ACS Data Users Group

Hi all - I'm trying to determine countries of origin for international migrants to my state in 2022. Is there a pre-made ACS table that I could find in the data explorer that would show this? I haven't been able to find one, but thought I'd post here in case I'm missing something.

I turned to the PUMS data, analyzing the MIGSP variable, but the margins of error are too high even when I aggregate by continent. I think even if I pooled two years of PUMS data the MOE would still be too high for any valid analysis and I can't use the 5 year data because it's not recent enough.

Thought if I could find a pre made ACS table at least the MOEs would be a bit better since it's a larger sample.

Thanks for any thoughts you may have.

You can get the total population who moved from outside of the U.S. from table B07204 but I can't think of any pre-tabulated disaggregation by origin--PUMS is the best option. Would pooling 3 years of PUMS data work?

Also, if you're willing to share, I'm curious what state you're working in that has large MOEs for estimates by region/continent...

There is a "special product" on migration moves; it's not a base "B" table; it's always a few years older than the most recent ACS.

Documentation is here www.census.gov/.../2016-2020-Migration-Flows-Documentation.pdf

Data table (Excel) is here www2.census.gov/.../county-to-county-2016-2020-current-residence-sort.xlsx

The bad news is: All of the international origins are glommed together into just 9 regions of the world. These are (roughly) continental origins, not country origins.

--todd graham

Yes! Thank you - I should have included this in my list. (I was thinking if 2022 5-year ACS was too old, the 2020 migration flows custom tab would be waaay too old for this request.)

Thanks, this will be too old for my current purposes, but I'm glad you pointed me to it so I can file it away as a resource for other analysis.

Thanks, I'm in DC, hence the large MOEs.

I'm investigating a recent surge in international migration that we saw in the 2022 and 2023 population estimates so farthest I could go back would be 2021, but I think even then MOEs would be too high. With the one year's worth of data I've got a Z of .6 when testing if the number of migrants from one continent is statistically different from the number of migrants from another continent. I did a quick calculation to simulate adding another year's worth of data and got Z=0.9

I'm no expert in statistics, so if anyone has any other ideas for hacks, I'm all ears.