variationalDCM: An R package for variational Bayesian inference in diagnostic classification models

Abstract

This paper introduces variationalDCM, an R package that performs recently-developedvariational Bayesian (VB) estimation methods for diagnostic classification models (DCMs).DCMs are a class of discrete latent variable models that reveal respondents’ current knowledge statuses and perform clustering based on these statuses mainly in educational measurements. This package enables computationally efficient Bayesian estimation for various DCMs in large-scale datasets. The paper describes the parsimonious and generalizedDCMs that variationalDCM supports and provides the empirical illustration of the functions in this package.