BROOKE GLADSTONE:
This is On the Media. I'm Brooke Gladstone.
BOB GARFIELD:
And I'm Bob Garfield. Since the first of the year, more than a billion dollars has been spent on political and issue ads. According to a recent study, the Obama campaign was on track to spend more than a hundred million dollars on TV ads in October alone. Then, of course, there are the direct mail campaigns, the email blasts, the door-to-door efforts and the robo-calls, like this one:
MALE ROBO VOICE:
Hello. I’m calling for John McCain and the RNC because you need to know that Barack Obama has worked closely with domestic terrorist Bill Ayers whose organization bombed the U.S. Capitol, the Pentagon, a judge’s home and killed Americans. And Democrats will enact an extreme leftist agenda if they take control of Washington.
BROOKE GLADSTONE:
No matter the method, the point is first to lure people to vote and then to affect how they vote. But are these pricey appeals really doing the trick? Donald Green, director of the Institution for Social and Policy Studies at Yale, has spent years assessing the power of all these methods to both mobilize and persuade voters.

Green found that campaigns generally aren't getting much bang from their bucks.
DONALD GREEN:
We've conducted many, many studies, more than a hundred in all, and the basic message of the studies that we've done is fairly clear, and that is that by and large the more personal, the more authentic, the more effective things are. When things are impersonal and mass produced, as in the case of direct mail or robotic phone calls, they tend to be relatively ineffective.
BROOKE GLADSTONE:
But exactly how ineffective [LAUGHS] are they, TV ads and robo-calls? Do they have no effect at all on the outcome?
DONALD GREEN:
Well, we've studied robo-calls by celebrities, by governors, by local officials, by local clergy, partisan, non-partisan. We just don't find any effect on whether people vote. That we're pretty sure about, that their effects on mobilization are very meager.
BROOKE GLADSTONE:
What about TV advertising? I mean, that’s where the lion’s share of money is spent.
DONALD GREEN:
Only one randomized study, of which I know, has looked at the effects of TV advertising on vote choice. We did find a momentary effect. The effect lasted a few days, but a week later it was all gone.
BROOKE GLADSTONE:
And you found that person-to-person pitches, even door-to-door appeals, are really the effective way to go, but it takes a lot more time and money. Can you break down the different methods into a dollar-per-vote figure?
DONALD GREEN:
We do try, so the most personal might be a door-to-door visit, conversation at one’s doorstep. You know, that produces votes that could be quite costly if you’re paying well-paid, well-trained volunteers, but if you've got a truly volunteer effort, it can be quite inexpensive.

So the typical figure is something in the order of, you know, 20 to 30 dollars per vote, which sounds like a lot, but you have to remember that a direct mail campaign might cost maybe three times that, five times that.

Phone calls - volunteer phone calls are quite cost-effective, particularly volunteer phone calls where a caller will call you up and say, Brooke, can I count on you to vote? And if you say yes, maybe you get a return phone call the day before the election saying, you know, now Brooke, can I count on you to make good on your promise tomorrow?

That has a huge effect, whereas a call from a telemarketing firm tends to have a fairly minimal effect and a robo-call seems to have no effect at all.
BROOKE GLADSTONE:
And yet, going against this trend is some new research that shows that text messages really are quite effective, especially when you take into account how cheap they are to send. I understand a text message costs about 6 cents per contact and it comes to about $1.50 per [LAUGHS] vote. How come texts pack more punch than emails or robo-calls?
DONALD GREEN:
Now, it might have to do with the nature of the lists, in that these are opt-in lists and people have agreed to be reminded.
BROOKE GLADSTONE:
If it is an opt-in list, doesn't that kind of cast a deep shadow over this research? Because if people are saying, go ahead, contact me, then they're already in the tank, so to speak.
DONALD GREEN:
Maybe so. Maybe there’s a behavioral readiness on their part, and when they finally get the reminder they're ready to, to run with it. And we'll see whether that works for a generic list of people.

What’s fascinating about it is maybe the way in which people relate to their cell phones. The cell phone commands, and I must obey. And if it reminds me to vote, I go vote.

But we really don't fully understand why it is that this impersonal tactic seems to work with the effectiveness of a personal tactic.
BROOKE GLADSTONE:
Now let's move to another study you did on less traditional ways to get out the vote.
DONALD GREEN:
Well, let me give you a little bit of background on these studies. They are inspired by our reading of 19th-century American history. Americans turned out at, you know, 80-plus-percent rates, sometimes 90-plus-percent rates, despite the fact that it was a very poor period in American history, people were relatively uneducated, and it was very difficult to register to vote. Usually there was only one day when you could register to vote. Despite all those things, turnout was very high.

And so, a lot of this research is inspired by an effort to recreate some of the circumstances of 19th-century voting, and one of those was the fact that one would vote in public, often hanging around the polling place for hours, imbibing a little whiskey, maybe arguing, maybe brawling – after all, a men-only affair - and people were there to see and be seen as voters.

With the progressive reforms of 1889 and after, voting became a kind of solitary affair, something you did in relative anonymity.

And so, the experiments that you’re referring to were an attempt to remind people that voting is a matter of public record and that they would been seen as to whether they voted or not.

BROOKE GLADSTONE:
Tell me how the study was structured.
DONALD GREEN:
The study was designed to measure the effects of increasing amounts of social surveillance, if you will. The first group got nothing. That was the control group. Another group was told, you know, voting is a matter of civic duty, so please go out and vote. Another group was told, voting is a matter of civic duty and researchers will be studying whether you vote or not, but they won't contact you in any way.

The next treatment examined a mailing that said voting is a matter of civic duty, do your civic duty, it’s a matter of public record, and showed people whether or not they and their housemates had voted in the last two elections.

And, finally, the strongest treatment was one that encouraged people to do their civic duty, reminded them that voting is a matter of public record, showed them whether they and their housemates had voted in the last two elections and showed them whether 10 of their neighbors had voted in the last two elections, and presumably showed their neighbors whether they had voted.

And the last two treatments, the ones that revealed whether people had voted in the last two elections, had enormously strong effects, relative to direct mail, which almost never has an effect.
BROOKE GLADSTONE:
It boosted voting in the studies by what, 26 percent?

DONALD GREEN:
Yeah, 26 percent. And so, the question is, you know, what, what’s going on there in terms of a social/psychological mechanism? And I think that part of it is people realize, oh if it’s a matter of public record, well, then I can't say I voted and not really vote. I actually have to go vote.
BROOKE GLADSTONE:
Donald Green is a director of the Institution for Social and Policy Studies at Yale and author of the book, Get Out the Vote. Donald, thank you very much.
DONALD GREEN:
My pleasure.