Keys to a Future Majority: P2P Contact, Social Groups, and Voting

I came onto Music for America’s forums to tell their Communications Director, Mike Connery, about the research that I had done, and to try and see if they’d be willing to help me apply it.

My idea was to create a multi-media CD that would replicate the agenda-setting and priming effects that Iyengar had demonstrated. I thought that if we could convince, trick, or bribe young people to take a look at a CD that contained multimedia relevant to the campaign, that we could have a huge effect on youth turnout. MfA had a huge outreach operation, was extremely well funded, and I thought that it had the ability, if it desired, to get such CDs into the hands of tens of thousands of young people.

Mike wasn’t convinced. He challenged me to justify a project on the scale I was talking about, forced me to consider production time and cost, and pushed me to solidify my ideas on how something like this could work. He did not dismiss me, even though he had never met me and didn’t know me at all. He engaged me, he challenged me, and he encouraged me to continue on, which I did.

Mike had two main criticisms of my ideas. First- he noted that I didn’t understand how MfA worked—the logistics of distributing materials, how volunteers worked at shows, etc.—and so I decided that I would start volunteering with MfA, if only to prove to Mike that the project I was advocating for could work within MfA’s operations.

Mike’s second criticism was that TV/broadcast media are not the only methods available to convince people to do something or think a certain way. Broadcast media may be powerful, but having someone talk to their friends or peers in a social context which they identified with and felt comfortable, he insisted, was not only more cost effective, but also much more effective in general. This seemed like pretty solid common sense to me.

Let’s say that two people are asking you to do something, it could be voting, choosing a brand of a product, or really just about anything, but for our purposes let’s say that they are asking you to volunteer for an issue-based campaign. The first of these two people is a well known celebrity, whom you don’t know personally, and whom you have very few, if any, social connections to. Those celebrities will try to communicate with you via Television, Radio, and direct mail, and you will not be able to communicate back with. The second person asking you to help out is someone your age, and is talking to you in some social space that you go to on your own. This person may or may not be a friend, but they clearly are someone “like you,” and you can ask them questions and there is some chance that you will see this person at some point in the future. Who do you think you’d be most likely to listen to and help? It seems obvious that you’d be much more likely to help those people whom you see face-to-face and have some social connection to, rather than some celebrity, whom you can be fairly certain you will never meet and never interact with in any meaningful way.

But we need not rely solely on common sense here - there is an abundance of sociological research which demonstrates the superiority of peer-to-peer (P2P) outreach as well as the importance of social context in the political decision making process.

The sociologists Donald P. Green and Alan S. Gerber have demonstrated in a number of studies and reviews that targeted P2P canvassing is both extremely effective as well as cost effective. In one of their reviews they found that not only did contact with a canvasser increase voter turnout by almost 11%, at an average cost of about $15 per new voter, they also found a few other interesting facts:

• Among voters 25 and under, face-to-face canvassing had slightly stronger effects.



• Face-to-face canvassing produced .spillover. effects. Other registered voters living at the same address with voters in the treatment group voted at significantly higher rates than adults living with voters in the control group. In the five cities where face-to-face canvassing proved effective, voter turnout increased by 5.7 percentage-points in households that were contacted by canvassers.
• Since canvassers were able to make contact with voters in only a small minority of the housing units they visited, the direct effect of canvassing, ignoring spillover effects, was to raise turnout by at least 2.1 percentage-points among voters 25 and under. This figure could be greatly increased by more intensive canvassing.
• Appeals that were coupled with issue advocacy were slightly more effective in getting voters to the polls than straight Get Out The Vote (GOTV) messages.
• Providing voters with voter guides had no discernible influence on turnout.
• Raising turnout among 100,000 eligible voters by 3 percentage-points costs approximately $45,000, assuming that canvassers cost $10 per hour, contact 8 voters per hour, and mobilize one additional voter per 12 contacts.

(Green, Gerber, and Nickerson. Getting Out the Youth Vote in Local Elections: Results from Six Door-to-Door Canvassing Experiments, 2002) PDF

The “spillover” effect that Green and Gerber mention is also pretty well established, and shows that P2P messaging has a very significant multiplier effect. For example, Political Scientist James Fowler looked at voting habits within social networks and found that an initial increase of one voter, do to GOTV activities, can increase overall participation by, at the very least, 2 additional voters.

One way to understand why there is a significant spillover or multiplier effect for GOTV activities is to look at the ways in which ones social groups influence patterns of political participation. To put it simply, we are most likely to participate when we belong to a social group that encourages participation. Similarly, we are most likely to change our political views and behaviors when someone within our immediate social networks changes theirs. As the researchers PA Beck, RJ Dalton, S Greene, and R Huckfeldt demonstrated, our political conversations within our existing social groups are far more influential than any media interactions. This is also a validation of the thesis behind Robert Putnam’s Bowling Alone: America’s Declining Social Capitol , that the decline in participation in social groups, such as social clubs and bowling leagues, is related to the general decline of civic participation in America today.

And so, while I still wanted to conduct my persuasion research and outreach with Music for America, I was becoming increasingly intrigued by MfA’s model of outreach. Not only did MfA’s main outreach consist of face-to-face interactions with young people, but these interactions were carried out at the concerts and parties where most of the young people socialized at. Instead of trying to create a political movement of young people from whole cloth, MfA was politicizing social networks that were already existing, which, the more I pondered it, seemed like a brilliant idea.

And so I signed up for a bunch of concerts and I decided to see of these theories worked in practice for my own participation.