[EL] EAC registration data is just as faulty as the Census Bureau's data

[Charles H Stewart]

For the record, the Elections Performance Index (originally published by Pew, now being updated by the MIT Election Data and Science Lab) uses the Census Bureau registration rate (calculated the correct way, not the way the Census Bureau calculates it) for its comparison of election rates across states.  The main thing I’ve learned by studying these statistics for the past decade is that one needs to approach them with humility. -cs

-------------------------------------------------------------------------------------------------------------
Charles Stewart III
Kenan Sahin Distinguished Professor of Political Science
The Massachusetts Institute of Technology
Cambridge, Massachusetts   02139
617-253-3127
cstewart at mit.edu<mailto:cstewart at mit.edu>

From: Richard Winger <richardwinger at yahoo.com>
Sent: Monday, June 18, 2018 12:11 PM
To: Charles H Stewart <cstewart at mit.edu>; Election Law Listserv <law-election at uci.edu>
Subject: EAC registration data is just as faulty as the Census Bureau's data

Thanks to Charles Stewart for the link to the EAC report, showing the number of registered voters in each state in November 2016.  I am completely puzzled that this data is so much at variance with the official data reported by each state.  For example, the EAC report says there were 24,486,638 registered voters in California, yet the California Secretary of State has never released any report that shows as many as 20,000,000 registered voters.


I obtained the number of registered voters from each state's elections office, and for November 2016 it adds up to 196,144,094.


Yet the EAC says the national total of registered voters was 214,109,367.  And the Census Bureau says it was 157,596,000!


Curtis Gans and his Committee for the Study of the American Electorate always did a great job of compiling registration data, but Gans died and his organization apparently died with him.


America Votes, the election returns reference books published every 2 years, depended on Curtis Gans' data thru 2012.  For 2014 and 2016, it used the Census data, which is clearly wildly inaccurate.

Richard Winger 415-922-9779 PO Box 470296, San Francisco Ca 94147

________________________________
From: Charles H Stewart <cstewart at mit.edu<mailto:cstewart at mit.edu>>
To: Richard Winger <richardwinger at yahoo.com<mailto:richardwinger at yahoo.com>>; Election Law Listserv <law-election at uci.edu<mailto:law-election at uci.edu>>
Sent: Monday, June 18, 2018 6:07 AM
Subject: RE: [EL] does anyone know where to find no. of registered voters in each state as of 2016 general election?

While there are issues with the data (as there always is with voter registration stats), the one place I know of with (roughly) comparable registration statistics for all states is the EAC NVRA biennial report.  Here’s the link to the link to the most recent report: https://www.eac.gov/assets/1/6/2016_EAVS_Comprehensive_Report.pdf

Charles

-------------------------------------------------------------------------------------------------------------
Charles Stewart III
Kenan Sahin Distinguished Professor of Political Science
The Massachusetts Institute of Technology
Cambridge, Massachusetts   02139
617-253-3127
cstewart at mit.edu<mailto:cstewart at mit.edu>

From: Law-election <law-election-bounces at department-lists.uci.edu<mailto:law-election-bounces at department-lists.uci.edu>> On Behalf Of Richard Winger
Sent: Monday, June 18, 2018 12:36 AM
To: Election Law Listserv <law-election at uci.edu<mailto:law-election at uci.edu>>
Subject: [EL] does anyone know where to find no. of registered voters in each state as of 2016 general election?

Although I regularly gather data on the number of registered voters in each state, I only do it for the states that have partisan registration.  But now I need to know for every state, including the 19 without registration by party, and I am surprised that I can't find it in one place on the internet.

The Census Bureau purports to have this, but its data is not at all accurate.  I am very surprised that it shows California as 16,096,000 (Census rounds off this data).  But every California registration tally during 2016, from the Secretary of State of California, ranged from 17,300,000 to 19,000,000.  Never was California as low as the census figure.  Nor was California as low as the census data in 2015 either.

Professor Michael McDonald has data on the number of votes cast in each state, and the number of individuals who could potentially have registered and voted.  But unless I am not seeing it, there is no registration data there.

Richard Winger 415-922-9779 PO Box 470296, San Francisco Ca 94147

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20180618/a3324840/attachment.html>