German ancestry is #1, English ancestry is #2

[King of Kensington]

Very few ethnic Russians emigrated.

In the 1930 census, 64% of those born in Russia had a Yiddish mother tongue.

https://www2.census.gov/prod2/decennial/documents/16440598v2ch08.pdf

There were also ethnic Germans.  Many Germans from Russia went to the Great Plains, especially North Dakota.  Norwegians and Germans from Russia were the main settlers of North Dakota.

[Builder Refused]

Very few ethnic Russians emigrated.

In the 1930 census, 64% of those born in Russia had a Yiddish mother tongue.

https://www2.census.gov/prod2/decennial/documents/16440598v2ch08.pdf

There were also ethnic Germans.  Many Germans from Russia went to the Great Plains, especially North Dakota.  Norwegians and Germans from Russia were the main settlers of North Dakota.

The movement inside of Europe and existence of different polities is interesting in and of itself. Like I've heard amongst Jews the German ones were considered more urbane because they reformed their way of giving mass or whatever, and they lived in cities rather than mostly rural places.

You also get the Scotch-Irish and French-Canadians from this, and I guess the Hispanic population is basically this too. Also, it's really weird they grouped together English and Celtic that's like grouping German and Polish like just, why?

[jimrtex]

I wonder what’s driving the recent uptick in English identification? More people taking DNA tests and realizing they’re not Irish? Less people identifying as American? I answer English on the census and I know most people from my hometown have English ancestry but the ACS shows self English ID at less than 15%.

The census form has changed over time, as well as the presentation.

Up until 1970 the Census asked for place of birth for the person, and place of birth of parents. Instead of identifying as "American" you would say that you were born in Kentucky, your mother was born in Kentucky, and your father was born in Virginia.

The distinction was really between foreign-born and foreign-stock and native born.

With the cutoff of large-scale immigration after WWI, fewer and fewer persons were actually born outside the United States. Most persons migrate as an adult in their 20s or perhaps as a small child. If you are in your 30s, you are more likely to be established in your employment, your housing, your children's schooling.

So let's say you were 25 in 1915 when you immigrated from Italy. In 1980, if still living, you would be 90. If you had small children when you immigrated, they would be 70 years old. Their younger siblings, would report they had been born in the USA with foreign-born parents.

By 1980 the younger generation would have been educated in the USA, worked in the USA, and possibly been in the USA armed forces.

But this would just be for the cutoff. A median immigrant might have migrated in 1905.

So the 1980 Census replaced the parents born in question with an ancestry question. I don't know that the Census Bureau is all that interested in the question. It gets ambiguous responses. Justification is that helps with enforcement of anti-discrimination laws. Note that while Korean, Mexican, and Afro-Amer are given as examples, they are not reported by the Census Bureau. In fact there is a huge number of "other responses". Apparently the Census Bureau (DOJ actually) believes that race and ethnicity are more reliable for discrimination practices.

Examples given were: Afro-Amer, English, French, German, Honduran, Hungarian, Irish, Italian, Korean, Lebanese, Mexican, Nigerian, Polish, Ukrainian, Venezuelan, etc.

The list is alphabetical. It is possibly coincidental that English and German were near the front of the list, which included the three most common historical origins: English, German, and Irish.

In 1990 the list was: German, Italian, Afro-Amer, Croatian, Cape Verdean, Dominican, Ecuadorian, Haitian, Cajun, French Canadian, Jamaican, Korean, Lebanese, Mexican, Nigerian, Irish, Polish, Slovak, Taiwanese, Thai, Ukrainian, etc.

I don't know why the changes were made, but English was left off the list. What does a respondent infer? Note there was a big increase in reporting of Croatian and Slovak ancestry. The list is still alphabetical, except German, Italian, and Irish are out of order.

In 2000, the list was: Italian, Jamaican, African Am, Cambodian, Cape Verdean, Norwegian, Dominican, French Canadian, Haitian, Korean, Lebanese, Polish, Nigerian, Mexican, Taiwanese, Ukrainian, etc.

