Studying Voter Turnout with Google
Taking a break from March Madness to share one of this week’s student submissions about politics in the United States.
Ms. Claudia Valencia shared a recent Monkey Cage article about the impact of early registration deadlines.
According to Alex Street, an assistant professor of political science at Carroll College, early registration deadlines needlessly prevent many people, perhaps millions of people from voting.
Turnout is higher in states that allow voters to register on Election Day. Despite fears over administrative difficulties, surveys show that polling place wait times are actually shorter in states with election-day registration than in the rest of the country. Fears of voter fraud are sometimes cited as a reason against allowing election-day registration. But all of the research shows that voter fraud is extremely rare.
Perhaps just as interesting as Street’s conclusion is how he reached it. To estimate the number of additional voters who might have participated in the 2012 election, he used the number of Google searches for the phrase “voter registration” in the period leading up to election day 2012.
To estimate the relationship between searching online, and actually registering, we turned to state records of registered voters. The data confirm that, in the period leading up to voter registration deadlines, the daily number of Google searches in each state was closely related to the daily number who registered. If the same pattern had been allowed to continue up to Election Day, millions more Americans would have registered in time to vote.
Technological advances make it possible effectively to extend voter registration deadlines to election day, and technology, in the form of Google, helps us see that doing so would likely improve voter turnout.