The Official 2020 Census Thread

[jimrtex]

The Census Bureau has finished the Census Unedited File (CUF) now and is applying the final ingredients (Imputation, Unduplication etc.) to release the population figures (planned for April 16-30).

https://www2.census.gov/about/partners/cac/sac/meetings/2021-03/presentation-post-enumeration-data-processing-status.pdf

https://www2.census.gov/about/partners/cac/sac/meetings/2021-03/presentation-2020-census-operational-review.pdf

https://www2.census.gov/about/partners/cac/sac/meetings/2021-03/presentation-post-enumeration-survey-update.pdf

The results of the PES won't be released until late this year or perhaps next.

It is a sample where they interview people in person to see what sort of errors they made when they responded (e.g. forgetting people, adding people, etc.) or why they skipped questions or misunderstood them.

The slides suggest that some people can't remember where they were on April 1, 2020; or who lived there. They are also encountering different residents.

As part of their reply brief in the Ohio case, they provided a schedule of the steps. Completing the CUF means that they are running about one week ahead of schedule.

[jimrtex]

The Census Bureau is fast-tracking re-districting data for states to ca. 1.5 months before their deadline, by giving them the data early - but not in a user-friendly format:

Is the legacy format the same as was used in 2010?

YES:

So they were responsive to the Ohio lawsuit.

The State of Ohio and Department of Commerce (Gina Raimondo) have agreed to let the judge decide on jurisdiction and standing without a hearing. If he decides that they do, there will be hearings on the redress.

I wonder what will happen to the Alabama lawsuit.

[jimrtex]

Here is an article about so-called „differential privacy“:

https://www.ncsl.org/research/redistricting/differential-privacy-for-census-data-explained.aspx

Data for sub-state entities such as small towns or small counties get deliberately falsified by the Census Bureau because of privacy reasons.

For example if you have small census tract where only 50 people are enumerated and 1 of these people are black, you can draw conclusions out of the data and what other characteristics apply for this black person.

Therefore the CB will falsify the data for example to show that 2 or 3 black people are living there, not just one.

State counts will not be impacted or edited though.

This procedure is also applied here in Austria so that no conclusions can be drawn on individual people in small towns.

It’s also a reason why postal votes are not counted on the precinct level here because one could draw conclusions about voting behavior of an individual in very small precincts ...

The Census Bureau presented 2010 Census data with the 2020 techniques applied to the states for analysis. Colorado found systematic biases such as shifting population to rural areas. Some legislative districts would have up to 3.5% variance in population.

Colorado Response

This process requires a top down process from the national level - which means all states have to be fully processed before it can be applied, which may delay state data.

This shifting of people and delay is the basis for Alabama's second lawsuit.

[jimrtex]

Sen. Portman (R-OH) being a clueless dick:

Lawmakers frustrated over delay in Census redistricting data

Republican senators expressed frustration Tuesday that data used for redrawing congressional and legislative districts won't be available until August at the earliest, but the U.S. Census Bureau's acting director told them the schedule was driven by a goal of releasing complete and accurate numbers.

At a hearing of the Homeland Security & Governmental Affairs Committee, Republican lawmakers told acting Census Bureau director Ron Jarmin that the delay was upending their states' redistricting plans.

The statistical agency recently said the redistricting data would be ready in an older format by August and in a more user-friendly format by September, months after the redistricting deadlines for many states. By law, the redistricting data is due by March 31, but the bureau said it needed the extra time because of delays caused by the COVID-19 pandemic.

Under questioning, Jarmin said prioritizing some states like Ohio over other states in getting the redistricting data would actually cause a bigger delay. He also said hiring more staff wouldn't speed up the process since that would take staff experts away from processing the data so they could train the new hires.

"That sounds nonsensical to me," Portman responded.

https://mycbs4.com/news/connect-to-congress/lawmakers-frustrated-over-delay-in-census-redistricting-data

Portman has probably never worked a real supervising job like I do.

If you currently have your hands full with a big workload and additionally have to train newbies, you get distracted and your own work remains undone.

You will only get a reward out of it if those newbies are trained and can get the job done on their own.

