History of Racial Classification in the US Census: 1790-2010

Every census from 1790 to the present has recognized the racial or color category of “white.” But there has been interesting variation over time in how census enumeration methods have placed people in this category. There also are interesting variations in how the data for whites are reported. The following are some selected points of interest I think are worth noting. (For more detail on these points, I list two sources I drew on at the end; they may be helpful for getting started.) While the census has alway recognized the race or color category of “white”, here are a few variations over time that I am aware of and that may be interesting to some:

1. Early censuses 1790-1820 had no instructions or set categories for how to enumerate “color” and there were no formal “schedules” or “forms” on which enumerators were required to list color. Enumerators used their own judgment about listing color in their records with no guidelines from the census. When the data were coded and tabulated, persons were treated as white by default if no other specific color was provided.

2. 1830 brought the first form in which there was a specific place for enumerators to list color. Still no guidelines from the census on how to enumerate people on color.

3. 1890 was the first census in which “white” was not the default coding if race was not mentioned in enumerator records (1790-1820) or was left blank on the form (1830-1880). In this census, white had to be specified on the form.

4. 1790-1950 the census enumerators made the judgment on color. Starting in 1960, the respondent made the judgment.

One interesting trend here is how Hispanics have ultimate been classified on race after 1950. The census instructions are clear that most persons of Hispanic background should be coded white under census coding guidelines. Through 1950, census enumerators followed those instructions. In the transition to self report, Hispanic respondents sometimes disregarded the census instructions and classified themselves as something other than white. The rate of doing so jumped sharply in 1970 and by 1980 exceeded 50%. It has stayed that high since.

This caused an appreciable drop in the white population in the Southwestern states from what it would have been had census practices of 1950 been continued.

5. Around the decades of high immigration — 1880-1930 – distinctions among whites were of great concern. Consequently, tabulations for whites were regularly broken out by native and foreign born due to the concerns about differences among native whites and new immigrants of the era. That practice was discontinued after 1940.

So, the word “white” has remained a constant. But how people are assigned to the category and how those data are reported has varied quite a bit over time.

Mark Fossett, Texas A&M University

For further information and other sources relating to these issues, see the following works.

Reynolds Farley. 1970. The Growth of the Black Population. Markham.

Joel Perlmann and Mary C. Waters (eds.). 2002. “The New Race Question: How the Census Counts Multiracial Indivuals. Russell Sage Foundation.