US Census Bureau report says population was underestimated by 18.8 million

The report released Wednesday by the Census Bureau (

GREENSBORO, NC (WGHP) – We just finished drawing new electoral districts based on tons of data from the 2020 census, and now we find out that a bunch of those statistics were incorrect.

The US Census Bureau on Thursday released what it calls the “2020 Post-Census Survey Estimate Report,” which is a 19-page document that serves as an audit for the initial population report released in October. 2020 which indicates that the US population was underestimated by 18.8 million.

To put it in layman’s terms, that’s about equal to anyone living in North Carolina over 80% which were not even counted. Or it’s about twice the population of New York that has been neglected.

The total population is now 323.2 million, and while the report says the standard variance rate for one error is 0.25%, the overall population change hasn’t changed much from the original count. .

What we learned

But when you look at the trends in the various demographic distributions that are so intrinsic to understanding population change, you see larger fluctuations. Check out these key findings:

Residents of the Rio Grande Valley were invited to participate in the 2020 census, as advocated by these English and Spanish signs posted in Alton, Texas. The number of Hispanics was underestimated in the first version of the data. (Sandra Sanchez/Border Report File Photo)
  • The report says the census undercounted blacks and various indigenous groups, but overcounted whites, non-Hispanic whites and Asians.
  • There was an undercoverage in the number of people aged 50 or younger and an overcoverage of older people. The number of children from birth to 4 years old was underestimated, but there was no real variance in the age groups 5-9 years or 10-17 years.
  • Adult females were overcounted and adult males were undercounted.
  • Even the number of people who said they owned their homes was overestimated and renters were underestimated, which the report says previous audits have found.

Recalculations revealed that the number of black or African Americans was underestimated by 3.3% and Hispanics by 4.9%, and that the number of whites was 0.66% higher than what was estimated. ‘he should have been. The report says the undercount of Latinos was three times greater than in 2020, NPR reported.

There were 3.05% more males aged 30-49 than reported, but there were 2.63% more females aged 50 and over. The 4 and under age group was underestimated by 2.79%.

Cropping problem

The Census Bureau has performed these checks and balances every year since 1980, and they are based on comparing counts and models that study the data and assess their accuracy by accounting for missing features.

The report cited the COVID-19 pandemic, which necessitated changes in data collection processes and the time needed to do so, and debates over the methodology that politicians wanted to adapt as factors that affected the count.

The Census Bureau released its distribution results in April 2021, when North Carolina won a 14th seat in the House of Representatives. But those delays delayed the initial release of the data by about seven months and thus delayed the 10-year redistricting that is taking place in every state.

The North Carolina General Assembly, for example, did not approve new voting cards until Nov. 4, and subsequent legal challenges dragged on in court until late last week. The primary election was postponed from March to May.

“States to check”

But what we have now is a new set of data that may have informed discrepancies in the information used to draw and assess these district lines, particularly in relation to race, a factor protected by national law and often used in debates about gerrymandering. Disenfranchised black voters were a key argument for plaintiffs challenging the General Assembly maps. The state Supreme Court ultimately dismissed them and demanded a re-draw.

This new census report did not break down the discrepancies by state, but it did show that about 1.025 million people lived in the same state but in a different county than originally planned. An additional 3.711 million were recorded in the correct county, but were counted incorrectly based on other factors.

“We have a lot more reports to check, review and go through,” Timothy Kennel, deputy division chief for statistical methods, said in a webinar ahead of Thursday’s release.

The state-by-state breakdown affects the distribution of approximately $1.5 trillion. North Carolina estimated that census-based programs brought in $44 billion in fiscal year 2017.


Comments are closed.