Is requiring that a vote be overturned in cases where the evidence shows over a 95% chance that the election system failed the voters and democracy a radical solution? When should there be a re-vote? Or simply overturn an election?

Two important articles in the spring issue of CHANCE, from the American Statistical Association present some stark evidence and raise some necessary questions: <read both>

Florida 2006: Can Statistics Tell Us Who Won Congressional District-13?
by Arlene Ash and John Lamperti

Statistical Solutions to Election Mysteries
by Joseph Lorenzo Hall

The first looks at what statistics can tell us about the missing 18,000 votes in Florida-13 in the 2006 Congressional election. Some of its conclusions:

Read the rest of this entry »

I’m no fan of the Readers’ Digest*, but even they have finally recognized there is some concern about electronic voting: Quick Study: Voting Machines <read>

Some of the definitions and references seem to be a good start. One item that I would disagree with is:

Computer Scientists
From MIT to Caltech, computer science professors have lined up on both sides of the DRE debate. Stars include Avi Rubin of Johns Hopkins (anti-DRE) and Michael Shamos of Carnegie Mellon (favors e-voting but critical of DRE manufacturers).

This is like comparing a dusty teaspoon to the sand in the universe. The line of Computer Scientists behind Avi Rubin is almost infinite. The line behind Michael Shamos has perhaps two or three others. In fact, its not just a DRE debate.

* I gave up my subscription to Readers’ Digest several years ago. They keep sending it and I keep trashing it. Every year they ask me to renew and I refuse. I’d change my address to send the free subscription to another, but I just would not want to feel responsible if they actually read it.

Bysiewicz Urges Passage of S.B. 444 - We Urge Amendment #6141

Many of the changes necessary to start Connecticut on the road to stronger, effective audits that were in H.B. 5888 have now been offered as an amendment #6141 by Reps Caruso and Urban to S.B.444:

• An Independent Audit Board
• A Stronger Chain-of-Custody for Memory Cards, Optical Scanners and Ballots
• 100% Independent Pre-Election Testing of Memory Cards

Read the rest of this entry »

Update:  Most of the provisions are now included in an amendment by Reps Caruso and Urban,  House LCO Amendment #6141 to S.B. 444.  

Update: If you have not done so already, please contact your Connecticut House Representative.

Thanks for your calls and e-mails. H.B. 5888 passed the Planning and Development Committee with unanamous support of the members present, with some good words in support by Chairman Feltman. The only reservations were expressed around the 100% ballot printing provisions and the mandatory privacy spaces — both of which we believe will be improved in an amendment to be offered shortly. Now on to the full House.

Read the rest of this entry »

A new addition to our blogroll, www.NoVBM.org (No Vote By Mail), provides more, much more, on the risks of vote by mail that we cover from time to time.

Are you wondering why voting by mail or absentee is a bad idea? Are you looking for information about the problems encountered by all absentee voting systems? Well then, welcome to The No Vote By Mail Project’s Website. There’s hundreds of articles now in the archives, so I thought I’d put together a giant list of the best articles I’ve written so far about voting by mail and the problems I’ve found. It might take a few days to put together, but here’s what I have so far:

Whenever you think voting by mail or the Internet might be a good idea, visit www.NoVBM.com.

Consider the risks of Vote By EMail or Vote By Fax!

We are not convinced that vote by mail can be made safe and private. The purpose of the secret ballot is not just to protect the privacy of the voter, but to protect us all from the votes of others being bought or coerced. For more read our earlier post <read>

Today, the Hartford Courant carried an AP story, largely supportive of Email Voting and FAX voting by the military, Few Troops Vote By E-Mail, Most States Prefer Low-Tech Balloting <read>

I appreciate the caveat in the middle of the article:

Adding an electronic boost to the voting process would ease problems for service members, but it raises security and privacy concerns.

But, that is not the thrust of the article:

That still leaves tens of thousands of service members in far-flung military bases struggling to meet voting deadlines and relying largely on regular mail to get ballots and cast votes — often at the last minute because of delays in ballot preparations in some states. Connecticut is among the 37 states that don’t allow balloting by e-mail.

