Why relational?

People are more likely to vote when their friends tell them to. Our meta-analysis estimates that a GOTV contact from a friend increases a voter’s chance of turning out by 2.5-3 percentage points. That’s larger than the average impact of nearly any other turnout method — larger than phone banking, texting, mail, pledge cards, etc. While there are many ways to leverage relational organizing, we designed the vote tripling pledge collection as a scalable way to create these powerful friend-to-friend contacts. There are several possible reasons why relational contacts are so powerful. Friend-to-friend contacts may be better able to cut through the noise of an election cycle, messages from peers may induce more pressure to turn out to vote, or friends may be able to contact voters who are typically ignored by traditional turnout operations!

Who are the three friends that triplers are picking?

In short, triplers tend to pick friends who are demographically similar to themselves and have similar partisanship. The Analyst Institute describes this as “homophily”, where essentially, birds of a feather tend to flock together!

We conducted two rigorous studies in early 2020 that matched triplers’ friends to the voter file. Both studies showed that triplers’ friends are, on average, similar to the triplers — in age, demographics, and partisanship. Critically, if the tripler is a Democrat, there is at least an 80% chance the friends they remind will be as well. The studies also showed that there is a correlation between the voter propensity of triplers and their friends, but it is not as absolute as one may think; even the highest-propensity triplers select likely non-voters 12-35% of the time. In other words, most triplers, regardless of their voting propensity, pick worthwhile friends to remind.

Does vote tripling work during primaries?

Evidence and our on-the-ground experience both suggest that the vote tripling suite of tactics is effective in primary elections. 

In the August 2020 Florida Primary Elections, we found that sending vote tripling texts to early voters increased voter turnout of copartisan housemates by 0.95 percentage points. Our polling place vote tripling experiment in Ohio also took place during a primary election and increased turnout in the treatment group by 4.5-10 percentage points. 

For campaigns seeking to use vote tripling in primaries, we do not have research assessing whether triplers accurately assess the candidate preferences of their friends. However, our demographic results about friend selection do offer a solution: primary campaigns can simply recruit triplers from communities/demographics where they have strong support, knowing that most friend-to-friend reminders will remain within that community.