Software Development

Featured Package

GDINA

The GDINA R Package

A comprehensive R package for estimating, evaluating, and applying Generalized DINA (G-DINA) models and a wide variety of other cognitive diagnosis models.

The GDINA package is one of the most widely used tools for cognitive diagnostic assessment. It supports dozens of CDMs within a unified estimation framework, provides model selection and fit evaluation tools, and includes Q-matrix validation routines.

Key Features

Model Estimation

Fits G-DINA, DINA, DINO, ACDM, LLM, RRUM, and many more CDMs using EM algorithm estimation.

Model Evaluation

Provides absolute and relative model fit indices, item-level fit statistics, and residual analysis tools.

Q-Matrix Validation

Data-driven methods to detect and correct misspecified Q-matrix entries, improving diagnostic accuracy.

Classification & Scoring

Estimates attribute mastery probabilities and provides posterior-based classification of examinees.

Quick Start

# Install from CRAN

install.packages("GDINA")

# Load and fit a G-DINA model

library(GDINA)

fit <- GDINA(dat = sim10GDINA$dat, Q = sim10GDINA$Q)

summary(fit)

Get the Package

Available on CRAN and GitHub

CRAN GitHub Documentation

Citation

Ma, W., & de la Torre, J. (2020). GDINA: An R Package for Cognitive Diagnosis Modeling. Journal of Statistical Software, 93(14), 1–26.

Additional Resources

GitHub

Browse all open-source code, simulation studies, and supplementary materials associated with lab publications.

View on GitHub →

Tutorials & Workshops

Dr. Ma regularly presents workshops on CDMs and the GDINA package at national conferences. Contact the lab for workshop materials.