ABSTRACT
Currently, the gold standard measure to assess loneliness is the University of California, Los Angeles Loneliness scale version 3 (UCLA v3). Objective. The aim of this study was to analyze the psychometric properties of UCLA v3 for the European Portuguese population. Method: A sample of 282 participants was surveyed in Portugal. Results: Confirmatory factor analysis (CFA) supported a model portraying a global loneliness bifactor solution for positive and negative wording items, which achieved optimal fitness. Multi-group CFA indicates scalar and metric invariance across gender. Loneliness test scores (global score, positive items and negative items) correlated with well-established mental health indicators such as psychological stress, depressive and anxiety symptomatology, or psychological inflexibility. Internal consistency of the loneliness test scores was optimal for the global measure (α =.91;ω =.91) as well as for the positive (α =.87;ω =.87) and the negative factors (α =.86;ω =.88). Conclusions: Results support UCLA v3 as a reliable and valid measure of loneliness for future research studies interested in examining the prevalence of loneliness and impact in health in the context of Covid-19 in the Portuguese population and as a health indicator in health promotion and clinical interventions. © 2023 Sociedad Española para el Estudio de la Ansiedad y el Estrés - SEAS. Colegio de la Psicología de Madrid. Todos los derechos reservados.