It’s been nearly two weeks since John McCain accepted the Republican nomination and a full two weeks since Sarah Palin was officially made his running mate and it appears that the polls are starting to stabilize.
Consider the poll released today by The New York Times:
Polls taken after the Republican convention suggested that Mr. McCain had enjoyed a surge of support — particularly among white women after his selection of Gov. Sarah Palin of Alaska as his running mate — but the latest poll indicates “the Palin effect” was, at least so far, a limited burst of interest. The contest appeared to be roughly where it was before the two conventions and before the vice-presidential selections: Mr. Obama had the support of 48 percent of registered voters, compared with 43 percent for Mr. McCain, a difference within the poll’s margin of sampling error, and statistically unchanged from the tally in the last New York Times/CBS News Poll in mid-August.
Other polls show a similar shift. Gallup, Hotline and Reuters/Zogby show Barack Obama with a narrow lead after a couple of weeks when McCain was leading. Rasmussen still shows a narrow McCain lead.
I am not a huge believer in polls — it is too easy to game them and so much depends on the sample, the questions and the sequence of the questions. Plus, as Les Payne and Brooke Gladstone pointed out on Bill Moyers’ Journal last week, the polls tend to oversimplify the electorate.
BROOKE GLADSTONE:Right. Well, and, you know, there’s always going to be a number of, a large section of the public that feels that way. But as you know, if we want to talk about something that’s happening in this campaign that bears heavily on the media, it’s the role of polls. And the fact of the matter is because every poll asks the question “Who would you vote for if the election were today?” instead of “Who are you going to vote for in November?” the number of genuine undecideds is hugely reduced.
Because if there were 30 percent undecided as there may well be even in the electorate today, nobody would be interested in the polls. So they ask this other question, forcing them to present their slight lean as a decision, so, therefore, the undecideds go into the single digits because the question is “Who would you vote for now?” instead of “Who will you be voting for in November?”
There are a lot of people out there that can be affected by this information.
LES PAYNE:I think that media, and I use that term advisedly, too often go to ask the polling question as opposed to doing the reporting. We have to inform our readers first, as opposed to asking them what they think about something we have not told them about. So, to the question of if the election was held today, I mean, the answer is, ‘I would be very surprised because I thought it was in November.’
I would add that I have some questions about how the samples are developed: Are the pollsters taking into account the huge influx of newly registered Democrats? Are they balancing the two parties in their samples?
Thurman Hart, on Blue Jersey, offers some interesting insight into the polling process by putting into black and white numbers that indicate that early polling downplays the extent to which the New Jersey electorate breaks for Democrats. In nearly every statewide election since 2004, the 3-4 percent lead held by the poll leader became a 7-9 point win for the Democrats. That’s a remarkable number that, I think, pretty much encapsulates the flaws with polling.
Hank Kalet 