In 2000, there may have been a resistance being labeled as a hyphenated American. The 2000 list implies that the historic dominant immigrant groups British (English, Scotch, Welsh), German, and Irish had been integrated into an American identity.

2000 was the last time the question was asked as part of the decennial census. It was part of long form sent to a 5% sample. There might have been more commentary and general grousing about the census forms. If Debbie Downer had not actually filled out the long form, she might "remember" that she had skipped the questions about whether they had indoor plumbing and that she had said she was an American.

Beginning in 2006, the long form was replaced with the American Community Survey (ACS) where the 5% sample (actually less than 5% since the percentage varies based on a calculated need for accuracy) was spread out over a five-year period. Instead of every 20th house being sent a long form in census years, each month every 1200th house would be sent an ACS form. These could be aggregated over a 5-year period. The results might be smudgy, but if there was a consistent trend, the results for 2011-2015 might represent the values for July 2012 fairly accurately; while those for 2012-2016 might represent July 2013, and we could detect trends.

In addition with a much smaller sample each month, it would be easier to follow up with non-respondents. In a small city with 24,000 person (10,000 household) only 8 households would be surveyed each month. If 2 households failed to respond a census worker can make direct contact in the next month. Compare to when the long form was used and there would be 500 long forms and 9500 short forms. You would have to hire a lot more temporary workers, and the followup might take months.

From 2010 to 2017 the ACS trend was downward for almost all European ancestries. It appears to be sharper for groups that were not large enough (or exotic enough) to maintain a distinct community (e.g. Dutch and Swedish). If your grandmother had migrated from Rockford or Worcester or Chicago to California would you still identify as Swedish, particularly if there were multiple generations of intermarriage.

There was a sharp dropoff between 2017 and 2018 for many ancestries (English 24.0M to 22.8M, German 46.2M to 41.2M, Irish 32.6M to 30.8M). There was also an increase in composite identities (European from 4.2M to 6.4M, and British from 1.5M to 2.5M).

And then between 2019 and 2021 (English from 23.6M to 31.8M, German from 40.4M to 42.2M, Irish from 30.4M to 31.5M) and drops in composites (British from 2.5M to 1.2M, and European from 6.9M to 4.7M).

I doubt that there were shifts like that in how people identify. The Census Bureau may have changed how they have interpreted responses. In addition, they have pushed for online responses. They currently only invite people to respond online. If they fail to do so, the followup might involve sending a paper form. When they first started using online responses they sent the paper form along with the alternative.

[jimrtex]

Yes, the increases were in the South, while elsewhere more or less the same.  American responses declined.

2021 ACS:

Alabama  12.2%
Arkansas  11.8%
Kentucky  15.1%
North Carolina  13%
South Carolina  12.4%
Tennessee  13.5%

2000 Census:

Alabama  7.8%
Arkansas  7.9%
Kentucky  9.7%
North Carolina  9.2%
South Carolina  8.2%
Tennessee  9.1%

Add in the 2017 and the 2019 ACS.

[jimrtex]

While most Southern whites are primarily of English descent, I've heard that the more educated ones specifically identify as "English" while less educated ones identify "American".

For instance, Williamson County, TN is by far the most college-educated county in the state and also the only one where a plurality of residents specifically identify as English as opposed to American. Is there something distinct about the ancestry of people living in that particular county? I don't think so.

It's just one data point, but it's something to consider.

Possibly correlated with Episcopalian. There may be an Episcopal church in large Southern cities and not in more rural areas.

[Buffalo Mayor Young Kim]

German immigrants don’t get talked about nearly as much as their Irish and Italian peers, but for most of the 19th century they were the leading place of origin for immigrants arriving in the US.
So this isn’t surprising at all. In rural Illinois, and I’m guessing Indiana, Ohio, Wisconsin, etc., Brauns and Schneiders vastly outnumber Browns and Taylors.

