A Comprehensive Guide to Item Recovery Using the Multidimensional Graded Response Model in R

TL;DR

This paper provides a detailed, step-by-step guide with R code for item recovery in the Multidimensional Graded Response Model, including simulation design, parameter estimation, and visualization, aiding researchers in implementation.

Contribution

It offers a comprehensive, practical tutorial with R scripts for item recovery in MGRM, filling a gap in accessible methodological guidance.

Findings

Parameter estimates were successfully obtained from simulated datasets.

Bias and RMSE were calculated and visualized for different conditions.

The guide facilitates replication and adaptation of MGRM item recovery procedures.

Abstract

The purpose of this study is to provide a step-by-step demonstration of item recovery for the Multidimensional Graded Response Model (MGRM) in R. Within this scope, a sample simulation design was constructed where the test lengths were set to 20 and 40, the interdimensional correlations were varied as 0.3 and 0.7, and the sample size was fixed at 2000. Parameter estimates were derived from the generated datasets for the 3-dimensional GRM, and bias and Root Mean Square Error (RMSE) values were calculated and visualized. In line with the aim of the study, R codes for all these steps were presented along with detailed explanations, enabling researchers to replicate and adapt the procedures for their own analyses. This study is expected to contribute to the literature by serving as a practical guide for implementing item recovery in the MGRM. In addition, the methods presented, including data generation, parameter estimation, and result visualization, are anticipated to benefit researchers even if they are not directly engaged in item recovery.