The Takeaways

• Vote Forward letters had a big impact on both absentee ballot requests and voter turnout in Maine’s 2020 Senate primary
• Letters that included a request form were slightly more effective than letters without one

The Backstory

We always knew that 2020 was going to be an interesting election year. With the COVID-19 pandemic, this election year got even more “interesting”, as states scrambled to adapt their electoral systems to offer more low-risk voting options. During the summer, Maine Democrats were looking for ways to encourage some of their most at-risk voters to participate safely in the state’s July Senate primary, so they turned to Vote Forward.

We designed an experiment focused on Democratic-registered Maine voters aged 60 years and up.* These voters had voted in previous primaries, but never by mail - making them great candidates for a vote-by-mail initiative. One third of this group was held out as a control group (see FAQ on control groups here), and the remaining voters were split into two groups: a group who received a Vote Forward letter describing how to enroll in vote-by-mail for the July primary election, and a group who received a similar letter along with a printed enrollment form.

Our highest priority was to encourage these voters to request absentee ballots for the Senate primary. We succeeded! Absentee ballot request rates were an estimated 4.2 percentage points higher among voters who received a Vote Forward letter (of any type), compared to voters in the control group.

This effect averages across the different letter types. When we break it down, we see that the letters that included a vote-by-mail enrollment form were more effective (+5 points compared to controls) than the letters without a form (+3.5 points compared to controls) in raising ballot request rates. We had included these forms because we believed they might make it easier for voters to get started right away on their ballot requests, and based on our evidence, it appears they did.

But ballot requests, of course, do not decide elections. We also hoped that our letters would encourage these voters to go on and cast their ballots. And they did! Our letters boosted overall voter turnout (regardless of voting method) by an estimated 2.8 percentage points. Our letters also increased the proportion of those voters who voted by absentee ballot, by an estimated 3.8 percentage points, suggesting that we did indeed encourage voters to follow through and vote via the absentee ballots they had requested.

This means that our letters not only affected how these voters chose to cast their ballots - by mail, rather than in-person - but also, whether they voted at all. Since our primary goal was to shift voters towards voting by absentee ballot, the fact that we also encouraged more people to vote overall was a pleasant bonus.

Voters in 2020 have had to navigate rapidly changing election processes, new options for voting, and the health risk of showing up at the polls in person. Our letters were able to support some of the most at-risk voters, helping them switch over to a safer method of voting and turn out to vote in spite of those challenges. These positive findings encourage us to keep pushing forward with absentee ballot request campaigns, in the 2021 Georgia Senate runoff election and beyond.

**(Note that this is an example of one of our Political campaigns; however, most voters who receive Vote Forward letters are part of our nonpartisan, Social Good campaigns.)*