Israelis immigrated there from all over. A lot came from Germany and Russia. But some were already there, and some came from nearby. And after the state of Israel was declared, and the nearby nations went to war with it, the Jews in those nations often became persona non grata, and a lot of them decamped to Israel. So there’s are quite a lot of Sephardic Jews in Israel. And my understanding is that they are culturally distinct from the Ashkenazi Jews who are still a majority of Israeli Jews. So… I think some Israelis are and some aren’t “MENA”. At least, that’s my best guess.
That one’s additonally complicated because even Jews who identify as white may be identified as non-white by some antisemites. Even the ones with blonde hair, blue eyes, and ash-pale skin.
And as you say, unless “mixed” is an option, there are going to be a lot of people who won’t want to jam themselves into a category that doesn’t, on its own, fit them.
– if the demographics are useful, and they may well be, how about this: list everything they can think of (including Ashkenazi and Sephardic and the smaller groups); add “other” because no matter how long the list somebody will be left out, and also because some people don’t know – and then let people pick up to, say, four or five choices? @puzzlegal can pick Ashkenazi and white, I can pick Ashkenazi and Mediterranean, somebody else can pick Hispanic and Black and Chinese, somebody who doesn’t know all their ancestry can pick, say, white if most people think they look white and Other for whoever else might be in there, and so on.
ETA: that would lead to complicated work for the statisticians. But the results they got, while requiring a lot more work, would also be a lot more accurate.
But is it good for the census bureau to keep track of our demographics in such minute detail?
“Israeli” is not a “race” any more than “New Yorker” is. As for where it is located, Israel is pretty obviously in Arabia and the Middle East, isn’t it?
My opinion is that if the Census Bureau has or wants a (non-anonymous) list of Jews, or a list of Blacks, list of Chinese… then that is not so great. In fact, even anonymised statistics can be misused, and I am sure someone will find a way to do that.
The checkboxes are suggestions while the text boxes are for precision. Either way there’s no way anyone can hammer out an accurate system.
I’m looking at the sample and some might balk at South Asians grouped in with East Asians or point out that Sudan (a place who’s full name literally means “(land of the) Black people” in Arabic) more then qualifies for two of those categories.
We are measuring current sociological behaviour here, there’s really no objective facts involved; as long as this is acknowledged (above all else), sure, let the census make an attempt at accommodating people’s reflections on their self-identity, personal history, and ethnic presentations.
I don’t recall the census we filled in as being quite that granular. This is the output of the ethnic grouping question from 2021 and it seems that it doesn’t quite go down to that level of detail.
AFAICT the level of detail isn’t really an issue, as long as it’s not specific enough to pick out individuals from aggregate data? I don’t see how it is worse for individual data privacy for census data to indicate that, say, X percent of those surveyed identify as Chinese and Hispanic than that Y percent identify as “Other”.
And from the point of view of government policy to serve the public, more demographic detail is better:
I suppose the question isn’t about the aggregate data as much as about the individual detail; which is, if I understand it right, kept, but not made public (or supposed to be used as individual data by the government) until 72 years after the census was taken; by which time, they thought when they originally set that 72-year limit, nearly everyone in it would be safely dead.
What if the government decided to change the law, because some future administration did want, for nefarious purposes, a list, say, of everyone identifying as Jewish, or as atheist, or who had changed their gender or identified with a non-binary gender?
I’m not all that worried about this, personally; because I figure that if things get that bad I’m screwed anyway. But I can understand if somebody is worried about this.
The Canadian census has “Ethnic or Cultural Origin” with a few hundred different response groups, some more useful/specific than others (“British Columbian” isn’t terribly specific unless you read it as a sub-class of British-origin Canadian). There is a lot of detail breakdown for various indigenous groups, including Metis, which I don’t think the US census does, as this is likely of more significance for Canada.
Census Profile, 2021 Census of Population
Do they let you choose more than one?
I can’t remember all the options they showed as check boxes on my form (I ticked “Canadian”), but there was a place to enter your own self-identification, which is where the varied responses come from. It’s interesting that StatsCan made no attempt to “clean up” the responses.
As a practical matter, I wonder what purpose this actually serves, beyond trivia? Yes, I have read the descriptive paragraphs provided here, but I’m still left with the question, “to what end?”
In a binary sense, does inclusion in these finer categories put one in a “protected class” vs. not a protected class? If so, then why not just have a yes/no answer to “Are you in a protected class?” And does the protected status of an individual change with these new designations?
Otherwise, is this about where taxpayer dollars and manpower are divvied up? If so, is the result different due to these additional categories?
Without clarity on these questions, I’m left with the sense that this will provide additional colored pie pieces on a chart, but not much more.
As the most simple example, If someone says “the X community is disproportionately affected by Y condition,” and Y is perhaps a public health matter, or a civil rights matter, or an employment rights matter, or an information deficiency problem, and if Congress enacts funding for resources to address that matter, then, if nothing else, then the federal government has some basis for starting to allocate where those resources might be spent. What’s the point of allocating resources to Houston, for example, if no significant population of the X community lives there?
Let’s take a more traditional example (numbers are made up): Congress decides that chinchilla ranchers deserve government subsidies, and appropriates funding for that program, and then allocates them by state. Say $50 million is appropriated, and each state gets $1 million. Well, it turns out that Delaware, Maine, and Rhode Island have zero chinchilla ranches, but they now have $3 million to spend on chinchilla subsidies. Meanwhile, Arizona and New Mexico have 80 percent of the country’s chinchilla ranches, but are getting only $2 million–4 percent–of the subsidies. If you had data on where the chinchilla ranches were, you could have made a more logical allocation decision.
I think the option to choose more than one is necessary, but it might really water down the results so much they lose meaning. My daughters are approximately equal parts Native American, Mexican American, Eastern European Jew, and Scottish. * I’m sure many families have similar blendings.
*The Scottish part I just learned recently. All my life I thought my mother’s family were Irish in origin. Turns out they left from Ireland, but were not Irish. Then went from Scotland, to Ireland, to Canada, and then to the US. I probably answered questions like this wrong all my life. So, the stats are off people!
Believe it or not, I grew up on a chinchilla “ranch.” Had to feed and water those vicious things until I was 13 years old.
Carry on.
They’re adorable!
As someone who worked the last census, I’d say it doesn’t really matter much. The white folks identified themselves as white and the not-white folks pretty much “preferred to not answer.”