Sleep Statistics: How Americans Spend the Night
Published September 15, 2026 · StatsGalore
Average daily time recorded as sleeping [1]
Americans recorded an average of 9.03 hours a day in the sleeping category in the 2025 American Time Use Survey. That figure can look surprisingly high beside personal estimates of a normal night. The survey's definition explains part of the difference: it includes naps and spells of sleeplessness, and it covers the entire week and everyone age 15 and older. [1]
The numbers
| Measure | Daily average | Data year | Source |
|---|---|---|---|
| All people age 15 and older | 9.03 hours | 2025 | [1] |
| Men | 8.93 hours | 2025 | [1] |
| Women | 9.13 hours | 2025 | [1] |
| All personal care activities | 9.80 hours | 2025 | [1] |
What the sleep average actually measures
The sleeping category records time described in a diary of the previous day. It is a measure of how time is allocated, rather than a laboratory measure of unconsciousness. A respondent can spend time trying to fall asleep that belongs in this category even though a wearable device might record part of the same interval as awake.
The population also matters. The estimate includes students, employed people, retirees, and people outside the labor force. It combines weekdays and weekends. It should not be read as the typical sleep duration of an adult working a conventional weekday schedule.
[1]Comparing men and women
Women averaged 9.13 hours and men averaged 8.93 hours in the published table. Subtracting the rounded averages gives a difference of 0.20 hours, or about 12 minutes per day. This is a comparison between population averages; it does not predict the habits of an individual woman or man.
The table alone does not establish why the averages differ. Age composition, work arrangements, household responsibilities, and other characteristics can vary between the groups. Treating the difference as a biological explanation would go beyond what these descriptive numbers show.
[1]Hours in decimal form
Decimal hours are fractions of an hour, not hours followed by minutes. The 0.03 in 9.03 hours is about two minutes, so the rounded average is roughly nine hours and two minutes. A value of 8.93 hours is approximately eight hours and 56 minutes, not eight hours and 93 minutes.
Converting the figures can make them easier to picture, but it does not add precision. These values are survey estimates that have already been rounded. A difference of a few displayed minutes may not represent a statistically reliable difference between groups.
[1]Using this statistic fairly
This survey is useful for comparing sleep with other uses of time because the diary follows the same overall accounting framework. Sleeping is included in personal care; adding the sleep figure to the personal care total would double-count it.
For a question about sleep quality, diagnosed sleep disorders, or recommended sleep duration, a different source and measurement method are needed. Time recorded in a sleeping category does not, by itself, establish whether someone slept well or obtained enough restorative sleep.
[1]Sources & notes
BLS Table 1 defines sleeping to include naps and spells of sleeplessness. Figures are population means for all days of the week, not medians or medical recommendations. Minute conversions are calculated from the displayed, rounded hours.
The 2025 ATUS did not collect diaries for September 30–November 11 during the federal shutdown. Under the existing weighting method, later fourth-quarter diaries represent the missing days as well. See the BLS collection notice before making year-to-year comparisons.
Source review: 2026-09-15. Article publication and underlying data dates are separate. How we use statistics · Suggest a correction