[EL] Impact on the census

[Levitt, Justin]

Doug’s right, of course.  The article (and most of the other studies) only estimates anticipated changes in self-response.  For those who do not initially respond to the survey, the Census will attempt to follow up, including with door-to-door enumerators.

It’s just that there’s reason to think that the same population that declines to respond to the census by Internet or mail because of the citizenship question may be even less likely to respond to personal door-to-door follow-up from the federal government.  And in communities feeling fearful, follow-up with neighbors may be no more successful or accurate. The precise efficacy of the non-response follow-up operations in this climate is really, really difficult to assess through historical study (like the study in the paper).  Which is why a real test under real-world conditions might have been informative.  Perhaps even before the decision itself was finally made.

Justin

From: Douglas Johnson <djohnson at ndcresearch.com>
Sent: Sunday, June 23, 2019 10:41 PM
To: Levitt, Justin <justin.levitt at lls.edu>
Cc: Election Law Listserv <law-election at uci.edu>; Rick Hasen <rhasen at law.uci.edu>
Subject: Re: [EL] Impact on the census

Just so there’s clarity (since so many general news media articles have completely missed this difference):

what this article studies is the change in self-response rates, which if they decline (to quote the article) “would predict an overall 2.2 percentage point drop in self-response in the 2020 census, increasing costs and reducing the quality of the population count.”

This article does NOT measure expected undercount, and the 2.2 percentage point drop estimate is self-response rates,NOT a predicted undercount.

As the article clearly states, there are follow-up efforts that make significant strides toward filling in the data for non-responsive households, with or without the citizenship question. According to this Census study those follow up efforts are likely to be more expensive if there’s is a citizenship question, and this and other articles assert that if the follow up efforts have to rely on neighbor- or administrative records the data may be less precise.

 (This goes back to the often-missed fact that final census records for each household are never partial responses - if a partial form is received, it will get completed one way or another.)

I am not weighing in either way on the study’s topic (plenty of other sources are already doing that).

My only goal in this message is to clarify for those not deeply immersed in Census methodology that the study does not quantify the expected undercount, just the drop in the expected self-response rate, which is related but very different.

- Doug

Douglas Johnson
Rose Institute of State and Local Government
Claremont McKenna College

On Sun, Jun 23, 2019 at 9:56 PM Levitt, Justin <justin.levitt at lls.edu<mailto:justin.levitt at lls.edu>> wrote:
The paper Rick flags below makes three assumptions – two acknowledged and one unacknowledged.  First, it assumes that the citizenship question affects only households with at least one noncitizen, despite fears in the broader community.  Second, it assumes that response in 2020 will be no worse than response in 2010, despite the change in climate.  Third (this is the unacknowledged one), it assumes that the impact of a question on the 10-question enumeration delivered to every household in the country is no more salient than the impact of a question on a 70-question survey delivered to a tiny fraction of the country each month.

If any of those assumptions doesn’t hold, the impact will be worse.  If several of those assumptions don’t hold, that impact will be compounded.

Oh, and the paper tallies its final results in terms of household response.  If the households of those who don’t answer the census are larger than those who do, the error rate of the total count will be larger still.

None of that is a critique of the paper.  It’s just that there’s plenty of reason to think that these are still low-ball estimates of damage.  Don’t be surprised when the oil spill<https://www.britannica.com/event/Deepwater-Horizon-oil-spill> is a lot bigger than the initial estimates.

Justin

From: Law-election <law-election-bounces at department-lists.uci.edu<mailto:law-election-bounces at department-lists.uci.edu>> On Behalf Of Rick Hasen
Sent: Sunday, June 23, 2019 9:21 PM
To: Election Law Listserv <law-election at uci.edu<mailto:law-election at uci.edu>>
Subject: [EL] ELB News and Commentary 6/23/19

