Errors Re: checking into the 2004 general election

From: Edmund R. Kennedy <ekennedyx_at_yahoo_dot_com>
Date: Wed May 25 2005 - 10:06:01 CDT

Hello Charlie and Teresa:

Yes, I had something similar happen during the
November 2004 elections in my polling place. None of
the totals; poll book signatures, ballots, and the
counter on the scanner, reconciled. The error I
noticied could be characterized as significant as it
was around 15% of the votes cast. I pointed this out
to County Registrar and he still hasn't responded. I
guess this is the time to mention the registrar's
prayer, "Lord, please don't let the election be

At one time we had a discussion as to what was a
reasonable expectation for errors especially in terms
of a percentage of votes cast. I know that in many
states, a election with a margin of victory of one
tenth of one per cent automatically requires a
recount. Should we take this as an acceptable
general error rate for elections? IMHO it seems a
little strict but I am deliberately mixing apples and
oranges here. I think I would propose that an error
rate of one and one half percent of the votes cast
(including provisionals) should be considered an
acceptable error rate for an election. I don't have
any particular reason for this number to base this on
but we have to start somewhere.

I will note that this number is the combined error
rate for errors generated by both the technology and
the adminstrative process. As conducting an election
is something that we typically do only one day a year,
or less frequently, and is usually staffed by
volunteers, some errors are likely to be inevitable.

Anyway, I think this is an important basic screening
number to discuss and I'm hoping that some of the
studies underway can put a better handle on this.

Thanks, Ed Kennedy
--- Charlie Strauss <> wrote:

> yeah new mexico has several counties with more votes
> than cast
> ballots in the official certified record. In one
> case the race for
> county clerk, of all races(!), showed 300+ votes
> cast and the winner
> (the incumber clerk) getting over 2000 votes in that
> precint. Elen
> has it on her web site, and you can see it
> yourselfon the SOS web
> site. I pointed it out to the election director and
> he did not deny
> ot or offer any explanation.
> I note that in many places, notably ohio, there is
> this rather dumb
> practice of attributing all the absentee votes for
> an entire
> legislative district to a single precint while
> (inconsitently)
> assinging the numbers of ballots cast to the proper
> precincts. As I
> understand it this is an artifact of the limited
> memory capabiltiy of
> the scan systems. the systems are able to count
> totals in each race
> and totals of ballot types, but it cant break down
> the race totals by
> ballot type. hence the seemingly odd way of
> reporting that always
> produced a few precints where the number of absentee
> votes exceeds
> the number of absentee ballots. (they should simply
> report the
> absetnee votes without attributing them to a precint
> but they dont).
> However in the New Mexcio case this is not the
> issue. those are
> actual day-of-voting totals that grossly exceed the
> number of voters.
> they are not absentee ballots. This seems to happen
> in every
> election in some county here.
> On May 24, 2005, at 10:21 PM, Teresa Hommel wrote:
> > Ellen Theisen of found some very
> strange info when
> > she did this. Have you checked her web site?
> > Teresa Hommel
> >
> > Stephanie Frank Singer wrote:
> >> Hey gang, this is off-topic, but I'd appreciate
> your input.
> >>
> >> Has anyone done the following check on the
> validity of the 2004
> >> general election? Each precinct reports a total
> number of voters,
> >> presumably read off of the voting machines or
> ballot boxes. This
> >> number should be the sum of all the votes for
> each candidate, plus
> >> the number of any undervotes.
> >>
> >> On the other hand, each state has a database of
> voters, including
> >> information about which elections they voted in,
> presumably taken
> >> from the sign-in sheets at the polls. So, if one
> had that data,
> >> one could have an independent count of how many
> voters voted in a
> >> particular precinct in the 2004 general election.
> >>
> >> Is the comparison of these two numbers practiced
> by any Boards of
> >> Elections?
> >>
> >> Has any independent entity checked these numbers?
> >>
> >>
> >> Stephanie Frank Singer, Ph.D.
> >> Campaign Scientific
> >> 215-715-3479
> >>
> >>
> >> _______________________________________________
> >> OVC discuss mailing lists
> >> Send requests to subscribe or unsubscribe to
> >>
> >>
> >
> > _______________________________________________
> > OVC discuss mailing lists
> > Send requests to subscribe or unsubscribe to
> >
> > _______________________________________________
> OVC discuss mailing lists
> Send requests to subscribe or unsubscribe to

10777 Bendigo Cove
San Diego, CA 92126-2510
Work for the common good.
My profile:  <>
I blog now and then at:  <>
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 Tue May 31 23:17:48 2005

This archive was generated by hypermail 2.1.8 : Tue May 31 2005 - 23:17:53 CDT