Fitting item response unfolding models to Likert-scale data using mirt in R

Area of Science:

• Psychometrics
• Statistical Modeling
• Data Analysis

Background:

• Decades of development in unfolding models for Likert-scale data.
• Limited practical application due to a lack of suitable estimation software.

Purpose of the Study:

• Demonstrate the utility of the R package mirt for estimating unfolding models.
• Provide practical guidelines and R syntax for applying these models.
• Evaluate the performance of mirt through simulation studies.

Main Methods:

• Utilizing the mirt package in R for parameter estimation.
• Applying unidimensional and multidimensional unfolding models.
• Conducting parameter-recovery simulation studies.

Main Results:

• The mirt package effectively estimates parameters for various unfolding models.
• Simulation studies confirm the package's potential effectiveness.
• Practical guidelines and R syntax are provided for real-world application.

Conclusions:

• The mirt package significantly lowers the barrier to applying unfolding models to Likert-scale data.
• Researchers can now confidently estimate these models with real datasets.
• Facilitates broader adoption of advanced psychometric techniques.