The census taken in the United States every 10 years is supposed to count everyone. But that doesn’t really matter to everyone.
After each census, the US Census Bureau reports how well it did in counting every person in the country. In 2020, as in previous years, the census didn’t get a completely accurate count, according to the bureau’s own reports. The official census number reports more non-Hispanic whites and people of Asian descent in the United States than there actually were. And it was reporting too few blacks, Hispanics, and Native Americans who live on reserves.
The Conversation US asked Aggie’s Yellow Horsesociologist and demographer at Arizona State University, to explain why and how the census misses people, and how it is possible to assess who has not been counted.
1. Who is missed during the census?
the most often missed people are low-income people, people who rent or don’t have a house at all, people who live in rural areas, and people who don’t speak or read English well. Often they are people of color – black Americans; Indigenous peoples; or people of Hispanic, Asian, or Pacific Islander descent.
Because of their living conditions, these people can be difficult for enumerators to find. And they may be more reluctant to participate due to concerns about confidentiality, fear of repercussions and distrust of government.
Nevertheless, the US Census Bureau tries to count everyone, aiming targeted public relations campaigns in specific communities to encourage members to participate. Additionally, Census Bureau employees are knocking on doors in person across the country, trying to follow up on those who have not responded to mailings, announcements and events.
However, the pandemic has made this process more difficult for the 2020 census, both by making people uncomfortable with in-person visits and by shorten data collection time.
2. Who missed?
Official estimates show that the 2020 census was indeed highly accurate, capturing 99.8% of all residents of the country. But the census missed count 3.3% Black American, 5.6% American Indian or Alaska Native who live on reservations, and 5% Hispanic or Latino. This could mean that approximately 1.4 million black Americans are missing; 49,000 Native Americans or Alaska Natives who live on reservations; and 3.3 million people of Hispanic or Latino origin.
This performance is much worse than in the previous two censuses, when smaller proportions of these populations were missed.
The 2020 census also counted 1.64% more non-Hispanic white people than there actually are in the country. For example, college students could have been counted twice – at their residence halls and at their parents’ homes.
3. How can they count missed people?
It can be confusing to understand how the Census Bureau knows how many people it missed. Efforts to measure census accuracy started in 1940. Census officials use two methods.
First, the Census Bureau uses demographic analysis to create a population estimate. This means that the office calculates how many people could be added to the population counts, through birth registrations and immigration records, and how many people could be removed, through death records or emigration reports. Comparing this estimate with the actual number may reveal a global scale the number of people the census missed.
As a second measure, the Census Bureau runs what it calls a “post-census survey,” taken after the initial census data collection. The survey is conducted independently of the census and randomly sent to a small group of households from the census blocks of each state, the District of Columbia and Puerto Rico. The results of this survey are compared to the census results for these households and can reveal how many people were missed, or if some people were counted twice or counted in the wrong place.
4. Can the Census Bureau correct its data?
The Census Bureau has determined that its 2020 data is not accurate and has measured the amount of this inaccuracy. But in 1999, the Supreme Court ruled that the office cannot adjust the numbers it was sent to Congress and the states for the purpose of allocating seats in the United States House of Representatives and, therefore, votes in the Electoral College. This is because federal law prohibits the use of statistical sampling in distribution decisions and requires that such changes be made solely on the basis of the number of people actually counted. This means that political representation in Congress may not accurately reflect the constituencies served by the representatives.
Corn numbers may be adjusted when used to allocate federal funding for essential services in communities around the nation. More … than $675 billion a year is provided to tribal, state and local governments in proportion to their number of inhabitants.
However, this adjustment only takes place if requested by tribal, state or local authorities. The Census Bureau Counting Question Solving Program can correct 2020 census data up to June 2023. After the 2010 census, the program received requests from 1,180 governments, out of approximately 39,000 nationally. As a result, approximately 2,700 people were newly added at the census, and about 48,000 household addresses have been corrected.
This approach can mitigate the damage to communities where the census has missed people. But that doesn’t stop the Census Bureau from missing them – or others – on the next census.[Get the best of The Conversation, every weekend. Sign up for our weekly newsletter.]