You are not logged in. LOG IN NOW >

No, That Research Does Not Suggest Online Voter Registration Could Reduce Turnout

BY Nick Judd | Wednesday, August 29 2012

Research described yesterday as indicating that online voter registration could reduce turnout actually says nothing of the sort.

Writing yesterday for TechCrunch, Gregory Ferenstein suggested that research indicates when someone uses TurboVote, a tool to start the voter registration process, they might be less likely to finish the process of voter registration than if they had engaged in the whole thing offline. He also suggested, more broadly, that the research — published in Political Research Quarterly — indicates that online voter registration might decrease turnout.

He put it this way:

... experimental research into the impact of such online registration systems [PDF] finds that they actually decrease registration. Apparently, the ease of the online process lulls citizens into complacency and they forget to follow through with the rest of the process. The unfortunate drawback can be offset with SMS reminders, which TurboVote encourages. So, depending on the number of people comfortable giving Google their digits, this well-intentioned experiment could backfire.

He's wrong for three very simple reasons.

The first reason is that he seems to misunderstand the research. The study was focused on email outreach, not online voter registration. (The title of the paper is "The Cost of Convenience: An Experiment Showing E-Mail Outreach Decreases Voter Registration.") The study he cites examined whether emailing college students with various appeals to register to vote would make them more likely to download a voter registration form online and then mail it in. People sent those emails were three-tenths of one percentage point less likely — that's 0.3%, if you're nasty — to be registered to vote after the election than the rest of the population tracked in the study.

The second reason is that he seems to misunderstand what TurboVote does. TurboVote asks users for their information and does give them the option to print and mail in forms themselves, if they want. But it also offers, for a small fee, to mail users a pre-filled voter registration form for their state that they can then sign, place in a provided stamped envelope, and send in — a significantly lighter lift and the reason why they call themselves "Netflix for voting." Ferenstein also notes that a follow-up experiment by Rock the Vote in 2008 found that sending text-message reminders to people who began the voter registration process makes them four percent more likely to be on the voter rolls after the election. TurboVote also sends texts like these, and email reminders. It's fair to say that the research shows TurboVote's email reminders might not do a lot of good. But TurboVote is already also using text messages that are likely to increase registration rates.

On that point, users aren't "giving Google their digits," although TurboVote does use Google Wallet to handle payment. Users are giving those digits to TurboVote, a 501(c)3.

The third reason Ferenstein is wrong is that he misuses "online voter registration" as a term. TurboVote is not online voter registration. In online voter registration, the user goes from zero to voter entirely over the Internet, with no offline component necessary unless there's a complication.

California is about to get online voter registration. An online voter registration system in New York was launched Aug. 16 and has so far been used by more than 3,474 people, including 1,028 first-time voters. In Washington State, not only can you register to vote online, you can register to vote through Facebook.

Just because Ferenstein took a swing at criticizing online voter registration and missed doesn't mean it isn't worthy of critique. A colleague tells me his son tried to use TurboVote a few days ago and it crashed on him; he also wound up printing out the forms to mail in on his own time later. I would be astonished to learn that Washington's Facebook system worked perfectly or that other online voter registration systems worked without problems.

If you want a fundamental criticism of online voter registration, here's one. The systems that have been adopted so far work by checking for digital signatures on file at a state's Department of Motor Vehicles. If you already have a signature on file there, then you can register online. If you don't, you can't finish the process online, and have to suffer through some rigamarole with paper forms.

Generally, you only have a signature on file with the state DMV if you have a driver's license or other form of state-issued ID.

So online voter registration is helpful to everyone except for the same people at the center of the voter ID law debate — folks who for whatever reason do not have a driver's license or other state ID and as such would be denied access to the polls, even if not only eligible but also registered to vote. One could argue that this means the people most in need of increased ballot access are unaided by online voter registration. But if Ferenstein wants to suggest that online voter registration might actually decrease the number of registered voters, and by extension the number of participants in an election, he'll have to find other research supporting his point.