Explore the most frequently asked questions about the American Community Survey (ACS) Experimental Data. If your question is not covered here, please feel free to Contact Us.
The 2020 ACS 1-year experimental data tables are now available on the ACS Experimental Data webpage in Excel format. The tables will not be available in data.census.gov or the Census Bureau’s Application Programming Interface (API). We released a research paper with the data tables, titled Addressing Nonresponse Bias in the American Community Survey During the Pandemic Using Administrative Data, which discusses the methodology used to create the experimental estimates.
The 2020 ACS 1-year experimental data release includes 54 data tables covering social, economic, housing, and demographic characteristics, similar to the format of the ACS 1-year Supplemental Tables. We also released the 2020 ACS 1-year Public Use Microdata Sample (PUMS) files with experimental weights. You can find both of these releases on the ACS Experimental Data Page.
The 2020 ACS 1-year experimental data tables are available for the nation, states, and the District of Columbia. The experimental weighting methodology does not apply to Puerto Rico (household or group quarters populations), so there will not be experimental estimates released for Puerto Rico. The 2020 ACS 1-year Public Use Microdata Sample (PUMS) files with experimental weights are available for the nation, regions, divisions, states, and Public Use Microdata Areas (PUMAs).
The Census Bureau does not recommend data users compare the 2020 ACS 1-year experimental estimates with our standard ACS estimates. Also, Census Bureau experimental data should be used with caution because they may not meet all of our quality standards. We have released a limited set of 2019 and 2020 ACS estimates with the experimental weighting methods applied in a working paper to help data users understand the impact of the methods on the weights in a non pandemic year to provide context for year-to-year comparisons.
The Census Bureau urges data users to exercise caution when using the 2020 experimental data and to determine whether the data are suitable for their particular use. The experimental methodology makes some adjustments for the nonresponse bias, but it does not improve the resulting estimates for all topics. We have released a limited set of 2019 and 2020 ACS estimates with the experimental weighting methods applied in a working paper to help data users understand the impact of the methods on the weights in a non-pandemic year to provide context for year-to-year comparisons.
The pandemic most affected ACS data collection operations from April through June 2020. We resumed in-person interviewing in some areas in July 2020 and for all areas without stay-at-home orders in September 2020. We were able to return to our standard five-mailout strategy in April 2021. We do not anticipate problems with the 2021 data, but will know more when we process and review the data in the spring of 2022.