Btw, ‘American’ ancestry pretty much means Ulster/Protestant Irish settlers that dotted the southern frontiers and stopped referring to themselves as Irish when the other kind of Irish people started showing up in large numbers.

[King of Kensington]

In rural Illinois, and I’m guessing Indiana, Ohio, Wisconsin, etc., Brauns and Schneiders vastly outnumber Browns and Taylors.

I doubt it (except maybe in Wisconsin)

Btw, ‘American’ ancestry pretty much means Ulster/Protestant Irish settlers that dotted the southern frontiers and stopped referring to themselves as Irish when the other kind of Irish people started showing up in large numbers.

Most Southern whites are of English ancestry.  The Scotch-Irish presence, while not insignifcant, is overstated.  There's a weird cottage industry claiming that white Southerners are actually "Celtic."

[King of Kensington]

Most common last names by state:

https://blogs.ancestry.com/cm/whats-the-most-popular-surname-in-your-state/

Interestingly no German surnames show up in the top three in any state, not even in Wisconsin.  Besides the obvious Spanish dominance in California, Texas and the Southwest, and Asians in Hawaii - you see a few ethnic identifiers (Sullivan in Massachusetts, Olson in North Dakota, more Johnsons and Andersons in the Upper Midwest).

[Builder Refused]

Most common last names by state:

https://blogs.ancestry.com/cm/whats-the-most-popular-surname-in-your-state/

Interestingly no German surnames show up in the top three in any state, not even in Wisconsin.  Besides the obvious Spanish dominance in California, Texas and the Southwest, and Asians in Hawaii - you see a few ethnic identifiers (Sullivan in Massachusetts, Olson in North Dakota, more Johnsons and Andersons in the Upper Midwest).

Reading up on this it seems it's because names were spelled in many different ways, that or, less often they were Anglicized like Busch to Bush to Braun to Brown

[Roll Roons]

Most common last names by state:

https://blogs.ancestry.com/cm/whats-the-most-popular-surname-in-your-state/

Interestingly no German surnames show up in the top three in any state, not even in Wisconsin.  Besides the obvious Spanish dominance in California, Texas and the Southwest, and Asians in Hawaii - you see a few ethnic identifiers (Sullivan in Massachusetts, Olson in North Dakota, more Johnsons and Andersons in the Upper Midwest).

Reading up on this it seems it's because names were spelled in many different ways, that or, less often they were Anglicized like Busch to Bush to Braun to Brown

Anglicized surnames may be more common than you'd think. In the Midwest/Plains states, I wouldn't be surprised if a lot of the Smiths were originally Schmidt and the Millers were originally Mueller.

A lot of these people probably changed their names either at Ellis Island or when the US entered WWI.

[Biden his time]

I wonder what’s driving the recent uptick in English identification? More people taking DNA tests and realizing they’re not Irish? Less people identifying as American? I answer English on the census and I know most people from my hometown have English ancestry but the ACS shows self English ID at less than 15%.

-SNIP-

All the smudging in recent years is why we can especially look forward to the point-in-time numbers we'll be getting from the 2020 census. They didn't re-add the ancestry question, but next to the race category checkmark a printed origin is requested for White and Black (as well as Asian, Am. Indian, but those were already provided for in the past).

For White Americans the examples are:

German, Irish, English, Italian, Lebanese, Egyptian, etc.

For Black Americans the examples are:

African American, Jamaican, Haitian, Nigerian, Ethiopian, Somali, etc.


These are much more likely to be accurate, and even with differential privacy they will be much more reliable than any numbers out of the American Community Survey.

[justfollowingtheelections]

Anyway, nearly every White American* has at least some English ancestry

lol no we don't.  That's the most ignorant comment ever.  Not all white Americans are anglosaxons.

