AP NEWS

AMY CHOZIK: Can peer pressure defeat Trump?

February 24, 2019

EDITOR’S NOTE: Amy Chozick is a writer at large for The Times covering business, politics and media.

Twelve years ago, social scientists cracked the code on how to get Americans to vote. Before a special election in Michigan, 100,000 households received slightly different mailers: One reminded them that voting was their civic duty. Another applied gentle social pressure by including the voting history of everyone in that particular household (“Who votes is public information!” it reminded them). The final flyer — and by far the most powerful — revealed the voting history of the recipients’ neighbors.

“What if your neighbors knew whether you voted?” it asked, along with a warning that after the election, researchers would “publicize who does and does not vote.”

“It proved to be the most effective intervention ever uncovered by an order of magnitude,” said Todd Rogers, a professor of public policy at Harvard who specializes in education and voting behavior.

The findings had the power to transform political organizing, especially for Democratic candidates who rely on high voter turnout. There was just one problem.

“It made people crazy and super irritated and offended,” Rogers said. “The underlying psychology is that when people feel like they’re going to be held accountable, they’re more likely to do it, but they also get really mad about it.”

So political groups backed away from the idea. For the next decade, even as we abandoned our privacy with a swipe of the opt-in button and as apps that rely on social pressure proved effective in dieting, parenting and saving for retirement, political organizers largely ignored the power of peer pressure. Then came 2016 and the election of Donald Trump.

“Not only did progressives lose, they were surprised they lost,” said Sangeeth Peruri, founder and chief executive of VoterCircle, a platform that allows users to tap into their address books and easily identify and encourage (that is, nag) eligible voters. Users have the option to see whether their contacts voted in past elections and whether they are registered Republicans or Democrats.

“Losing is one thing, but being surprised you lost is a failing of the system,” Peruri said. His platform, one of the first, started in 2015. But others soon followed.

The shock of Trump’s election galvanized several developers who had worked on Barack Obama’s presidential campaigns. Steeped in Silicon Valley debates about privacy, they wondered if they’d been too timid about harnessing the enormous trove of publicly available voter data. After all, party affiliation and voting history had long been used internally by campaigns. Maybe political groups needed to stop caring about people’s feelings if it helped get them out to vote.

Mikey Dickerson, executive director of the New Data Project and a former Google engineer who was chief of the U.S. Digital Service in the Obama administration, had learned about the Michigan survey in 2008, but he didn’t come up with his VoteWithMe app until 2017. The app outs anyone who didn’t vote in previous elections. (Alyssa Milano and Alexandria Ocasio-Cortez, looks like you both sat out the 2014 midterms.)

It identifies friends in competitive districts (with a fire emoji) and provides handy text-message reminder templates (“You gonna vote?”). “We looked around and didn’t think it had ever been done before — putting voter-file data directly in front of you as the end user,” Dickerson said.

Many in the technology industry initially scoffed at his idea. (“VoteWithMe is a creepy new app that checks your contacts’ voting history,” one article declared.) But several hundred thousand users downloaded VoteWithMe before the midterms, and Dickerson, who plans to step back and hand the technology over to like-minded groups, said lots of similar apps are springing up ahead of 2020.

“I am worried about the foundations of the systems of government surviving and an administration that puts kids in cages and separates families,” he said. “I’m not going to feel bad that some of my friends, unaffected by any of this, have a mild amount of discomfort because some jerk — probably me — can see whether or not they voted.”

The weight of peer pressure has a particular pull on millennials. They represent more than 30 percent of eligible voters, about on a par with baby boomers, but have the lowest voter turnout of any age group. Only 49 percent of voters ages 18 to 35 voted in the 2016 presidential election, according to the Pew Research Center. Democratic pollsters predict that increasing turnout among millennials by 10 percentage points would all but guarantee they defeat Trump in 2020.

In an era when privacy feels like a nostalgic notion and our political leanings can be more or less gleaned from where we live, how we dress and what we watch, is there even such a thing as the sanctity of the voting booth? After several celebrities learned the hard way that most states don’t permit photos inside the voting booth, last year California legalized “ballot selfies” (thank you, Kim Kardashian). The day of the midterm elections, 1,000 people a minute were posting Instagram stories with “I Voted” stickers, according to the company.

Developers said these turnout apps aren’t intended to shame anyone. As Debra Cleaver, chief executive of the San Francisco-based Vote.org, a nonprofit group that works to increase voter participation, put it: “We call it social pressure or social validation. ‘Vote shaming’ sounds like it was coined by a reporter because it makes you want to click.” (Fair.)

They say their primary purpose is “relational organizing,” or tapping into your social network to tell contacts about a candidate or election (as opposed to the old-fashioned and less effective “operational organizing” that involves dispatching volunteers to cold-call strangers). In other words, coastal liberals can make a difference while fiddling with an app on their sofas (“I know you’re going to vote on Nov. 6, duh, but make sure to remind your friends!” one text template reads), rather than flying to Iowa to knock on strangers’ doors. “I haven’t knocked on a door since 2014,” Cleaver said.

Buffy Wicks, a community organizer and former Obama campaign aide who was recently elected to the California state Assembly, said it doesn’t have to be “an either/or.” She hosted 239 house parties at which she encouraged supporters to download VoterCircle, but she also knocked on 115,559 doors. More than 100,000 people voted for her.

The rush to design apps to increase voter turnout is part of a wider push in Silicon Valley — trying to shake the taint of peddling fake news and Russian propaganda — toward “civic tech,” or innovations designed in the public interest. “There’s obviously a PR aspect,” Cleaver said. “But there is no way relational organizing apps can undo the damage that Facebook and Twitter have done.”

