For several weeks I’ve been exploring an argument increasingly heard in the press and some political circles: Persuasion is wasted effort, at least in a midterm race; increasing turnout is the only way to win.
A future column will bring it all together, but here I’ll focus on the impact of persuasion, as last week focused on the effects generated by turnout efforts. Those who denigrate persuasion make two kinds of arguments: first, that there are not many persuadables, and second, that they cannot really be persuaded.
Emblematic of the former was a post on The New York Times’s The Upshot site, which communicated its thrust in its title, “The Myth of Swing Voters in Midterm Elections,” and noted that fewer than 6 percent of 2008 presidential election voters cast ballots for a congressional candidate from the other party in 2010. That’s one definition of persuadables, though it fails to distinguish the persuadable from the persuaded.
No doubt, most of the time, the number persuaded is small. But in the unusual races, the upsets, the ones that run counter to presidential type, those persuadables exert a huge influence. President Obama garnered 38.9 percent of the vote in North Dakota while our client, now senator, Heidi Heitkamp, got 50.5 percent. Sadly there was no exit poll in North Dakota (tsk, tsk, National Election Pool!), but it’s clear Heitkamp won not by bringing additional Democrats to the polls but by persuading some voters for GOP nominee Mitt Romney to split their tickets. Absent persuasion, Heitkamp would have lost.
In Indiana, where exit polls reveal a clearer picture, 21 percent of Romney voters defected from the GOP Senate nominee, while just 5 percent of Obama voters failed to support now-Sen. Joe Donnelly. Had he not persuaded those defectors, he too would have lost. More dramatically, Democratic Sen. Joe Manchin did 35 points better than Obama on the same day in West Virginia. He persuaded a good many voters.
Are there enough voters to persuade? Certainly not everywhere and always, but where tough races are won, there often are.
The second line of argument against persuasion suggests that even if there are persuadable voters, they cannot actually be persuaded. As The New Republic contended, “Mobilization is the only proven way for a campaign to close that gap [between the number of votes Democrats have and the number they need to win].” Unless you buy unproven pigs-in-pokes, no one following the magazine’s advice would invest in persuasion.
But it turns out the writer is demonstrably wrong on the facts. In addition to the strong evidence from exit polls and election results, a series of communication experiments reveals that persuasion does take place.
Persuasion experiments are hard to conduct and are far outnumbered by those testing get-out-the-vote (GOTV) strategy. No one has done the kind of meta-analysis of data that I cited on GOTV, but I calculated a simple mean of the effects recorded in 26 studies I located and found an average impact of 3.87 percentage points added to the Democrats’ vote. While some of these experiments revealed a short half-life for persuasion, others yielded more persistent effects. Of course, some studies showed zero impact, as did some of the GOTV experiments. Some also suggested much greater impacts, as did some of the GOTV studies. But on average, the persuasion experiments added more votes to the bottom line than did the turnout experiments.
As promised, a future column will bring this all together, but none of it is to suggest that GOTV is not important. Rather, it is to say that those who see GOTV alone as Democrats’ salvation make a grave and costly error by ignoring facts and evidence. It should never be either/or, but rather both/and — both GOTV and persuasion.
Mellman is president of The Mellman Group and has worked for Democratic candidates and causes since 1982. Current clients include the majority leader of the Senate and the Democratic whip in the House.