New Estimate from Census Department Researchers Forthcoming in Peer Reviewed Journal Finds Citizenship Question Could Depress Response Rate by 8 Percent in Households That May Have Noncitizens<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2felectionlawblog.org%2f%3fp%3d105741&c=E,1,8R6RtcVVlv13r6DI0CzsAgz98xhdi6TiQCxhF_nQ7UhwRM90zAXlAkmJnPwv2CfjgCEeHYOeFIUvdQ-1z-gfpYqffJ_EpEfyrbdRiEg77ie-Cw,,&typo=1>
Posted on June 23, 2019 9:15 pm<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2felectionlawblog.org%2f%3fp%3d105741&c=E,1,uV-ltDXc0GyNguK3VMqyayOF8jSEyM2gun3TjHlz5ST0opKX7u-pbWVowcNw3gJcZQIIMQ2pOGWW7MYMZJSfnoxRsphOdsGhLbAsdZQhPlzV7fvB1h0,&typo=1> by Rick Hasen<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2felectionlawblog.org%2f%3fauthor%3d3&c=E,1,HnXH0N8WWwq6lw4NWloH6sw_3KcrptEbEliK2G-EpG2eQ0lGnx7ML6t1ZMqiR245on9EI-J34KyhVQTAWbhcs-CDIk7c7eUiTx_hlOfIfmCHFxIkYfZF5l3s&typo=1>

This paper<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fassets.documentcloud.org%2fdocuments%2f6165808%2fU-S-Census-Bureau-Working-Paper-Understanding.pdf&c=E,1,WuWh_ix14K6jbrLQevvw8qrq-hzZm1NqPoTaQ-VCgeTXLQ0XYMuo1LjyT7Yi4rntjhGUWM5ECJLm6LeVA2UFF7Vff_veVj1jxvoifBb_WAZX5H4t6DCjYQ,,&typo=1> is forthcoming in Demography (h/t Hansi Lo Wang<https://twitter.com/hansilowang/status/1142936853575852032>). Here is the abstract:

he addition of a citizenship question to the 2020 census could affect the self-response rate, a key driver of the cost and quality of a census. We find that citizenship question response patterns in the American Community Survey (ACS) suggest that it is a sensitive question when asked about administrative record noncitizens but not when asked about administrative record citizens. ACS respondents who were administrative record noncitizens in 2017 frequently choose to skip the question or answer that the person is a citizen. We predict the effect on self-response to the entire survey by comparing mail response rates in the 2010 ACS, which included a citizenship question, with those of the 2010 census, which did not have a citizenship question, among households in both surveys. We compare the actual ACS-census difference in response rates for households that may contain noncitizens (more sensitive to the question) with the difference for households containing only U.S. citizens. We estimate that the addition of a citizenship question will have an 8.0 percentage point larger effect on self-response rates in households that may have noncitizens relative to those with only U.S. citizens. Assuming that the citizenship question does not affect unit self-response in all-citizen households and applying the 8.0 percentage point drop to the 28.1 % of housing units potentially having at least one noncitizen would predict an overall 2.2 percentage point drop in self-response in the 2020 census, increasing costs and reducing the quality of the population count.

Recall that at oral argument in the census case, Justice Gorsuch sought to minimize census department studies which predicted (a smaller) response rate drop.
[Share]<https://www.addtoany.com/share#url=https%3A%2F%2Felectionlawblog.org%2F%3Fp%3D105741&title=New%20Estimate%20from%20Census%20Department%20Researchers%20Forthcoming%20in%20Peer%20Reviewed%20Journal%20Finds%20Citizenship%20Question%20Could%20Depress%20Response%20Rate%20by%208%20Percent%20in%20Households%20That%20May%20Have%20Noncitizens>
Posted in census litigation<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2felectionlawblog.org%2f%3fcat%3d125&c=E,1,lvWepiDyt5wvptovl_lCmSahdnxrPkLL2r-Zs1BPQ_RT7q9VVOlC5fFwjLiiai7bHnZgOdrOtqwB8AypEmzcyFxduR2ou5tXaru8K5hszTrCadHagYWhOJhG&typo=1>

_______________________________________________
Law-election mailing list
Law-election at department-lists.uci.edu<mailto:Law-election at department-lists.uci.edu>
https://department-lists.uci.edu/mailman/listinfo/law-election<https://linkprotect.cudasvc.com/url?a=https%3a%2f%2fdepartment-lists.uci.edu%2fmailman%2flistinfo%2flaw-election&c=E,1,su7NfoCRe0_zMVfwUncD-h_xTq4Iv-Svb0y5_z6w0pYCaRWpd9NxvY1d4YxAfffgsG45W4dRRWonhfTlUN0i0PKNkz901O-05NySwEpKbMLU7QRtgqT0BfQWqnwI&typo=1>
--
- Doug

Douglas Johnson
National Demographics Corporation
djohnson at NDCresearch.com<mailto:djohnson at NDCresearch.com>
phone 310-200-2058
fax 818-254-1221
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20190624/ec6ac268/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: image001.png
Type: image/png
Size: 2021 bytes
Desc: image001.png
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20190624/ec6ac268/attachment.png>