FAQ: The 2020 Census Redistricting Data and Related Materials Release
Since releasing apportionment data from the 2020 Census in April, the U.S. Census Bureau has been hard at work on the next set of 2020 data – detailed counts used by states and localities for redistricting. These data, which include the first sub-state population counts and demographic characteristics from the census — information that states and localities use to redraw electoral district boundaries — play an illuminating role in the advancement of our democracy. While the process of congressional line-drawing is a state-based function, the Census Bureau provides data to the states derived from the decennial census that states can use in their individual redistricting processes.
Why are redistricting data important?
Redistricting data allow states to redraw congressional districts, as well as state legislative district lines, to account for changing demographics every ten years. While the U.S. Constitution does not mandate the use of decennial census data for creating congressional districts, these data are widely recognized as the best data to use for such purposes. Additionally, many states are required by their own constitutions or statutes to use official census data for redistricting. Counties and cities often use census data to draw their own electoral boundaries, such as for city councils and school boards.
These data also provide the first opportunity to assess the 2020 Census quality and accuracy at the local level, including indications of potential undercounts in historically undercounted communities.
What is the timeline for release of redistricting data?
The COVID-19 pandemic significantly delayed the release of 2020 Census redistricting data, which was originally scheduled to be released on a rolling basis from mid-February through April 1, 2021. During 2020 Census data processing, the release of apportionment data was prioritized, slowing the more intricate processing of the redistricting data. However, in order to provide some relief to states awaiting redistricting data to draw boundaries for upcoming elections, the Census Bureau will have two releases of identical redistricting data. On August 12, 2021, the Bureau will release legacy format data, which will be timelier in its delivery, but not very “user friendly.” The Census Bureau will transmit the official Redistricting Data Files (also called Public Law 94-171 files) to the states, Washington D.C., and Puerto Rico on September 16, 2021; this release will be easier for states and other interested parties to access and use.
What is the difference between the legacy format files released in August and the files being released in September?
The only major difference in the two releases will be the format in which the redistricting data are presented. The legacy format file data, which provides raw data, are being released earlier in the process, in order to allow states to begin their redistricting process as soon as possible. The data released on September 16 will be easier for stakeholders to access and analyze. The legacy files contain untabulated “census block” level data, which will ultimately be tabulated – that is, turned into user-friendly data tables – for the official September data release under P.L. 94-171.
Why not use American Community Survey (ACS) data for redistricting purposes?
Data produced from the American Community Survey (ACS) are useful for many purposes, but not as the primary basis for congressional line-drawing. The redistricting data produced from the decennial census and delivered to states is based on a count of all the nation’s residents and is produced at levels and with demographic data designed to allow for fair allocation of political representation, as mandated by Congress’s enactment of Public Law. 94-171 in 1975. The ACS is based on a small (though robust) sample of households and group facilities, is not as current as decennial census data, and is not available at the same granular level (the census “block” level) as decennial census data.
What materials will accompany the legacy format files?
The legacy format files will be accompanied by materials to help those with technical skills to properly utilize the data. Redistricting mapping software vendors have already committed to providing a tabulated version of the legacy files shortly after their release, which could mean that all states could have access to the tabulated data they need for redistricting purposes by the end of August (without waiting for tabulated data in the form of official Redistricting Data Files being released by the Census Bureau this September).
What materials will accompany the late September data release?
The second release of redistricting data on September 16 will be delivered in the form of DVDs and flash drives to governors (or highest elected officials) and legislative leaders of each state and Tribal government, the District of Columbia, and Puerto Rico, alongside an integrated software browsing tool. This dataset will also be posted to the Census Bureau’s data.census.gov website, which will allow for public access in an easily accessible user interface.
What will the redistricting data tables contain?
For each respective state, the redistricting data will contain the following information:
- Population counts for counties, cities, census tracts, and census blocks
- Population counts for state-specified congressional and legislative voting districts and school districts
- Demographic Data: Race, Hispanic/Latino origin, & voting age population (18+)
- Housing unit counts and occupancy status (occupied or vacant)
- Group quarters (GQ) counts by major GQ type (e.g., college dorms, prisons, military barracks and skilled nursing homes) to help states interested in addressing prison-based gerrymandering in their redistricting process.
What is differential privacy and how is it being used in the release of redistricting data?
Differential privacy is a modern method of “disclosure avoidance” — the Census Bureau’s way of protecting the confidentiality of all personal census responses, through the careful addition of statistical “noise” so that no one can identify individuals or households by looking at the anonymous statistics the Census Bureau publishes. The 2020 Census redistricting data will be the first dataset to be protected using this scientific method.
Alongside the release of redistricting data on August 12, the Census Bureau will also be releasing demonstration data (Privacy-Protected Microdata Files) that will show the impact of the disclosure avoidance system chosen for the 2020 Census on 2010 Census data. With the release of the late September data, the Census Bureau will also be providing the disclosure avoidance system production code base, related technical papers, and an “Introduction to Differential Privacy” handbook.
When will we see additional operational metrics?
The Census Bureau will release two more sets of operational metrics in August.
- On August 18, the bureau will release summary county- and tract-level metrics by state for certain previously released operational quality metrics. For each state, it will release the average, median, and standard deviations for counties and tracts. Individual county and tract-level operational metrics will not be available.
- On August 25, the bureau will release item nonresponse rates (questions left unanswered) for population count, age or date of birth, race and Hispanic origin questions. These rates will be available at the national and state levels as well as for the District of Columbia and Puerto Rico.
Coupled with the redistricting data, these metrics will support further analysis of the quality of 2020 Census data, as well as a deeper understanding of how individual neighborhoods and communities were counted.
Please direct questions about the information in this fact sheet to Meeta Anand, census consultant to The Leadership Conference Education Fund, at [email protected].