ACS Data Users Group

We are pleased to announce the release of new American Community Survey (ACS) estimates for detailed race, Hispanic origin, ancestry, and tribal population groups. These statistics, available from the 2017-2021 ACS 5-year Selected Population Tables (SPT) and American Indian and Alaska Native Tables (AIANT), provide the greatest level of statistical information for detailed population groups.

The ACS is the most relied-on source for up-to-date social, economic, demographic, and housing characteristics every year. Estimates from today's release provide these characteristics for detailed population groups using lower minimum population requirements compared to the standard ACS 1-year release. Highlights include:

• SPTs: Provide estimates for 399 race, Hispanic origin, ancestry, and tribal population groups at numerous geographic areas, including census tracts.
• AIANTs: Provide estimates for 1,059 detailed tribal groups at numerous geographic areas, including Alaska Native Regional Corporations and American Indian and Alaska Native Areas.

Data and Additional Documentation Released

The Race, Ethnicity, Ancestry and AIAN Tables page includes links to access the data in data.census.gov and the Application Programming Interface. Documentation was also posted to browse the available tables, population groups, geographies, comparison guidance, and accuracy of the data.

Press Kit

The 2017-2021 SPT and AIANT Press Kit includes a wide range of resources, including the press release, webinar, and access to background materials for this release.

Hi all, I'm wondering if anyone can help me out. I've been eagerly awaiting this data release because I'm working on a data project focused on Hmong people in Dane County, Wisconsin. These Selected Population Tables are the only ACS tables that identify Hmong people, so we've been working with the 2015 5-year estimates which are quite outdated at this point. However, I was greatly disappointed to see that this release does not give any data on Hmong people in Dane County. Among Asian people, the 2015 data (table B01003) gives the total population for Asian Indian, Chinese, Filipino, Hmong, Korean, Japanese, and Vietnamese people. However, for 2021, the only detailed Asian races available are Asian Indian and Chinese. This is a dramatic reduction in the amount of data that is available. What's the reasoning for this? Did that many fewer people fill out the ACS from 2017-2021 compared to 2011-2015? Our county population has grown dramatically, so it doesn't really follow to me that there wouldn't be a large enough sample now as compared to 2015 (for example, in 2015, the Asian Indian alone population was 5,817 and the Chinese alone population was 7,376, while in 2021 the Asian Indian alone population is 9,274 (a 59% increase) and the Chinese alone population is 9,997 (a 36% increase)).

Brittany, use PUMS data with the three Dane County PUMAS for geo from 2017-21 ACS. You'll find Hmong in variable RAC2P- code 46. I just ran this recently, so I know it's available. Please contact me if you need help doing this. eric

Brittany--

I expect there are just as many or more Hmong in Madison. The numbers won't have declined.

Eric Grosso suggests crafting your own crosstabs from PUMS or IPUMS -- this is good advice.

RE: the "detailed tables" that USCB itself prepares -- and why is there less summary data available? USCB's disclosure review board rewrote its standards about 5 or 6 years ago. They made a sharp turn toward frugality and reluctance in publishing tables with many categories or levels. Specifically they became unwilling to publish for categories that are numerically infrequent locally and/or nationally. Most affected were the "detailed tables" on ancestry and on languages spoken at home.

These decisions by USCB were described as legally necessary to reduce personal identifiability risk. There were briefings by ACSO to explain all this (I remember attending a briefing by Tori Velkoff of ACSO at the time). The reduction in categories though was extremely blunt.

good luck,

Todd Graham