Nevertheless, the apps are gaining traction. Revolution Messaging and Phone2Action and other liberal websites allow users to pressure their representatives, raise money and support candidates. Republicans have developed apps to, for instance, bolster the National Rifle Association or donate to Trump’s re-election.

All of these, plus the individual apps that the armada of Democratic candidates running for president will soon offer, are enough to give anyone app fatigue. That’s why, in the coming months, Peruri of VoterCircle plans to rebrand his platform as OutreachCircle, making it a liberal one-stop shop to plan house parties, call your representative, advocate for a cause, give money to a candidate … and, if you’re so inclined, snoop on your friends’ voting history.

The goal is to eventually build habits so that rather than just ranting on Twitter, people can use OutreachCircle to keep their communities engaged, essentially performing the function of an old-fashioned neighborhood precinct captain or PTA president. “Our influencers aren’t Katy Perry. It’s your church leader or your high school coach,” Peruri said.

The apps are limited by often incomplete voter data (not even Silicon Valley can make the Board of Elections efficient). But Naseem Makiya, chief executive of Outvote, an app that worked with MoveOn.org and Beto O’Rourke’s Senate campaign, said preliminary results show that a single text from a friend makes people roughly 10.2 percent more likely to vote. Other developers said peer-to-peer messaging led to a 2 to 3 percentage point increase in turnout in November — enough to swing races in tight districts. Robin Wolaner, 64, a retired executive in San Francisco, used VoteWithMe to identify friends and family in competitive districts in Pennsylvania. “Frankly, most of my friends don’t need a reminder to go vote for Nancy Pelosi,” Wolaner said. As for vote shaming (sorry, I mean social validation), she tried to gingerly remind contacts that their voting history was public.

“In some cases I sent messages saying: ‘Can this be right? Did you really skip the last midterm election?’” Wolaner said. “My kids would tell you I am a natural born nag, so it sort of fit my personality.”

Natasha Baker, 30, a lawyer in Washington, said she used VoteWithMe to encourage distant friends and family in Indiana and Virginia to vote. “It’s kind of creepy, I have to admit,” she said. But, she added, “If you have a friend who you can see voted in every election, you don’t have to spend your afternoon on that.”

Then there is the voyeuristic urge to snoop. Dickerson was surprised to see that the VoteWithMe app was trending in the Apple store days after the midterm elections. “It doesn’t make a ton of sense — we’ve had no ads or promotions since voting,” Dickerson said.

But to behavioral scientists, the downloads made perfect sense. “We are intensely social creatures and need to situate ourselves inside the collective,” said Robert Cialdini, a professor emeritus of psychology and marketing at Arizona State University and the author of several books on peer influence. He cited a study he did at a Holiday Inn in Tempe, Arizona. His team compared the usual cards requesting that guests reuse their towels to protect the environment with several alternatives: One added that most guests of the hotel had reused their towels; another said that most guests “who stayed in this room” had reused their towels.

The last spurred a compliance rate of 49 percent, the largest spike in towel recycling the hotel had ever seen. “We think we’re free-standing individuals,” Cialdini said. But we “believe that the choice of our peers will work well for us, too.”

Maybe that was why I felt compelled one Sunday afternoon to sit in a coffee shop with a friend and scroll through VoteWithMe with the giddy enthusiasm of a sample sale. I learned that one friend who lives in the West Village, owns a Beto T-shirt and has been known to insufferably quote the Pod Save America bros didn’t vote in the past midterm elections. (You know who you are.) My uncle in Texas who watches Fox News, however, was designated a “strong voter.”

There may, however, be a fundamental flaw in the theory driving the apps. As developers work to expand them from hundreds of thousands of downloads during the midterms to mainstream use by millions before the 2020 election, they could be leaving out the very voters Democrats most need to reach.

The tendency of people to mimic their social networks — what behavioral scientists call homophily — could backfire in this case. Politically engaged people who download voting apps, and the friends they nag, will probably show up at the polls in greater numbers. But the opposite could be true for those in poor and disenfranchised communities where voting isn’t the norm.

Rogers, the Harvard public policy professor, pointed to a 2015 study that observed social pressure on students by making an SAT prep session sign-up sheet public. Students in the AP class signed up in greater numbers when they knew their friends would see the list. Sign-ups in the remedial class, however, where studying wasn’t as socially accepted, decreased. “The increased transparency could have unintended consequences,” Rogers said, “if you look at your network and see no one votes.”

Cleaver of Vote.org said she worried that this was a blind spot typical of Silicon Valley, an industry dominated by male engineers who are generally from privileged backgrounds. “Everyone wants the solution to increasing voter turnout to be an app, but turnout is low in this country because of decades of racism, sexism and voter suppression,” she said.

Cleaver, a self-described “tech person” who is backed by the startup accelerator Y Combinator, urged me to write about anything other than these apps. She proposed an article about the perils of voter registration still being tied to the Department of Motor Vehicles when a growing number of young people don’t have driver’s licenses. Or what about the problem with mail-in ballots that require a signature when young people don’t know cursive? Don’t even get her started on new voters being required to print forms. (“I mean, who owns a printer?”)

She said that the tech industry, once hailed as a savior, is undergoing an identity crisis for good reason. “The logical conclusion of technology is that it all goes terribly awry,” she said. For the 2020 election, her group, Vote.org, plans to spend millions of dollars on billboards.