2020 Post Enumeration Surveys

Quote:

---

The 2020 Census did not have a significant net coverage error rate. At the national level, the PES estimated a net coverage error rate of 0.24 percent (0.25 percent standard error) or 782,000 (821,000 standard error) people. The PES population estimate was not significantly different from the 2020 Census count. However, there were statistically significant undercounts or overcounts for specific subgroups.

We estimated undercounts for the groups Black or African American, American Indian or Alaska Native, Some Other Race, and Hispanic or Latino. Overcounts were estimated for White, Non-Hispanic White Alone, and Asian.

The 2020 Census undercounted renters and overcounted owners. This trend was seen in previous censuses.

The 2020 Census undercounted some age-sex groups and overcounted other groups.
The 2020 Census undercounted people under the age of 50 and overcounted people over the age of 50.
Adult males were undercounted, and adult females were overcounted in the 2020 Census.
Young children (aged 0 to 4) were undercounted, while the estimated net coverage error rates for older children (ages 5 to 9 and 10 to 17) were not statistically significant.

---

BusterAg said:

---

Why did you quote an unrelated portion of the study. It also finds: The 2020 PES identified statistically significant overcounts in New York, Massachusetts, Rhode Island, Hawaii, Delaware, Minnesota, Utah, and Ohio, while finding undercounts in states like Texas, Florida, Mississippi, Arkansas, Tennessee, and Illinois," he wrote. "Of the eight states overcounted in the 2020 census, six states have typically voted for electors for the Democratic Party candidate in presidential elections for the last three decades."

Is this OK to you? Should we attempt to correct this?

---

WHO the census was undercounting is just as important as WHERE the undercounts were happening. I'm not opposed to updating apportioning on updated data. Based on the WHO of the undercounts, I'm not convinced the new districts would be majority Republican anyway.

I think it came out a year or so after the census was released that New York should have lost an additional congressional district and Texas should have one more. California may also be over in terms of their districts.

I'm not convinced that a 0.24% error in a census count of a nation of 331 million people is somehow malicious.

AtticusMatlock said:

---

I think it came out a year or so after the census was released that New York should have lost an additional congressional district and Texas should have one more. California may also be over in terms of their districts.

---

If I remember correctly, Biden and Company in Jan 2021 delayed the release of the Census Data by a few months and that was when some adjustments to the states numbers were implemented. They blamed the Census delays on Covid.

Not sure if they can fix that now.

Something something about water being wet

None of our govt agencies are apolitical and they sure as hell haven't been "captured" by the right

Rocag said:

---

I'm not convinced that a 0.24% error in a census count of a nation of 331 million people is somehow malicious.

---

Where those discrepancies occur, absolutely indicate a pattern of maliciousness.

For example, it might not matter if only 0.25% of the election counting was "erroneous"
If that 0.25% occurred in Pennsylvania, it absolutely makes a difference.

Do you seriously think it was coincidence that number in blue states were overestimated, and numbers in red states were underestimated? You know fully well, that this is what is used to decide the seats in the house.

Rocag said:

---

I'm not convinced that a 0.24% error in a census count of a nation of 331 million people is somehow malicious.

---

Yeah that's a super small percentage...only amounts to about the same amount of people as the census count for San Francisco or Seattle. That would be easy to screw up when counting.

Sims said:

---

Rocag said:

---

I'm not convinced that a 0.24% error in a census count of a nation of 331 million people is somehow malicious.

---

Yeah that's a super small percentage...only amounts to about the same amount of people as the census count for San Francisco or Seattle. That would be easy to screw up when counting.

---

BTW this is the same argument libs use against voter fraud with respect to registration versus actual voting.

Dilute the pool of possible errors so that the actual errors look less fishy.

And do you seriously think it was a coincidence that non-whites were undercounted while whites were overcounted? I mean, we can both do this if we want. Let's treat that as a conspiracy, too! Why don't you answer for that one. After all, the 2020 census was conducted by the Trump administration. He'd been in office for 3 years at that point, plenty of time to set up his nefarious plans to undercount minorities and steal House seats away from their populations.

That one elector who should be allocated to Texas but is instead with New York is the difference between Trump winning the presidency without PA, MI, WI and needing one of those 3. I seem to recall from a few years back that NY kept that elector but an incredibly small margin.

I personally believe only American citizens should be counted in the census - even if it hurts Texas. I hope to see this policy enacted going forward.

sounds like the same group that counts our votes.

How would you categorize the who? What distinctions are most important, and why?

Should we make sure that all people with size 8 shoes are counted? Or, do you have a different type of error in mind
And, why does that error matter?

On a state by state level, it matters very much. If you had one electoral college voter move from each of the overcounted states to each of the most indercounted states, the election in 2024 would be very different.

Why are the errors you are talking about so important?

Rocag said:

---

And do you seriously think it was a coincidence that non-whites were undercounted while whites were overcounted? I mean, we can both do this if we want. Let's treat that as a conspiracy, too! Why don't you answer for that one. After all, the 2020 census was conducted by the Trump administration. He'd been in office for 3 years at that point, plenty of time to set up his nefarious plans to undercount minorities and steal House seats away from their populations.

