The Occupational Score Index (OCCSCORE) is a measure of occupational reward that is available across decennial census datasets from 1850 to 2000. It was developed at the Minnesota Population Center as part of the Integrated Public Use Microdata Series (IPUMS). The index is based on 1950 occupational classifications. As a first step in constructing the index, occupation codes for each decennial census dataset were re-categorized into the 1950 classification structure. This yielded one, common occupational classification across all decennial censuses. The IPUMS staff then quantified this common occupational structure using median total income for each occupation as reported in a 1956 Special Report published by the Census Bureau. The values for OCCSCORE are presented in hundreds of 1950 dollars. As an example, if median total income for economists was $20,000 in 1950, OCCSCORE would equal 20 for economists in all decennial census datasets.
OCCSCORE has two major uses. First, it provides a proxy for income in decennial censuses prior to 1960 that lack individual income data. Second, it provides a consistent measure with which to compare labor market outcomes from 1850 to 2000. However, OCCSCORE has four major shortcomings. First, it does not account for changes in occupational hierarchy across time. For each decennial census dataset, OCCSCORE conveys the rank ordering of occupations by income in 1950. However, the actual rank ordering of occupations likely changes over time. Second, the index does not account for variation in income within occupations. For example, in a given decennial census year, two individuals working in the same occupation have the same OCCSCORE value. However, based on items such as tenure and education, individuals with the same occupation may have very different incomes. Third, the index does not account for cost of living differences. An individual working as an economist in San Francisco, California, has the same OCC-SCORE as an economist in Omaha, Nebraska. However, their respective purchasing powers are clearly different. Finally, although the IPUMS staff has taken great care when constructing OCCSCORE, re-categorizing occupations into the 1950 classification is problematic. New occupations evolve over time, and the U.S. Census has periodically changed the occupational classification system.
The primary alternative to the OCCSCORE is a set of prestige indices. Prestige indices are typically based on regression models where income and education are used to predict responses to an occupational prestige survey. The predicted values from these models are then used to predict the prestige of an occupation. Unlike prestige indices, OCCSCORE only measures the monetary return to occupations, not status or prestige. Overall, prestige scores have been more widely used in social sciences research, because occupation augmented by information on income and education is thought to be a better indicator of socioeconomic status than just the monetary return to occupation. One disadvantage of prestige indices is uncertainty about models’ predictions of prestige.
Overall, OCCSCORE is a useful measure of occupational reward for research using censuses prior to 1960. For research focused only on 1960 or later, however, individual income measures are superior.
