Jury still out on effect of 'early calls' on voters

The Democratic national chairman is perturbed. Once again, he argues that television network exit polls and early projections are damaging the US election process.

''They have a great deleterious impact. One of the greatest disservices to our American democracy is a commercially owned network calling our elections off early,'' says Charles Manatt, head of the Democratic National Committee (DNC).

What had him complaining was the outcome of the Kentucky contest for the United States Senate. CBS-TV declared at about 6:05 p.m. that President Reagan had apparently won in Kentucky - even though polls in the western part of the state remained open for another 55 minutes.

''I think that one cost us a US senator,'' Mr. Manatt says. Kentucky's incumbent senator, Walter D. Huddleston, ended up losing the race to a Republican challenger by only 4,714 votes out of more than 1.2 million cast.

In fact, the race was so close for the Senate spot that CBS computers were unable to ''call'' that race until 11:40 p.m., more than four hours after all the Kentucky polls had closed.

Is Manatt right? Did the CBS ''call'' that Mr. Reagan had won in Kentucky keep thousands of potential voters for Senator Huddleston away from the polls?

It's difficult to answer such a question with certainty. But Huddleston's campaign chairman, Ed Logsdon, says he doubts that the networks had anything to do with the Democratic loss. He spells it out: ''There was just one thing that beat us, and it wasn't N-B-C, A-B-C, or C-B-S. It was R-E-A-G-A-N. Without the presidential race, we wouldn't have had a contest here in Kentucky.''

Although there were early network projections that Reagan would carry Kentucky, Mr. Logsdon observes that the networks held off on making any projections about the Senate race itself until well after the polls closed. The first network projection, by ABC, came at 7:31 p.m., 31 minutes after the final polls had closed.

Even so, the complaints continue about network use of exit polls to make early projections. The most voluable grumbling comes from the West (chairman Manatt is from California), where the polls are open later in the day because of the difference in time zones. Western voters often will hear a winner declared in the presidential race before the sun has set.

William C. Adams, a professor of public administration at George Washington University, attempted to take a scientific look at the effect of exit polls last Tuesday night. Using public polling techniques, Dr. Adams's team of researchers contacted 1,256 people in Oregon to measure the impact of TV polls and projections.

Adams's two major findings surprised some critics.

First, he concluded that there was ''no evidence that the early announcement depressed turnout anywhere near the extent that has been popularly believed. Our research was unable to uncover any significant damage in Oregon due to projections.''

Second, those few who were discouraged leaned heavily toward Reagan. Sixty-one percent of them said they would have backed the President if they had bothered to vote. So the effect, even though tiny, appears to have hurt the winner - rather than the loser, as assumed by many critics.

The critics have charged that in a number of cases, early projections ''wildly distorted the outcome'' of elections, Adams notes. In fact, he says, the study indicates the effects are ''negligible'' and ''do not have anywhere near the power the critics have claimed.''

Even so, Adams found that 39 percent of those who didn't vote in the Portland , Ore., area that day heard about Reagan's victory before the polls closed there.

So do projections not matter?

That's harder to answer, says Adams. His study indicates that early projections probably don't really have any impact, unless a race is extraordinarily close - perhaps 200 votes. Careful study might indicate that it really doesn't make any difference in more than 1 race out of every 2,000.

In Idaho, for example, US Rep. George Hansen (R) apparently has lost his reelection bid by 68 votes - 101,099 to 101,031. It doesn't get much closer.

A race like Mr. Hansen's could be affected by an early call in the presidential contest, Adams says. On the other hand, it could just as well be affected by an afternoon rain shower or 10,000 other things. It's impossible to say what made the difference for Hansen.