In 1980, the television networks caused a bit of a ruckus when they declared, before polls had closed in the West, that Ronald Reagan had won the presidential election.
Soon after that announcement was made, with still more than an hour of voting time left on the West Coast, Jimmy Carter conceded defeat on national television. The scenario was repeated in 1984, when the networks correctly named Ronald Reagan the victor at eight-thirty Eastern Standard Time, voting rush hour in the West.
In Congress, legislators from some western states raised the hue and cry, complaining that the practice of reporting results from the East before the polls had closed in the West disrupted voter turnout in the Pacific Time region. The truth is that the networks did not even have to wait for the polls to close in the East before knowing the results.
Journalists depend on exit polls, accurate to within a few percentage points, for their election numbers. Each network employs thousands of exit-pollsters to sample a fraction of the eighty million or so people who cast votes. The information collected by the pollsters, including party affiliation, candidate voted for, and reason for the vote, is fed into network computers, which prorate the data to arrive at state-by-state figures.
The exit poll was first used in 1967 by CBS in three separate state elections. During the 1972 presidential election, CBS conducted the first-ever national exit polls. And by the 1980 election ABC, NBC, The New York Times, the Los Angeles Times, and the Associated Press had all joined CBS in carrying out exit polls.
The big difference between an exit poll and pre-election tracking polls is that the exit poll measures a fait accompli, a cast vote, which is obviously more accurate than measuring potential votes before an election.
By 1985, the controversy surrounding exit polls, not whether they were accurate, but whether early forecasts deterred voters from turning out by creating in their minds the impression of a foregone conclusion, prompted the networks to sign a pledge to forswear “projections, characterizations, or any use of exit polling data that would indicate the winner of the election in a given state before the polls were closed.”
Some Western states had already taken matters into their own hands: Washington and Wyoming passed legislation prohibiting pollsters from interviewing voters within three hundred feet of a voting place.
None of the opposition to exit polls has prevented the media from conducting them; it merely stanched the flow of information.
Whereas the networks once would have begun releasing figures the minute they became available, usually with the disclaimer, “Remember, this is not a result, just a trend,” before going on to say, “but if the trend continues Hart looks all but unbeatable,” the networks now wait until the polls close before stating what they have known for hours.