Re: [ElectionIntegrity] Taskforce on Federal Legislation

From: Kathy Dopp <kathy_dot_dopp_at_gmail_dot_com>
Date: Fri Nov 17 2006 - 13:17:45 CST

On 11/17/06, Barbara Simons <simons@acm.org> wrote:
>
> Dear friends,
>
> Holt is currently working on the version that he will be introducing in
> the next Congress. Several computer scientists who have been fighting
> DREs, myself included, have been asked to submit suggestions. If you
> have concrete politically viable recommendations on the following two
> issues or some other aspects of the legislation, please send them to me
> and I will forward them to Holt's office.

Thanks Barbara.

>
> 1. Who should be responsible for auditing oversight? HR550 called for
> the EAC to have oversight, but we all recognize the major shortcomings
> of the EAC as it is currently composed. However, if states have the
> responsibility, then we are confronted with the likes of Blackwell in
> Ohio and Katherine Harris in Florida. Can you suggest a way around
> this problem?

In our "election integrity audit" paper, we suggest any of three
possibilities for groups who could be independent auditors in any state.
See p. 14 & 15 of
http://www.electionarchive.org/ucvAnalysis/US/paper-audits/ElectionIntegrityAudit.pdf

Audits must be performed by independent auditors who do not work for
election offices or for voting machine equipment or service companies or
otherwise have access to voting equipment. Auditors could consist of:

1. state audit officers (many states have audit branches); or
2. election judges or poll workers who did not work during the election to
be audited; including persons whose names are supplied by candidates and
authors of ballot measures; or
3. independent accounting firms.

2. What should be the percentage of ballots that are audited? Should
> it be an absolute number, eg 2%? If not, how should the percentage be

Absolutely NO FIXED audit rate should be used. The language should simply
say something like:

"audit an amount of vote counts sufficient to detect any amount of vote
miscount which could have wrongful altered an election outcome" using the
minimum margin between candidates in the state to calculate.

Audit amounts must be calculated on at minimum, the state level, but even
better at the county level and definitely not, under any circumstances, on
the federal level like the current Holt bill.

Charlie Strauss is working on the wording for new legislation in New Mexico.

Please run any legislative proposals by me - i.e. let me read them to look
for flaws or omissions that could cause the audit to be wrongly manipulated.

arrived at? If yes, what do you think that figure should be? Please
> keep in mind that most election officials don't even understand the
> notion of randomness and certainly won't be able to deal with a complex
> mathematical formulation.

No mathematical formulation is needed in the legislation as long as it
simply says, detect any amount of vote miscount that could wrongly alter
an election outcome.

Requiring a state audit committee with a qualified PhD mathematician or PhD
degreed statistician (not someone with a degree in another field who calls
him/herself a statistician) to evaluate or calculate the audit amounts would
not hurt.

If you are concerned about "overhauling" HR 550, now is the time to
> speak out on those concerns. But please think strategically. We could
> push for a "perfect" bill that has no chance in hell of passing, or we
> could push for a very good bill that significantly improves the current
> situation and that might actually become law. To me the choice is a
> no-brainer.
>
> Also, please keep in mind that if some version of HR 550 becomes law,

It HR550 became law as is, it could do more to deter efforts at sufficient
state-wide audits than it does to ensure election outcome integrity. As you
know, VVPBs are useless unless manually audited and the current HR550 audit
would do more to interfere with sufficient audits and could be a complete
sham given the way HR550 is currently written.

>
>
> P.S. It can be counterproductive to attempt to get legislation
> modified AFTER it has been introduced. Representatives and Senators
> endorse a specific version of a bill. If you start monkeying with the
> wording, you can end up losing key endorsements, and you also look
> indecisive. This is not good politics. Now, not next month and
> certainly not next year, is the time to propose modifications to HR
> 550.
>

Thank you Barbara. PLEASE ask Holt's office to run a copy of the proposed
audit legislation by me now, prior to when they release it.

If, for instance, the public release of an auditable report of detailed vote
counts on each machine or in each precinct is not made public prior to the
selection of the machine vote counts to audit, then the audit is not
verifiable and is easily manipulated.

I would say that the one document that it would be critical for Michelle
Mulder in Holt's office to read before writing any audit proposal is this:
http://www.electionarchive.org/ucvAnalysis/US/paper-audits/ElectionIntegrityAudit.pdf

Also, I am in the process of rewriting a legislative vote count audit
proposal for Utah but am not finished yet:
http://electionarchive.org/ucvAnalysis/US/paper-audits/VoteCountAudit-UT.pdf

Here is a list of flaws that the Utah Lt. Governor's audit policies have
that would make an audit easy to manipulate:
http://utahcountvotes.org/ltgov/Response2LtGov-Audit-Recount.pdf

Verified Voting also has an excellent checklist of what a manual vote count
audit should look like in order to be transparent and verifiable.

The nice thing re. requiring audits to be verifiable, is that it kills two
birds with one stone because in order for an audit to be verifiable by the
public, the election officials must, prior to the random selection of vote
counts, release a detailed list of what the vote counts for every machine
and every precinct should be for every vote type (election day, absentee,
military, early, provisional, and so forth - whatever exists in each
state). This means that not only is the audit verifiable by the public, but
also the data for independent election forensics analysis is publicly
available prior to making any vote counts official.

Again, please ask me to review the actual text of the audit legislation
prior to releasing it to the floor.

Thank you very much Barbara.

Best,

Kathy

_______________________________________________
OVC-discuss mailing list
OVC-discuss@listman.sonic.net
http://lists.sonic.net/mailman/listinfo/ovc-discuss

==================================================================
= The content of this message, with the exception of any external
= quotations under fair use, are released to the Public Domain
==================================================================
Received on Thu Nov 30 23:17:08 2006

This archive was generated by hypermail 2.1.8 : Thu Nov 30 2006 - 23:17:19 CST