Why the Polls Don’t Add Up
by Andrew Kohut, Director
Special to the New York Times
You can hardly blame the Democrats if they seem a bit confused. After all, as the situation in Iraq has worsened over the past six weeks and national polls have shown a steep decline in President Bush’s job-approval ratings (some, including the latest CBS/New York Times survey, have him registering well below the 50 percent mark), John Kerry can’t seem to pull ahead of the president the national horse-race polls.
Last week’s Gallup, Fox News and NBC/Wall Street Journal surveys — all taken well after the revelations of prisoner abuse at Abu Ghraib — continued to show registered voters split about evenly between the president and the senator. New surveys by CNN/USA Today/Gallup and by my colleagues at the Pew Center did show the senator gaining a small lead, but that edge disappeared in the Gallup poll when the sampling was narrowed from registered voters to ”likely” voters, and in the Pew poll when respondents were asked to also consider the candidacy of Ralph Nader.
Understandably, many Democrats have begun to despair — if Mr. Kerry can’t gain ground when the president is in trouble, when can he? His defenders suggest that the evenly divided, highly polarized electorate is so dug in that neither candidate can break away. Others attribute Mr. Kerry’s lack of progress to the multimillion-dollar Bush advertising blitz in swing states.
These explanations may have some merit, but the data show there is still a sizable independent swing vote that could drive the election one way or the other. And the declines in the senator’s favorable ratings have been modest — even in the swing states, where the Bush-Cheney advertising hit him hardest, polls show that most voters still hold positive or neutral views of him.
The real reason that Mr. Kerry is making so little progress is that voters are now focused almost exclusively on the president. This is typical: as an election approaches, voters first decide whether the incumbent deserves re-election; only later do they think about whether it is worth taking a chance on the challenger. There is no reason to expect a one-to-one relationship between public disaffection with the incumbent and an immediate surge in public support for his challenger.
We saw the same dynamic in the 1980 race. President Jimmy Carter’s favorable rating in the Gallup surveys sank from 56 percent in January to 38 percent in June, yet he still led Ronald Reagan in Gallup’s horse-race measures. For much of the rest of the campaign, voters who disapproved of Mr. Carter couldn’t decide whether Mr. Reagan was an acceptable alternative.
Through the summer and early fall, the lead changed back and forth, and CBS/New York Times and Gallup polls showed conflicting results — at one point in August, Gallup found Mr. Reagan ahead of President Carter by 16 percentage points, yet just two weeks later it registered a dead heat. It was not until the two men held a televised debate eight days before the election that Ronald Reagan gained legitimacy in the eyes of the electorate.
Similarly, in May 1992 President George H. W. Bush had only a 37 percent approval rating according to a Times Mirror Center survey, but the same poll showed him with a modest lead, 46 percent to 43 percent, over Bill Clinton.
Only the Democratic convention and the debates brought about an acceptance of Mr. Clinton (even though his negative ratings were higher than Mr. Kerry’s are now). It took a long time for him to be seen as an acceptable alternative to Mr. Bush.
Should the voters’ disillusionment with the current President Bush continue, they will evaluate John Kerry and decide whether he is worth a chance. But, as in the past, the focus at this stage is on the man in the White House — and given the events in Iraq, it is unlikely to come off him any time soon. Mr. Kerry’s lack of progress should not, for now, be cause for concern to Democrats. Public opinion about Mr. Bush is the far more important barometer — and if it remains low, Mr. Kerry will have a chance to make his case.