Historical comparability of occupation and industry data from the Current Population Survey
BLS publishes both employment and unemployment data by occupation and industry from the CPS.
The occupational classification reflects the type of job or work that the person does, while the industry classification reflects the business activity of their employer or company. The occupational and industry classifications are based on a person’s sole or primary job, unless otherwise specified. For the unemployed, the occupation and industry are based on the last job held.
Changes in the way industries and occupations are defined over time affect the comparability of historical data. As indicated below, some of the changes in classification represent complete breaks in the time series.
Current occupational and industry classifications
The Current Population Survey currently uses the 2010 Census occupational classification and, beginning with data for January 2014, the 2012 Census industry classification. These classifications were derived from the 2010 Standard Occupational Classification (SOC) and the 2012 North American Industry Classification System (NAICS), respectively, to meet the special classification needs of demographic household surveys. The Census classifications use the same basic structure as the SOC and NAICS, but are generally less detailed.
Most recent changes in the occupational and industry classifications
The 2010 Census occupational classification was introduced with CPS data for January 2011 and replaced an earlier version that was based on the 2000 SOC. Historical data were not revised. As a result, CPS occupational data beginning with January 2011 are not strictly comparable with earlier years. Although the names of the broad- and intermediate-level occupational groups in the 2010 Census occupational classification remained the same, some detailed occupations were re-classified between the broader groups, affecting comparability with earlier data.
At the detailed occupation level, there are numerous differences between the 2010 Census occupational classification and the earlier classification. With the introduction of the 2010 classification, dozens of new detailed occupations were introduced, the scope of many other occupations was redefined (although the titles did not always change), and some occupations were discontinued—all creating breaks in series comparability. For more information on the changes, see the XLS file “2010 Census Occupation Codes with Crosswalk” available from the Census Bureau.
The 2012 Census industry classification was introduced with data for January 2014. It replaced an earlier version based on the 2007 NAICS. The differences between the 2012 and 2007 industry classifications are minor and do not involve re-classification of industries between the broader industry sectors. Several industry titles were revised in the 2012 classification with no change to the industry definitions. In the Manufacturing, Wholesale, and Retail sectors, there were some minor definitional changes to a few detailed industries. Historical data were not revised with the introduction of the Census 2012 industry classification.
The 2007 Census industry classification was introduced with data for January 2009 and used through December 2013. It replaced an earlier version based on the 2002 NAICS. The differences between the 2007 and 2002 industry classifications were relatively minor and did not involve re-classification of industries between the broader industry sectors. Several industry titles were revised with no change to the industry definitions. In the Information sector, there were some minor definitional changes to a few detailed industries. Historical data were not revised with the introduction of the Census 2007 industry classification.
There also were minor differences between the 2000 and 2002 Census industry classifications. The differences were confined to the Wholesale, Retail, and Information sectors, primarily Information.
Earlier data comparability and availability
Prior to the introduction of the 2002 Census occupational and industry classification systems in 2003, CPS data used the 1990 Census occupational and industry classifications. The composition of detailed occupations and industries was substantially different in the 1990 systems, as was the structure for aggregating them into major groups. As a result, any comparisons of data on the 1990 classifications with data using later classifications are not possible without major adjustments (see the section on conversion factors below).
CPS data for 2000–2002 were originally produced and published using the 1990 Census classifications. When the 2002 classifications were introduced in 2003, BLS retroactively revised the January 2000–December 2002 data to extend the history available on the new classifications. The 2000–2002 industry data had to be classified with the 2000 Census industry classification, rather than the 2002 classification, but there are only minor differences between the two (see above).
Historical employment series on the 2002 Census classifications are available at broad levels of occupational and industry aggregation back to 1983. BLS constructed these historical series based on the distribution of employment for the years 2000–2002 re-coded from the 1990 classifications to the 2002 classifications. Further information and instructions for accessing the reconstructed series are available on the BLS website.
Historical employment series at the detailed occupational and industry levels on the 2002 classifications are available back to 2000 only.
For more information on the significant differences between the 1990 and 2002 classification systems, see the February 2003 article on changes introduced to the CPS in January 2003. (PDF)
Also available: Summary information about the occupational and industry classifications used from the 1970s through 2003. (PDF, page 8)
Conversion factors for the 1990 and 2002 Census occupational and industry classifications
BLS created conversion factors between the 1990 and 2002 Census classifications. These factors are based on 3-year average survey microdata (2000–2002) that were dual-coded to both the old and new classification systems. They are tabulated at the major group level as well as by the detailed classifications.
The conversion factors, provided in Tables 1-8 below, show the percent distribution of employment re-coded from one classification system to another. To illustrate, Table 1 shows employment estimates for each major occupational group under the 1990 classification and the percent of employment that was re-coded into the major groups based on the 2002 classification. These percentages could be applied to historical employment data on the 1990 classification to approximate employment on the 2002 classification. The resulting series provides a general employment trend over time but—given that it is only an approximation—the data are not likely to be useful for precise point-to-point comparisons or measuring actual change over time.
Important notice to users of the conversion factors: Occupation and industry coding of CPS microdata records is carried out by the U.S. Census Bureau. Detailed occupational or industry classifications will sometimes have to be imputed or "allocated" when sufficient information to assign a classification cannot be obtained. In addition, despite best practices, some classification error will occur. The conversion factors incorporate classifications that were allocated or, on occasion, mis-assigned. As a result, users of these conversion factors may notice atypical occupations and industries appearing in the detailed distributions, but they generally represent a very small portion of employment. Users also should note that at the detailed classification level, all occupations and industries with less than 10,000 average employment are excluded from the tables. Within the distributions, percentages less than 0.05 will not be shown separately at either the detailed or major group level.
Last Modified Date: March 30, 2018