Seven practical recommendations for IRT-based scale development¶

Strategy · All strategies
Evidence · no claims cited

Description¶

The article closes with seven recommendations for scholars embarking on IRT-based scale development: ensure a sufficient sample size, test unidimensionality with multiple methods, check local independence with Yen's Q3, select the IRT model by response categories and model-data fit, examine model and item fit, report estimation methods and interpret item parameters, and report item and test information functions with standard errors. As the review states, "For IRT analysis, a sample size of at least 500 examinees is recommended".

Design Implications¶

Context¶

Requirements¶

• Researchers must have access to IRT software and knowledge of IRT models and estimation techniques, which the review notes are common barriers.

Constraints¶

• The recommended minimum sample size of 500 may vary depending on the number of parameters and ability distributions in the model.

Target Learners¶

• Educational measurement researchers
• Graduate students in psychometrics

Target Learning Goals¶

• Conducting sound IRT-based scale development
• Checking and reporting IRT assumptions and information functions

Related Strategies¶

Examples¶

-

Key Sources¶

• Kılıç, A. F., Koyuncu, İ, & Uysal, İ. (2023). Scale development based on item response theory: A systematic review. International Journal of Psychology and Educational Studies, 10(1), 209-223. https://dx.doi.org/10.52380/ijpes.2023.10.1.982