Selecting the "Best" Factor Structure and Moving Measurement Validation Forward: An Illustration

Area of Science:

• Psychometrics
• Quantitative Psychology
• Mental Health Assessment

Background:

• Factor analysis best practices are often overlooked in psychometric evaluations.
• This leads to inaccurate factor structures, complex competing models, and biased results.
• The Posttraumatic Stress Disorder Checklist (PCL-5) requires rigorous factor structure validation.

Purpose of the Study:

• To present a practical, actionable process for factor analysis in structure validation.
• To summarize updated PCL-5 factor models and provide a rationale for validation.
• To conduct a comprehensive statistical validation of the PCL-5 factor structure.

Main Methods:

• Overview of six key statistical and psychometric issues in factor analysis.
• Development of a flowchart for recommended procedures in latent structure analysis.
• Statistical and psychometric validation of the PCL-5 using a sample of 1,403 U.S. Air Force operators.

Main Results:

• Previously proposed factor structures for the PCL-5 were not supported by the data.
• A bifactor model demonstrated greater statistical appropriateness for the PCL-5.
• The study highlights common pitfalls in psychometric evaluations and offers solutions.

Conclusions:

• Rigorous factor analysis is crucial for accurate psychometric property evaluation.
• The bifactor model is a statistically superior approach for the PCL-5.
• Findings provide guidance for researchers evaluating measurement instruments.