[jimrtex]

I wonder what’s driving the recent uptick in English identification? More people taking DNA tests and realizing they’re not Irish? Less people identifying as American? I answer English on the census and I know most people from my hometown have English ancestry but the ACS shows self English ID at less than 15%.

-SNIP-

All the smudging in recent years is why we can especially look forward to the point-in-time numbers we'll be getting from the 2020 census. They didn't re-add the ancestry question, but next to the race category checkmark a printed origin is requested for White and Black (as well as Asian, Am. Indian, but those were already provided for in the past).

For White Americans the examples are:

German, Irish, English, Italian, Lebanese, Egyptian, etc.

For Black Americans the examples are:

African American, Jamaican, Haitian, Nigerian, Ethiopian, Somali, etc.


These are much more likely to be accurate, and even with differential privacy they will be much more reliable than any numbers out of the American Community Survey.

It will be interesting what they report. The categories for White were added after the OMB rejected adding a Middle Eastern and North Africa (MENA) ethnicity category. That is why the example includes a couple of MENA groups.

I just looked through the 2021 ACS PUMS documentation and there does not appear to be a secondary white characteristic. It does include some mixed race categories such as White-Vietnamese which in most of the reports is aggregated as White-Asian.

[Sol]

Anyway, nearly every White American* has at least some English ancestry

lol no we don't.  That's the most ignorant comment ever.  Not all white Americans are anglosaxons.

Most white Americans aren't WASPs of course, but most white Americans have some English ancestry unless they're part of a highly concentrated and recent diaspora community or some sort of group with heavy endogamy, which tbh is a pretty small percentage of white people. You almost certainly have some English or at least British Isles ancestry unless you fall in that camp!

[King of Kensington]

English ancestry is common among whites everywhere except the NYC area and Wisconsin/Minnesota/Dakotas.

[kcguy]

Anyway, nearly every White American* has at least some English ancestry

lol no we don't.  That's the most ignorant comment ever.  Not all white Americans are anglosaxons.

Most white Americans aren't WASPs of course, but most white Americans have some English ancestry unless they're part of a highly concentrated and recent diaspora community or some sort of group with heavy endogamy, which tbh is a pretty small percentage of white people. You almost certainly have some English or at least British Isles ancestry unless you fall in that camp!

I'm not sure whether I'm arguing for or against, but I'll add an anecdote.

For fun once, I traced the ancestry of one of my friends.  His family has lived in the Dakotas for over a century, and every one of his lines came to the U.S. between 1865 and 1900.  His ancestry splits neatly into 4 equal parts:
1.  Germans from modern Ukraine
2.  Germans from modern Moldova
3.  Germans from actual Germany
4.  Northern Ireland -- the surname was Kelly

[Death of a Salesman]

It's kind of ironic that the vast majority of Americans with German ancestry vote Republican while that party wouldn't get 20% of the vote here. The Republican Party is pretty much seen as a joke here, even among a lot of conservatives.

It’s even more extreme with Dutch-Americans.

At the time of Dutch immigration to the US, Dutch culture was very religious and conservative, and Dutch-Americans have avoided the cultural revolution that swept over the Netherlands in the intervening decades. For instance in 1922, the vote broke down as follows
Catholic Parties: 29.86%
Protestant Parties: 25.52%
Socialist/Communist Parties: 21.21%
Classical Liberal Parties: 14.81%

[LabourJersey]

While most Southern whites are primarily of English descent, I've heard that the more educated ones specifically identify as "English" while less educated ones identify "American".

For instance, Williamson County, TN is by far the most college-educated county in the state and also the only one where a plurality of residents specifically identify as English as opposed to American. Is there something distinct about the ancestry of people living in that particular county? I don't think so.

It's just one data point, but it's something to consider.

Possibly correlated with Episcopalian. There may be an Episcopal church in large Southern cities and not in more rural areas.

