hplieninger/mpt2irt: Bringing Multinomial Processing Tree Models To Item Response Theory To Investigate Response Styles

Multinomial processing tree (MPT) models are used to differentiate between different cognitive processes in categorical data. Those can be lifted to hierarchical MPT models to account for person- and/or item effects (Matzke et al., 2015): This is done by reparameterizing each MPT parameter by means of an item response (IRT) model. Boeckenholt (2012) developed a model to disentengle two reponse styles, namely midpoint (MRS) and extreme responding (ERS), and the target trait. This package allows to fit an extension of this Boeckenholt Model to acquiescence (ARS). It makes use of Bayesian hierarchical models requiring Stan or JAGS.

Getting started

Package details

Package repositoryView on GitHub
Installation Install the latest version of this package by entering the following in R:
hplieninger/mpt2irt documentation built on Aug. 4, 2018, 10:52 a.m.