Buncombe County voters wait in line to cast primary ballots outside Leicester Elementary School on the evening of Tuesday, March 15, 2016. Colby Rabon / Carolina Public Press

Support nonprofit news that’s accountable to you

Give today and NewsMatch will match your new monthly donation 12x or double your one-time gift, all up to $5,000.

$
$
$

Your contribution is appreciated.

Elections are unpredictable. Election administration does not have to be.

For example, the Bipartisan Policy Center and MIT have jointly conducted research on polling-place lines in the 2016 and 2018 elections.

Their findings, which are consistent with other research, show that voters of color wait in line, on average, longer than white voters and that election jurisdictions using electronic voting machines had longer wait times than those using hand-marked paper ballots.

The factors that cause long lines on election day – outside of unforeseeable problems like storms or power outages – are known weeks if not months in advance.

Charles Stewart III is a professor at MIT and an expert on election administration. He has used the predictable nature of things like line length at a polling place to help counties prepare for elections.

“If you know how long it takes to serve somebody, so, like how long it takes to check somebody in or how long it takes for somebody to vote, you can estimate the average wait time,” Stewart said.

Rather than waiting for the election to see if there were long lines, Carolina Public Press investigated the distribution of voting equipment in Mecklenburg County to detect red flags ahead of primary election day on March 3.

CPP chose Mecklenburg because it will be the largest of the seven counties using new voting machines for all voters in the coming primaries.

The rest of the state will use hand-marked paper ballots that are counted with a scanner.

CPP’s analysis had two questions:

  1. Did each precinct have enough voting equipment to avoid long lines on election day?
  2. Were resources distributed fairly among precincts?

Best practices to avoid long lines?

CPP input data from Mecklenburg County into models from ElectionTools.org and MIT to simulate line length. The results predict that Mecklenburg County should be well within the best practice recommendations by the Presidential Commission on Election Administration that no voter should wait longer than 30 minutes to cast a ballot.

In fact, except for the early morning and evening rushes, voters should not have to wait in line at all.

Stewart’s years of analyzing voter turnout schedules gave him some insight that could help voters avoid lines and keep line length down for others.

“If you can avoid the beginning of the day, don’t vote at the beginning of the day,” Stewart said. “Wait till, you know, two or three hours after the polls have opened.”

North Carolina’s early voting options also help keep lines under control on election days. Since it was established in 2000, early voting in North Carolina has become increasingly popular. Even so, the early days in early voting are underutilized, Stewart said.

If voters have already made up their minds, the first few days of early voting might be the best bet to avoid crowds.

Mecklenburg’s director of elections, Michael Dickerson, said his board overestimates the number of voters who will show up on election day. With early voting being so popular, that’s pretty easy to do.

“If you look through history, you’ll see that my overall turnout for election day is going to continue to drop,” Dickerson said. “And early voting will continue to increase.”

An equitable distribution of machines?

CPP also found that Mecklenburg distributed its election technology equitably among precincts. For this analysis, CPP compared the ratio of voting machines to expected voters showing up on March 3. While the ratio was not exactly the same across precincts, the amount of change among precincts was small. When the assignments were compared to the demographic makeup of voters in each precinct, there was no correlation. That means demographics were not considered in the assignment of voting machines, nor should they be, according to the N.C. Board of Elections.

Instead, county boards of elections primarily “consider total voter registration, voter participation history, one-stop early voting turnout, and the size and layout of the voting enclosure,” according to Pat Gannon, public information officer for the N.C. Board of Elections.

This means, in theory, that voters of color should not see longer wait times in Mecklenburg than white voters in the same county.

Around the country, that is not a guarantee. Most recently, in the 2018 midterm elections in Georgia, precincts with high proportions of black voters in Fulton County – part of the Atlanta metropolitan area – had the longest wait times in the country.

Like Mecklenburg, the entire state of Georgia uses electronic voting machines for all voters.

There have been numerous complaints in the last decade that some precincts have been shortchanged on equipment, said Duncan Buell, a computer science professor at the University of South Carolina and an elections security expert who serves on the Richland County, S.C., Board of Elections.

Buell and Stewart gave the same reason for why some precincts were shortchanged – voting machines cost a lot of money.

That makes counties that have limited budgets more likely to buy as few machines as they think they need. Unlike with hand-marked paper ballot systems, in which more tables can be set up to accommodate more voters when turnout is high, voting machine-dependent systems lack that flexibility.

In addition, counties or precincts serving majority voters of color often have less wealth and political influence, also resulting in receiving less equipment.

“There have been lots of challenges to the effect that not enough hardware has been allocated and the misallocation has skewed based on demographics or other sorts of things,” Buell said. “This is a genuine issue that needs to be addressed.”

In Mecklenburg County, it has been.

