The researchers report that adult sleep averages have increased about three hours per week over the last decade, up from 56 to 59 hours. They based their analysis on data from time diaries collected jointly by the U.S. Bureau of Labor Statistics and U.S. Census Bureau. Respondents aged 18 to 64 were asked to report all their activities the previous day in chronological order.

The Maryland study, Not So Deprived: Sleep in America, 1965-2005, identifies long working hours as the main 'thief' of sleep, but concludes that most Americans still manage to average at least the 'proverbial' eight hours of rest each night.

"Many Americans work too much, but most don't seem to be cutting corners on their sleep to do so," says University of Maryland sociologist John P. Robinson, a pioneer time-use researcher and the lead author of the study. "Lots of people may feel like they're on a 24/7 treadmill. But the picture of the typical American as sleep-starved is not consistent with what they report in their time diaries."

Even so, working hours remain the prime predictor of sleep averages. Heavy work schedules, for example a second job, can reduce sleep averages by up to 10 hours per week. Age and gender differences disappear when working hours are taken into account.

Time Diaries

The Maryland team analyzed data collected jointly by the U.S. Bureau of Labor Statistics and the U.S. Census Bureau annually between 2003 and 2005. More than 37,000 adults aged 18 to 64 spent about 10 to 15 minutes on the phone describing their activities the previous day, hour-by-hour.

Also, the Maryland researchers compared these findings with smaller-scale time diary research conducted by Robinson between 1965 and 2001 at roughly ten-year intervals. The samples in these earlier studies varied in size from 1,200 to 10,000.

Specific Findings

Sleep Patterns 1965-1995: There was little change in sleep averages during this period, particularly in comparison to the far larger shifts in time spent on housework, child care and watching TV. "The proverbial figure of eight hours per day (56 hours per week) has remained close to the diary norm for those aged 18 to 64 in each national study between 1965 and 1995," the report says.

Sleep Patterns 2003-2005: The time diaries collected by the federal government on an annual basis between 2003 and 2005 showed rising sleep averages - 8.2 hours on weeknights, 8.9 on Saturday and 9.5 on Sunday, a total increase of about three hours per week.

"While these recent increases are statistically significant, we're approaching them with some caution," says Maryland sociologist Steven Martin, co-author of Not So Deprived. "The numbers didn't change for more than 30 years. We want to see if these increases hold up in the long-run."

Recent parallel diary data from Canada show a similar increase in sleep averages of about three hours a week.

Television: Compared to time diaries collected in 1965, TV viewing no longer eats into adults' sleep. The latest studies show that people who watch more TV also get more sleep. "This original thief of sleep time has now evolved into its main companion or partner in providing relaxation and escapism from our waking life," says the report.

It's How You Collect the Data

The Maryland findings contrast with a series of annual surveys conducted by the National Sleep Foundation, which put adult sleep averages at less than seven hours per night. By comparison, the Maryland findings exceed these estimates by almost 10 hours per week, or more than an hour per night.

Robinson says methodological differences may account for this. The Foundation research relies on answers to more general survey questions - how much sleep respondents get each night, or times they go to sleep and get up.

"Feeling tired and stressed can color your answers to a survey," Robinson says. "Rather than asking about vague and ambiguous time periods, the diary is more precise, requiring respondents to recall only what they did on a single specific day."

Also, Robinson says the Foundation sampled only employed adults, which results in lower sleep averages.

"Obviously, there's a significant gap in these findings, but both sets of figures are consistent with pressured lifestyles," Robinson concludes. "Sleeping more may itself be a sign that waking hours are increasingly hectic and tiring. Also, these sleep results should not be used to minimize the severe problems many people face from insomnia, truncated sleep or inadequate 'down times.'"

An earlier version of the University of Maryland study, Not So Deprived: Sleep in America, 1965-2005, was presented at an academic meeting, the International Association of Time-Use Researchers (IATUR) in October 2007.

Story Source:

Materials provided by University of Maryland. Note: Content may be edited for style and length.