The United States Census stands apart from those of other nations on questions of race and ethnicity, according to research by New York University sociologist Ann Morning. Her findings were presented to the United Nations in late July.

Morning, who examined census data from 138 countries, found the following anomalies between the U.S. Census and those compiled by other nations:

  • The U.S. is the only country that devotes an ethnicity question to identifying only one (non-indigenous) ethnic group: Hispanics. No other country targets a single ethnic group in this way.
  • The U.S. is the only country that treats “ethnicity” and “race” as separate concepts that merit separate questions.
  • The U.S. is among a small proportion of the studied countries that uses the concept of “race” on the census (14 percent).

“The United States’ unique conceptual distinction between race and ethnicity may unwittingly support the longstanding belief that race reflects biological difference and ethnicity tracks cultural differences,” said Morning. “In this framework, ethnicity is perceived as socially produced, but race is an immutable facet of nature. Consequently, walling off race from ethnicity on the census may reinforce physical interpretations of race and preclude understanding of the ways in which racial categories are also socially constructed.”

By contrast, Morning reports that other countries, such as Australia and New Zealand, distinguish general ethnicity questions (“What is the person’s ancestry?” and “Which ethnic group do you belong to?”, respectively) from those that refer to indigenous status (“Is the person of Aboriginal or Torres Strait Islander origin?” on the Australian census; the New Zealand census asks, “Are you descended from a Mäori [that is, did you have a Mäori birth parent, grandparent or great-grandparent, etc.]?”).

“Censuses from these and other countries that are demographically comparable to the United States illustrate several ways in which ethnic enumeration could be streamlined or expanded for future U.S. Censuses,” Morning said. “These approaches highlight the importance of clear conceptual goals in designing ethnicity questions that obtain the desired information, yet do not burden respondents with unnecessary overlap that may cause confusion.”

She also points to the United Kingdom’s census, which uses a racial framework (white, Asian, black) to structure its request for more detailed national/ethnic identifiers (British, Irish, Indian, Pakistani, Caribbean), as a model the U.S. could adopt to reduce the number of its census ethnicity items.

Of the 138 national census questionnaires Morning examined, 87 countries (63 percent) employed some form of ethnic census classification. North America (82 percent), South America (82 percent), and the South Pacific (84 percent) were the regions with the greatest propensity to include ethnicity on their censuses. Asia’s tendency to enumerate by ethnicity was close to the sample average (64 percent), whereas Europe and Africa, both 44 percent, were much less likely to do so.

The 138 states and territories surveyed included 18 in North America (49 percent of the region’s nations, including Central America and the Caribbean); 11 in South America (79 percent); 18 in Africa (32 percent); 36 in Europe (73 percent); 36 in Asia (72 percent); and 19 in the South Pacific (76 percent). However, not all countries conducted a national census in the 1995-2004 period under study.

Press Contact