In the 2016 presidential election, 48.3 percent of college students voted, according to a Tufts University study. That figure, though low, was still better than the 45.1 percent who voted in the 2012 election. Here is some of the breakdown:

  • Students majoring in the social sciences voted at the highest rate, followed by those studying the health professions, the humanities and business.
  • Students majoring in the STEM fields — science, technology, engineering and math — voted at the lowest rate in both election years. 
  • On average at the state level, voting rates at schools in New York, Massachusetts and Pennsylvania tended to increase the most from 2012 to 2016, while schools in Georgia, Wisconsin and Mississippi had the largest decline.
  • Students at public four-year institutions vote more frequently than students at community colleges and private four-year colleges.

Some schools, though, have a student voting participation rate much higher than the national average for 18- to 24-year-olds, and it’s not an accident.

In this post, Michael Peshkin, an engineering professor at Northwestern University, explains how his school gets students to register and vote. Peshkin is one of the inventors of cobot, a robot designed to physically interact with humans in a shared workspace; his area of study is the interaction of people with automation.

By Michael Peshkin

My freshman engineering-class students are seated and quiet when I arrive. It’s their first day of college. They’re excited and anxious. Subject-labeled spiral notebooks are at the ready. Some students instead have the tablet equivalent, with a special stylus, soon to be misplaced. My class is called “Makerskills” and it’s an elective. I’ve promised students who enroll that they will acquire wizard-level capabilities that they can use in project-based courses and even in dorm-room start-ups.

They are wondering whether they are will discover themselves as “makers” — or even whether they are destined to be engineers at all. Outside of academics, they are navigating how they want to present themselves to a new social milieu. Identity formation is an important part of college.

Identity as a politically engaged citizen, including participation in elections, can also be acquired at college, and Northwestern University is an exemplar of this. Our students' voting turnout is 64 percent, far above the 48.3 percent average for college students in the 2016 presidential election and even above the turnout rate for U.S. citizens of all ages.

It’s not an accident. Over years of trial and error, we’ve found methods that work in encouraging college-age voting. Many of the lessons apply to voting at all ages. Technology is important, but the starring role is old-school: person-to-person contact.

Each year, when new students arrive on campus in those first days of sincerity and wonder, we connect with all of them. A peer talks to them, individually, about their voting choices: Vote here or vote back home? Vote in person or vote by mail? The subtext is this: You’re 18 years old, you’re at college, naturally you’re a voter. Research shows that voting is a habit. And it’s a habit that is best acquired young.

When greeted individually, 95 percent of our eligible students register to vote on the spot. If they want a vote-by-mail application for Indiana or New Jersey, we’ve got it, and we get it filled out right away. On the dorm packing list, postage stamps may not have made the cut; students get along very well without stamps; everything they want to do is online. Even printers are becoming scarce. We can’t send potential voters home to an office supply desert! If we do, their print-at-home form will never see the inside of a real-life envelope. So we finish it all up for them — stamp and mailing service included.

Small inhibitors such as stamplessness cut into follow-through. But immediacy — getting it all done right now — also counteracts one of the biggest pitfalls of using technology to encourage voting: the tendency to put aside and eventually forget about anything that lives online.

Email is a particular hazard. Researchers have documented a statistically significant decrease in voter registration among students invited to register via email, compared to a control group not contacted. That’s because once a useful message is received and safely stored in the inbox, it loses any specific sense of urgency. Over time, the message drifts down the queue and is forgotten. Meanwhile, real-life opportunities to register are passed by.

It would be strange if our engineered approach to voting engagement did not have a technological component, and it does. We make heavy use of fast, efficient for all the needed forms and instructions. Turbovote is another service that works similarly, although it’s more prone to being used in join-and-forget mode. And we send all our future voters on with a referral to Ballotready, so they can plan their ballot and bring their voting decisions with them when they vote — as a scrollable list on their phone, of course. Good technology!

Broadly, though, the use of technology to drive turnout is a history of disenchantment. Ten years ago, campaigns made heavy use of text messaging to get out the vote. Text messaging had novelty at that time, but no more. A phone call from another voter, even a sincere volunteer, has long since lost its charm. Banner ads and social media buys are past their prime, and today’s young people use ad blockers, so they don’t even see them. Studies show that door-to-door canvassing is far more effective than technologically mediated outreach, and has stood the test of time.

We’ve known that our method — social engineering with a dose of technology — accomplishes close to universal registration among eligible students. However, registration is not necessarily voting. Only recently was a study at Tufts University able to measure the actual voting rate at hundreds of schools. Tracking student voting records is a technical challenge, given that so many students vote back home by mail in another state.

The results of that study illustrate that while motivation is unquestionably a big factor in turnout, so is mechanics.

Northwestern’s very high voting rate resulted mostly from making a two-step process into a one-step process. Our students just have to decide to vote. They don’t have to decide to register and then decide to vote.

Ours is not the only way. Edgewood College in Madison, Wis., beat our voting record. Its secret? Several weeks with an early-voting site by the cafeteria.

I know how my first-year students will be as Election Day in November approaches. They will be a little late to class. I’ll have to shush them to get started. Their note-taking, spiral-bound or tablet, will have tapered off. They will have become seasoned college students; “tender weeks” gone by. They will all be registered to vote.

We’ll be building the Simon game, in which most of the difficulty of programming is coping with the wild card: human behavior.