Technical Documentation
This page contains technical documentation and related information on the Current Population Survey (CPS).
Concepts and methodology of the CPS
Summarized documentation on the concepts and methodology of the CPS.
- How the government measures unemployment (HTML) (PDF)
- Quick Guide to Methods and Measurement Issues in the monthly Employment Situation report (HTML)
- Handbook of Methods, chapter 1, labor force data derived from the Current Population Survey
(HTML) (PDF)
- Technical notes to household survey data published in Employment and Earnings
- Introduction (Relationship and comparability with establishment and other surveys) (PDF)
- Collection and Coverage, Concepts and Definitions, Historical Comparability and Estimating Methods (PDF)
- Seasonal adjustment (PDF)
Comprehensive documentation on the design and methodology of the CPS, including a history of the survey (links to the Census Bureau website).
- Current Population Survey, design and methodology (Technical paper 66) (October 2006) (PDF 3.2MB)
- Previous versions:
- Technical paper 63, revised March 2002 (PDF 3.1MB)
- Technical paper 63, March 2000 (PDF 2.2MB)
- Additional information about the CPS, including documentation for survey microdata, is available from the Census Bureau
Historical comparability is affected by revisions to population controls, changes in occupational and industry classification, and other changes to the survey.
- Discussion of historical comparability; see pages 3-11 (printed pages 184-192) (PDF)
- Changes to data collected on unemployment duration introduced in 2011
- Revisions to the Current Population Survey Effective in January 2003, including new questions on race and Hispanic ethnicity, updated population controls, and new occupational and industry classifications (PDF)
- Expansion of the Current Population Survey Sample Effective July 2001, from Employment and Earnings, August 2001 (PDF)
- CPS redesign implemented in January 1994, including questionnaire changes and updated population controls.
CPS and CES employment differences
There are two monthly surveys that provide sample-based estimates of employment: the CPS, also known as the household survey, and the Current Employment Statistics (CES) survey, also known as the establishment or payroll survey. The establishment survey employment series has a smaller margin of error on the measurement of month-to-month change than the household survey because of its much larger sample size. However, the household survey has a more expansive scope than the establishment survey because it includes the self-employed, unpaid family workers, agricultural workers, and private household workers, who are excluded by the establishment survey. The household survey also provides estimates of employment for demographic groups.
- Report: Recent trends in employment from the BLS household and payroll surveys (updated monthly) (PDF)
- Quick Guide to Methods and Measurement Issues in the monthly Employment Situation report (HTML)
- Chart: Employment trends from two surveys (April 2012)
- Article: Understanding the employment measures from the establishment and household surveys (February 2006)
(Abstract) (Excerpt) (PDF)
How the government measures unemployment
Description of how the national unemployment statistics are developed from the Current Population Survey (CPS), written in non-technical language.
- How the government measures unemployment (HTML) (PDF)
Occupational and industry classification used in the CPS
The Current Population Survey currently uses the 2010 Census occupational classification and the 2007 Census industry classification. These classifications were derived from the 2010 Standard Occupational Classification (SOC) and the 2007 North American Industry Classification System (NAICS), respectively.
- Documentation on the Census occupational and industry classifications, with crosswalks to the SOC and NAICS. (HTML)
- Information on the availability of selected historical employment data constructed on the 2002 Census occupational and industry classifications (HTML)
- Documentation of the revisions to the Current Population Survey in January 2003, including the conversion to the 2002 Census classifications (PDF)
- Discussion of historical comparability, including changes in the occupational and industry classifications over time (PDF, page 8)
- More information on the Census occupational and industry classifications is available from the Census Bureau.
Population control adjustments to the CPS
In order to produce estimates from survey data, statistical weights must be applied. Population controls are independent estimates of population used to weight the CPS sample results.
The population controls are developed by the U.S. Census Bureau. They are based on decennial census information, and between decennial census years they incorporate administrative data,
such as birth and death statistics, along with the Census Bureau's estimates of net international migration.
The Census Bureau reviews and adjusts the population controls every year. BLS introduces the annual population control adjustments into the CPS estimates beginning with the January data.
The adjustment can either increase or decrease the population level, depending on whether the latest information indicates the population estimates have trended too high or low.
Conceptually, the population control adjustments represent the cumulative over- or under-estimation of the population since the last decennial census point.
The results from Census 2010 were incorporated into CPS population controls with the release of data for January 2012. CPS estimates for January 2000 through December 2011 reflect
population controls based on Census 2000.
- Description of recent adjustments to CPS population controls
- Article: Revisions to the Current Population Survey effective in January 2003 (PDF)
- Description of earlier adjustments to CPS population controls
- Discussion of historical comparability, including adjustments to the population controls over time (PDF, page 5)
The level shifts in the CPS labor force and employment series resulting from annual population adjustments can make it difficult for data users to compare changes over time periods that include these adjustments. As a convenience to its data users, BLS created research series that smooth out the level shifts in the labor force and employment estimates resulting from the January 2000 and subsequent population control adjustments.
- Labor force and employment research series, smoothed for population control adjustments made since January 2000 (PDF)
- Table: Civilian labor force smoothed for adjustments in population controls, seasonally adjusted (HTML) (XLS)
- Table: Employment smoothed for adjustments in population controls, seasonally adjusted (HTML) (XLS)
- Documentation on the methodology used to create the smoothed labor force and employment research series (PDF)
Reliability of estimates from the CPS
Statistics from the CPS are subject to both sampling and nonsampling error. When a sample rather than the entire population is surveyed, there is a chance that the sample estimates may differ from the "true" population values they represent. The exact difference, or sampling error, varies depending on the particular sample selected, and this variability is measured by the standard error of the estimate.
- Household data, reliability of estimates; see pages 12-19 (printed pages 193-200) (PDF)
Research articles
Browse recent BLS analyses of CPS data by topic: unemployment, labor force characteristics, earnings, and demographic characteristics.
Seasonal adjustment of CPS estimates
Over the course of a year, the size of the labor force, the levels of employment and unemployment, and other measures of labor market activity undergo sharp fluctuations due to such seasonal events as changes in weather, major holidays, and the opening and closing of schools. Because these seasonal events follow a more or less regular pattern each year, their influence on statistical trends can be eliminated by adjusting the statistics from month to month. These adjustments make it easier to observe the cyclical and other nonseasonal movements in the series. BLS regularly produces seasonally adjusted series for selected labor force data from the CPS.
- Seasonal adjustment overview (PDF)
- Description of recent modifications in the seasonal adjustment of national labor force statistics from the CPS
- Modifications for earlier years
- Article: Model-based seasonally adjusted estimates and sampling error (September 2005)
(Abstract) (Excerpt) (PDF)
- Article: An Evaluation of Concurrent Seasonal Adjustment for the Major Labor Force Series (August 1987)
(PDF)
Last Modified Date: August 9, 2012
