CINCINNATI—By the time Mika Darley-Emerson stepped out of the light drizzle and into a doorway’s modest refuge, she’d already forgotten about the instructions she had been given about how to talk to a prospective voter.
Four years ago, as she was beginning her first year as a graduate student at the University of Cincinnati, Darley-Emerson had coordinated a system of early-vote shuttles for the Obama campaign. This past Saturday, she returned to the city from the suburbs, where she now has a baby and a partially completed dissertation, to help get out the vote for the president’s re-election. “We go where the volunteers are needed,” the winsome 34-year-old said. Darley-Emerson had a script that was supposed to direct her interactions with voters, but it was buried under the list of names and maps of the middle-class Westwood neighborhood she had been assigned. She’d also received training that morning emphasizing that at the start, volunteers were to find out which candidate the voter supported before moving ahead to other questions. The purpose of a GOTV canvass is to mobilize only those known or highly likely to be supporters.
The fourth house Darley-Emerson visited was home to an African-American man in his 30s who came to the door in a sweatshirt emblazoned with the word ETHIOPIA. Darley-Emerson introduced herself, and immediately asked him if he had taken advantage of Ohio’s early-voting window. He explained that he was a nursing student and that a trip to the downtown board of elections wasn’t easy to schedule. Soon Darley-Emerson was talking him through the various transportation possibilities, and the relative merits of garage and street parking. Together they made a plan—he would drive downtown to vote mid-afternoon Sunday—and Darley-Emerson moved to the next item on her mental checklist: pushing this likely voter to move down his ballot to support Ohio’s Democratic senator, Sherrod Brown.
In so doing, Darley-Emerson had the type of psychologically meaningful interaction that years of randomized-control trials had shown would increase a citizen’s likelihood of casting a ballot. But she had overlooked the opening, threshold question on her script, the one she’d been trained to start with: Who do you support for president? “I forgot to ask that first,” Darley-Emerson acknowledged after retreating to the sidewalk, and then joked that she was as “an absent-minded teacher” getting an early start on her career. “Part of why I felt a little more comfortable making that mistake,” she reflected, “is most of the people on our list at this point are supporters.”
Darley-Emerson’s rounds—and those of hundreds of thousands of other canvassers and callers in the closing hours of the election—may look like the basic work of campaigns, the slog of door knocks and repetitive phone calls. But as is the case with much of Obama’s campaign, the dutiful fieldwork is undergirded by sophisticated analytics unmatched by his Republican opponents. The houses on Darley-Emerson’s list were not only likely to contain supporters, but supporters for whom a visit from a canvasser could make all the difference.
Matt Reese, a legendary Democratic consultant who played a crucial role in organizing West Virginia for John F. Kennedy’s presidential primary campaign, had a simple explanation when asked what field operatives do. “I wish God gave green noses to undecided voters, because between now and election eve, I’d work only the green noses. I wish God gave purple ears to nonvoters for my candidate on election eve, because on election day I’d work only the purple voters,” Reese would say when asked about his methodology. “The ones we go after are nonvoters who are for us and the undecided voters.”
The challenge for a campaign, of course, is to sort voters into those two categories without the benefit of such visible markers. In an ideal world, a campaign would interview every voter to discern his or her likelihood of casting a ballot and the candidate he or she supports. But no campaign has the money or manpower to track down and survey even a sizeable fraction of the voting public, either through volunteers or paid operations. So campaigns use data about the electorate to increase the quality of their guesses about which voters are likely to have green noses and which ones purple ears. For decades, political targeters had to use geographic or demographic heuristics (classifying precincts based their past vote performance or Census tracts based on their complexion) to sort voters into those categories en masse. With the individual-level voter data and statistical analysis available to today’s campaigns, they can now sort voters one by one.
Over a two-week stretch starting at the end of July, the Obama campaign’s analytics department contacted 54,739 voters from paid call centers and asked them how they planned to vote. Obama’s databases already knew a lot about the approximately 180 million registered voters in the United States (and even a bit about those who weren’t registered, in a way that could help guide the campaign’s efforts to enroll them). The goal was to collect intelligence about potential voters’ 2012 intentions and distill that down to a series of individual-level predictions. The most important of these scores, on a range from 0 to 100, assessed an individual’s likelihood of supporting Barack Obama and of casting a ballot altogether.