OBJECTIVE: to describe morale among the oldest old, and to investigate which social, functional and medical factors are associated with morale in this population. DESIGN: a cross-sectional study. SETTING: a population-based study in the municipality of Umeå, a city in Northern Sweden. SUBJECTS: half of the 85-year-old population, and the total population of 90-year-olds and > or = 95-year-olds (95-103) were asked to participate (n = 319) and 238 were interviewed. METHODS: structured interviews and assessments during home visits, interviews with relatives and caregivers and review of medical charts. The 17-item Philadelphia Geriatric Center Morale Scale (PGCMS) was used to measure morale. Participants were assessed with the Barthel Activities of Daily Living (ADL) Index, Geriatric Depression Scale (GDS-15), Mini-Mental State Examination (MMSE), Mini Nutritional Assessment (MNA), and a symptom questionnaire. Multiple regression analyses were conducted to find independent factors to explain the variation in the PGCMS score. RESULTS: eighty-four per cent (n = 199) of those interviewed answered the PGCMS. Three-quarters had middle range or high morale. GDS score, type of housing, previous stroke, loneliness and number of symptoms, adjusted for age group and sex, explained 49.3% of the variance of total PGCMS score. CONCLUSIONS: a large proportion of the oldest old had high morale. The most important factors for high morale were the absence of depressive symptoms, living in ordinary housing, having previously had a stroke and yet still living in ordinary housing, not feeling lonely and low number of symptoms. The PGCMS seems applicable in the evaluation of morale among the oldest old.