Census Codes

A Brief Introduction to U.S. Census Bureau Occupation Codes and the Myth of Vocational Testimony

The United States Department of Commerce, U.S. Census Bureau has replaced the Dictionary of Occupational Titles (DOT) with the Standard Occupational Classification (SOC), which is located on the Occupational Information Network (O*NET). It is also possible to access the SOC directly.

 The Census Bureau's job codes are based upon SOC codes, but they are not identical to them. Census 2000 codes and their corresponding SOC codes. 

 The U.S. Census Bureau does not produce a standard table that shows occupations by full or part-time status. Furthermore, census codes do not report information on seasonal job categories such as agriculture and forestry.

 A list of the standard decennial census tables that the U.S. Census Bureau generates from occupational data, as well as the categories of information in each table is at Chapter 5 of File 3.

 Source: Letter from Charles Louis Kincannon, Director, Economics and Statistics Administration, U.S. Census Bureau.

 After reading Director Kincannon's letter, you should consider that some purveyors of reports allegedly addressing the numbers of jobs that might exist in the economy use "Census codes" instead of Dictionary of Occupational Titles codes. The Department of Labor does not obtain information about the existence of jobs based upon DOT codes and instead uses Census codes

Some VE's still testify that Census codes allegedly contain data about groups of DOT definitions (often called DOT numbers or DOT codes). Surprisingly to some, Census codes currently do not happen to relate to such information. There no longer is a link between Census Codes and the DOT.

The United States takes the census every ten years and also changes the census codes every ten years. The Census 2000 classifications were completely revised compared to 1990. Census Codes are now based on the 1997 NAICS (North American Industry Classification System) and the 1998 SOC (Standard Occupational Classification) coding structures. The 1990 system was based on the 1987 SIC (Standard Industrial Classification Manual) and the 1980 SOC (Standard Occupational Classification Manual). Source - Department of Labor. There is no longer a link between the 2000 census codes and the revised Fourth Edition of the Dictionary of Occupational Titles (DOT), which was last updated in 1991.

 The SOC was revised because it had not been updated since 1980. The revision was long overdue because of changes in the labor force and in the way economists view the labor force. Once the revision process was started, the SOC Revision Policy Committee quickly determined that, due to the extent of the changes being proposed, it was necessary to redesign the entire SOC. Go to http://stats.bls.gov/soc/home.htm for more details. Source - Source - Department of Labor.

The 2000 Census Bureau's job codes are different than the codes used in 1990. Which census data is the vocational expert using in your case? Look at a few of the important differences in the codes and compare to the codes on the report the VE brought to the hearing. Ask, are the VE's data are up-to-date? Where did the VE's data come from? When were they last updated? How were they updated? Did the VE know the census codes she brought to the hearing were 13-years old? If the report on the VE's lap says it is for the most recent quarter of this year, why is that report using 1990 Census codes? How did the publisher of that report use 1990 Census codes to produce statistically valid and reliable data about today's jobs? 

How does your VE explain this inconsistency and how does your VE know which DOT codes are associated with the 2000 Census Codes? See SSR 00-4p ("When vocational evidence provided by a VE or VS is not consistent with information in the DOT, the adjudicator must resolve this conflict before relying on the VE or VS evidence to support a determination or decision that the individual is or is not disabled. The adjudicator will explain in the determination or decision how he or she resolved the conflict. The adjudicator must explain the resolution of the conflict irrespective of how the conflict was identified")

The witness may assert that there was a "crosswalk"done between the 1990 Census codes and the 2000 Census codes, and thus, all is well. Ask, "who did the crosswalk?" "What assumptions were made during the crosswalk?" "Has this crosswalk been approved by the Department of Labor?" "How good was the fit in individual DOT codes transferred from one Census year to another?" How would you know, since the Department of Labor abandoned the DOT and no longer validates DOT-related information?" "How strong is the correlation between the physical demands reflected in the DOT for specific jobs, and the data reflected in the Census codes, now that 20 years has passed?"

In addition to the Department of Labor's BLS, the Census also explains there is no meaningful one-to-one relationship between the 1990 the 2000 census occupational data:

 For industry and occupation data, however, you cannot compare the categories directly across the two census years. The wording of the categories is different, and, even when the words appear to be the same, the definitions of the categories are sometimes different. For example, “Managers” and “Farming occupations” were defined differently in 1990 than in 2000, even though these words appear in tables from both censuses. These differences were caused by the revisions to the Standard Occupational Classification (SOC) and to the North American Industry Classification System (NAICS). The Census 2000 industry and occupation (I&O) categories were derived from these two standard classifications.

This chart gives a few examples of how the coding system has changed:

The Department of Labor frequently updates the SOC.  For example, the Department of Labor reported these changes for the 2010 SOC.  Which version is the VE using in your case?

The 2010 SOC system contains 840 detailed occupations, aggregated into 461 broad
occupations. In turn, the SOC combines these 461 broad occupations into 97 minor groups and
23 major groups. Of the 840 detailed occupations in the 2010 SOC, 359 remained exactly the
same as in 2000, 452 had definition changes, 21 had a title change only, and 8 had a code change
without a change in definition. Most of the definition changes (392) were editorial revisions that
did not change occupational content. Therefore, no substantive changes occurred in occupational
coverage for about 90 percent of the detailed occupations in the 2010 SOC.
Occupational areas with significant revisions and additions included
    o Information technology (minor group 15-1100 Computer Occupations)
    o Healthcare (major groups 29-0000 Healthcare Practitioners and Technical Occupations
        and 31-0000 Healthcare Support Occupations)
    o Printing (minor group 51-5100 Printing Workers) and
    o Human resources (minor group 13-1000 Business Operations Specialists)

See Department of Labor, What's ne in the 2010 SOC, at http://www.bls.gov/soc/soc_2010_whats_new.pdf