Table of Contents | Previous | Next |
CURRENT POPULATION SURVEY (CPS)
The survey background information for the CPS covers two surveys: 1) the basic monthly survey, administered each month to collect primarily labor force data, and 2) the Annual Demographic Survey (ADS), formerly known as the March CPS Income Supplement, conducted annually—mostly in March—to collect data on work experience, income, and migration.
Purpose
The Current Population Survey (CPS) of about 50,000 households has been conducted monthly by the Bureau of the Census for the Bureau of Labor Statistics for more than 50 years. It is the primary source of information on the characteristics of the U.S. labor force. The sample provides estimates for the nation as a whole and is part of model-based estimates for individual states and other geographic areas. Government policymakers and legislators use CPS data as important indicators of the U.S. economic situation and as a tool for planning and evaluating many government programs. The press, students, academics, and the general public also use CPS data.
The Annual Demographic Survey or March CPS supplement is the primary source of detailed information on income and work experience in the United States. Each year, the Bureaus of Labor Statistics and Census issue publications based on this survey. “A public-use microdata file is available for private researchers, who also produce many academic and policy-related documents based on these data. The Annual Demographic Survey is used to generate the annual Population Profile of the United States, reports on geographical mobility and educational attainment, and detailed analysis of money income and poverty status. The labor force and work experience data from this survey are used to profile the U.S. labor market and to make employment projections” (http://www.bls.census.gov/cps/ads/adsdes.htm).
For more information:
http://www.bls.census.gov/cps/
Agencies/Institutions
The CPS is cosponsored by the Bureau of Labor Statistics (BLS) and the U.S. Census Bureau. The Census Bureau administers the survey.
Research/Survey Design
The CPS uses a research design with both cross-sectional and longitudinal elements because households are interviewed over time before being replaced by other households. The nature of the rotation pattern allows 75% of sampled households of the basic CPS to remain the same month to month, with 50% of the sample in common between one month and the same month a year later.
Households for the CPS come from a probability sample of housing units in each state and the District of Columbia, updated after each Census and with measures taken to include newly constructed units. Counties in each state are divided into primary sampling units (PSUs) of one or more contiguous counties; housing units are drawn from the most populous PSUs in each state as well as from PSUs randomly selected among PSUs in each state with similar characteristics. Following data collection, the weighting procedure includes steps that ensure that the proportion of people and households by state, age, sex, race, and Hispanic origin matches current Census projections.
One person (the household respondent) usually answers for all members of the household. However, telephone call-backs commonly occur to obtain certain types of information known only by someone else in the household.
For more information:
http://www.bls.census.gov/cps/
Date(s)/Periodicity
Respondents to the CPS are surveyed eight times: two periods of 4 consecutive months that are 1 calendar year apart. (There are thus 8 months between the fourth and fifth surveys of each household.) Questions in the ADS are asked once a year, primarily in March, although certain households in the expanded sample are interviewed in February or April. Different supplements are typically administered in other months. Administration of the CPS and ADS according to this regular schedule is ongoing.
For more information:
http://www.bls.census.gov/cps/
Population/Sample
“The CPS is the primary source of information on the labor force characteristics of the U.S. population. The sample is scientifically selected to represent the civilian noninstitutional population. Respondents are interviewed to obtain information about the employment status of each member of the household 15 years of age and older. However, published data focus on those ages 16 and over. The sample provides estimates for the nation as a whole and serves as part of model-based estimates for individual states and other geographic areas” (http://www.bls.census.gov/cps/bovrvw1.htm). The basic monthly survey does not include members of the Armed Forces, but the ADS collects data for military personnel who are living with at least one civilian adult. Because the housing unit is the sampling unit, all people age 15 and over currently living in the housing unit are surveyed in later months, even if they are new to the housing unit. People who move out of the housing unit are no longer followed, so it is possible that the respondents to the survey one month are entirely different from the respondents in a later month in which the housing unit is still in the sample (if, for example, one family moves out and another moves in).
