Virginia 2019
Letters to Virginia Voters in 2019 boosted turnout by over a percentage point.
We asked volunteers to send letters to over 300,000 voters in Virginia's 2019 general elections. Our goal was to test (again) whether handwritten letters increase voter turnout, and beyond that, to test whether sending multiple letters has even more impact.
The Takeaways
Our letters boosted turnout by over a percentage point in both campaigns
We didn't find strong evidence that sending two letters is any more effective than sending one
The Backstory
Our usual, tried-and-true approach to voter turnout is to send handwritten letters to voters - one to each voter. But of course, for some people, that one letter doesn't get through. Maybe it gets lost in a stack of mail, or quickly forgotten, or just plain ignored. So we wondered: would sending two letters, instead of just one, help overcome those barriers?
For Virginia's general elections in 2019, Vote Forward set up experiments to test this question in two get-out-the-vote campaigns, an A (core social good) campaign and a B (political) campaign. ½ of the eligible voters in each campaign were randomly held out as a control group and received no letters. ¼ were randomly assigned to receive one handwritten letter, and ¼ were randomly assigned to receive two letters. Our questions were:
Can we show, once again, that Vote Forward letters boost voter turnout?
Did two letters boost turnout even more, versus one letter?
To the first question, our results showed that the handwritten letters did boost voter turnout. Compared to the control group, we estimate that people who received any Vote Forward letters (regardless of how many letters) voted at a rate that was 1.9 percentage points higher in the A campaign, and 1.1 percentage points higher in the B campaign.
A Campaign (Social Good)
B Campaign (Political)
Those might seem like small numbers, but since getting non-voters to vote is pretty hard, even small increases are big wins. And they're right in line with our previous findings in Ohio and Alabama.
This adds to our conviction that our handwritten letters are working to boost turnout, even in relatively high profile elections. But can we make them work even better, by sending even more letters?
Here, our results are less conclusive. We don't see solid evidence that the voters who received two letters turned out at a higher rate, compared to those who received just one. We did see a small boost (0.5 percentage points) in our A campaign, but it's not statistically significant - in other words, it's too small for us to feel confident that it's real.
A Campaign (Social Good)
B Campaign (Political)
It's worth remembering that, in this experiment, we didn't vary the letter template or the timing at all. From a voter's perspective, perhaps receiving two very similar letters, around the same time, isn't too different from receiving one letter. In future, we might explore other questions around multiple letters, but for now, we can't say that two is better than one. What we can say for sure is that we put handwritten letters to the test one more time - and once again, they proved effective.
We've shown that handwritten letters - from voters, to voters - can motivate people to turn out in elections. With this year's Big Send, we're aiming to send 15 million of them. To join us, sign up at votefwd.org/thebigsend.
