Front Page Magazine
Dont Trust the Exit Polls
by Howard Fienberg and Iain Murray
NOT TO BURST YOUR BUBBLE, but you might be up late on November 7 if you are waiting to find out which politician won what position in which precinct.
In fact, you may be awake for days, because pronouncements of victory and defeat derive from exit poll results. For those results, while historically more accurate than regular poll numbers, are actually little more than educated guesses. Pundits love them, and frequently they are right. But not always. These gauges often spring leaks.
Exit polls involve face-to-face polling of voters as they leave the voting booths on election day. For the most part, they are conducted by the Voter News Service (VNS), a consortium of TV networks and the Associated Press, which also sells its data to other news organizations. Traditional telephone polling, when it's not horrendously flawed, nets such low response rates these days (30-40 percent on average) that the simplicity of exit polls is quite refreshing: unlike pre-election polls, there is no difficulty sorting out registered or likely voters, since only confirmed voters are polled. But exit polls rely on a small sample of voting precincts and, like any other survey, can run afoul of non-response or false responses. For example:
* VNS overstated the final Democratic vote in the 1992 presidential race. Many believe that this was due to Perot voters being relatively suspicious and unwilling to cooperate with the exit polls.
* In the 1989 Virginia gubernatorial election, exit polls showed African-American Democratic candidate Douglas Wilder winning by 10 percent. Since he actually won by less than 1 percent, many feel that some white voters, feeling perhaps a bit guilty, lied about voting for Wilder.
* In the Republican primaries in New Hampshire in 1992 and in Arizona in 1996, exit polls overestimated the vote for Pat Buchanan. The most likely reason for the mistakes, which resulted in misleading news coverage? Zealous Buchanan voters were more willing to participate in exit polls. In 1992, exit polls predicted a George Bush win by only a small margin over Buchanan -- Bush actually won by 16 points. In 1996, exit polls put Bob Dole a distant third after Steve Forbes and Buchanan - Dole actually came in a close second to Forbes.
With strong and passionate third-party pushes coming from both Buchanan and Ralph Nader this year, expect their voters to at least tilt the exit polls, if not the ballot box.
Of all the pitfalls in the path of exit polling, the largest is the absentee ballot. In 1993, Washington state first let everyone vote by mail and over half of the state's voters are expected to vote this year by absentee ballot. Voters in Oregon's tight vote this year will vote almost entirely by mail.
In 1996, according to the Associated Press, absentee and early balloting encompassed more than ten percent of the total vote in 18 states, more than 20 percent in nine states, and more than 30 percent in three states. In 1998, the total vote included more than 10 percent in absentee and early ballots in 15 states, more than 20 percent in eight states, more than 30 percent in five states and more than 40 percent in three states.
This year, Washington states expects over half of its votes to come via mail, while voters in Oregon will vote almost entirely by mail.
Empirical evidence of impact on exit polling arose as early as the 1982 California election, when exit polls predicted Democrat Tom Bradley would defeat Republican George Deukmejian for the governorship and that Democrat Jerry Brown would defeat Republican Pete Wilson for U.S. Senate. Both predictions were wrong because so many Republican voters cast absentee ballots. As restrictions on absentee, electronic and mail ballots lessen, exit poll projections inevitably suffer, since many voters are not available at a real voting booth.
Major news media have agreed not to announce exit poll results for each state until its polls have closed, in order to avoid influencing voters. But it is still important for us to remember that the results at which journalists are grinning or grimacing during election coverage are not the real deal.
Historically, exit polls have been more reliable than regular polls and the news media treat them as gospel. But if the poll result is close, anyone who tells you that they know who has won is lying. If you want to know the real result, ignore the polls, pour yourself a cup of strong coffee, and stay up all night - or longer.
Howard Fienberg is research analyst and Iain Murray is senior analyst with the Statistical Assessment Service (STATS), a nonprofit nonpartisan organization dedicated to improving public understanding of scientific and social research.
[see the original at FrontPageMagazine]
return to Howard Fienberg's page