How are jobs, jobless rates calculated?

Pasadena Star News

By: Cynthia Kurtz

Posted 8/08/2013 

On the first Friday of every month, information about the number of jobs created or lost in the prior month and the current unemployment rate is announced by the federal government. The numbers last week were disappointing - just 162,000 jobs were created in July far below the number anticipated or needed. The national unemployment rate still slipped slightly lower to 7.4 percent, probably because of more people leaving the workforce rather than finding work. 

The health of our economy is tracked using these two figures. They drive public policy and consumer confidence. Where does this important data come from and how is it collected? It comes from surveys of households and businesses across the country, probably including a few in the San Gabriel Valley. 

Unemployment rates are computed from the Current Population Survey of 60,000 households. The households are carefully chosen to represent the U.S. population based on geographic sampling and reflecting urban and rural areas.

Each month one-quarter of the sample households are removed from the interview list and not sampled again for eight months. After the second year of participating, a household is removed for good. Therefore, there is both continuity and change built into the sample. About 75 percent of the sample is the same from month to month and about 50 percent from year to year.

The U.S. Census Bureau has been conducting this survey every month since 1940. It takes 2,200 Census Bureau employees to gather the data each month. 

Information about whether a person worked, searched for work, or was absent from work in the past week is collected about every sample household member over the age of 15. The survey gathers information on about 110,000 persons each month.

The data on jobs gained or lost is computed from the Current Employment Statistics Survey - a survey of 145,000 businesses and government agencies covering approximately 557,000 job sites. 

The establishment survey counts paid employees whether they are full-time or part-time. The sample includes 900 industries, businesses from every state and from 400 metropolitan areas. About 40 percent of the sample comprises businesses with less than 20 employees. When compared with numbers from the previous month, the number of new jobs can be determined by industry type.

The survey is conducted by the Bureau of Labor Statistics. Data is collected through a variety of methods tailored to meet the needs of the surveyed establishment - telephone, touch-tone data entry, internet, fax and mail are used. Some months you might hear there is a revised “new jobs” number from a prior month. Revised data is released when additional surveys are received from sampled employers.

Both surveys have strengths and limitations, but together they give us a gauge to measure the nation’s monthly economic changes. If you are ever asked to be a participant in either survey, say “yes.” It’s important information for setting long-term economic policy and that is important for everyone.