Convergence, Admissibility, and Fit of Alternative Confirmatory Factor Analysis Models for MTMM Data

Area of Science:

• Psychometrics
• Quantitative Psychology
• Statistical Modeling

Background:

• Multitrait-multimethod (MTMM) data analysis is crucial for understanding construct validity.
• Existing models like CTCM and CTCU have limitations in convergence and admissibility.
• Evaluating novel MTMM models is essential for advancing psychometric research.

Purpose of the Study:

• To compare the performance of six analytic models for MTMM data.
• To identify models that overcome convergence and admissibility challenges.
• To assess the accuracy of parameter estimation in MTMM models.

Main Methods:

• Analysis of 258 previously reported MTMM datasets.
• Comparison of the correlated trait-correlated method (CTCM) and correlated trait-correlated uniqueness (CTCU) models.
• Evaluation of four understudied models, including Rindskopf's reparameterization (CTCM-R).

Main Results:

• The CTCM-R model demonstrated superior performance in addressing convergence and admissibility problems.
• Constrained models yielded admissible solutions but exhibited significantly poorer model fit.
• A simulation study confirmed the CTCM-R model's accuracy in parameter estimation.

Conclusions:

• The CTCM-R model is a robust alternative for analyzing MTMM data, mitigating common estimation issues.
• Constrained MTMM models may not be suitable for real-world data due to poor fit.
• The CTCM-R model provides the most accurate parameter estimates for confirmatory factor analysis of MTMM data.