Thursday, October 23, 2008

What Landslide?

For all the talk about how far ahead Barack Obama is supposed to be in the polls, now comes the first significant poll since the last presidential debate from the Associated Press. What does it tell us, Obama in a landslide? Hardly:

The contest is still volatile, and the split among voters is apparent less than two weeks before Election Day.

"I trust McCain more, and I do feel that he has more experience in government than Obama. I don't think Obama has been around long enough," said Angela Decker, 44, of La Porte, Ind.

McCain has posted big gains among likely voters earning under $50,000 a year; he now trails Obama by just 4 percentage points compared with 26 earlier. He has surged among rural voters; he has an 18-point advantage, up from 4.

When rural and suburban America vote in unity, we've shown in 2000 and 2004 that we can effectively veto the large Democratic turnout in big, high-tax, welfare state cities. If John McCain continues to gain among rural voters and those earning under $50,000, what you end up with is a very close election-one that John McCain becomes increasingly likely to win. If McCain does end up pulling off what could be the biggest upset in American electoral history, he will have the Republican base to thank for it. Rural and small-town America will have turned out in droves for the Republican ticket.

In addition to new polls showing the race is within the margin of error, Michael Barone in The Wall Street Journal questions whether such a wide and varied assortment of polls can be trusted in the first place.

To start with, political polling is inherently imperfect. Academic pollsters say that to get a really random sample, you should go back to a designated respondent in a specific household time and again until you get a response. But political pollsters who must report results overnight have to take the respondents they can reach. So they weight the results of respondents in different groups to get a sample that approximates the whole population they're sampling.

Another problem is the increasing number of cell phone-only households. Gallup and Pew have polled such households, and found their candidate preferences aren't much different from those with landlines; and some pollsters have included cell-phone numbers in their samples. A third problem is that an increasing number of Americans refuse to be polled. We can't know for sure if they're different in some pertinent respects from those who are willing to answer questions.

Yet there was a curious anomaly: In most primaries Mr. Obama tended to receive higher percentages in exit polls than he did from the voters. What accounts for this discrepancy?

For the record, Michael Barone still believes that Barack Obama has a slight advantage, as do I. However, as in 1948, opinion can change in the final week of the campaign, and as we saw in 2004, exit polls are no longer an accurate barometer because an increasing number of people do not wish to share their voting preferences with pollsters no matter what those happen to be.

Those of you in the party opposite who were hoping for a quick election night victory may be in for another very long evening.



At Thursday, October 23, 2008 11:39:00 PM, Anonymous Anonymous said...

I'm starting to feel really bad for you. One outlying poll and you're off to the races. What's more likely? Every single other poll is wrong or that this one happened to get some skewed results? The good news is that there are only 11 more days for you to continue the descent into madness until you are forced to face reality...

At Sunday, October 26, 2008 2:03:00 PM, Anonymous Anonymous said...

I know there is a temptation on the part of some to believe polls that offer favorable results, and ignore polls that don't. However, it's not an intellectually honest way of measuring a campaign, despite the fact that it's not uncommon. It is also a fact that I didn't think you would be one of those that did.
You think the AP poll is a very significant campaign development? You really think so?
Consider this reason for skepticism: 44% of those in
the poll's voter sample were self-identified evangelical Christians, who tend to be conservative and vote Republican.
However, in the last presidential race, evangelicals constituted only 23% of voters -- about half
the number used in this AP poll!
Now, I suppose, that evangelical turnout might be higher this year than in 2004, but a jump from
23% to 44%?!?
That's practically impossible, and rather stupid to assume as part of the basis for a national campaign poll!
Com'on David bong down; Can you say "polling bias"? Com'on, I know you can!
You're making a big, big mistake getting too excited based on this Transparently crummy AP poll.



Post a Comment

<< Home

Locations of visitors to this page
Profile Visitor Map - Click to view visits
Create your own visitor map