The basic CPS consists of a sample of 60,000 occupied housing units and all nonmilitary people 15 and over within them. If the original residents of a housing unit move out while that unit is still in rotation, the new residents of the unit will be surveyed in subsequent months. The ADS surveys the 60,000 occupied housing units scheduled to receive the monthly survey in March as well as two additional groups of people: 4,500 Hispanic households identified the previous November (to improve Hispanic estimates of ADS constructs) and 34,500 households who form what is known as the State Children’s Health Insurance Program (SCHIP) sample expansion. The latter expansion, designed to allow the ADS to better measure the number of children in each state without health insurance coverage, involves oversampling non-Hispanic non-white households, non-Hispanic white households with children younger than 19, and households in states whose estimates of children’s health insurance coverage were the most unreliable. In total, then, the ADS sample includes about 99,000 households. Additional weighting is done to obtain estimates for households and families.
The response rate for the monthly survey is about 93% and for the ADS averages 80% to 82%.
“The effect of nonresponse cannot be measured directly, but one indication of its potential effect is the nonresponse rate. For the March 2002 basic CPS, the nonresponse rate was 8.3%. The nonresponse rate for the March supplement was an additional 8.6%, for a total supplement nonresponse rate of 16.2%” (U.S. Census Bureau, 2002, p. G-3). Nonresponse for the basic CPS in March 2002 was higher than normal, as the basic CPS usually enjoys a 93% response rate. Typically, about 55% to 60% of nonresponse owes to refusals, with noncontact taking up the vast majority of the remainder (BLS and Census, 2002, p. 16-3; http://www.bls.census.gov/cps/basic/perfmeas/typea.htm).
For more information:
http://www.bls.census.gov/cps/
Content Covered
The monthly CPS allows estimates of employment, unemployment, earnings, hours of work, and other indicators. These indicators are available by a variety of demographic characteristics, such as age, sex, race, marital status, and educational attainment, and by occupation, industry, and class of worker. Various other topics, including school enrollment, income, previous work experience, health, employee benefits, and work schedules, are estimated through supplemental questions.
The supplemental questions about previous work experience, income, and migration are asked in the Annual Demographic Survey. “Today, information is gathered [through the ADS] on more than 50 different sources of income, including noncash income sources such as food stamps, school lunch program, employer-provided group health insurance plan, employer-provided pension plan, personal health insurance, Medicaid, Medicare, CHAMPUS or military health care, and energy assistance. Comprehensive work experience information is given on the employment status, occupation, and industry of persons 15 years old and over”
(http://www.bls.census.gov/cps/ads/shistory.htm).
For more information:
http://www.bls.census.gov/cps/
Availability of Data for Public Use
Both the Census Bureau and the BLS regularly produce reports summarizing data from the CPS. The dataset can be searched or downloaded via FERRET (Census) and LABSTAT (BLS). See http://www.bls.census.gov/cps/datamain.htm.
The CPS website is http://www.bls.census.gov/cps/cpsmain.htm. The questionnaire for the Basic Monthly Survey is available at http://www.bls.census.gov/cps/bqestair.htm. The questionnaire for the 2002 Annual Demographic Survey is in Appendix D (D-1 to D-109) of the technical documentation for that survey, which is available at http://www.census.gov/apsd/techdoc/cps/cpsmar02.pdf.
Reference List for Users’ Guide, Codebooks, Methodology Report(s)
Bureau of Labor Statistics & U.S. Census Bureau. (2002). Current Population Survey: Design and methodology. Technical Paper 63RV. Washington, DC: Authors. URL: http://www.census.gov/prod/2002pubs/tp63rv.pdf
Technical Paper 63 Revised (63RV), on the design and methodology of the CPS, was issued in March 2002 and is available at http://www.census.gov/prod/2002pubs/tp63rv.pdf.
Technical documentation specifically for the March 2002 Annual Demographic File (issued November 2002) is available at http://www.census.gov/apsd/techdoc/cps/cpsmar02.pdf.
U.S. Census Bureau. (2002). Technical documentation: Current Population Survey. March 2002.
Washington, DC: Author. URL: http://www.census.gov/apsd/techdoc/cps/cpsmar02.pdf
A data dictionary for the March Income Supplement is available for household, family, and person variables; the dictionary, at http://www.bls.census.gov/cps/ads/sdatadic.htm, provides numeric codes for possible question responses and frequency distributions from the March 1995 survey.
Two different data dictionaries for the basic monthly survey are available online: one that provides variable descriptions by topic at http://www.bls.census.gov/cps/bdatadic.htm and one for the public use file that lists variables in a variety of ways at http://www.bls.census.gov/cps/basic/datadict/199801/bdatdict.htm. Neither data dictionary for the basic CPS includes frequency distributions.
| Table of Contents | Previous | Next |