[EL] do long lines occur more in black precincts?

Mon Oct 22 12:44:10 PDT 2012

I sent the below to Michael earlier and he asked that I share it with the list (despite it being a war story):

Michael,

I am just a practitioner, but I have not seen this anywhere in the past.  What I do know exists are turnout models based on past performance.  AA models (correctly in my experience) show a much more consolidated turnout between 3 pm and closing.  White models (and to a lesser extent the Hispanic models) tend to have three bumps - before work, lunch, and early evening.  Black voters, for whatever reason, and no matter how hard campaigns try to get voters out early, overwhelmingly don't vote before work.  As a result, in my experience, AA-heavy precincts tend to end up with substantial afternoon lines, when most other precincts do not, largely because they have balanced out their voters.

This certainly matches the turnout modeling which the consultants use to predict turnout.

Interestingly, this appears to be cultural and not class.  Middle/upper class white, working class white, and Mexican-American-dominated precincts all have a three bump pattern, while both poor and middle/upper class black communities (and Puerto Rican-dominated precincts, to a lesser extent) seem to vote late.   At least in my experience.

I have also seen excessively long lines caused by translation issues where there is a poor quality translator.  And long lines anywhere there is a new judge of elections.  Anecdotally, I suspect (and based on the Ohio footage, there may be good evidence for) similar issues occurring on/near college campuses.

I know - all of the above are massive generalizations.

A personal anecdote:

In 2004 I was a temporarily-recovering lawyer and was running a number of wards that were pre-dominantly white in Philadelphia for the Kerry campaign (can't remember the other guy on that ticket, for some reason).  We had 97% of expected turnout at our 3 pm count and essentially were out of voters to knock-and-drag.  Did not hear of substantial lines anywhere in my area of responsibility, except very early in the morning.

There were five of us across the city of 1.5 MM and we had a soft protocol for lines over 30 minutes which would trigger coffee, donuts, and "celebrities" to work the lines.  Neither I nor the other person with white-majority wards used it at all.  The two folks who covered the more African-American portions of the city ran out of resources, even after we re-tasked ours to assist them.

Kevin

Kevin Greenberg
Flaster/Greenberg PC<http://www.flastergreenberg.com/>

1600 John F. Kennedy Boulevard, Second Floor, Philadelphia, PA 19103
Phone: 215-279-9912 Fax: 215-701-1151
Email: kevin.greenberg at flastergreenberg.com<mailto:kevin.greenberg at flastergreenberg.com>

email<mailto:kevin.greenberg at flastergreenberg.com> l bio<http://www.flastergreenberg.com/people-Kevin_Greenberg.html> l offices<http://www.flastergreenberg.com/locations.html> l v-card<http://www.flastergreenberg.com/vcard-197.vcf>

From: law-election-bounces at department-lists.uci.edu [mailto:law-election-bounces at department-lists.uci.edu] On Behalf Of Doug Spencer
Sent: Monday, October 22, 2012 3:31 PM
To: Rausch, Dave
Cc: law-election at uci.edu
Subject: Re: [EL] do long lines occur more in black precincts?

Michael --

To my knowledge there is currently no good measure of voter elasticity for waiting in line, but there are some good papers that look at ballot length and voting machine allocation and find correlations to voter turnout (with the assumption that the mechanism for decreased turnout is long lines and tired voters). See, for example, see papers by Ben Highton<http://faculty.psdomain.ucdavis.edu/bhighton/copenhagen2012/readings/Highton_lines_ps_2006.pdf> and Allen & Bernshteyn<http://www.amstat.org/publications/chance/2006/CHANCE%2019_4.pdf> (in addition to the ones Lori cites to above).

With regard to racial statistics, Charles is correct that independent observers can reliably collect this information -- we did in 2008 in California's Bay Area (I am one of the authors of the Berkeley study). We did not see significantly different line lengths in Black or Latino precincts, but I should note that we did not see very long lines in any precinct so our findings are not particularly instructive on this point.

However, as we point out in our paper, three things contribute to the formation of polling station lines: (1) rate of arrival, (2) interaction with poll worker, and (3) casting the ballot. Any one of these factors can lead to long lines independent of the others, and to the extent that we really want to diagnose the problem and solve it, we need to look at correlations between each of these factors and other relevant statistics (e.g., race). This is obviously outside the scope of the SPAE and CCES, but very important for understanding not just when/where lines form, but also why.

Doug

-----
Douglas M. Spencer
Jurisprudence and Social Policy Ph.D. Program
University of California, Berkeley, School of Law
2240 Piedmont Avenue
Berkeley, CA 94720-2150
(415) 335-9698

URL: http://www.dougspencer.org

On Mon, Oct 22, 2012 at 9:09 AM, Rausch, Dave <jrausch at mail.wtamu.edu<mailto:jrausch at mail.wtamu.edu>> wrote:
A student paper from the University of California Berkeley looks at waiting times during the 2008 presidential primary election in California:

http://www.vote.caltech.edu/sites/default/files/Lines%20at%20Polling%20Stations.pdf

Dave Rausch

Dave Rausch, Ph.D.
Teel Bivins Professor of Political Science
Faculty Athletics Representative
Dept. of Political Science and Criminal Justice
West Texas A&M University
Canyon, TX 79016-0001
(806) 651-2423<tel:%28806%29%20651-2423>    Fax: (806) 651-3610<tel:%28806%29%20651-3610>

My webpage:
                http://www.wtamu.edu/~jrausch

From: law-election-bounces at department-lists.uci.edu<mailto:law-election-bounces at department-lists.uci.edu> [mailto:law-election-bounces at department-lists.uci.edu<mailto:law-election-bounces at department-lists.uci.edu>] On Behalf Of Michael Peshkin
Sent: Monday, October 22, 2012 10:38 AM

To: law-election at uci.edu<mailto:law-election at uci.edu>
Subject: [EL] do long lines occur more in black precincts?

Back in '08 I published a study suggesting that long lines at polling places may especially afflict black voters.

Rick asked on this list "I'd be interested to hear from others if the methodology of this study is sound (especially in its reliance on news reports of long lines)."   After which I got pilloried.

Now we're again about to see news footage of those determined black voters in Ohio waiting in line for hours -- and nothing about the ones who went home to feed the kids.

Is it even possible to measure the wait time - vs - race statistics?   Are there practical yet methodologically sound sampling techniques?

Has anyone ever done it?   What do you guess they would find?

And would it matter?   Is the elasticity known, for turnout - vs - wait time?

--

Article: http://inthesetimes.com/article/4068/are_long_lines_the_new_poll_tax/

Graph: http://peshkin.mech.northwestern.edu/longlines/longlines-vs-black.pdf

Discussion on this list: http://mailman.lls.edu/pipermail/election-law/2008-December/date.html

_______________________________________________
Law-election mailing list
Law-election at department-lists.uci.edu<mailto:Law-election at department-lists.uci.edu>
http://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/20121022/866b74d5/attachment.html>