Loneliness is a subjective and unpleasant experience associated to a variety of physical and mental health problems. Older people face an elevated risk of loneliness. It is measured through self‐reported questionnaires, among which the most widely used one is the University of California, Los Angeles (UCLA) Loneliness Scale. The latest version of this scale is called UCLA‐3, and it has not been validated in Spanish elderly population. This article aims to present a validation study of the Spanish version of the UCLA‐3 for its use with older people, by establishing its reliability and validity. Data were obtained from the second wave of a longitudinal research conducted with older adults attending Lifelong Learning programs. The sample consisted of 335 people aged 55-years old or older. Sociodemographic data were collected, and loneliness was measured with the UCLA‐3 and the de Jong Gierveld Loneliness Scale. Results have partially replicated a three factor structure already found in the literature in other populations. The present work provides relevant and innovative evidence, as this is the first time that the UCLA‐3 was analysed using ESEM statistical approach in a Spanish elderly sample.