[EL] what if the Census is not defensibly accurate?

Mon Aug 3 20:12:13 PDT 2020

Can anyone explain to me why it is that, if we stop counting a month early,
we "risk a severe undercount of people of color?"  I'm being serious; I
really don't know how we would know this.  How do we know that we do not
risk undercounting a greater percentage of Whites than people of color?  It
seems to me that, in order to make these types of categorical statements,
we would have to know how many people of each race have been counted
already, as well as how many remain to be counted.  Do we know that?  I'm
really not trying to be difficult, I'm just trying to understand how these
types of statements get made.

Joseph
___________________
*Joseph E. La Rue*
cell: 480.737.1321
email: joseph.e.larue at gmail.com

CONFIDENTIALITY NOTICE:  This email may be protected by the attorney-client
privilege or the attorney work-product doctrine.  If you are not the
intended recipient, please delete all copies of the transmission and notify
the sender immediately.

On Mon, Aug 3, 2020 at 6:36 PM Jeff Hauser <jeffhauser at gmail.com> wrote:

> "BREAKING: Census Bureau confirms all #2020Census counting efforts will be
> cut a month short, risking a severe undercount of people of color.
> Collecting responses online, over the phone & by mail, plus door knocking
> at unresponsive homes, to end on Sept. 30"
> https://twitter.com/hansilowang/status/1290454281632522241
>
> So, yeah. Most likely we're going to have an intentionally wrong census in
> a few months. Pretending that whatever Wilbur Ross has done is a real
> census seems, to me, wrong.
>
> On Thu, Jul 16, 2020 at 12:14 PM Brunell, Thomas <tbrunell at utdallas.edu>
> wrote:
>
>> Hopefully non-response follow-up will be able to reach standard levels of
>> coverage, though that may be impossible.  The Bureau will have to rely on
>> administrative records and whole household imputation to the extent NRFU
>> falls short.  Hot deck imputation usually accounts for around 2 percent of
>> all households in recent censuses.  This time it might be  significantly
>> higher.  In order for imputation to be reasonably  accurate at higher
>> levels of geography we still need people in the field to certify which
>> houses are unoccupied and those that are in fact occupied.  Combine really
>> high levels of imputation with the new differential privacy methods and it
>> will be all but impossible to know how good the census data really are.
>>
>>
>>
>> Tom Brunell, Ph.D.
>>
>> Professor of Political Science
>>
>> Program Head, Political Science and Public Policy & Political Economy
>>
>> 800 W. Campbell Drive, GR 31
>>
>> Richardson, TX 75080
>>
>> (972) 883-4963
>>
>>
>>
>> *From: *Stephanie Singer <sfsinger at campaignscientific.com>
>> *Sent: *Thursday, July 16, 2020 10:51 AM
>> *To: *Jeff Hauser <jeffhauser at gmail.com>
>> *Cc: *Election Law Listserv <law-election at uci.edu>
>> *Subject: *Re: [EL] what if the Census is not defensibly accurate?
>>
>>
>> This is an important hypothetical to consider. It’s within the realm of
>> possibility. As is the parallel question for the election.
>>
>> And if there is no constitutional Plan B, what would be likely to happen
>> next?
>>
>> On Jul 15, 2020, at 3:53 PM, Jeff Hauser <jeffhauser at gmail.com> wrote:
>>
>> My hypothetical is what happens if the results are trash. We are amidst
>> an international pandemic while we have also just seen the twitter accounts
>> of half of the world's most powerful people hacked nearly simultaneously.
>> We also have a USPS being driven into bankruptcy by decades of ideological
>> assault.
>>
>> I  think saying that it is inconceivable that the Census might end up
>> broken beyond repair is... not in good faith. It's a possibility, and
>> experts should be thinking about what a Plan B might be within the
>> constitutional mandate of a census under the terms specified by statute.
>>
>> On Wed, Jul 15, 2020 at 6:48 PM Richard Winger <richardwinger at yahoo.com>
>> wrote:
>>
>>> The Census Bureau has plenty of tools beyond just receiving responses
>>> from individuals.  The Post Office has records; there are utility records;
>>> property tax records including various homestead exemptions that show who
>>> lives at a particular address; credit reporting records; drivers license
>>> records.  All of those records are held by organizations cooperate with the
>>> Census Bureau.
>>>
>>> Richard Winger 415-922-9779 PO Box 470296, San Francisco Ca 94147
>>>
>>>
>>> On Wednesday, July 15, 2020, 3:44:10 PM PDT, Jeff Hauser <
>>> jeffhauser at gmail.com> wrote:
>>>
>>>
>>> What if, due to malice or circumstance (including but not limited to a
>>> pandemic), the census that occurs in 2020 does not provide a result in
>>> which any person of good will can have confidence?
>>>
>>> I'm not talking about litigation around the edges -- I am asking what if
>>> the whole thing appears rotten. What interpretative standards would be
>>> applied by courts, what would be the best work arounds....
>>> _______________________________________________
>>> Law-election mailing list
>>> Law-election at department-lists.uci.edu
>>> https://department-lists.uci.edu/mailman/listinfo/law-election
>>> <https://nam02.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdepartment-lists.uci.edu%2Fmailman%2Flistinfo%2Flaw-election&data=02%7C01%7Ctbrunell%40utdallas.edu%7Cbf073f877abc4bd7271508d829a01415%7C8d281d1d9c4d4bf7b16e032d15de9f6c%7C0%7C1%7C637305114814575454&sdata=%2BiqFeox4MM5SPTflYVk2HiHqRmmcMIQfuu3b1GoxP%2FA%3D&reserved=0>
>>>
>> _______________________________________________
>> Law-election mailing list
>> Law-election at department-lists.uci.edu
>> https://department-lists.uci.edu/mailman/listinfo/law-election
>>
>>
>> _______________________________________________
> Law-election mailing list
> Law-election at department-lists.uci.edu
> https://department-lists.uci.edu/mailman/listinfo/law-election
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20200803/50634bdb/attachment.html>