Internal consistency, retest reliability, and their implications for personality scale validity.

Area of Science:

• Psychology
• Psychometrics
• Personality Science

Background:

• The NEO Inventories are widely used for personality assessment.
• Understanding the psychometric properties of personality scales is essential for accurate measurement.
• The generalizability of these properties across different populations and measurement methods requires investigation.

Purpose of the Study:

• To examine the differential reliability and validity of facet scales within the NEO Inventories.
• To assess the generalizability of psychometric properties across age, culture, and measurement methods.
• To determine the association between different forms of reliability and validity criteria.

Main Methods:

• Analysis of a large dataset (N = 34,108) from the NEO Inventories.
• Evaluation of composite estimates for facet scale stability, heritability, and cross-observer validity.
• Statistical examination of the predictive relationship between various reliability estimates (retest, internal consistency) and validity criteria.

Main Results:

• Composite estimates of facet scale stability, heritability, and cross-observer validity demonstrated broad generalizability.
• Two measures of retest reliability independently predicted three validity criteria.
• None of the three internal consistency measures predicted the validity criteria.

Conclusions:

• Retest reliability is a more important predictor of personality scale validity than internal consistency.
• Internal consistency is useful for data quality checks but not for assessing scale validity.
• Findings suggest similar patterns for other personality inventories, highlighting the need for further research into retest reliability.