Technology

Texts You Can Believe In

Forget robo-calls—Obama’s text messages are this campaign’s secret weapon.

Over the last couple of months, John McCain has launched at least a dozen automated phone campaigns that question Barack Obama’s alleged ties to terrorists, among other charges. McCain, who was famously targeted by ugly robo-calls in the 2000 presidential primary, defends his effort as “totally accurate.” Several Republicans have criticized the calls. Even Sarah Palin says she doesn’t much like them. The Obama campaign has scolded McCain to stop the phone campaign; Obama has even launched his own robo-calls to denounce McCain’s robo-calls.

With all this Sturm und Drang, you might think that automated phone calls will make a difference in the presidential race. They won’t. Robo-calls are the pyrotechnics of politics: They create a big disturbance, but they don’t have a prolonged effect. Numerous studies of robo-call campaigns show that they’re ineffective both as tools of mobilization and persuasion—they don’t convince voters to go to the polls (or to stay away), and they don’t change people’s minds about which way to vote. So why do campaigns run robo-calls? Because they’re cheap and easy. Telemarketing firms charge politicians between 2 and 5 cents per completed robo-call; that’s as low as $20,000 to reach 1 million voters right in their homes.

Compared with TV advertising, door-to-door canvassing, and mega-rallies, automated phone calls are seductive because they harness modern telecommunications technology in the service of political persuasion. That being said, it’s Obama’s campaign, not McCain’s, that has hit upon the cheapest effective way of contacting voters via the phone: text messaging. During the last two years, Obama has collected hundreds of thousands, if not millions, of cell phone numbers from loyal supporters and new registrants. Now his campaign is sending out text messages to people across the country—the texts remind people to register to vote, to go to the polls, and to organize others on behalf of the campaign.

On the surface, these texts don’t seem that different from robo-calls—they’re both automated messages and both easy to ignore. But for reasons that aren’t completely understood, text messaging is different: We pay attention to short messages that pop up on our phones.

These conclusions arise out of work by Donald Green and Alan Gerber, two political scientists at Yale whose book, Get Out the Vote: How To Increase Voter Turnout,is considered the bible of voter mobilization efforts. Green and Gerber are the product of a wave of empiricism that has washed over political science during the past decade. Rather than merely theorizing about how campaigns might get people to vote, Green, Gerber, and their colleagues favor randomized field experiments to test how different techniques work during real elections. Their method has much in common with double-blind pharmaceutical studies: With the cooperation of political campaigns (often at the state and local level), researchers randomly divide voters into two categories, a treatment group and a control group. They subject the treatment group to a given tactic—robo-calls, e-mail, direct mail, door-to-door canvassing, etc. Then they use statistical analysis to determine whether voters in the treatment group behaved differently from voters in the control group.

Political scientists have run dozens of such studies during the past few years, and the work has led to what you might call the central tenet of voter mobilization: Personal appeals work better than impersonal ones. Having campaign volunteers visit voters door-to-door is the “gold standard” of voter mobilization efforts, Green and Gerber write. On average, the tactic produces one vote for every 14 people contacted. The next-most-effective way to reach voters is to have live, human volunteers call them on the phone to chat: This tactic produces one new vote for every 38 people contacted. Other efforts are nearly worthless. Paying human telemarketers to call voters produces one vote for every 180 people contacted. Sending people nonpartisan get-out-the-vote mailers will yield one vote per 200 contacts. (A partisan mailer is even less effective.)

Meanwhile, pinning leaflets to doors, sending people e-mail, and running robo-calls produced no discernible effect on the electorate. Green and Gerber cite many robo-call studies, but the most definitive is a test they ran during the 2006 Republican primary in Texas. Gov. Rick Perry recorded a call praising a state Supreme Court candidate as a true conservative. The robo-call was “microtargeted” to go out only to Perry supporters—people who’d be most open to his message. But as Green and Gerber show, Perry supporters who received the call reacted no differently from those who’d been kept off the list. They were no more likely to vote, nor, if they voted, to vote for Perry’s candidate.

