A preliminary study of state voter ID laws helps allay concerns about whether communicating with voters about these laws could suppress turnout, finding that there seems to be no negative effects, especially when information is disseminated spelling out just what forms of identification suffice and where to turn for more detail.
Research by UC Berkeley professor Jack Citrin, director of the campus’s Institute of Governmental Studies, political science professor Donald P. Green of Columbia University in New York City and University of Southern California political science professor Morris Levy, that was recently published in the Election Law Journal, applies empirical analysis to estimate the impact of different ways of notifying registered voters about new ID rules.
Of various strategies applied in the research experiment to inform and assist voters to comply with the identification requirements, those that provided more detailed information and pointed potential voters to other sources of explanatory information yielded the best returns.
The study’s authors caution that more work needs to be done to understand how states communicate their laws to the public and how those communications affect turnout..
The researchers conducted extensive examinations of voter ID regulations instituted in both Tennessee and Virginia for general elections in 2012. Tennessee in 2011 adopted a requirement for photo identification. Virginia in early 2012 approved new ID rules that allowed a range of identification options, including simply showing utility bills containing the voter name and address.
With the cooperation of the League of Women Voters, Citrin, Green and Levy conducted field experiments in the 2012 general election with 30,000 voters in 12 Tennessee and Virginia counties and cities bordering each other and that also are demographically similar. The sampling also included some heavily minority communities with large African-American populations.
Those registered voters who had cast ballots in 2008 and 2010 were excluded, and the research team selected one member at random from the remaining households with whom to proceed in their testing, partly because they wanted to measure any spillover impacts to other family members after the 2012 general election.
The study featured slightly different messages in postcards sent from the League of Women Voters to the voters, with a control group receiving no mailings at all.
One group received a reminder to vote because it’s important for democracy.
A second group received a warning postcard identical to the basic get-out-to-vote reminder, with additional information. Those in Virginia were told that the state now requires voters to show proof of their identity at the polls voter casting ballots. Tennessee voters in the second group were advised that they would need photo IDs in order to vote.
A third group received what the researchers called the “help treatment.” The Virginia help postcard listed the types of ID material that would suffice, and listed bank statements, utility bills, driver’s licenses or a voter registration card. The Tennessee version advised that documents with a voter’s name and photo – such as a driver’s license, federal or state employee ID, passport or military identification – would suffice.
Both the second and third messages contained a help line phone number.
Voters in each group were randomly selected. Postcards were dispatched prior to the Nov. 6, 2012 election. Then in February 2013, the political scientists obtained voter turnout information from registrars in each county in the study. This enabled them to analyze turnout rates for each treatment group in each geographic area. The results:
- The control group that received no special postcard had a 39.95 percent turnout rate.
- The reminder group turned out just a bit more, at 40.08 percent.
- Those in the warning group posted a 40.9 percent turnout.
- Voters in the help category recorded a 41.08 percent turnout.
Turnout for the warning group, about one percentage point higher than the group that received no postcards, “runs counter to the conjecture that warnings about the need for proper identification demobilize voters,” the researchers wrote in their journal article. They also noted that the highest turnout, for those in the help group, suggests that extra encouragement may have nudged a few more people to the polls.
UC Berkeley’s Citrin said it might be helpful for states with voter ID laws to give out identification cards when people register to vote, and thus expedite the process on Election Day, or to tell those registering just what they will need when they go to vote.