Do self-report instruments allow meaningful comparisons across diverse population groups? Testing measurement...

Area of Science:

• Public Health
• Psychometrics
• Quantitative Research Methods

Background:

• Comparative public health research frequently utilizes self-report instruments to assess personal attributes like attitudes and behaviors.
• Accurate interpretation of health disparities relies on the assumption that self-report measures function equivalently across diverse demographic groups.
• Direct observation of personal attributes is often infeasible, making validated self-reports crucial for understanding health differences.

Purpose of the Study:

• To outline a framework for establishing the psychometric defensibility of self-report instruments in cross-group comparative research.
• To demonstrate how confirmatory factor analysis (CFA) can be employed to rigorously test measurement invariance.
• To provide guidance on ensuring that group comparisons reflect true differences in constructs rather than method effects.

Main Methods:

• The study details the application of confirmatory factor analysis (CFA) to assess measurement invariance.
• A hierarchical testing approach is described, evaluating configural, metric, strong (scalar), and strict factorial invariance.
• The methodology is explained in nontechnical language, supported by a worked example and technical appendices.

Main Results:

• Confirmatory factor analysis (CFA) provides a robust framework for evaluating the cross-group invariance of psychometric properties in self-report instruments.
• Testing a nested hierarchy of invariance hypotheses (configural, metric, scalar, strict) allows researchers to systematically assess measurement equivalence.
• Evidence of factorial invariance is essential for defensible comparisons of means and variances across groups in public health research.

Conclusions:

• Establishing measurement invariance through CFA is critical for the valid use of self-report data in comparative public health research.
• Rigorous testing of invariance hypotheses ensures that observed differences between groups are attributable to the construct of interest, not measurement artifacts.
• The presented CFA framework and invariance tests offer a standardized approach to enhance the quality and interpretability of cross-group health disparity research.