new buzzword - "MMRA" - Mandatory Manual Random Audits

[garybeck]

the Voter Verified Paper Ballot buzzword seems to have really caught on. people are saying VVPB and we all know what it means. I think that use of the term has effectively conveyed a message and it would be good to duplicate this method.

I think there's another element to election reform that is just as important as VVPB and it would be good if we all agreed on a name for it and started using it more consistently.

The concept is the AUDIT and these key points are important:

1) must be Mandatory ("auditable" is NOT ENOUGH!)
2) must be Random
3) must be Manual (not just a "redo" of the machine count)

so how about we call it MMRA

Mandatory Manual Random Audits

If there are other key elements of the audit that I'm missing please post them here and we can iron out an improved acoronym. then let's start talking it up!

-gary

[BeFree]

1. If the idea is

To make it impossible to use DREs, then I'm all for it.

The increased pressure and observance of those corruptible machines leads down but one path: Obliteration of DREs.

If the mass of voters begins to demand a vote counter as good or better than their ATMs, the trail blazes are well set, and the mob begins to move in our desired direction.

If the sole intent is to put cosmetics on e-voting, then it's just a waste of time and effort.

[Stevepol]

2. I think if it's mandatory FOR EVERY ELECTION, that's it.

I actually think if the audits are random and conducted FOR EVERY ELECTION, this would solve all the problems. Of course, there should be other safeguards as well, e.g., no throwing away of the paper, the paper trumps the machines in case of a recount, companies that produce machines caught cheating should be subject to prosecution and severe penalties, etc.

In fact, a really good machine like the Avante for example, should be superior to all other methods I believe. It would eliminate over- or under-votes, and the paper ballot would be easily readable by anybody recounting or auditing (no messy circles that could be interpreted in two ways, no hanging or pregnant chads, etc.)

Still, equally acceptable would be all paper.

[garybeck]

3. no doubt. every election.

that's why it's MANDATORY. "auditable" is not good enough. not even close. if legislation calls for mandatory audits, it would apply to all national elections. I feel strongly, and legislation without MMRA is weak and will not protect us from fraud. a paper ballot does no good sitting in a pile somewhere waiting for a candidate to call for a recount.

gary

[Fly by night]

4. I like it very much -- I want to make this part of the law retroactive...

... because my county in middle Tennessee used paper ballots in 2004 that were opti-scanned by an ES$S machine, and the county election commission is required to hold those paper ballots for another 20 months. I am anxious to perform an ole' MMRA-ectomy on at least 5% of my county's ballots and at least 5% of the ballots in as many other Tennessee counties as possible for the 2004 election to see if there are any smoldering embers of the election theft here. In middle Tennessee, I wouldn't be surprised ....

MMRA -- all the way!!

[Wilms]

5. May I add...

"Access with Privacy and Security" APS

[Bill Bored]

6. SSMMRA

Statistically Significant Mandatory Manual Random Audits.

The problem is what percentage do you audit randomly?

What do you do when there's a discrepancy, or who decides what to do?

A lot of this is being left up to BOEs, the EAC, etc. I say put a minimum standard in the Law!

Obviously, if there's a discrepancy, you go by the VVPB count, but then do you just stop or do you do additional MMRAs, and if so how many before you go to a 100% hand count?

I've recently proposed that if a discrepancy is found, you audit additional machines in the jurisdiction at double the original percentage, and you keep doubling it until there are no more discrepancies. Leave the BOE free to exceed this standard, but never audit any less than the standard.

For example, say you audit 5% at random and there's a 1-vote discrepancy. Then you have to audit ANOTHER 10%, chosen randomly, and go by the paper count for any discrepancies. If you find an error in the 10% audit, audit ANOTHER 20%, and so on. If the errors keep coming, you keep auditing until you count all the votes by hand. If the initial problem was a "glitch", then you stop, pending a recount.

Note that the Ohio recount law was even more strict but I'm talking about the initial count here. An audit still is not a recount. It's a check to see that the first count was correct. After that, recount laws go into effect but the above audit is paid for by the taxpayers.

[Land Shark]

7. What about catastrophic error?

It seems that when catastrophic "error" is possible such as boarding planes or entering courtrooms with weapons, checking only a 2% or 4% sample is not adequate to assure us of the character of the poulation as a whole. If one voting machine is capable of catastrophic error (and with Washington state's 129 vote gubernatorial race, it would seem that 130 votes (or 260 votes) would be catastrophic) as seems likely, then we may need more than just small samples.

Or, the sample size increases when the margin of apparent victory decreases, until a trigger point (1%?) gets an automatic recount of 100%.

These trigger points for recounts are deceptive with machines, because they are based on the paper ballot paradigm of moderate numbers of errors of small magnitude, while machines when working properly have a lower rate of error but the errors are larger in magnitude. Thus, a single error or fraud in a central tabulator could easily put any race outside the margin for an allowable recount, if there is such a number in the state in question. Alternatively, it may increase or shift the cost of who pays for the recount.

