Reassessing the fitting propensity of factor models.

Area of Science:

• Statistics
• Psychometrics
• Quantitative Psychology

Background:

• Model complexity is crucial for statistical model evaluation.
• Fitting propensity (FP) quantifies a model's ability to fit diverse data patterns.
• Existing research on FP is limited, focusing on theory over practical application.

Purpose of the Study:

• To practically examine the fitting propensity (FP) of commonly used factor analysis models.
• To provide a historical context for statistical model evaluation, challenging parameter count as the sole complexity measure.
• To offer recommendations for future research on FP in latent variable modeling.

Main Methods:

• Historical review of statistical model evaluation methods.
• Analysis of exploratory and confirmatory factor analysis models using analytic examples.
• Characterization of findings in relation to existing claims about factor model FP.

Main Results:

• Fitting propensity (FP) offers a more nuanced understanding of model complexity than parameter count alone.
• Analytic examples illustrate the FP of widely used exploratory and confirmatory factor analysis models.
• Findings are contextualized against prior research on factor model FP.

Conclusions:

• FP is a vital metric for assessing statistical model complexity in practice.
• Further research on FP in latent variable modeling is recommended.
• This study contributes practical insights into evaluating factor analysis models.