These findings create an obvious difficulty for campaigns: It’s expensive and time-consuming to run the kind of personal mobilization efforts that science shows work best. Green and Gerber estimate that a door-canvassing operation costs $16 per hour, with six voters contacted each hour; if you convince one of every 14 voters you canvass, you’re paying $29 for each new voter. A volunteer phone bank operation will run you even more—$38 per acquired voter. This is the wondrous thing about text-messaging: Studies show that text-based get-out-the-vote appeals win one voter for every 25 people contacted. That’s nearly as effective as door-canvassing, but it’s much, much cheaper. Text messages cost about 6 cents per contact—only $1.50 per new voter.

Not much is known about the specifics of Obama’s text-messaging operation (the campaign did not respond to my request for comment). We do know that the campaign compiled its list of cell numbers in two main ways. First, the campaign has requested mobile numbers of new voters at registration drives. Then, late in the summer, the Obama camp got a huge haul of mobile numbers through a clever gimmick surrounding Obama’s V.P. pick—if you texted the campaign, Obama promised to text you back as soon as he’d made his choice.

I joined Obama’s text list around that time. (I would have joined McCain’s text message list as well, but he doesn’t have one.) Since then, I’ve received two or three messages a week from the Obama campaign. A typical one: “Help Barack. Tell your friends & family the last day to register to vote in CA is this Monday, Oct 20th! Visit VoteForChange.com to register NOW. Please forward.”

The texts reminded me to watch the convention and the debates and to donate money to the Red Cross when Hurricane Gustav hit. In September, Obama asked me to text him my ZIP code. I did, and now I get location-specific messages—alerts to phone banks and debate-watching parties in my area, reminders of registration deadlines in my state, and appeals for me to volunteer in neighboring states. The messages are rendered in a friendly, professional tone (they refer to the candidate as Barack) and have been free of both fundraising appeals and any kind of negative campaigning.

The beauty of text messaging is that it is both automated and personalized. This is true of e-mail, too, but given the flood of messages you get each day (no small amount from Obama), you’re probably more attuned to ignoring e-mail. Text messages show up on a device that you carry with you all day long—and because you probably get only a handful of them each day, you’re likely to read each one.

This is especially true when the message seems to have been tailored to you specifically—Obama’s often are. The campaign knows a lot about me: At the least, it knows that I live in California, and because I joined the text-message list in order to learn the V.P. pick, that I’m fairly interested in politics (and therefore likely to vote). It’s possible that they might know even more; given my ZIP code and my phone number, they could potentially have tied my text-message account to my voter registration file, allowing the campaign to send me messages based on my party registration, whether I usually vote by mail, and whether I sometimes forget to vote. (It doesn’t appear that the campaign knows what’s in my registration file, though; I’m registered as a permanent absentee voter, but the campaign hasn’t asked me to mail in my ballot yet.)

Because text messages allow for such precise targeting, it seems likely that over the next week the Obama campaign will direct its appeals to voters in battleground states, especially first-time voters that the campaign has registered during the past year. In 2006, political science grad students Aaron Strauss and Allison Dale studied how newly registered voters responded to text-message reminders sent out just before the election. The text messages increased turnout by 3.1 percentage points. Strauss says there’s a simple reason why: “The most prevalent excuse for registered voters who don’t cast a ballot is, ‘I’m too busy’ or ‘I forgot.’ Texting someone is a convenient, targeted, and noticeable reminder for them to schedule their Election Day activities with a block of time set aside for going to the polling place.” In a post-election survey, Strauss and Dale asked voters whether they found the text messages helpful; 59 percent said yes.

Obama’s campaign seems to know these lessons well. During the primaries, the campaign sent out multiple messages to supporters during Election Day; they’ll do the same next week. There’s some question about whether text messages will continue to be effective beyond this election—if telemarketing companies can get ahold of our cell numbers and we get barraged by political spam, text-based mobilization efforts may eventually become as useless as robo-calls. At the moment, though, we’re in thrall to our cell phones—and when Obama texts you next Tuesday, you’ll have a hard time saying no.