Objectives:To investigate the effect of a social robot intervention on depression, loneliness, and quality of life of older adults in long-term care (LTC) and to explore participants' experiences and perceptions after the intervention.Design:A mixed-methods approach consisting of a single group, before and after quasi-experimental design, and individual interview.Participants:Twenty older adults with depression from four LTC facilities in Taiwan were recruited.Intervention:Each participant participated in 8 weeks of observation and 8 weeks of intervention. In the observation stage, participants received usual care or activities without any research intervention. In the intervention stage, each participant was given a Paro (Personal Assistive RobOt) to keep for 24 hours, 7 days a week.Measurements:The Geriatric Depression Scale, the UCLA Loneliness Scale Version 3, and the World Health Organization Quality of Life Questionnaire for older adults were administered at four time points. Individual qualitative interviews with thematic analysis followed.Results:A repeated multivariate analysis of variance and Friedman's test showed no significant changes during the observation stage between T1 and T2 for depression and quality of life (p >.5). For the intervention stage, statistically significant changes in decreasing depression and loneliness and improving quality of life over time were identified. Three themes emerged from the interviews: (i) humanizing Paro through referring to personal experiences and engagement; (ii) increased social interaction with other people; and (iii) companionship resulting in improved mental well-being.Conclusions:There were significant improvements in mental well-being in using Paro. Further research may help us to understand the advantages of using a Paro intervention as depression therapy.