The InterModel Vigorish as a Lens for Understanding (and Quantifying) the Value of Item Response Models for...

Area of Science:

• Psychometrics
• Statistical Modeling
• Data Analysis

Background:

• Statistical models require indices to assess appropriateness for specific data contexts.
• Item response theory models are widely used but need robust evaluation metrics.

Purpose of the Study:

• Introduce the InterModel Vigorish (IMV) as a novel index for quantifying model accuracy.
• Evaluate the IMV's utility for comparing dichotomous item response models.

Main Methods:

• The IMV quantifies accuracy by measuring prediction improvement between two sets of predictions.
• Comparison of IMV with existing indices like AIC and RMSEA using simulated and empirical data.
• Analysis of 89 dichotomous item response datasets to demonstrate IMV's practical application.

Main Results:

• The IMV demonstrates desirable features including portability and generalizability across diverse data contexts.
• Qualitative differences in behavior were observed between IMV and other common indices.
• Empirical applications confirmed the IMV's utility in practical psychometric analyses.

Conclusions:

• The InterModel Vigorish (IMV) is a valuable new index for assessing statistical model accuracy in psychometrics.
• IMV facilitates easy comparison of predictions across various models and contexts.
• The index's portability and generalizability enhance its utility for researchers.