If long lines do occur there on March 3, any investigation can begin by knowing that at least it was not due to poor or inequitable distribution of voting technology.

New voting equipment in use for the first time

Mecklenburg County will be one of 10 counties using new voting machines for early voting and one of seven on election day, according to data provided by the N.C. Board of Elections.

Mecklenburg has used voting machines since 2006, but the previous style of equipment was retired by state law at the end of last year due to the deterioration of the machines and serious concerns about their inability to be used in a secure election.

In August, the state Board of Elections certified new equipment by three vendors: Elections Systems & Software; Hart InterCivic; and ClearBallot. ES&S will serve 96 counties, Hart will serve four.

ClearBallot, which seemed likely to sign a contract with Onslow County, withdrew from the state.

Since 2006, ES&S had a monopoly in the state.

ES&S is offering two options to counties. The majority of counties opted to have voters mark their ballots by hand, then feed them into a scanner to be counted. Mecklenburg is among the handful of counties that decided to stick with a voting-machine-for-all system.

On these new machines, called ExpressVotes, voters will still use a touch screen as they have for the last 14 years in Mecklenburg.

The key difference is that the new ExpressVotes system creates a new kind of paper record.

While the old machines printed something like a receipt tape that voters never touched, the ExpressVotes will print a “ballot summary card” that voters will then take across the room and put into a scanner.

Dickerson has been adamant about telling voters that they cannot take their ballot summary card home. It is not a receipt. It must be put in the scanner to have its votes counted.

In addition, voters will be told to verify their ballots before casting them in the scanner, according to a memo sent to county election boards by the state board. Counties will be required to put up signage and will have a dedicated poll worker stand next to the tabulator and ask voters, “Have you carefully reviewed each selection on your ballot?”

This is an attempt to nullify the security concerns that have fired up election integrity advocates and split the field of academics and experts in election security.

ExpressVotes are a type of voting machine called a ballot-marking device. A voter uses the touch screen interface to choose the candidates or issues to vote for, then the machine prints the choices on a ballot card.

But because ballot-marking devices are computers, they can, like any computer, be subject to hacking, error or other manipulation.

Given federal reports that Russia, Iran and other foreign nations are already actively interfering in the 2020 elections, the possibility of hacked voting machines has advocates for secure elections on edge.

The key here, and why voters will be asked to verify their ballots, is whether or not the machines will print different votes on the printed ballot card than the voter selected on the touch screen.

“There seems to be virtually no evidence that voters in large groups actually look at what is printed,” Buell said.

Indeed, the first report of its kind on this security threat, done by researchers at the University of Michigan, showed that voters are both unlikely to check their ballots and to detect errors even if they do check their ballots.

This has been the key point of voting integrity advocate Marilyn Marks, a Mecklenburg voter and the executive director for the Coalition for Good Governance. Marks has used her nonprofit to fund lawsuits in several states, most notably in Georgia, to challenge alleged poor election administration and issued a 114-page protest that both the Mecklenburg and state boards of elections declined to take up.

Mecklenburg will be one of four pilot counties to run a new kind of election audit in the state. This audit, called a risk-limiting audit, is an improvement over the state’s current procedures. But advocates like Marks and academics like Buell say that, absent enough voters verifying their ballots, the audits still will not be able to verify the outcome of the election.

The landscape of election security and potential threats is confusing and multifaceted. But election experts in hearings before Congress, in federal reports and in publications from nonprofits and think tanks make one thing clear: The only guaranteed way to control a vote is if voters stay away from the polls.

The best thing a voter concerned about election security can do is go vote.


You can strengthen independent, in-depth and investigative news for all of North Carolina

Carolina Public Press is transforming from a regionally focused nonprofit news organization to the go-to independent, in-depth and investigative news arm for North Carolina. You are critical to this transformation — and the future of investigative and public interest reporting for all North Carolinians.

Unlike many others, we aren’t owned by umbrella organizations or corporations. We are an independent and nonpartisan 501(c)(3) not-for-profit organization, founded and operated in North Carolina. And we haven’t put up a paywall — we believe that fact-based, context-rich watchdog journalism is a vital public service. But we need your help. Carolina Public Press’ in-depth, investigative and public interest journalism takes a lot of money, persistence and hard work to produce. We are here because we believe in and are dedicated to the future of North Carolina.

So, if you value independent, in-depth and investigative reporting in the public interest for North Carolina, please take a moment to make a tax-deductible contribution. It only takes a minute and makes a huge difference. Thank you!

Jordan Wilkie

Jordan Wilkie is a Report for America corps member and is the lead contributing reporter covering election integrity, open government, and civil liberties for Carolina Public Press. Email jwilkie@carolinapublicpress.org to contact him.

Leave a comment

Your email address will not be published. Required fields are marked *