So, now that I've spent 30 minutes talking to you, will you vote for my guy?

April 22, 2009 1:43 PM

For the politicos out there: how effective is phone banking / canvassing, statistically speaking?

So I'm working on a political campaign right now on the communication side of things, and I was curious as to some of the inner-working of the field program. I'm relatively new to all this, so a lot of it is a mystery to me.

We are running a particularly intense field program, and I was curious about the literature and history behind it. Some particular questions pop up in my head, like how likely is voter going to support you if a volunteer talks to them face-to-face? Over the phone? To the candidate directly? Where did this technique of politicking originate? What are the major pieces of literature behind these techniques? Things like that.

Sometimes I wish I had majored in Political Science rather than English...

posted by Weebot to law & government (8 comments total)

Judging by the current occupation of the white house I'd say very effective. ;)

Having volunteered for that campaign, both over the phone and door to door, I think the latter is better. More people are likely to ask questions they have to a real person on their front step than a stranger on the phone. And people are less likely to slam a door in someone's face than they are to hang up on a phone call. It's the whole "asshole on the internet" thing in a new form... it's easier to be rude to a disembodied voice than a smiling face in front of you.
posted by Kellydamnit at 2:10 PM on April 22

Good field work among those you already know will, or would likely, vote for you (voter registration / absentee / get out the vote) can add 1% to 3% to your tally, maybe ramping to as much as 5% in optimal conditions: opponent with weak field himself/herself, low turnout elections, your candidate's having a big pool of voters with a low propensity to register or show up, etc.

1% to 3% may not seem like much, but it wins a heck of a lot of elections.

For what it's worth, most political consultants don't think highly of field work by proxy as a means of persuasion. However, the candidate's own shoe leather is a totally different thing: every political consultant will tell you stories of candidates who won shocking upsets simply because they knocked on an absurd (thousands, tens of thousands) number of doors, and by election day had 20% or 30% of the electorate feeling personally acquainted with them and saying so their friends and family.
posted by MattD at 2:41 PM on April 22

I remember reading articles similar to what you want in Public Opinion Quarterly back when I was a polisci major, but that would require JSTOR access to dig any of it up.

As for seminal names, you're looking for V. O. Key. I can't tell you how readable he is, but it was always the go to name when looking at the history who started quantifying public opinion and political marketing.

I'd say the important thing to keep in mind is that you aren't there to win minds or change opinions. The ground troops are there for one reason and one reason only: name recognition. So if you get access to any academic journals, try that in the search engine.
posted by politikitty at 3:44 PM on April 22

It's not just about name recognition; all the recognition in the world is useless if people don't show up to vote. Studies show "supporters" are more likely to actually show up and vote if there's face-to-face contact.

Weebot, I recommend Get Out The Vote by Green and Gerber.
posted by faster than a speeding bulette at 4:19 PM on April 22

I agree, phone banking is extremely important even with the rise of cellular phones. Many more people unexpectedly volunteered to be phone banked by signing up for support President Obama. I never experienced a phone bank (I do fundraising) as long as 30 minutes. Most run three to five minutes and you can clear 60 calls in three hours.
posted by parmanparman at 4:48 PM on April 22

I was going to suggest books or articles by Green and Gerber and their students, as well. They do field experiments testing the effectiveness of different get out the vote techniques. They or their followers have tested such things as the effectiveness of community festivals or the difference between a phone bank run by volunteers from the community and one staffed by professionals. It's fairly recent literature, but it's done the best job of answering these questions.

If you're looking for very early work on the effect of campaigns, the earliest I can think of is Berelson, Lazarsfeld, and McPhee, Voting: A Study of Opinion Formation in a Presidential Campaign, which is about a panel study that tracked citizens in a town in New York during the 1948 presidential election.
posted by capsizing at 4:56 PM on April 22

Look through the archives of fivethirtyeight.com. Nate Silver (the guy who runs 538) does a lot of regression analyses that seemed pretty spot on during the primaries and general election last year(s) and, as I recall, he had several posts about the effectiveness of door knocking and phone banking.

The most important aspect of voter contact is getting supporters to the polls rather than changing minds. When I volunteered for the Obama campaign we would mark down whether a person supported or opposed his candidacy. Even early on we would immediately stop contacting people who outright opposed him. As election day drew nearer we began to limit contact with neutral voters in order to focus time and energy on voters who had expressed some degree of support. Then, the final week, we concentrated all our efforts on contacting voters who strongly supported Obama in order to make sure they actually voted. In other words, the beginning phases were more about voter identification and the end game was Get Out The Vote that had been previously identified.
posted by Quizicalcoatl at 5:59 PM on April 22

The aforementioned Green & Gerber have published a lot of their work on their website at Yale. Plenty of studies and statistics there.
posted by DavidNYC at 12:20 PM on April 23