Item Parameter Recovery Using the Multidimensional Graded Response Model in R
Journal of Psychometric Research · 2026
DOI: 10.62425/jopres.1906236 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Psychometric Research, 2026. Verfasst von Yeşim Beril Soğuksu und 2 weiteren Autor:innen. Die Arbeit nennt 59 Literaturangaben.
Abstract (Originalsprache)
The purpose of this study is to provide a step-by-step demonstration of item recovery for the Multidimensional Graded Response Model (MGRM) using R. For this purpose, an illustrative simulation setup was employed, including test lengths of 20 and 40 items, interdimensional correlations of 0.3 and 0.7, and a sample size of 2000. Parameter estimates were obtained from datasets generated under the MGRM, and bias and Root Mean Square Error (RMSE) values were calculated and visualized as part of the demonstration process. Rather than evaluating recovery performance across conditions, the study focuses on explaining each stage of the analysis, data generation, parameter estimation, evaluation metrics, and visualization, through clearly annotated R code. By providing reproducible and accessible implementations, the study aims to serve as a practical guide for researchers interested in conducting item parameter recovery analyses within the MGRM framework. In addition, the presented workflow may be useful for researchers seeking guidance on specific steps such as data generation, estimation, or visualization.
Zitieren
APA
Soğuksu, Y. B., Güneş, A. B., & Gürdil, H. (2026). Item Parameter Recovery Using the Multidimensional Graded Response Model in R. Journal of Psychometric Research. https://doi.org/10.62425/jopres.1906236
Vancouver
Soğuksu YB, Güneş AB, Gürdil H. Item Parameter Recovery Using the Multidimensional Graded Response Model in R. Journal of Psychometric Research. 2026. doi:10.62425/jopres.1906236
BibTeX
@article{Souksu2026,
author = {Soğuksu, Yeşim Beril and Güneş, Ayşe Bilicioğlu and Gürdil, Hatice},
title = {Item Parameter Recovery Using the Multidimensional Graded Response Model in R},
journal = {Journal of Psychometric Research},
year = {2026},
doi = {10.62425/jopres.1906236}
}
EUVIMED