New Hampshire Polls: What The Press Missed
With thirty-six hours to go before the New Hampshire primary, the press had already set the story of an Obama win and a Clinton epitaph in stone. Barack Obama was the anticipated victor and was poised for a major momentum boost in the race for the nomination. The Clinton campaign, on the other hand, was said to be staring into the face of imminent campaign shake-ups, as pundits began writing Hillary Clinton's political epitaph in advance. This was all based on polls that showed a sizable Obama lead as the voting commenced. And every single bit of that was wrong.
In the wake of the New Hampshire primary, the press is busy asking questions about how the numbers could have been so off. All sorts of contentions have been offered: Hillary's tears may have goosed female votes in her favor, McCain may have siphoned away too many of Obama's independents, the Iowa bounce never materialized. One thing everyone could agree on was that the poll numbers led the press to arrive at a premature consensus.
That's what's so ironic. Because the hidden story of New Hampshire is that even as the press began arriving at that consensus, the Democratic electorate was nowhere near doing the same thing.
As the last polls came in the Sunday before the primary, one important number failed to register: 47 percent. That's the percentage of the Democratic electorate in New Hampshire that hadn't made up its mind on who to vote for. Jennifer Donahue of the New Hampshire Institute of Politics at Saint Anselm's College, told the Huffington Post, "I always look for that number first because historically, New Hampshire voters make up their minds at the last minute." On Sunday night, that was the number that stood out in the CNN/WMUR poll. That large number of Democratic undecideds so close to the vote was unprecedented. And it was no outlier: the pollster at CNN/WMUR said that number was consistent throughout all the polls being taken.
"That, right there, is cautionary," Donahue says, adding that the news should have been, "Half the democratic electorate still undecided...and that would be the headline until that number hits 30 percent."
The pollsters are partly to blame, Donahue relates, because they didn't offer enough guidance on how to assess the results. The polls should have led, in "bold, red letters" with a headline noting that the electorate had largely not made up their minds and that the contest was still a dead heat in spite of Obama's lead in the horse race figures. Still, the press missed this, despite compelling reasons (the shortened schedule, the New Hampshire tradition of voters who make up their minds at the last minute) to be on the lookout. Had they done so, the story on Monday morning might have been that Obama's lead was largely illusory, and the race, too close to call.
That's not what happened. USA Today led off its coverage on Saturday speaking of the Democratic race as a "dead heat." By Monday morning, the story was an Obama "leap." Reuters said on Monday that Obama had "rocketed" past Clinton, and quoted John Zogby calling it "a breathtaking movement in Obama's direction...a surge for Obama and movement away from Clinton." The Washington Post called Clinton an "underdog" after the Saturday debate, and on Monday morning, citing the very poll Donahue found to be "cautionary," said that Obama had opened "a significant lead in the state, suggesting a major bounce in support following his win in the Iowa caucuses."
None of the above articles mention the fact that 47 percent of the Democratic voters had yet to make up their minds.
The New York Times seemed to arrive at the premature consensus prematurely. As early as January 5, they were already writing about "career-threatening scrapes" and the need for a "retooling" on the Clinton side. By Monday morning, the Times was saying the same thing about the overnight polls as everyone else. Interestingly, however, a Sunday story from Jeff Zeleny and David D. Kirkpatrick, noted that Hillary Clinton had begun a "door-to-door effort to appeal to undecided voters." Had they noted the 47 percent figure, they might have been able to point out what a shrewd strategy this was -- right before the primary, hours before opinions began to firm, Clinton was hitting the most advantageous bloc of Democratic voters that were available to her, while Obama was fighting McCain over the same group of "undeclared" independents.
So right about now you're asking yourself, "All right, so the press ignored a key finding in the polls and captured a dead heat as a horse race that had Obama several lengths ahead. So what? The results are the results." Well, by failing to note the large number of late undecided voters (and, if I might usurp the office of ombudsman for a moment, I'd point out that this institution has no claim of immunity on this score, either), the press had a critical effect on the dynamics of the Democratic race. Obama must contend with the perception that a bubble of momentum has burst in New Hampshire - but there was never a bubble in the first place. And on Clinton's side, an entire day of news stories, documenting internecine shake-ups, dire strategic pivots, and rumors of hirings and firings could have been avoided had the race been correctly captured as anybody's game.
That said, all this polling data was available to the candidates as well. Perhaps on some hidden level, strategic considerations were made. On the surface, however, this doesn't seem to be the case, and the two candidates are stuck with the perceptions that came out of New Hampshire. Given the fact that the Democrats have a pair of compelling frontrunners, this miscue only amplifies the advice given by a stern Tom Brokaw after the primary results were in: "I think that the people out there are going to begin to make judgments about us if we don't begin to temper that temptation to constantly try to get ahead of what the voters are deciding, in many cases, as we learned in New Hampshire when they went into the polling booth today or in the last three days. They were making decisions very late."
Loading comments…
First Posted: 01-10-08 04:10 PM | Updated: 03-28-08 02:45 AM