Background Our aim was to describe and explore older adults' device-measured sedentary behavior and physical activity (PA) pattern by sex, age, education, marital status, body mass index, and physical function; and to assess agreement regarding fulfillment of PA recommendations, i.e. 150 min/week of moderate-to-vigorous intensity PA (MVPA), between device-measured and self-reported PA. Method We included 656 older adults (64% women), aged 66, 81-87 or >= 90 years from a Swedish population-based cohort study. The activPAL3 accelerometer provided information on sedentary behavior (sedentary time, sedentary bouts, sit-to-stand transitions) and PA. Stepping >= 100 steps/min was considered MVPA; standing and stepping < 100 steps/min were considered light-intensity PA (LPA). Self-reported PA was compared with min/week in MVPA and steps/day. Results On average, 60% of wear time was spent sedentary, 36% in LPA, and 4% in MVPA. Relative to men, women, had significantly (p < 0.05) more sit-to-stand transitions, spent 33 min/day less sedentary and 27 min/day more in LPA, and were more likely to report meeting PA recommendations, but showed no difference in steps/day, MVPA, or sedentary bout duration. Older age was associated with more sedentary time, lower MVPA and fewer steps/day. The prevalence of meeting PA recommendations was 59% device-measured and 88% by self-report with limited agreement between methods (Cohen's Kappa = 0.21, Spearman's rho = 0.28). Age differences were much more pronounced with objective measures than by self-report. Conclusions We found significant sex differences in sedentary behavior and time in LPA in older adults, but not in MVPA, in contrast to previous findings. Sedentary time increased with age, with small differences in accumulation pattern. MVPA time was lower with older age, obesity, and poor physical function. A majority of the participants > 80 years did not meet the PA recommendations. Given the strong relationships between sedentary behavior, PA and health in older adults, programs are needed to address these behaviors. Agreement between device-measured and self-reported fulfillment of PA recommendations was limited. Device-based measurement adds value to PA studies, providing richer and different data than self-report.