ACS Data Users Group

Due to the impact of the COVID-19 pandemic, the Census Bureau is changing the 2020 ACS release schedule. Instead of providing the standard 1-year data products, the Census Bureau will release a series of experimental estimates from the 1-year data no later than November 30th. This will include a limited number of data tables for the nation, states, and District of Columbia.

Census Bureau staff have posted several resources to help data users prepare for this change on an Experimental Data web page. These resources include new a new list of table IDs, titles, and shells for the 2020 ACS 1-year Experimental Estimates.

We are starting this new discussion thread/topic in the Online Community so that ACS data users can post comments or questions about the 2020 ACS 1-year Experimental Estimates.

Here's a link to a new PRB blog about the 2020 ACS Experimental Estimates:

Why the 2020 American Community Survey Is Different and Why It Matters

I'm interested to hear how other data users are planning to fill in the gaps this fall, in the absence of standard ACS 1-year estimates.

There is nothing we can do as the end users. This is the best that the Census bureau has been able to produce, and none of us have the resources nor the expertise to produce any alternative estimates.

Can someone familiar with both ACS and BRFSS provide some insight? Both surveys appear to have similar data collection processes. However, BRFSS released its 2020 data. Did BRFSS experience the same challenges as ACS in 2020? If yes, how did BRFSS resolve the issues that caused ACS to withhold the regular 2020 data and to issue experimental data instead? If no, what might explain the different decisions these two surveys made with regard to the release of the 2020 data? Thanks.

OMG they aren't NEARLY NOR REMOTELY similar. BRFSS has always been a phone survey, so the impact of COVID was close to none -- if anything, response rates crept up a little bit for the late spring / summer of 2020 when people stayed home and were bored to death and would rather talk to the government interviewer than to nobody. ACS was hit VERY hard as it lost the ability to follow up in person for about half of the year from March to September, and with self-completion rates about 70% (off the top of my head), and lower for hard to reach / hard to convince to respond populations (ethnic, racial, language minorities; rural populations). I am sure there are pockets where the annualized response rates were as low as 40%.

The Census Bureau just posted a list of Frequently Asked Questions related to the 2020 ACS 1-Year Experimental Data.

You can find the FAQs here: 2020 ACS 1-Year Experimental Data FAQs. You can also navigate to them from the main ACS Experimental Data page under the "Explore More" section.

The Census Bureau has released a new analytical report and quality measures supporting the 2020 ACS 1-Year Experimental Data release.

The report, An Assessment of the COVID-19 Pandemic’s Impact on the 2020 ACS 1-Year Data, describes how the pandemic disrupted ACS data collection in 2020 and gives data users insights into why the Census Bureau decided to release the 2020 ACS 1-year estimates as an experimental product. The Census Bureau ultimately determined the standard 2020 ACS 1-year estimates did not meet our statistical quality standards.

The Census Bureau released a blog, Pandemic Impact on 2020 American Community Survey 1-Year Data with the report, providing an overview about the contents of the report, and quality measures showing the impact of the pandemic on the annually-released measures of survey quality.

The Census Bureau also posted the 2020 ACS 1-year Sample Size and Data Quality page, as well as supporting documentation to help users prepare for the release. These quality measure tables include sample size, coverage rates, response rates, and item allocation rates.

Continue to visit the 2020 ACS 1-Year Experimental Data Release page over the coming months for the most up to date information about this release, including FAQs addressing top questions related to this release.

Here's how we wrote about it for our users, if it's helpful for anyone! What you need to know about the 2020 American Community Survey

The Census Bureau announced today that the 2020 ACS 1-year experimental data will be released on Tuesday, November 30. This release will also include the 1-year PUMS with experimental weights, a technical working paper, and a blog. The Census Bureau will also hold a webinar on November 30 at 2 p.m. (EST) to explain and answer questions about the release.

To learn more: https://www.census.gov/newsroom/press-releases/2021/acs-webinar-experimental-data.html