Re: A 3-Step Audit Protocol w/ 99%confidence

From: Kathy Dopp <kathy_dot_dopp_at_gmail_dot_com>
Date: Fri Jan 26 2007 - 18:35:34 CST

> Date: Fri, 26 Jan 2007 12:43:55 -0700 (GMT-07:00)
> From: charlie strauss <>
> Subject: Re: [OVC-discuss] [EILeg] [ei] Re: A 3-Step Audit Protocol w/
> 99%confidence

> I agree with arthur, this fudge factor is the achilles heel of the recount strategy. The good news however is two fold.

That is simply not true Charlie.

1. Any analysis without a maximum vote shift per vote count would be
logically incoherent since

a. there are usually not 100% of votes available for target within any
one vote count, and

b. if even if there were, it would be a pretty bad strategy to steal
every possible vote, since there has to be atleast a few voters that
might notice and complain, and

2. The assumption is merely for the amount of vote shift that is
considered to be "NOT IMMEDIATELY NOTICEABLY SUSPICIOUS" (read all our
literature - beginning with Roy Saltman), so that any amount that was
shifted over that amount is immediately noticeable and suspicious. So
any vote count that has over the amount of 2 times the margin shift
from either:

a. a prior election - but that method has its drawbacks, or

b. relative to partisanship of active voter registration relative to votes, or

c. relative to available pre-election or exit polls

is then automatically thrown into the audit in addition to the
randomly selected precincts.

So, pray tell, where is the weakness?

I realize that we haven't sufficiently stressed this point yet, but it
is mentioned in footnotes in one or more of my papers on audit
mathematics and I have had email conversations about it with Ron
Rivest when he asked how to decide if the amount of shift in a vote
count were over the assumed amount. And Yes, this area of the
decision-making process still needs more research, like a lot of the
election audit field.


OVC-discuss mailing list
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Tue Jan 1 14:12:49 2008

This archive was generated by hypermail 2.1.8 : Tue Jan 01 2008 - 14:12:51 CST