The correlation with Episcopalians is true in smaller towns in the South, but in the Northern cities that isn't the case, as Episcopalian congregations are increasingly made up of converts from other denominations (Catholic, Evangelical, etc.)

[King of Kensington]

It's kind of ironic that the vast majority of Americans with German ancestry vote Republican while that party wouldn't get 20% of the vote here. The Republican Party is pretty much seen as a joke here, even among a lot of conservatives.

It’s even more extreme with Dutch-Americans.

At the time of Dutch immigration to the US, Dutch culture was very religious and conservative, and Dutch-Americans have avoided the cultural revolution that swept over the Netherlands in the intervening decades. For instance in 1922, the vote broke down as follows
Catholic Parties: 29.86%
Protestant Parties: 25.52%
Socialist/Communist Parties: 21.21%
Classical Liberal Parties: 14.81%

There was Dutch immigration post-WWII to Canada, Australia and California as well and they tended to be socially and politically conservative and many went into farming.

[Bismarck]

It's kind of ironic that the vast majority of Americans with German ancestry vote Republican while that party wouldn't get 20% of the vote here. The Republican Party is pretty much seen as a joke here, even among a lot of conservatives.

It’s even more extreme with Dutch-Americans.

At the time of Dutch immigration to the US, Dutch culture was very religious and conservative, and Dutch-Americans have avoided the cultural revolution that swept over the Netherlands in the intervening decades. For instance in 1922, the vote broke down as follows
Catholic Parties: 29.86%
Protestant Parties: 25.52%
Socialist/Communist Parties: 21.21%
Classical Liberal Parties: 14.81%

There was Dutch immigration post-WWII to Canada, Australia and California as well and they tended to be socially and politically conservative and many went into farming.

Yep northwestern Iowa and Western Michigan are dominated by very conservative Dutch folks. These areas do seem to do better than average rural areas socioeconomically as well.

[Skill and Chance]

Most common last names by state:

https://blogs.ancestry.com/cm/whats-the-most-popular-surname-in-your-state/

Interestingly no German surnames show up in the top three in any state, not even in Wisconsin.  Besides the obvious Spanish dominance in California, Texas and the Southwest, and Asians in Hawaii - you see a few ethnic identifiers (Sullivan in Massachusetts, Olson in North Dakota, more Johnsons and Andersons in the Upper Midwest).

This isn't the best way to measure ethnicity, though.  Some ethnic groups have a wider distribution of surnames than others.  In particular, Spanish surnames are much more concentrated than English surnames, so while 1/50 people of primarily English ancestry might have the surname Smith, closer to 1/20 people of primarily Latin American ancestry might have the surname Rodriguez or Hernandez.

[𝕭𝖆𝖕𝖙𝖎𝖘𝖙𝖆 𝕸𝖎𝖓𝖔𝖑𝖆]

Most common last names by state:

https://blogs.ancestry.com/cm/whats-the-most-popular-surname-in-your-state/

Interestingly no German surnames show up in the top three in any state, not even in Wisconsin.  Besides the obvious Spanish dominance in California, Texas and the Southwest, and Asians in Hawaii - you see a few ethnic identifiers (Sullivan in Massachusetts, Olson in North Dakota, more Johnsons and Andersons in the Upper Midwest).

It is very interesting that Jones is in the top three in Kentucky and Tennessee (and Maryland and Virginia but no other state). As Al has mentioned, the only remaining trace of older Welsh settlement in Appalachia.

[King of Kensington]

Ancestry map of the US:

https://i.redd.it/tmhr7ne1avt51.png

[Bismarck]

Ancestry map of the US:

https://i.redd.it/tmhr7ne1avt51.png

That map has at least one glaring error. Dubois County, Indiana, one of the most German counties in America, is listed as being English. That’s Mike Braun’s home county and it historically was a German catholic enclave. 

[King of Kensington]

Was hoping to find a good map that doesn't separate English and American.  But I guess that ain't it.

I would also quibble with Finns being put under Scandinavian category.