Advertisement

MEDIA POLITICS : Questions About Accuracy Arise as Dukakis Denounces Media Polls

Share
Times Staff Writer

Take your pick. On NBC this week, the network’s newest poll showed that “Dukakis may be gaining some,” anchorman Tom Brokaw said.

But in polls this week on CBS and ABC, Democratic presidential candidate Michael S. Dukakis was losing ground to Vice President George Bush.

Does NBC’s research really suggest that Dukakis is closing the gap? How should the public reconcile these differing poll results?

Advertisement

And what should Americans make of Dukakis’ newest rhetoric on the stump, implying that polls are somehow influencing the race? “I think people are getting a little tired of being told how they’re going to vote,” Dukakis was heard saying Friday on network news, the second night in a row.

The answers:

No, NBC’s research does not suggest that Dukakis is closing the gap, as much as it does that its previous poll showed Bush with an inflated lead, according to NBC’s pollster.

But while experts say the proliferation of media polls tends to minimize the impact of any one errant survey, they concede that the public may be left confused.

Nonetheless, most dismiss out of hand Dukakis’ new criticism of the polls. Little or no evidence suggests that polls influence how people vote in presidential elections.

The media began conducting polls originally because they were tired of relying on campaigns themselves, which tended to dole out poll results to reporters in deliberately misleading ways. But media polls may have created other problems.

“Media organizations have an interest in hyping and defending these polls,” argued Robert Lichter, director of the Center for Media and Public Affairs. “I think there is a problem when journalistic credibility becomes tied to the corporate product, which is what these polls are.”

Advertisement

“There are times when results raise questions,” said Washington Post pollster Richard Morin, and “I think we strengthen and improve our credibility when we note . . . any questions we have.”

The NBC case may be more a problem of unclear explanation than corporate interest. The network’s poll this week showed Bush up by nine percentage points, 51% to 42%. The NBC-Wall Street Journal poll taken two weeks ago, just after the second presidential debate, showed Bush up 17 points. According to anchorman Brokaw on Thursday, Dukakis had gained eight points. Good news for Dukakis?

Accident of Timing

In fact, not really. According to NBC polling consultant Laurily Epstein, the earlier poll was an accident of timing. In the three days after the second debate, “You had an entire weekend of positive news for Bush,” and the previous poll “seemed to pick up the very peak of that feeling.” Had NBC polled a few days later, Epstein suspects Bush’s lead would have been smaller.

NBC’s new nine-point Bush lead now resembles the eight-point spread in the ABC-Washington Post poll released Wednesday, or the 13-point lead in the CBS-New York Times poll released Tuesday.

But Lichter is particularly troubled by ABC’s handling of an unusual state-by-state poll revealed on the eve of the last presidential debate. The poll interviewed 10,000 people over several weeks and enabled the network to make Electoral College projections.

ABC devoted half of its evening broadcast and all of “Nightline” to the poll that day, even chasing down the campaigns for on-camera reaction before the results were public. Peter Jennings concluded that evening: “George Bush has support in enough different states to win, in a landslide.”

Advertisement

“Nightline” that night pitted the Dukakis campaign not against Bush but against ABC and the methodology of its poll.

“I’m not happy that the poll was put on the air,” Dukakis aide Jack Corrigan said.

“What would you have us do with it?” answered anchorman Ted Koppel, as much in the role of advocate as moderator.

ABC pollster Jeff Alderman, who was challenged by Corrigan on “Nightline,” notes that while Dukakis’ campaign complained, it didn’t present one of it pollsters on the air to rebut the poll.

And he is skeptical of campaigns getting angry at pollsters.

“They can’t get mad at the public for not liking them, so they get mad at the people who found out the public didn’t like them.”

But Dukakis now is angry: “It’s not the pollsters who vote on Election Day,” he argues from the stump, trying to exhort a backlash.

Actually, said polling expert William Schneider and others, “people take presidential races in particular very seriously, and there is no evidence that people look at the polls or that there is much of a bandwagon effect.

Advertisement

“If there were,” said Schneider, a political scientist at the American Enterprise Institute and a consultant to The Times, “then in races where polls predicted landslides from the outset, you would see some drop-off in turnout.” That has not occurred.

Pollsters seem to rally instead to this cry, offered anonymously Friday by one media pollster: “If patriotism is the last bastion of scoundrels, bashing the polls is the last bastion of the trailing candidate.”

Advertisement