Validation of an abbreviated Big Five personality inventory at large population scale: Psychometric structure and associations with common psychiatric and neurological disorders
Weixi Kang,Jeggan Tiego,Peter J Hellyer,William Trender,Jon E Grant,Samuel R Chamberlain,Adam Hampshire
DOI: https://doi.org/10.1016/j.comppsych.2024.152514
Abstract:Background: The five-factor model of personality, as quantified using instruments such as the Big Five Inventory, consists of broad personality domains including Extraversion, Agreeableness, Conscientiousness, Neuroticism (emotional instability), and Openness. Such instruments typically include >40 items. However, instruments with many items can be unwieldly and a cause of measurement error in clinical and cohort studies where multiple scales are sequenced. Conversely, established 5- and 10-item versions of the Big Five Inventory have poor reliability. Here, we developed and validated an abbreviated 18-item Big Five Inventory that balances efficiency, reliability and sensitivity. Method: We analysed three datasets (N = 59,797, N = 21,177, and N = 87,983) from individuals who participated in the online Great British Intelligence Test (GBIT) study, a collaborative citizen science project with BBC2 Horizon. We applied factor analyses (FA), predictive normative modelling, and one-sample t-tests to validate the 18-item version of the Big Five and to investigate its associations with psychiatric and neurological conditions. Results: The 18-item version of the Big Five Inventory had higher validity and retest reliability compared to the other previously shortened versions in the literature, with comparable demographic associations to the full Big Five Inventory. It exhibited strong (i.e. large effect size) associations with psychiatric conditions, and moderate (small-medium) associations with neurological conditions. Neuroticism (emotional instability) was substantially higher in all psychiatric conditions, whereas Conscientiousness, Openness and Extraversion showed differential associations across conditions. Conclusion: The newly validated 18-item version of the Big Five provides a convenient means of measuring personality traits that is suitable for deployment in a range of studies. It retains psychometric structure, retest reliability and clinical-group sensitivity, as compared to the full original scale.