[EL] Only 33, 000, Not 240, 000, South Carolina Voters Lacking ID?

[Doug Hess]

I found the article on the SC numbers rather oddly written or incomplete.
For instance, it states that some number "allowed their ID cards to expire,
probably have licenses with names that didn't match voter records or were
dead."  Hmmmm....allowed to expire, but are they still useable for voting?
What does "probably have" mean? And is "probably" also meant to modify
"dead"?  See link for text:
http://www.postandcourier.com/news/2011/dec/23/election-agency-data-called-flawed/


A lesson for advocates here is that when using numbers, it's well worth
using a range and emphasizing, in cases like this, that even a smaller
amount of harm makes your point. Granted that does not help with creating
soundbites, but it is analytically fair. If you know, even vaguely, what
the threshold is on some criterion, you just need to pass it by a
reasonable margin.  I.e., instead of stressing an exact number, which then
distorts how the number is perceived when people quibble over it, it would
be better to stress that a back-of-the-envelope estimation is that tens of
thousands of people will face this obstacle to voting. Add to that
the number people that forget to bring their ID, headaches that
implementation problems will bring about, and confusion from the public
(e.g., not knowing if you have the right ID), and the law looks bad enough
in light of the "non-problem problem" it is meant to address.

The issue for the DOJ clearance decision, I believe, is about another
aspect of the data, but the danger of stressing the size of the problem
with a rough estimate is the point: Now some pundits, politicians, and
others (judges?) will ignore just how big 30,000 or 90,000 is and focus on
it being a smaller number than 240,000.

-Doug
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://webshare.law.ucla.edu/Listservs/law-election/attachments/20111223/c1b44664/attachment.html>