Choosing candidates by pop political quiz
Online surveys match voters with candidates who best share their views. Is this good for democracy?
SAN FRANCISCO — High school senior Kevin Donnelly knew something was wrong when the results popped up on his computer screen: Bill Bradley was his "choice" for president.
Mr. Donnelly had done his homework and was already leaning toward either Ralph Nader or Socialist Party USA candidate David McReynolds. But the online political quiz he took at the GoVote Web site - designed to tell which candidate most reflects a person's views - paired him with Mr. Bradley, a Democrat.
Donnelly was puzzled until he discovered why. The online quiz's field of selections didn't include third-party candidates. "It was a very bad match," says Donnelly, who will vote for the first time this year.
Online surveys meant to match voters and candidates are proliferating this election season. And they offer, say a range of analysts, a microcosm of the good and bad of what the Internet Age is bringing to American political life.
As Charles Lewis of the Center for Public Integrity put it earlier this year, the Internet can be a "significant and positive avenue for accountability and information that promotes civic awareness and participation." But only, he added, if "used appropriately."
Politics can be ... fun?
Many political analysts give these new online surveys high marks for engaging potential voters in a new way. In short, these voter quizzes add a dash of fun to the reams of political history and data now widely available on the Web.
But because they are fledgling, they are often incomplete. Further, some worry about potential privacy problems when these surveys become large enough to have commercial value to campaigns, consultants, and others in the political process.
Most of these online quizzes ask 20 or so issue-related questions and then rate candidates according to the answers. The results make good water-cooler chatter, given that most voters attach more weight than they realize to personality and character, which are stripped away in an issues quiz.
"People get very different results than they expect when they just focus on the issues," says GoVote chief executive Paul Hrabal. For instance, GoVote's quiz has been taken by about 150,000 users since it was posted in January, and the candidate who has scored first most often is Steve Forbes. Yet in the real world, Mr. Forbes was forced from the race in mid-February for lack of interest.
"For me, it's a quick way to get people past preconceptions," says Curt Anderson, founder of SelectSmart, a Web site based in Ashland, Ore., that helps users make choices on everything from household pets to political candidates.
Mr. Anderson sees value in helping guide people to candidates who share their views, rather than approaching that decision according to "some knee-jerk decision based on what party you're in or what party your parents were in."
Also, says Anderson, "we put it together in a way to make it kind of fun." Indeed, he says, a common reaction is surprise when users discover matches they did not expect.
SelectSmart's candidate selector quiz has been licensed to a number of other Web sites, including Talk City, SpeakOut, and the online magazine Salon. AOL has its own quiz called PresidentMatch.
Online quizzes are logging hundreds of thousands of visitors, ranging from political junkies to novice voters seeking earnest guidance. While some users like Kevin Donnelly are surely disappointed, sponsors of the quizzes say that happy customers dominate and that the surveys are not meant to be the final answer on political choices. Just a guide.
Indeed, SelectSmart's Web site informs visitors that "this selector is an exploration of the issues only. Make your informed vote after you judge their honesty, leadership, electability, and other qualities."
Tool for voters - or politicos?
However, the issue that concerns some political analysts is not the accuracy of online survey results, but the privacy of the user. Will commercial Web sites offering the surveys build databases and sell them to parties, candidates, or others with an interest in pinpoint voter marketing?
None of the prominent selector quizzes now requires a user to register or disclose personal information. But GoVote, for instance, offers users the option of registering. And while Mr. Hrabal says results are kept only in aggregate, even that worries some.
Democracy Network, based in Los Angeles at the Center for Governmental Studies, offered a candidate-selector service in 1996 but dropped the program for the 1998 election cycle.
"I have mixed feelings" about such online quizzes, says Tracy Westen, founder of Democracy Network, which was recently sold to the Grassroots Web site. His concern is that ultimately online quizzes could create profiles of a particular locale or zip code. That could be sold to marketers, and the result is a group of people being spoon-fed only what they want to hear, based on the data accumulated.
Hrabal says GoVote has no such plans, but he doesn't see a problem with certain uses of aggregated data, so long as users know up front through disclosure on the Web site.
Online quizzes are expected to spread because they are proving a real draw to Web sites on politics. GoVote.com, for instance, has more than 300 pages of political information, but Hrabal says "most people don't want to wade through all that."
At least not until they've been drawn into the topic of politics by the quiz.
(c) Copyright 2000. The Christian Science Publishing Society