The image “” cannot be displayed, because it contains errors.


The nation's clearinghouse for election audit information!


May 6 Snapshot: North Carolina PDF  | Print |  Email
By Verified Voting Foundation   
April 28, 2008
North Carolina moved to voter-verified paper records after an election meltdown in Carteret County in 2004. A Unilect Patriot electronic machine lost over 4000 votes, and sparked the eventual passage of the Public Confidence in Elections Law, which mandates at least a voter-verifiable paper record for all voting systems, and requires an audit of at least one statewide race. In Presidential election years, the audited race is automatically the Presidential race.

There are 5,794,294 registered voters in North Carolina, according to the State Board of Elections. Most of North Carolina's 100 counties use voter-marked paper ballots read by optical scanners. The state's counties have only one voting system vendor, Election Systems and Software (ES&S), which supplies optical scanners, the iVotronic direct-recording electronic (DRE) touch screen, and the AutoMARK ballot-marking device. The iVotronic is the only DRE system now used in the state.

It is probably fair to estimate that at least 30% of the May 6 primary ballots will be cast on the iVotronic, and that over two thirds of the votes will be cast on optically scanned paper ballots. Predicting statewide use of the iVotronic is difficult due to the use of the system for early voting in some of the optical scan counties. In recent elections, early voting has comprised about 30% of the overall turnout, but that figure includes both mail-in ballots and in-person early voting. In-person early voting, called One-Stop voting in North Carolina, began on April 17, and will continue until May 3.

The paper trail for the iVotronic, known as the Real Time Audit Log (RTAL), has been the subject of some criticism for its design. Printer jams have also occurred with the RTAL, compromising about 9% of the machines in Guilford County, NC in the 2006 general election. The Ohio EVEREST review Academic Team report (p. 94) noted a risk that even voters careful to check the RTAL might not notice if their vote was cancelled quickly after printing. Combined with the severe security vulnerabilities of the machines, this is a serious concern.

Here is a summary of voting systems in use in the state:
  • 67 counties, with over 58% of the state's 5.8 million registered voters, use optical scan systems with one accessible ballot-marker at election-day polling places.
  • Among the 67 optical-scan/ballot-marker counties, most use only paper ballots for all types of voting, but 4 counties, with over 270,000 registered voters (almost 5% of the total registered electorate), use the iVotronic at the One-Stop early voting sites.
  • 24 counties, with 29% of the voters, use the iVotronic equipped with the RTAL paper-trail printer as the voting system used for all but mail-in ballots.
  • 8 counties, with approximately 13% of the voters, use optical scanners as the principal voting system on election day, but use the iVotronic for accessibility on election day, and also use it at early voting sites (though 2 of these counties, Dare and Richmond, use both optically scanned paper ballots and the iVotronic at early voting locations).
North Carolina's post-election audit requirement has been in place since they passed their voter-verified paper ballot law in 2005. Currently, the selection of precincts to be audited involves a computer randomization process. To facilitate the ability of the public to observe that phase, using a non-computerized method would mean greater transparency. One method that is easy to observe and provides excellent randomness is the use of ten-sided dice, as discussed in a 2006 paper by computer scientists Arel Cordero, David Dill and David Wagner.

To offer better deterrent protection through the audit, it would be better if more than one contest were audited, especially one not known in advance. We hope to see improvements in the audit requirement going forward. But North Carolina is to be commended for having initiated this process over the past couple of years as it is one of the most important steps toward secure and verified elections.

The random selection of precincts for the post-election audit will be conducted at a public meeting on May 7 in Chapel Hill, NC. The meeting will be held at 9:00am in suite 104 of 730 Martin Luther King Blvd.

Comment on This Article
You must login to leave comments...
Other Visitors Comments
You must login to see comments...
< Prev   Next >