install this package from source:

1. Windows users may need to install the Rtools and include the checkbox option of installing Rtools to their path for easier command line usage. Mac users will have to download the necessary tools from the Xcode and its related command line tools (found within Xcode’s Preference Pane under Downloads/Components); most Linux distributions should already have up to date compilers (or if not they can be updated easily).

2. Install the devtools package (if necessary), and install the package from the Github source code.

# install.packages("devtools")
devtools::install_github("Wenchao-Ma/GDINA")

These binaries (installable software) and packages are in development.
They may not be fully stable and should be used with caution. We make no claims about them.