Pentagon officials have been urging more states to move into the electronic age before November, a move that could help reverse recent trends in which thousands of military members asked for ballots but either didn’t vote or had their ballots rejected for flaws.

Though e-mail continues to be a hotly debated topic among state voting officials, faxing is now broadly accepted as a way of getting ballots to overseas voters. And in as many as two dozen states, voters who sign a form waiving their right to privacy can send the ballots back by fax.

Once again its not just their right to privacy. Its our right not to have their vote bought or intimidated. As I recall our government now has the ability to read all of our e-mails and monitor calls, like faxes - employers such as the Pentagon also have that as a right - and as we all know the Internet is not that secure.

This is not such a small matter…as the article reminds us elections have been won and lost by a very few votes, much less than the votes involved here:

The push comes more than seven years after problems with overseas military voting set off an uproar in President Bush’s narrow 2000 victory. In Florida, where Bush squeaked out a 537-vote victory that gave him the presidency, questions were raised about several thousand overseas military votes that came in after deadlines and were counted in some districts but not counted in others…Of the roughly 1.3 million active duty military eligible to vote, about 500,000 are deployed overseas or permanently assigned there.

Of course this is an AP article.  While the Courant seems blind to larger voting issues, the editors do understand voter privacy. We recall an op-ed by a Courant Intern covering the similar problems with internet voting.

George Cody, President Registrars Of Voters Association Connecticut, responded to the Dead Voters Issue with a letter to the Hartford Courant <read>

Connecticut and the registrars are stuck between a rock and a hard place — aggressively clean up the database and risk dropping voters incorrectly — or demand clear evidence to drop a person from the rolls to avoid disenfranchisement.

Names are removed from the active rolls upon positive evidence of a change in status or residency. Proof positive is required to make changes in status. Registrars use several methods to seek out information. When notified of a voters death, they make every effort to verify the information through obituaries, death certificate records in the town clerk’s office or through local probate records. There is often a time gap in this information and, as the article points out, registrars do not have regular access to state and federal records. The time required to match and verify the information might lead to a delay in a voter’s final removal from the rolls, but the status of these voters is changed to inactive during the verification process.

When voters move out of state or town, their former town of residence is not necessarily notified, resulting in only anecdotal information…

The subsequent follow-up done by the students appears to have found clerical problems, rather than any evidence of potential fraud. Connecticut citizens can also rest assured that all voters are asked to present identification at all elections or to sign a statement about eligibility prior to voting. The Courant’s reporter has been kind enough to supply a list of the questionable names. This list will be disseminated to the towns for follow-up. We only wish we had this list earlier to make corrections and provide input to the study other than post-study comments.

Clerical problems do occur. When voters move out of town without notifying their former town, the chain of information is broken. They will often remain listed as active voters until we receive corrective information. If that move is out of state, notification is at best infrequent. Mistakes based on duplicate or similar names or incorrect post-election entry of who voted is another source of confusion.

We look forward to working with the list provided to further clean our lists. The centralized voter database has contributed to addressing in-state moves, but the real solution is with the voters and their families. If there is a change in any voter’s address or status, notify your registrar so corrective action can be taken.

We would like better data, but we are sympathetic to the registrars’ challenges on this issue. We are also interested in the Media’s and the Secretary of the State’s aggressive response to this issue in contrast to the much more significant issue of post-election audit integrity.

With the dead voter issue we are talking about 300 possible errors between 1994 and 2007. In just the last post-election audit we are talking about chain of custody lapses in several towns and 304 ballots missing in just one municipality in just one election. Perhaps it has to do with the media response to the two issues. While there has been a flurry of articles in the Connecticut press on the dead voters report, to our knowledge the mainstream media has yet to publish a single word on the results of the post-election audits of the November Election and the February Primary. Also missing is media concern with significant quality control issues in programming and testing elections also uncovered by researchers at the the University of Connecticut.

Update: More from Politico <read>. Billions for deficits and earmarks but nothing to provide voters confidence their choices were followed.

Larry Norden, director of the voting technology project at the Brennan Center for Justice at New York University’s law school, called the vote “a sad statement on how little Congress has done on the issue of making sure elections are as secure and reliable as possible.”

