We examined the relationship between the timing of habitual physical activities and sleep quality in older adults. The subjects were Japanese community-dwelling older adults (n=49, average age 70.1±3.5 years; men: 36.7%). We measured habitual physical activity using a 3-axis accelerometer (HJA-350 IT, Omron) for a week. Timing of physical activity was classified into the following three periods: (1) morning: waking to 11:59, (2) afternoon: 12:00 to 17:59, and (3) night: 18:00 to bedtime. We also categorized the intensity of habitual physical activity during 2 sessions as either (1) low (1.6-2.9 METs) or (2) moderate-to-vigorous (≧3.0 METs) intensity. The subjective sleep parameters were assessed using the Pittsburgh Sleep Quality Index (PSQI). We used a forced-entry multiple regression analysis to investigate the relationships between subjective sleep parameters and the timing of physical activities. Forced-entry multiple regression analysis revealed that sleep latency and PSQI global score were positively correlated with low-intensity physical activity at night. However, there was no significant correlation with moderate-to-vigorous activity. These results suggest that low-intensity habitual physical activity at night would be one useful and modifiable factor to improve sleep quality in the elderly.