What's Hot At APSA - 2000 conference of the American Political Science Association

Washington Monthly, Sept, 2000 by Michael Gerber, Rachel Marcus

Can the Internet get people to vote?

THIS AUGUST 30 THROUGH SEPTEMBER 3, many of the nation's top political scientists will converge on Washington for the annual conference of the American Political Science Association. They will present thousands of papers and participate in hundreds of panels on topics from "German Geisteswissenschaft" to military spending. But there is one topic that kept coming up in conversations with professors participating in the conference: Americans don't seem to vote much anymore. In fact, it's virtually certain that no matter who wins in November, neither Al Gore nor George W. Bush will receive the endorsement of a majority of eligible voters, and the winner might not even get 25 percent of the eligible vote. Fewer people voted in 1998 than in any midterm election since 1942, and participation in presidential elections has fallen from 62.8 percent in 1960 to an all-time low of 48.9 percent in 1996.

How can voter interest and participation be increased? Many reformers look to the twenty-first century's panacea to solve this problem: the Internet. Online technology has already proven its value in politics: 95 percent of gubernatorial candidates and 72 percent of Senate candidates from the two major parties ran Websites in 1998. More recently, Sen. John McCain raised more than $5 million over the Internet in his bid for the presidency. And many unions, schools, and businesses have found online voting to be the most efficient and convenient system for casting ballots.

There has also been one closely studied test of Internet voting in an election for national office. In March, when Arizona's Republican legislature set an inconvenient Democratic primary date, the Democrats decided to hold a private, independently-funded one. This gave them an opportunity to try something unprecedented: an election partly conducted over the Internet. While the primary did not produce any surprises--Gore won easily--it provided a chance to study the Internet's effect on voting.

Frederic Solop, a professor of political science at Northern Arizona University who examined Internet voting in Arizona, found that 48 percent of votes were cast online and that total participation more than doubled the turnout for Arizona's 1992 Democratic primary. Solop also found that, contrary to the predictions of some academics, race had no significant effect on public opinion of online elections.

Skeptics of Internet voting suggest that online elections appeal only to people who would vote anyway. Solop thinks his study shows otherwise. "Young voters put more faith in technology than older people," he found in his public opinion study. "[They] are the ones most excited about Internet voting." This does not come as a surprise, but it should make critics of Internet voting think twice. The youngest voters are the most disillusioned with politics and, presently, the least likely to participate; if they are enthusiastic about the Internet, maybe digital elections will reverse that trend.

But will the participation apparently stimulated by the advent of Interact voting be permanent? Will Internet voters still want to vote this November when the novelty of voting online wears off? Maybe yes. If you hook people to vote once, some political scientists argue, they will keep returning to the polls. A study by two Yale University researchers, Alan Gerber and Don Green, confirms the effect: Results from a 1999 municipal election show that participants stimulated to vote for the first time in 1998 had a 50 percent likelihood of voting again the next year. "Voting may be a habitual behavior," says Gerber.

Face to Face

Gerber and Green are now examining data on the behavior of over 25,000 Americans to determine the effects of three different types of appeals to voters: face-to-face interactions, mailings, and telephone drives. "We wanted to see whether or not these various stimuli had an effect on voter turnout," explains Gerber. In the study of nonpartisan voter drives, they found that the personal approach had the greatest success, resulting in a 10 percent boost in voter turnout. Mailings had a minimal effect--a turnout increase of only 0.5 percent--and phone calls were completely ineffective.

Candidates should look closely at the results of the experiment. "We have observed over the last several decades a changing character in American political campaigns, a movement away from campaign workers who are personally involved with the electorate ... to broadcast media," says Gerber. Door-to-door campaigns, these results indicate, would be more effective than expensive television ads at recruiting new voters. It is also more economically efficient than the more impersonal modes of communication, according to Green. He and Gerber analyzed different methods of appealing to eligible voters by comparing the amount of money spent to the actual number of people persuaded to vote on election day. While the study indicated a cost of $30 per voter for mailings, the price per voter of a door-to-door campaign is less than half that amount. "It's a puzzle as to why face-to-face campaigning is not used more," says Green. He blames the move away from personal campaigning on consultants. "Campaign consultants can run more campaigns with direct mail and media." For a door-to-door campaign, politicians do not need to consult with high-paid experts. A consultant who recommends door-to-door campaigning would potentially be making himself and his colleagues less essential.