Too Early To Obsess Over Horserace Polls?
This column was written by CBS News director of surveys Kathy Frankovic.
The question of which presidential candidate is ahead - the poll question that surveys "the horserace" - is hard to answer sometimes, but especially in the summer, when voters typically have other things to think about. But there has been such an intense focus on the horserace this summer that it has spawned some humorous but true criticisms of (what else?) the intense focus on the horserace.
One comes from Arianna Huffington who suggests a strategy for coping with the myriad of surveys; and that is: focus on just one. According to her Huffington Post blog, "Media insiders now talk about polling brands the way average Joes talk about their favorite beer. 'My producer is a Quinnipiac person, but I'm more into Gallup.' I suppose if they married," Huffington writes, "they'd have to raise the kids CBS/New York Times." For her, there are "freshly baked daily polls," just as there are "freshly baked daily baguettes" in France.
A New Yorker Shouts and Murmers essay by Bruce McCall parodies the subject by posing questions that no one can answer, because they are too complex or make no sense, by polling organizations that parody polling partnerships. As he writes: "A CBS/Pravda/Farmer's Almanac/Avatar: The Last Airbender" poll released today indicates that yesterday never happened for seventy-two per cent of all respondents, but, if it had, thirty-two per cent more Independents believe now than just last May that Barack Obama and John McCain are both leading in a race now too lopsided to call."
McCall and Huffington do, however, make a valid point: This is the wrong time of year to obsess about polls. Especially horserace polls.
The Gallup Daily tracking poll has gone from a nine-point Obama lead to a one point McCain lead, back to Obama ahead by four, and now by three points. Other polls show somewhat different patterns - or none at all. The latest CBS News poll, conducted July 31-August 5, has a 45 percent to 39 percent lead for Barack Obama among registered voters. The previous CBS News/New York Times Poll, conducted three weeks beforehand (July 7-14) was taken before Obama's trip to Iraq, the Middle East and Europe and before the most recent set of advertising and attacks between Obama and presumptive Republican nominee John McCain. Yet that poll had exactly the same results as this week's poll.
In fact, the poll that CBS News conducted just as Obama clinched the Democratic nomination, in early June, showed a very similar result, too. Obama led McCain 48 percent to 42 percent among registered voters, another 6-point lead, although in that poll, the percentage of undecided voters was actually half the size it became in July and early August.
Okay, it's summertime. Even though we have seen record levels of participation and of interest in this election in most polls, voters may be doing other things in July and August. And that's particularly the case for those who fall into the much-discussed categories of "undecided" or "uncommitted" voters - those who are either undecided between Obama and McCain, or who say their minds could still change. It's a big group - 13 percent of registered voters in the last CBS News poll were undecided, and about three in ten of those with a choice said their minds could still change, figures that are much higher than comparable responses four years ago.
These voters are looking for more information: for example, about half of undecided voters say that the choice of a vice presidential candidate will have a great deal of influence on their vote, compared with 30 percent of voters overall.
As part of its latest poll, CBS News re-interviewed 331 people who fell into the "uncommitted" category in the July poll. Most of them remain "uncommitted." And while nearly half of those who had been completely undecided in July now give a candidate's name when they are asked who they will vote for in the fall, two in three of them admit they could change again. And many of them can't articulate a specific reason for their current support.
One reason why few people's positions shifted in the last few weeks is that relatively few of these "uncommitted" voters had been paying close attention recently. When asked how much attention they had been paying to the campaign "in the last few weeks," only 18 percent said they had been paying a lot of attention. And these were not inattentive people: nearly half (45 percent) had said in the first interview that they had been paying "a lot" of attention to the campaign in general. But that question ("How much attention have you been able to pay to the 2008 presidential campaign - a lot, some, not much, or no attention so far?") is about the campaign overall. Voters could have paid a lot of attention during the primaries, but not necessarily in recent weeks.
We re-interviewed only previously "uncommitted" voters. Given that the overall level of non-commitment hasn't changed, there are certainly other voters who have become less committed or completely undecided in the last few weeks.
We need to be careful in what we expect from voters now, before the vice presidential choices are made and the convention addresses are given. We can be funny or serious, but the point is the same: Chill!
By Kathy Frankovic