---

Whites are overcounted due to their relatively higher socioeconomic status relative to minorities. This means that more whites have active credit ratings, own homes, long-term w2 jobs, etc. The reasoning is very technological, and unsurprising.

Now, why were 6 of the 8 most overcounted states Democrat? What is the technical explanation there?

Rocag said:

---

And do you seriously think it was a coincidence that non-whites were undercounted while whites were overcounted? I mean, we can both do this if we want. Let's treat that as a conspiracy, too! Why don't you answer for that one. After all, the 2020 census was conducted by the Trump administration. He'd been in office for 3 years at that point, plenty of time to set up his nefarious plans to undercount minorities and steal House seats away from their populations.

---

Please tell us the benefit of overcounting people by their race.

What a stupid "analogy."

LostInLA07 said:

---

That one elector who should be allocated to Texas but is instead with New York is the difference between Trump winning the presidency without PA, MI, WI and needing one of those 3. I seem to recall from a few years back that NY kept that elector but an incredibly small margin.

---

Only valid if either of those states were close to losing/adding a representative.

You can download the census data as an excel sheet off the website to get a better look. If you were to divide the population of a state by the national population divided by 435 you can get a basic approximation of how many representatives a state should have.

N = (state pop.) / (national pop. / 435)

This turns out to be pretty close to what the actual apportionment was.

Texas gets an N = 38.3 and was granted 38 representatives. If we were to add 1% to Texas' population it would push that up to N = 38.7 so might have gained one.

New York had an N = 26.6 with 26 representatives. Even with a 1%-2% overcount they're not going down to 25.

Rocag said:

---

And do you seriously think it was a coincidence that non-whites were undercounted while whites were overcounted? I mean, we can both do this if we want. Let's treat that as a conspiracy, too! Why don't you answer for that one. After all, the 2020 census was conducted by the Trump administration. He'd been in office for 3 years at that point, plenty of time to set up his nefarious plans to undercount minorities and steal House seats away from their populations.

---

And the Census results were delayed by Brandon and Crew.

Quote:

---

If I remember correctly, Biden and Company in Jan 2021 delayed the release of the Census Data by a few months and that was when some adjustments to the states numbers were implemented. They blamed the Census delays on Covid.

---

It was the Trump administration that first asked for delays in the reporting of census data.

From April 2020: Trump Officials Ask To Delay Census Data For Voting Districts, House Seats

Wow, so white libs are over representing themselves at the expense of minorities in their own state? Color me shocked?

Now to get back to the thread, it obviously doesn't matter to what the OP was discussing, that electoral power is being gamed by liberal states.

Rocag said:

---

It was the Trump administration that first asked for delays in the reporting of census data.

From April 2020: Trump Officials Ask To Delay Census Data For Voting Districts, House Seats

---

The article says the delay was to April 2021. I believe that Brandon and Crew delayed the reporting several months past that.

If I recall correctly the census numbers were released so late that numerous states really had to scramble to get their new district maps approved before the 2022 elections.

BusterAg said:

---

How would you categorize the who? What distinctions are most important, and why?

Should we make sure that all people with size 8 shoes are counted? Or, do you have a different type of error in mind
And, why does that error matter?

On a state by state level, it matters very much. If you had one electoral college voter move from each of the overcounted states to each of the most indercounted states, the election in 2024 would be very different.

Why are the errors you are talking about so important?

---

Yep.

Those apportioned House seats could even ALL be Democrats...

But, if Texas gets one more Democrat House Representative, it ALSO gets one more Electoral Vote...

Making Texas more valuable to the Republicans in the POTUS election.

The converse is true for NY...

So, there's a potential that red states would have had at least 6 more EVs

Rocag said:

---

I'm not convinced that a 0.24% error in a census count of a nation of 331 million people is somehow malicious.

---

The error was 0.01% in 2010.

Quote:

---

Last week, the Census Bureau released the results of our post-enumeration survey, called "Census Coverage Measurement (CCM)." The results showed that the 2010 Census had a net overcount of 0.01 percent, meaning about 36,000 people were overcounted in the census on the base of over 300 million. This sample-based result, however, was not statistically different from zero, after taking into account the sampling variability of the post-enumeration survey.

---

Odd that it is 24 times larger in 2020 AND skewed overwhelmingly to the blue states...

That's an entire order of magnitude larger error...

True, but even the link remarks on it being an exceptionally accurate census. Compare to previous years:



The 2020 census, at 0.25%, was still better than the 1980, 1990, and 2000 censuses.

Edit: I suppose you could take this to mean that the Obama administration was an order of magnitude better than the Trump administration at getting things done right. You know, if we're comparing 2010 to 2020.

Rocag said:

---

True, but even the link remarks on it being an exceptionally accurate census. Compare to previous years:



The 2020 census, at 0.25%, was still better than the 1980, 1990, and 2000 censuses.

Edit: I suppose you could take this to mean that the Obama administration was an order of magnitude better than the Trump administration at getting things done right. You know, if we're comparing 2010 to 2020.

---

So, is the Census keeping up with Moore's law? Because, they should be tracking that directionally at least, even if the impact is only like 10% of Moore's law.