An Item-Level Examination of the Factorial Validity of NEO Five-Factor Inventory Scores

The NEO Five-Factor Inventory (NEO-FFI) is often used in field-based research and clinical studies as it is designed to measure the same personality dimensions as the longer NEO Personality Inventory in a shorter time frame. In this study, the authors examined the reliability and structural validity of the NEO-FFI scores at the item level in a large sample of Jamaican young adults (N = 1,021; ages 17-24 years). Across different factor estimation and rotation methods, many NEO-FFI items performed poorly. Likewise, the estimated reliability of the NEO-FFI scores was poor (except Conscientiousness) across different estimations of reliability. These items and scores were then compared with other studies of the NEO-FFI that reported item-level pattern/structure coefficients or reliability estimates. Similarities in item performance and low reliability estimates across studies suggest that the items, rather than cultural differences, account for much of the poor performance of the NEO-FFI scores, especially in the domains of Extraversion, Openness to Experience, and Agreeableness.