The decennial census has a profound impact on how we are represented in our governing institutions — from Congress and state legislatures, down to city councils and school boards. The U.S. Constitution calls for two chambers in Congress — one, the Senate, includes two representatives from each state; the other, the House of Representatives, is composed of representatives distributed among the states based on their respective populations.

The size of the House is set by federal statute; each state is guaranteed at least one congressional district regardless of population. After each decennial census, the new population counts are used to reallocate the number of districts per state, according to a mathematical formula set by law. This process is called “apportionment.” After the states receive the number of districts allowed per state, it is each state’s responsibility every decade to draw the boundaries of those districts in their states. This process is called “redistricting.”

EXPLAINING THE RESULTS

Why did my state lose a Representative in Congress? The U.S. Constitution established one chamber of a bicameral Congress in which each state’s representation would be based on relative population size, as determined by a census every ten years. The number of seats in the House of Representatives — currently 435 — is set by law. With the relative growth or decline in state populations over a decade and the fixed size of the House, some states inevitably will gain or lose seats in Congress after each census. As noted above, each state is guaranteed at least one Representative, regardless of the size of its population.

My state gained/lost a seat in the U.S. House of Representatives. What happens next?

Whether your state gained or lost one or more seats in the House, or whether the size of your congressional delegation will remain the same for the next decade, your state will have to redraw congressional district lines (the redistricting process) to ensure that the population of all districts within each state is as equal as possible and that districts meet the requirements of the Voting Rights Act of 1965, as amended.

Are the “apportionment counts” the same as the “resident population” of each state? In other words, do the state population totals used for apportionment only include people enumerated in the 50 states during the roughly nine months of census operations (January 21 — October 15, 2020)?

No. The counts that are the basis for congressional apportionment also include members of the armed forces and federal civilian employees stationed abroad on Census Day, as well as any dependents living with them. This federally affiliated overseas population (FAOP) is only included in the 50 state population totals used to apportion House seats in Congress, and not in census data used for other purposes, such as redistricting and the allocation of federal funds. The FAOP, enumerated using administrative data from the Defense Department and federal agencies, is counted at the servicemember’s or federal employee’s ‘home of record,’ which is where the individual enlisted in the military or first joined federal service. Members of the armed forces who are stationed in the U.S. and deployed overseas on Census Day will be counted at their stateside address for all purposes — a new rule for the 2020 Census.

Could inclusion of the federally affiliated overseas population in apportionment counts change the congressional apportionment outcome?

Yes, it could. In 1990 and 2000, the last (435th) seat in the House shifted due to the inclusion of this population.

When will we see the first official counts from the 2020 Census?

The Census Bureau will send the state population totals used for congressional apportionment to the Secretary of Commerce by April 30, 2021, but no earlier than April 16 — four months later than originally planned due to the disruptions in the census schedule caused by the coronavirus pandemic. The bureau will simultaneously release those numbers publicly along with the number of House seats to which each state is entitled based on the apportionment formula.

What other 2020 Census data does the Census Bureau publish when it releases the apportionment results?

When the bureau releases the apportionment results, it will also release the national population total and the resident population totals of each state, Washington, D.C., and Puerto Rico. The “resident population” reflects the official counts from the census without including the federally affiliated overseas population. The other four U.S. territories (Guam, American Samoa, the U.S. Virgin Islands, and the Northern Mariana Islands) are enumerated as part of the separate Island Area Census; data from that operation takes longer to process, tabulate, and publish.

CENSUS QUALITY AND ACCURACY

How and when will we know how complete and accurate the 2020 Census was? The 2020 Census national population total can be compared with the bureau’s own independent estimate of the population derived from Demographic Analysis (DA). The bureau published the 2020 Demographic Analysis estimates (there are low, middle, and high estimates) in December 2020. DA estimates will not be available for states. However, a census that looks accurate at the national level might be less accurate at the state, city, or neighborhood levels. That’s because gross undercounts and gross overcounts tend to cancel each other out at the national level, thereby masking the often unequal distribution of undercounts (people missed), overcounts (people counted twice), and other incorrect enumerations (such as people counted in the wrong place or included by mistake).

For further information on how the Census Bureau measures census accuracy and when we will see that information, see Evaluating the Accuracy of the Decennial Census: A Primer on the Fundamentals of Census Accuracy & Coverage Evaluation, from the Georgetown Center on Poverty and Inequality.

What else will we know about the quality of the 2020 Census when the apportionment data are released?

For the first time, the Census Bureau will publish several operational metrics when it releases the first set of 2020 Census data in late April. These metrics will help experts and stakeholders understand the quality of the census. The indicators include how the bureau counted a household (or an address, if vacant) — through self-response, in-person or proxy interview, or using administrative records, for example — and how those methods are distributed across geographic areas. The metrics will be available for the nation, states, Washington D.C., and Puerto Rico. In May, the bureau will release a second set of similar quality metrics with more detail (such as by average household size). Visit the bureau’s website for more information on the quality indicators it will release in April and May 2020.

Will these data include information about the characteristics of the population, such as race?

No. The first data releases only include population counts. Data sets including population characteristics will be released later in 2021.

How will differential privacy, the disclosure avoidance system developed for the 2020 Census, affect the apportionment results?

Differential privacy will not be applied to the state population totals used for congressional apportionment.

EXPLAINING THE PROCESS

Why does the Census Bureau produce state population totals for congressional apportionment after the census?

Article I, section 2, clause 3 of the U.S. Constitution, as amended by the 14th Amendment, section 2, requires a census of the population every ten years in order to apportion seats in the U.S. House of Representatives among the states. The number of seats in the House is set by law, except for the constitutional requirement that every state receive at least one seat. The number of Electoral College votes each state receives also is based on its representation in Congress.

What happens after the Commerce Secretary receives the apportionment data?

The Secretary of Commerce will transmit the state population totals used for congressional apportionment and the resulting apportionment of seats to the president. The president will then transmit that information to Congress. The Clerk of the House will then have 15 days to send a certificate to the governor of each state that shows the number of House seats to which the state will be entitled for the next decade, starting with the 118th Congress in 2023.

What remedies do states have if they believe their apportionment results are unfair or not based on accurate counts?

Historically, states have had limited pathways to change the apportionment outcome. States have challenged their loss of a seat or failure to gain a seat in court without success in the modern era. Congress could increase the size of the House but is unlikely to do so in the short term.

Are the numbers the Census Bureau will release in late April 2021 used for redistricting?

No. The Census Bureau will continue to process data collected in the census to produce the detailed Redistricting Data Files (sometimes called the P.L., or Public Law, 94-171 files) for the 50 states, Washington D.C., and Puerto Rico. These subsequent tabulations of data down to the “census block” level will be broken out by race, Hispanic origin, and voting age. The files also will include housing unit counts by occupancy status and group quarters (GQ), the latter broken down by major types of GQs (such as college dorms, prisons, military barracks, and skilled nursing homes). When will numbers used for redistricting be released? The Census Bureau announced that it will transmit the official Redistricting Data Files (also called P.L. 94-171 files) to the states, Washington D.C., and Puerto Rico no later than September 30, 2021.