On testing factorial invariance: A reply to J.C.F. de Winter.

Area of Science:

• Psychology
• Transportation Research
• Psychometrics

Background:

• The Driver Behavior Questionnaire (DBQ) is widely used to compare driver subgroups.
• Ensuring the DBQ measures constructs identically across diverse groups is crucial but often overlooked.
• Previous factorial invariance analysis of the Finnish DBQ faced criticism regarding methodology and interpretation.

Purpose of the Study:

• To address criticisms of a prior study on the factorial invariance of the Finnish DBQ.
• To rigorously assess whether the DBQ measures the same constructs consistently across different respondent groups.
• To propose robust methodological approaches for establishing measurement invariance in self-report instruments.

Main Methods:

• Utilized stage-wise factorial invariance testing within Exploratory Structural Equation Modeling (ESEM).
• Responded to specific criticisms concerning measurement error, factor extraction, and invariance criteria.
• Employed advanced psychometric techniques to evaluate measurement invariance.

Main Results:

• The study refutes claims that previous findings were solely artifacts of measurement error.
• Demonstrates that appropriate methods can confirm factorial invariance for the DBQ.
• Highlights the importance of stringent criteria for establishing reliable measurement across groups.

Conclusions:

• The Driver Behavior Questionnaire (DBQ) can reliably measure driving constructs invariantly across respondent groups.
• Methodological rigor is essential for accurate cross-group comparisons using self-report measures.
• The study provides a framework for validating other self-report instruments for measurement invariance.