Hours: Two surveys compared

January 06, 2005

The Current Population Survey (CPS) measure of hours worked is, on average, fairly close to the American Time-Use Survey (ATUS) time-diary estimates of hours when the diary day is in the CPS reference week.

Comparison of ATUS time-diary estimates of weekly hours to CPS estimates
[Chart data—TXT]

However, the CPS measure of actual hours in the reference week is about 5 percent higher than the hours data collected in ATUS for the other 3 weeks of each month. The fact that the CPS reference week was selected, in part, to avoid holidays both explains some of the difference and simplifies the task of tracking trends in work hours using CPS data.

Analysts must keep in mind, however, that a measure of monthly hours worked constructed from CPS weekly hours data would overstate actual hours worked during the month.

The data used here are from the Current Population Survey and the American Time-Use Survey. To learn more about their concepts, definitions, and data on working hours, see Harley Frazis and Jay Stewart, "What can time-use data tell us about hours at work?" in Monthly Labor Review, December 2004.


Bureau of Labor Statistics, U.S. Department of Labor, The Editor's Desk, Hours: Two surveys compared on the Internet at http://www.bls.gov/opub/ted/2005/jan/wk1/art04.htm (visited September 15, 2014).


Spotlight on Statistics: Productivity

This edition of Spotlight on Statistics examines labor productivity trends from 2000 through 2010 for selected industries and sectors within the nonfarm business sector of the U.S. economy.  Read more »