Update: Added prospective on cost from Huffington Post <read>

“It all depends what you compare it to,” said [Princeton Professor Ed] Felten. “If you compare the amount of money we spend actually administering the election versus the amount that’s spent trying to convince people to vote this way or that way, it looks pretty small. It you compare it against the importance of getting the outcome right, it also looks relatively small. So, to me, it’s an investment that we should be willing to make.”

Article from Rush Holt’s state, New Jersey: Vote-count bill’s rebuff doesn’t add up <read>

You wouldn’t think anyone would be opposed to voting machines that can, if necessary, verify their precise ballot totals at the end of Election Day…”
It came out of committee two weeks ago unanimously supported by both parties,” Holt says over the phone. “We had compromised so there would be strong support when it came to the floor.”
So confident was House leader ship that it waived the usual voting procedure of requiring a simple majority and sought a two-thirds “super majority.”
That, explains Holt, would have demonstrated to the Senate and the White House that HR5036, otherwise known as the Emergency Assistance for Secure Elections Act, had veto-proof support.

But on the day of the vote, something beat it to the House floor: the Bush administration’s objections.
Bush did not specifically threaten a veto, but he made it clear he didn’t like the bill, which was enough for most of its Republican supporters to bail out.

The bill passed, 239-178. But that was way short of the super majority, which tells the White House that the House doesn’t have the votes to override a veto.

more on the Holt bill <read>

We favor paper filled out by voters, followed by optical scan, followed by a sufficient audit, just like 100%** of computer scientists.

* unless you want your vote actually counted and counted correctly
** very vary small margin of error, such as Michael Shamos.

Here is an example from Shamos’ state of the start of just an average day of paperless voting:

Bethlehem, PA:

In Bethlehem…When judge of elections Craig Hynes swung open the doors, 15 people rushed in, including some that had been their since 6:40 a.m….Hynes said. “This is going to be a long day.”

Moments later he realized how long it could be when one of his three voting machines malfunctioned. After four people had voted, he realized that it had only registered one voter. He had to reset machine.

“We lost three voters and there’s no getting them back,” explained Hynes, “and at this point we don’t even know who they were.”

At least one voter still has “hope” that their vote counted for some candidate, while we can hope that three will not “change”. Not the only problem <read more>

As they have said about Pennsylvania at Verified Voting:

Pennsylvania’s Presidential primary on April 22 will be essentially unrecountable, unverifiable, and unauditable - an irony, because state law requires manual audits of a statistical sample of ballots cast in elections.

Over 85% of Pennsylvania’s voters live in counties in which paperless electronic voting is the only method of voting at the polling place.

Update: Secretary of the State’s <press release>

A UConn study reported in the Hartford Courant demonstrates that the ultimate disability is not a barrier to voting in Connecticut: Dead Voters? Probe Finds Errors In Records, <read>. The deceased continue on the voter rolls and occasionally vote.

a list of more than 300 people across Connecticut who appear to have voted from the grave in elections dating to 1994, a two-month investigation of voting records by journalism students at the University of Connecticut has found.

The mysterious voters were identified by matching a statewide database of 2 million registered voters and their voting histories with two separate computer lists of dead people maintained by the state Department of Public Health and the federal Social Security Administration.

Following up on the matches, UConn students examined the records of nearly 100 of the suspect voters at 10 town and city halls among those with the most cases. Guilford led the state with 39, followed by West Hartford (17), Enfield (15), Stonington (13) and Norwalk (11).

Some people appeared to have voted frequently after death, the research found. In Hebron, for example, records show one man voted 17 times after he died in 1992.

The investigation also identified more than 8,500 people listed as dead who are still registered to vote in Connecticut, most long after their deaths. In Hamden, one woman remained a registered voter although she died in 1979…

All but nine of Connecticut’s 169 municipalities listed dead people on active voter rolls. At least 100 cases were identified in each of 28 cities and towns. In New Haven alone, 370 dead people were still registered; in Enfield there were 321; in West Haven, 310; in Hartford, 298; and in Bridgeport, 293.

While there are serious problems with Connecticut’s online voter registration system, concerns with deceased voters center around the antiquated clerical systems and local election administration:

Read the rest of this entry »