This weekend I canvassed for Qasim Rashid and Jess Foster. In this canvass we asked voters not just to commit to voting (via a Commit-To-Vote card) but to commit to asking three friends to vote. This idea, Vote Tripling, was developed by Robert Reynolds and was written up by Tina Rosenberg in the New York Times this March.
Asking people to reach out to three friends is more fun and more rewarding than asking people to fill out a Commit-To-Vote form. I feared that because this is a bigger ask that I would get more resistance. Not so. Indeed I found that 75-80% of supporters were willing to give me the first names of three people whom they were willing to talk to about voting for Qasim and Jess. Furthermore, all 17 voters who agreed to reach out to friends also agreed to give me their cell phone number. When collecting Commit-To-Vote forms I find that about 20% refuse to give me a cell phone number.
In short, not only does Vote Tripling work, it is also more fun than collecting Commit-To-Vote forms. I am not ready to say that we should abandon Commit-To-Vote forms in favor of Vote Tripling. In a months long campaign (like 2020) I could see a role for both. Perhaps using the Commit-To-Vote form over the summer and switching to Vote Tripling in the fall.
To learn more about vote tripling, visit www.VoteTripling.org
Ken Stanley 
Wow! I had not heard of ‘vote tripling’ before. I like the idea of something that makes canvassers feel comfortable. A confortable canvasser is a convincing canvasser. One thing I don’t quite get — the cell phone number? Do you plan to check up? Ask if they have talked to their three friends? And I wonder about making this expandible — so rather than vote triplers — vote multipliers. Ask How many friends could you tell? Two? Three? Five? Ten? Let the canvassee set the limit! But this sounds great. The more votes the better!
LikeLike
I have only used vote triplers in one campaign, so I am hardly an expert, but my personal view is that it is good to ask for only three names on the card. If you asked me for ten, I would probably not even try, but providing three seems like something I could do. It is good to set achievable goals. They let the canvasser can feel good about getting a commitment from the voter, and they make the voter feel good for having done something for the campaign.
An expert canvasser is not necessarily limited by the number of spaces on a card. If a voter filled in the card rapidly, the canvasser could say something like, “Wow, that was fast! Do you need more cards?”
Happy canvassing, and may you have many voters who fill in multiple cards.
LikeLike
We, too, found vote triplers to be easier to ask for than commit to vote forms. I highly recommend them.
LikeLike
Thank you, Ken! When you are not devising new canvassing techniques, you publicize others like vote triplers that increase effectiveness. And you validate your recommendations. We’re lucky to have your productive work.
LikeLike
Very interesting post. It will be interesting to see whether randomized control trials demonstrate that vote trippling does increase turnout. My concern would be that people will reply “yes” and provide 3 first names, but forget to reach out (or never have a true intention to reach out) to those 3 voters afterwards. The reason is that research has found that people lie routinely during surveys and when facing questions about what they should do vs what they actually do. One advantage of the apps mentioned in the NY Times Opinion piece that you refer to in your post is that the apps can track whether the person truly reached out to a friend and some can even show afterwards whether the person voted or not, and the Commit 2 Vote cards can be sent back and remind the voters to vote, but vote trippling relies on people’s assurances that they will, or have, reached out to other voters. I suspect in many cases those canvassed will reply that they have reached out when they actually have not.
By the way, vote trippling was used by the Dan McCready Campaign during the recent NC09 District 9 elections, as part of their texting and Outvote campaigns. Perhaps some of their data could be looked at to show whether it had some success in getting people to turn out to vote.
LikeLike
Votetripling.org did a randomized control trial with 50,000 voters (25K treated, 25K control). This was a text only intervention and showed a cost per net Democrat vote of about $27 using a very conservative calculation that only include those friends in the same household. Hence the impact is likely higher and the cost lower than that. Robert Reynolds of votetripling.org tells me that these findings will be published in the Analyst Institute in the near future. See http://bit.ly/TriplingRCTSummary for details.
LikeLike
I’m in! I am going to try to get “buy in” in the canvassing we do in my locale. Thanks for the tips Ken, you are always thinking about the next best way to make democratic campaigns better and more successful, which will make things better for middle class Americans.
LikeLike
Ken, that makes a lot of sense. However I am confused about the statistics. You say 75-80% agreed,and later you said 17 people agreed. Does that mean that the total number of people you talked to was 20 or 21? That seems improbably low knowing you. Am I understanding the figures correctly?
LikeLike
I knocked on 124 doors and spoke to 34 people. Of those 21 told me that would be voting Democratic and 17 of those committed to ask three people to vote on Nov 5th, gave me the names of those three and gave me their own cell phone number. (We don’t ask for contact info for their friends.) It is the 17 out of 21 that I refer to as 75-80% as those 21 supporters are the only people that I asked to contact others.
LikeLike
I do have one quibble. If one voter goes and gets three more voters, how many voters are there in total? Right! The cards would be more accurately described as vote quadruplers, though I do not expect that name to catch on.
Field organizers have never said this to me, but I think there might be limits to what can be achieved by canvassing. If there are too many canvassers and there is too much literature distribution, some voters might suffer “campaign fatigue” and become less likely to vote, or might at least start being less cordial to canvassers. It seems that vote tripling might help to stretch the limits of canvassing by harnessing a new and more personal approach, that is, the outreach by the person who completed the vote tripler card.
LikeLike
Good to know! It’s a simple idea but really clever in the way that it simultaneously reminds people to vote and gives them an easy opportunity to become small-scale volunteers themselves. This makes me wonder, when canvassers become more advanced or encounter particularly enthusiastic voters, is there any thought of having them ask the 3 friends to remind 3 of their friends to vote as well?
LikeLike