2

Many polls break national election polling down into White, Black, Hispanic, Asian, and none of those. However Hispanic is treated as an ethnicity on the Census.

In order to get data on the "some other race" category the race and ethnicity question on a poll would likely have the categories (White), (Black), (Asian, Pacific Islander or Native American), or (None of those). I grouped three into the last one because Native Americans and especially Pacific Islanders are too small to poll effectively without huge sample sizes or other techniques.

Have any polls been conducted that tried to measure "some other race" in the United States, in the context of a presidential election or congressional preference (generic ballot) preferably after or during 2016?

Note: this is asked because many Hispanic individuals select different things under the race box and I want to focus on one intriguing category. It also could be used to look at the nonwhite vote excluding white Hispanics. Analysis of Census data shows most who checked some other race selected yes on the separate Hispanic ancestry or origin question. There are also people who choose it who are Arab because the Census says they are white but they often don't identify as such. Leaving Hispanic off a poll would word it like the race question on the Census.

2
  • Is a census (which is a questionnaire) not a poll? The US doesn't have a voting database that records who voted for whom so I don't think this is answerable with real data. Is pre-election polling (which has clear problems, see 2016 election) acceptable?
    – uberhaxed
    Apr 25 at 19:03
  • 1
    FWIW, "some other race" is predominantly made up of Hispanics who don't identify as "white" typically viewing themselves as mestizo.
    – ohwilleke
    Apr 25 at 19:07

2 Answers 2

4

Have any polls been conducted that tried to measure "some other race" in the United States, in the context of a presidential election or congressional preference (generic ballot) preferably after or during 2016?

No. I've seen almost all of the relevant polls. None do so.

They generally separate out Hispanic as a race (sometimes also measuring both Hispanic and non-Hispanic parts of other races), and once you do that, the residual "some other race" number is too small to give statistically significant results.

It might be buried in non-reported cross-tabs somewhere, however, or discernible by calculation by backing out Hispanic white from a total for Hispanic. But none make that calculation themselves.

Another complication is that many polls in localities with low Asian-American percentages lump everyone who is not white, black, or Hispanic (without a racial cross-tab) into one pot, so you get Asian-Americans and Native Americans blended in too. Polling is typically less fine graded than census data.

-1

Hispanic is very broad and somewhat ill defined label. It includes Mexicans who arrived in the US as economic refugees, Cubans who arrived as political refugees, Nicaraguan and Salvadorians fleeing violence.

The press makes it seem like every Hispanic either came to US as an illegal immigrant or descended from illegals. In fact Hispanics came to US legally obeying our immigration laws.

As for race there is also a considerable difference from an indigenous person verses a someone of Spanish descent. Also there are large population groups such as people of Japanese decent in Peru. Would this population be considered Asian or Hispanic? They are definitely of Japanese descent, but were raised and adopt much of culture of Peru.

Being so diverse they don't reliably vote for any single group.

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .