When data files from census or other governmental surveys or records are used in research, codes for occupations often are already supplied. In general, the Integrated Public Use Microdata Samples (IPUMS) maintained at the University of Minnesota furnish the most common and most useful source of occupational data for constructing NPB scores and using them in research. Nam-Powers-Boyd occupational status scores are pre-calculated by IPUMS staff. These scores are available in data sets based on historical decennial censuses and on American Community Surveys which replaced the long-form census as the source of occupational data in the 21st century. (Pre-calculated NPB variables are not available in Current Population Survey data sets from IPUMS due to limited CPS sample size and consequent inability to provide reliable information for less common occupations.) To calculate your own Nam-Powers-Boyd scores for occupations from any downloaded decennial census or American Community Survey data from IPUMS, consult the links below; be sure to study the ReadMe instructions file before attempting to use the actual software programs provided.
Occupational Codes and Categories
If you prefer to use pre-calculated Nam-Powers-Boyd scores available directly as variables in downloaded IPUMS census data, the following explanation of those calculations may be helpful. Pre-calculated NPB scores available in IPUMS data sets deal with the difficult problem of continual evolution of the labor force (with some occupations disappearing from the work force and new ones taking their place) by using Census Bureau-assigned occupation codes for specific baseline years to create standard occupation categories and codes (see explanations of variables occ1950, occ1990 and occ2010 at the IPUMS USA web site for details). IPUMS uses these standard codes to create a harmonized framework of occupation categories to which actual occupations from each Census or ACS data source are assigned, to allow comparative study of such categories over time.
For historical research, occ1950 codes have been applied to occupations listed in census records reaching well back into the 1800s. However, the occ1950 codes are not well-suited to study of data for 1980 or later, after major reorganization of occupational classifications at the Census Bureau. The occ1990 codes, based on the 1990 Census Bureau classification of occupations, are available for IPUMS data sets from 1950 up to the present, including both decennial Census and American Community Survey (ACS) data. The occ2010 codes, designed primarily for use with ACS data after the start of the 21st century, collapse into other related categories some occupational categories with less than 10,000 workers counted in the 2010 American Community Survey. Since these collapsed categories are not available in census data beginning with the 2012 ACS, these categories also are collapsed for the occ2010 variable in earlier IPUMS data sets to create a uniform harmonized set of codes reaching back to 1950. Specific research projects must evaluate the different strengths and limitations of these alternate baseline occupational standards, as well as unharmonized occupational variables for specific years, in the light of the focus of contemplated analysis.
The link below includes a list of over 30,000 occupation descriptions given on census and survey forms and the detailed occupation codes assigned by the U.S. Census Bureau, to which we have attached the NPB-OSS scores for 2000.
You can download the coding aid file here:
https://cosspp.fsu.edu/popcenter/wp-content/uploads/sites/36/2025/06/namcoding.xls“
Ranking Occupations by Educational Attainment and Earned Income
To produce Nam-Powers-Boyd scores for occupations from any set of codes and categories, IPUMS researchers and other scholars follow several steps established by Charles Nam, Mary Powers, Monica Boyd and their colleagues over the second half of the 20th century. For additional details about the development of these scores, see citations in the subsection Origins and Development of Nam-Powers-Boyd Measures in the “References” section of this Archive.
The first step involves restricting analysis to the experienced civilian labor force of the United States. The IPUMS variable wkswork1 (or in some cases wkswork2) allows elimination of respondents who did not work at least one week during the year before data were collected. The IPUMS occupation variable (occ1950, occ1990, occ2010 or some other coding of occupation) allows elimination of respondents reporting military occupations. These two steps produce a sample representing the experienced civilian labor force of the United States.
The second step calculates an interpolated median educational attainment and a separate interpolated median earned income for persons in each occupational category. Census codes for education come from the IPUMS variable educ which has less than a dozen code categories, and so produces many tied values across occupations for an ordinary median due to massive heaping of responses at educational levels such as high school or college graduation. The interpolation process estimates fractional values for the educ variable within each occupational group, minimizing the problem of tied values across occupations. The same interpolation is applied to the IPUMS variable incearn for personal income earned from wages, salaries, farms and businesses.
The third step (taken first for education and then separately for earnings of each occupation group) rank-orders all occupations from lowest to highest interpolated median values. Properly-weighted counts (using person-weights defined by the Census Bureau for a specific microdata sample) for persons in each occupation group are cumulated, from the lowest-ranked to the highest-ranked occupation, so that the final total equals the size of the total experienced civilian labor force. An NPB education score for a given occupation divides the cumulated total for all occupations ranked lower on median education by the size of the total labor force. An NPB earnings score for a given occupation divides the cumulated total for all occupations ranked lower on median earnings by the size of the total labor force.
Importantly, occupations often will not be ranked in the same order in the education rankings as in the earnings rankings. Also importantly, at this step the actual interpolated education and earnings values disappear from the score results, which reflect only the sizes and rank ordering of the occupational groups. Thus these scores are measures of status relative to other occupations rather than reflecting any absolute measures of education or earnings. By convention, these two proportions reflecting the share of the experienced civilian labor force falling below the interpolated median education or earnings for a given occupation often are multiplied by 100 to represent percentage scores.
The final step averages together the education and earnings NPB scores to give the final Nam-Powers-Boyd Occupational Status Score for an occupational group. This step reflects the original intent of the scholars who developed the measure.
“If education is an indicator of the prerequisite required for an occupation and income is a measure of the reward that society will bestow on an occupation, then occupation may be viewed as an intervening activity between these two variables and hence a good single indicator of status.” (Nam & Powers 1983, 15)