That won’t help you right now, but in a few months to half a year only ... in case of the census data that would be too late.

This is a transcript of his questioning Portman questions Jarmin

That page also includes a video.

Portman was quite responsible.

He wondered whether their redistricting plans would be upheld given shifting populations by the Census Bureau.

Portman was director of OMB, which is likely a more responsible position than yours.

[jimrtex]

OH lawsuit struck down by a federal court:

Federal judge dismisses Ohio lawsuit seeking quicker release of census counts

A federal judge has dismissed a lawsuit filed by Republican Attorney General Dave Yost seeking to force the U.S. Census Bureau to release new population counts sooner for Ohio’s redistricting process.

U.S. District Judge Thomas M. Rose ruled Wednesday that Yost lacked standing in the case, and the Southern District of Ohio court lacked jurisdiction over the issue raised in the attorney general’s injunction request.

The state, the judge wrote in his ruling, has other options for completing redistricting – the process of drawing Statehouse and congressional district lines – absent the availability of new census counts earlier in the year.

“Ohio has not established that it cannot accomplish its redistricting in the time that remains between the unavoidably delayed results of the 2020 Census and its 2022 elections,” Rose wrote.

https://eu.dispatch.com/story/news/politics/state/2021/03/24/federal-judge-dismisses-ohio-lawsuit-seeking-quicker-release-census-redistricting/6987148002/

The judge is right: OH should simply change its constitution and don’t whine around.

That is not what the judge ruled.

[jimrtex]

Will we be receiving data on Puerto Rico and the other U.S. territories on the same schedule as the rest of the U.S. states and District of Columbia?

Technically, DC can have more than three electors, so it will have its apportionment population determined if not used.

The statute that provide for censuses for the insular territories say that their censuses are in addition to the US Census, though taken at the same time. Puerto Rico, Virgin Islands, Guam, and Northern Marianas, are specifically mentioned in this regard. American Samoa is not, but could be included at the Census Bureau's discretion.

13 U.S. Code § 191

It would appear that PL 94-171 only applies to the United States. Subsection (c) is PL 94-171.

13 U.S. Code § 141

[jimrtex]

This has probably been asked and answered on this forum before somewhere, but there's a question I've had for a while: Did this whole differential privacy thing originate within the Census Bureau (okay, maybe not within, but concerned citizens just mentioned the issue with no legal threats involved and Census Bureau officials decided on their own to adopt it; I understand that development began before the 2010 census but wasn't ready then), OR was there a lawsuit or specific threat of a lawsuit on 14th Amendment grounds (right to privacy) that led to the Bureau taking their current course?

These really don't answer why, but provide a little more information.

https://www2.census.gov/about/cic/Differential%20Privacy.pdf

https://hdsr.mitpress.mit.edu/pub/dgg03vo6/release/1

https://www.census.gov/programs-surveys/decennial-census/2020-census/planning-management/2020-census-data-products.html

I'm making all this up as I write ... don't read it if you are on a low salt diet.

There is a statute that says the Census Bureau can't release individual data. This might be rather old law that meant census forms were confidential.

In 2010, data swapping was employed. IIUC, this involves swapping (same-sized?) households between nearby census blocks. This would mean census block counts were accurate but racial characteristics would be off - but perhaps might be accurate at a block group or census tract level.

There has probably been academic research on the subject and likely a flow of people between academia and the Census Bureau.

The Census Bureau found that they were able to undo their 2010 swapping based on using all their published data.

I'm not sure I understand what information can be gleaned from the Census Data alone. Race, sex, age, household relationship, rent/owned. Maybe they can also add in the ACS data.

They have run a test using the 2010 Census Data but applying their differential privacy algorithm. But they then had to run various fix-up to make sure counts were not negative, etc. Those changes had a systematic bias against less populated areas.

They can also apparently set parameters such that more privacy means less accuracy, more accuracy means less privacy. But I have heard of an anecdote where some towns in Utah were 30% off.

Are they going to have to tweak these parameters for the 2020 Census?

Will their be sufficient accuracy for Hudson wards, or Maine House districts?