Freakonomics Blog summarizes a recent study finding no effect of organizational GOTV (get out the vote) e-mail campaigns.
“If an e-mail message from a campaign or non-profit group were to pop up in your inbox on election day asking you to please go down to your polling place and cast your vote, would you do it?
“Probably not, if the results of a study by Notre Dame political scientist David Nickerson are any indication. Nickerson conducted 13 field experiments during elections between 2002 and 2004, and found that aggressive e-mail get-out-the-vote campaigns have virtually no effect on voter turnout.
“What does seem to persuade people to vote is personal contact, with door-to-door canvassers and especially with co-workers, friends, and family.
“That latter point is driven home by another Nickerson paper, published in the American Political Science review, which asks: ‘Is Voting Contagious?‘
“Dubner and Levitt have written previously about the incentives to vote (or not vote). And Ian Ayres touched on the social aspect of voter persuasion in a recent post, saying that voting might be rational if you do it because you care about your fellow citizens. That squares with research showing that people are more likely to vote if they believe it is socially expected of them.
“Okay, now let’s say that election day e-mail is from your friend, or one of your parents: would it convince you to vote?”
See earlier Social Capital blog post about how face-to-face GOTV efforts in conjunction with technology do work. And there is evidence that talking about politics with others has a multiplier effect because you persuade others (if you are persuasive) and they in turn persuade others, etc.