Fitting diffusion item response theory models for responses and response times using the R package diffIRT

Abstract

In the psychometric literature, item response theory models have been proposed that explicitly take the decision process underlying the responses of subjects to psychometric test items into account. Application of these models is however hampered by the absence of general and exible software to fit these models. In this paper, we present diffIRT, an R package that can be used to fit item response theory models that are based on a diffusion process. We discuss parameter estimation and model fit assessment, show the viability of the package in a simulation study, and illustrate the use of the package with two datasets pertaining to extraversion and mental rotation. In addition, we illustrate how the package can be used to fit the traditional diffusion model (as it has been originally developed in experimental psychology) to data.