On the eve of the Republican convention, a new national poll suggests the race for the White House remains even.
A CNN/Opinion Research Corp. poll released Sunday night shows the Obama-Biden ticket leading the McCain-Palin ticket by one point, 49 percent to 48 percent, with the statistical margin of error.
The survey was conducted Friday through Sunday, after both the conclusion of the Democratic convention and Sen. John McCain’s selection of Alaska Gov. Sarah Palin as his running mate.
A previous CNN poll, taken just one week earlier, suggested the race between McCain, R-Arizona, and Sen. Barack Obama, D-Illinois, was tied at 47 percent each.
“The convention and particularly Obama’s speech seems to be well-received. And the selection of Sarah Palin as the GOP running mate, also seems to be well-received. So why is the race still a virtual tie? Probably because the two events created equal and opposite bounces assuming that either one created a bounce at all,” says CNN Polling Director Keating Holland.
Question. Is this dead cat bounce to be laid at the feet of Obama/Biden or has it been caused by McCain’s selection of Palin?
On the other hand, Gallop shows a post-convention bounce for Obama and a post-Palin drop for McCain:
Sun 24 Aug: Obama 45% v. McCain 45%
Mon 25 Aug: Obama 45% v. McCain 45%
Tue 26 Aug: Obama 44% v. McCain 46%
Wed 27 Aug: Obama 45% v. McCain 44%
Thu 28 Aug: Obama 48% v. McCain 42%
Fri 29 Aug: Obama 49% v. McCain 41%
Sat 30 Aug: Obama 49% v. McCain 41%
Sun 31 Aug: Obama 48% v. McCain 42%
[url=http://www.gallup.com/tag/Gallup+Daily.aspx]Gallop Daily Poll[/url]
Obama went from a tie before the convention to a statistically significant 6-8 point lead afterwards. (And it doesn’t look like McCain got any signficant bounce from naming Palin.)
Of course the polls at this point in the game are pretty much meaningless. And, in terms of the actual election outcome, broad national polls are even less meaningful. (Remember that Gore won the “national poll” of actual voters in 2000, but lost the all-important Electorial College vote.)
With respect to polls, it is always important to look at the polling sample. All polls are not created equal. A poll of 1,000 people walking down the street is not the same as a poll of 1,000 Americans is not the same as a poll of 1,000 Americans eligible to vote is not the same as a poll of 1,000 registered voters is not the same as a poll of 1,000 likely voters. And that doesn’t even get into the myriad other ways a statistical sample can be compromised. Even the best designed and constructed poll can’t defend against people like me who refuse to participate in polls when asked or who, like me, will occasionally give the opposite of my true intention when asked.
All that said, these poll results are pretty bad for a guy who way widely hyped as “The One” against an ‘old man’ and an ‘inexperienced’ woman. Looks like Obama/Biden could be handed their collective chapeau, derriere and Mackinaw in November.
#2 The Gallop daily tracking poll is a rolling three day average so it is hard to say there is no Palin bump based on the numbers you posted. The Democratic convention, for example, started on Monday August 25 but the bounce didn’t show up until Thursday/Friday August 28/29. I’d be interested to see the numbers from today and tomorrow before I conclude anything about a Palin bump.