The 2020 Census: I Think We Are Bring Lied To

Yesterday, I posted a short discussion on the US Supreme Court allowing the census to end early. That is despite the slowdowns in counting that the pandemic has caused. Apparently, the president and GOP strategists have concluded that ending the census early helps them to undercount democrats and minorities.

I have become aware of increasing evidence that the Census Bureau is lying to the American people by faking the count data and undercounting housing units in areas that it claims have been 99.9% counted.

Data point 1

The first piece of evidence came from the US Census Bureau itself. The US Census Bureau is reporting that the overall census is 99.9% complete for housing units in the US. Only two states for which data is shown have lower reporting rates, 98.3% for Louisiana and 99.4% for Mississippi. Data is shown for states listed alphabetically through Pennsylvania. Thus states such as Texas and Washington state are not listed. 

What this data does not indicate is how many people are still uncounted. And, why are nearly all states at 99.9% reporting? The data seems as if it is fudged to make the census look more complete than it really is.

Something smells over-ripe fishy. I think the Census Bureau, specifically Trump, is lying to the American people. The data consistency the Census Bureau reports is simply not believable. Data like this cannot be that consistent at 99.9%. This is fake data. This feels like amateurs with no knowledge of statistics or lying about data has faked the data.

Disclaimer: I could be wrong here. My suspicion is based on my decades of research and work with complex biological systems. Complex biological systems, including census counts, are usually noisy and usually rather imprecise at the level of three significant digits (99.9). Is there a statistician in the house? 

Data point 2

The San Diego Union Tribune (SDUT) reports today that the census response rate in some San Diego neighborhoods is less than 50%. The San Diego city population is about 1.1 million and the county population is about 3.5 million. The SDUT writes: 

“Most households in nearly 20 San Diego-area neighborhoods have not responded to the 2020 census and will no longer be able to following a U.S. Supreme Court decision Tuesday to end the once-a-decade count.

According to the U.S. Census Bureau, 20 census tracts in San Diego have a response rate below 50 percent. The northwest corner of Oceanside had the lowest response rate, with just 27 percent of households responding as of Tuesday.

In an area of Mission Beach and Mission Bay Park, about 32 percent of households have responded. 

They’re followed by College West with 33 percent and Borrego Springs with 35 percent.

The U.S. Supreme Court on Tuesday halted the head count to allow the Census Bureau time to produce an accurate count. The ruling is the latest in a roller coaster of decisions on the 2020 census, which lower courts previously argued should continue through the end of October. 

Activists, who have spent the last several weeks encouraging responses from San Diego’s “hard-to-count” communities, say they feel defeated by the court’s decision.

“We are very disappointed, to say the least, with the court’s decision because there’s no reason to stop it,” said Arcela Nuñez-Alvarez, co-director of the nonprofit Universidad Popular, a nonprofit that does census outreach to rural communities. 

Many advocates had hoped to have until the end of the month to continue encouraging people to fill out the census, especially in areas with low response rates, she said.”

Data point 3

Some other information I've seen suggests that the 99.9% response rate is a deflection from the people count to the “household count”, including this map from the US Census Bureau. Note the comment: “But that may not mean the count is complete or accurate.” 

I now believe that we are being lied to.

And, there is the map shown below that the Census Bureau has put out. The map is interactive and you can click on an area to see the accumulated data. It does not always match the Oct. 13 99.9% data. For example, Montana is shown below as 99.7%, not 99.9% as shown under that graphic. Or do such data discrepancies mean nothing? If it wasn’t the Trump administration, the benefit of a doubt might be warranted. But since it is the chronic liar Trump administration, in my opinion a benefit of any shred of doubt is not warranted. The burden is on the Census Bureau to prove their honesty, not on me to prove their dishonesty.