[Bill Bored]

12. These were not meant to be the only checks.

They were meant to serve as a minimum for the BOEs. The bill as it stands now only says to audit 3% and leaves the rest up to the BOE. I'm saying if the 3% isn't perfect, audit more. And at the BOEs discretion audit everything.

There are some laws on the books already requiring verification of the code in each machine and the new law mandates open source code. I have to read more. While not perfect, these are steps in the right direction.

I am concerned about the recount laws. At present, every lever machine is canvassed twice. I haven't had a chance to see how the new machines will be recounted, which if any sections of the law will be deleted, etc.

I agree the DRE is more trouble than it's worth! That's why we are pushing for OpScans. Both will probably be legal though.

[garybeck]

9. I agree, it's about the INITIAL count, not the recount

the initial count has to be audited

my 2 cents on your comments about statistically significant-

all well taken and good ideas. i've also dreamed up ways to figure out how many to include in the mandatory audit and what happens if a certain percentage are off.

for me, this is where I'd like to leave it to the experts. there are people, like JOhn Zogby comes to mind but there are many other experts on statistics. They should be consulted to come up with a foolproof plan. they know how many have to be sampled in the first audit and what to do if a problem is found. Not me.

I agree with what you say in principal, but not sure it needs to be in the name? I think the word 'audit' itself implies that statistical methods are used to verify accuracy.

I don't want it to seem like I'm proposing that the VVPB and MMRA together are all we need. There are still other issues, like open source code, and partisan election officials.

I just think now, with all the legislation being proposed, we need to focus in on our key issues. What is it that we're asking for? I think VVPB and MMRA are two good ways to get the point across

[Bill Bored]

14. I agree. nt

[Zan_of_Texas]

8. That's good. Now, kick around THIS new idea.

A publically-traded corporation -- has to get an audit from an independent accounting firm.

A publically-funded government entity --is subject to audit or oversight by the GAO.

An election -- is subject to audit by ...
....
....
the same people who gave you the original result.

Why is that?

It's just like the recount/reprint problem. The machine basically gives the same result. The election official may tend to give the same result too, rather than find errors. They may become sloppy or even criminal if no one is looking over their shoulder, ever.

WHY AREN'T ELECTIONS AUDITED/SUBJECTED TO OVERSIGHT BY INDEPENDENT ENTITIES WHICH DO NOT REPORT TO THE ELECTION OFFICIAL?

This is part of the problem that no one is talking about.

In the recount in Ohio, the election officials were in charge of every aspect. The independent observers were at the whim, timing, and conditions of the officials, who often selected "random" precincts to recount secretly.

Kick this around.

Here is one story illustrating -- note especially the officials who told Reuters to take a hike.

Election study finds widespread ballot-counting problems

By THOMAS HARGROVE
Scripps Howard News Service
December 20, 2004
http://www.knoxstudio.com/shns/story.cfm?pk=MISCOUNT-ELECT-12-20-04&cat=AN

- A review of election results in a 10-county sampling revealed more than 12,000 ballots that failed to record a vote for president, almost one in every 10 ballots cast. The unofficial audit by Scripps Howard News Service uncovered malfunctioning voting machines, improperly designed ballots and poor accounting procedures around the nation. The review of certified election returns led authorities to restore 662 votes for president in Louisiana and West Virginia that had been miscounted in easily detectable errors made by local officials. But most of the ballots discovered missing in the study will remain lost.

"I'm so upset over this that I can't sleep," said Sandy Campbell, clerk of Pike County, Ark., upon learning that a damaged optical scanning machine permanently lost nearly 700 votes. "We had no idea this had happened. But I'll know what to look for in the future. We'll try never to let this happen again."

The study - part of a yearlong project examining errors in America's election practices - checked the accuracy of the Nov. 2 election by comparing official results for president against the reported number of ballots cast in more than 2,400 counties nationwide.

Ten counties with some of the nation's worst voting record discrepancies were selected in the project. Local election officials were asked why their vote tallies didn't match their ballot counts.
All but one county official admitted they did not make this important crosscheck before reporting results that, most now concede, contained significant errors. State officials also failed to notice the discrepancies. <more>

[garybeck]

10. good point.

how about adding "open" to signify that all the audits have to be in public view, or observed by volunteers by all parties, ...

Mandatory Manual Open Random Audits

MMORA?

Are we getting to many words?

[MostlyAmused]

15. another option

Hi. Long time lurker here.

Wish we could come up with a perfect "L" synonym for one of the M's and shift it around to

MORAL vote counts.

That would stick in people's minds -- and in some people's craws, always a bonus.

[Bill Bored]

13. Good point, but don't forget, candidates also observe these audits. nt

[skids]

11. MRMA

Sounds better re-ordered.