Automating Simulation Research for Item Response Theory using R

Authors : Sunbok LEE, Youn-jeng CHOİ, Allan S. COHEN

Pages : 682-700

Doi:10.21449/ijate.472185

View : 10 | Download : 2

Publication Date : 2018-12-16

Article Type : Research Paper

Abstract :A simulation study is a useful tool in examining how validly item response theory insert ignore into journalissuearticles values(IRT); models can be applied in various settings. Typically, a large number of replications are required to obtain the desired precision. However, many standard software packages in IRT, such as MULTILOG and BILOG, are not well suited for a simulation study requiring a large number of replications because they were developed as a stand-alone software package that is best suited for a single run. This article demonstrated how built-in R functions can be used to automate the simulation study using the stand-alone software packages in IRT. For a demonstration purpose, MULTILOG was used in the example codes in the appendices, but the overall framework of a simulation study and the built-in R functions used in this article can be applied for a simulation study using other stand-alone software packages as well.
Keywords : IRT, Simulation, R