On Race and Ethnicity

John F. Early’s excellent discussion in Saturday’s Opinion section on the statistical establishment’s struggles with defining and describing race is a question we all need to address seriously. It seems that OMB, which tried and failed to introduce similar thoughts for inclusion in the 2020 decennial census, seems to be trying it again on perhaps a more hospitable administration. Their clumsy efforts to somehow make America divide comfortably into five easy to label World groups (and two smaller ones) is an immense struggle with reality. Few outside Washington could respond to their new race question categories or care about them. A large part of it comes from having failed to convince Central and South Americans that they are something called an “ethnicity” and so OMB created “Some Other Race” as a categorization in order to make the data collection process fit.

My first experience with racial data collection was for the New York metropolitan area to meet federal requirements for travel data to support public investment analyses. At first, the survey we designed included race. We were told that was divisive and ordered to take the questions on race out. About a year after completing that survey I arrived in the Washington metropolitan area, again working on developing transportation survey data, and when I presented my draft designs I was asked why there’s no race questions in there. How can we tell what’s going on with different people? So race was back in. That was all about 50 years ago.

Some current comic effects of the OMB struggle:

The OMB says Europeans are White, except Spaniards, who are given as an example of Hispanics. Hmmm! That makes it a little clumsy in that the little country next to Spain on the Iberian peninsula called Portugal will be white, but not Spain? In Spain the Catalans have never wanted to be Spanish anyway, so maybe they’re still white – – should we ask them? My wife is from Catalonia and was labeled Hispanic at work years ago to up the percentage for government reporting.

Oh, then there’s the small matter of a place in South America called Brazil which is Portuguese-speaking and constitutes the majority of South American population, so are they white like the Portuguese or are we making them Hispanics to fit the geography? So maybe we make the Portuguese Hispanics because they almost share the language, and White Europe stops at the Pyrenees.

The Census Bureau seems intent on creating a "race" or "ethnicity" out of MENA - Middle Eastern & North African populations, a compartmentalization of the North African Middle Eastern, sort-of-Arabic world. It includes old friends such as Israel, Syria, Iran and Saudi Arabia. It’s unclear where MENA begins and totally not clear where it ends -- around Iran, Iraq, Afghanistan? Who might voluntarily identify themselves as a MENA?

Then there’s the Asians in America, about 5% of our population, but more than half of the world’s so we’ll squeeze Indians, Chinese, Japanese, Indonesians, all into that meaningless but convenient phrase. When they describe Asian, Indonesia is never mentioned. I’m sure they won’t mind! We could shift the Indonesians to the Pacific Islanders and Hawaiians Race category and up their population by about a quarter of a billion people.

Finally, will African-Americans and Africans be comfortable with that single word appellation?

These have become tangled labeling tools to make summary reporting convenient for the US statistical system to which probably none of the respondent populations would recognize. Perhaps geography is the easier answer? As Mr. Early suggests, maybe just forget the whole thing that has so clumsily, been drawn again and again.

Alan Pisarski is a writer, analyst and consultant in the fields of transportation research, policy and investment.