Re: A generic best practice document for NewMexicolegislators

From: Ken Pugh <kpughmisc_at_pughkilleen_dot_com>
Date: Tue Jan 04 2005 - 10:31:00 CST

I think we may be payment for two different issues, but maybe not.

If an election is decided by a large margin (say 5%), then an error rate of
1% in the system should not affect the outcome. In that situation, to have
a recount of all the ballots is something that I would expect the losing
candidate to pay for.

Even if the margin of victory is large, the system should still be audited
for errors. That is something the state should pay for in every
election. There were proposals on this mailing list for an audit of 1% of
the ballots. I suggest that if that audit shows that the error rate is
greater than the estimated (e.g. 1%), then the margin of victory cutoff for
candidate payment should be increased.

I thoroughly agree with you that the error rate should be as low as
possible. As a computer guy, I would like to see it at zero. But I know
that it not practical. My gut feeling says it should be less than
.1%. That rate is for the whole process - from check-in through
tally. The computer portion should be less than .001%. The remainder of
the error comes from the human part of the process.

Just to give an example of a .1% error rate, assume that we are using paper
and computer. And that the computer ballots have an exact match with paper
ballots, although there may be one more or less. So therefore the ballots
that match have a 0% error. [That should be the goal, but perhaps a .001%
or .0001% might be reasonable, due to errors in doing the manual

Then we have the following possibilities. Which of the possibilities should
be considered innocent and which deserve investigation?

     Sign-ins Computer Paper Error-rate

a.) 1000 1000 1000 0
b.) 1000 1000 999 .1
c.) 1000 1000 1001 .1

d.) 1000 999 1000 .1
e.) 1000 999 999 .1
f.) 1000 999 1001 .2

g.) 1000 1001 1000 .1
h.) 1000 1001 1001 .1
i.) 1000 1001 999 .2

a.) is the goal.
Possible reasons for the others:

b.) means someone didn't drop their receipt in the box
e.) may mean someone walked in, registered, and didn't vote
h.) may mean that some voter didn't get marked as checked in
g.) and i.) could be a combination of b.) and h.)

c.), d.), f.) need some explanation.

The b.), e.) and h.) errors are the type of human errors I'm referring to.


At 12:21 AM 1/3/2005, you wrote:

>On Wednesday 29 December 2004 07:59 am, you wrote:
> > At 10:20 AM 12/29/2004, you wrote:
> >
> > For example, if the estimated error rate is 1% and the desired
> > probability of a wrong victor declaration was set at .5%, then
> > the maximum margin of victory for which the state would pay
> > for the recount would be maybe 5% (statisticians please help
> > out here with an exact number).
>I disagree. I believe that states should be under an obligation
>to perform and pay for investigation of any statistically
>significant anomaly, whether or not it could affect the current
>outcome. If we can catch them small, we can prevent them growing
>into something worse, with a quantifiable probability of
>improved outcomes that anyone in statistical quality control
>should be able to look up in a handbook.
>At a guess, I would say the state should pay for investigation of
>anything over half the average error rate, regardless of its
>impact on the result of the election, and not set an integer
>multiple at all. My theory is that current practice is
>remarkably sloppy, and should not be normative. And neither
>should the measured error rate at any given point in time. Why
>stop there, no matter where there is? Where we have measurable
>error, we should have an audit that tells us where most of the
>error comes from, and we should fix it.
>When we have not just numbers, but a reasonably accurate
>multivariate model of the error process, we can choose a more
>optimal target.
>Edward Cherlin
>Generalist & activist--Linux, languages, literacy and more

>"A knot! Oh, do let me help to undo it!"
>--Alice in Wonderland
>OVC discuss mailing lists
>Send requests to subscribe or unsubscribe to

OVC discuss mailing lists
Send requests to subscribe or unsubscribe to
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
Received on Sat Jan 7 22:28:55 2006

This archive was generated by hypermail 2.1.8 : Sat Jan 07 2006 - 22:28:59 CST