About a decade ago, I gave a guest lecture at the Illinois Institute of Technology about election security and the risks of hacking. (I don’t want to call myself prescient, but …) To get into one of the themes of the day, I started by asking students the average age of a poll worker. “Old,” said a student after a long pause. “How old?” I asked, and he replied, “Really old!” So I tried again. “How old is really old?” I was somewhat surprised to hear him say 35.
The real answer, according to national surveys of election offices by the U.S. Election Assistance Commission, is that the average poll worker is older than 60. Specifically, the commission says: “Of the age data reported for approximately 53 percent of poll workers who served in 2016, 24 percent of poll workers were 71 or older and another 32 percent were between the ages of 61 and 70.” Other reports suggest the average age is more than 70. And as a poll worker myself, I know that as I approach 60, I’m generally one of the younger poll workers in my precinct.
So why does age matter? Aside from some of the physical issues (the ability to tape signs to walls, stack chairs, lift tables, etc.), the older poll workers are less likely to be conversant with technology. Election offices strive mightily to make instructions that can’t be misinterpreted, and in many cases succeed. The instructions I get from Fairfax County, Virginia, are very well written—so long as you follow them extremely carefully. Even as a technology expert, I find myself befuddled—sometimes skipping a step and thereby increasing risk. In my most recent training, I forgot to check the bin used for ballots when the scanner is out of order. The bin should have been empty, indicating that no ballots had been cast before the polls opened—but there was one ballot in there. Because it was a training class, there was no harm done, but this demonstrates the importance of attention to detail. (I think the ballot was an accident, not a test to make sure, since none of the other people being trained found one.)
Poll workers have a long list of things to keep track of, although exactly what that means varies depending on the state and the technology in use. They need to be prepared to prevent equipment tampering, manipulation of paper ballots (both before and after being marked), theft, ballot box stuffing, repeat voting, and falsification of results (among many other threats). At the beginning of the day, poll workers need to check that the equipment hasn’t been tampered with, that sealed boxes of blank ballots haven’t been opened, and, yes, that ballot boxes are empty, for example. At the end of the day, poll workers need to make sure everything is closed up properly so tampering with equipment and marked ballots isn’t feasible.
Some tasks are relatively simple, such as ensuring that every voter is checked in and gets one and only one ballot, and that no one leaves the polling place without casting their ballot. Some are more complicated, such as checking that electronic poll books are properly synchronizing with each other or properly connecting to central databases (to ensure that voters can’t return and cast a second ballot) or verifying that scanners are operating properly and dealing with paper jams. But some of the things that a poll worker needs to do are much more subtle and technically difficult.
When it comes to security, what doesn’t happen is frequently as important as what does. How do you teach poll workers how to look for tampering with a voting machine when they don’t understand what a USB port is? How do you teach what a fake ID looks like (and therefore might be cause not to allow a voter to check in) without some sense of the capabilities of a modern printer? How do you teach the reasons for ensuring that security seals on scanners, boxes of paper ballots, and electronic poll books are in place (and that incorrect, missing, or broken seals can’t be ignored!) to people who don’t understand the notion of preventing manipulation, or the mindset of an attacker? How do you teach what a tampered seal even looks like when there are so many ways to manipulate seals almost undetected? Many of the seals commonly used in elections can be easily removed and reattached in seconds, so having many observers reduces the risks—but only if they know what to look for.
And most critically, how do you teach all of these things when the typical poll worker gets a few hours of training per year? (Precinct chiefs usually get more training, but it’s still fairly minimal. And increasing the amount of training makes filling these volunteer or minimally paid positions even harder.) Some localities have tried offering online training to provide additional information, but I doubt there are many people who take advantage of these opportunities.
Older poll workers aren’t necessarily any less capable of being responsible for election security than younger workers—and they’re frequently more dedicated and less likely to be distracted by their own phones. But we need to ensure that we’re providing enough training, and training that makes sense to them.
One of the challenges everywhere is the shortage of poll workers, so it’s very hard to say no to anyone who is willing. In Fairfax County, Virginia, where I serve as a precinct chief (the “chief officer of elections” in Virginia parlance), I am paid $200 for attending four hours of training (a 2½-hour annual class plus a 1½-hour refresher class right before the election) and typically working 16 hours on Election Day. In most cases, the only people who can afford to put in this much time for minimal (or no) pay are homemakers without children or retired people.
So what can we as individuals do? The answer is simple: help out. As a cybersecurity researcher and frequent critic of the state of election security, I believe it’s my obligation to help ensure accurate elections, which is why I started as a poll worker about a decade ago. If every American citizen who works in the cybersecurity field volunteered one day per year as a poll worker, we could largely solve the shortage and make a dent in risks to our elections—and we would see the realities of where the risks are (and are not) on Election Day. We’d also all be better informed, so when we talk to our elected representatives and media about security, we’d be speaking from a position of knowledge. Many of the threats to elections aren’t in the polling place, but it’s a great place to start.
Not a cybersecurity expert? If you’re someone who is reading an online magazine, then odds are you’re more comfortable with technology than many of the people working your polling station this election season. Most states allow anyone to observe key steps of the election process, including pre-election machine testing (called “logic and accuracy testing”), opening and closing the polls (sometimes limited to party representatives), decision-making for counting of provisional ballots, and numerous other aspects of the voting process. No matter your level of expertise, or the time you have available, there’s something you can do to help.
And the most important thing you can do? Vote, of course.
The opinions expressed do not represent those of the U.S. government.
