There’s a Zogby poll out today that actually shows John McCain and Sarah Palin beating the Obama/Biden ticket by two percentage points:
UTICA, New York - Republican John McCain’s surprise announcement Friday of Alaska Gov. Sarah Palin as his running mate - some 16 hours after Democrat Barack Obama’s historic speech accepting his party’s presidential nomination - has possibly stunted any Obama convention bump, the latest Zogby Interactive flash poll of the race shows.
The interactive online Zogby survey shows that both Obama and McCain have solidified the support among their own parties - Obama won 86% support of Democrats and McCain 89% of Republicans in a two-way head-to-head poll question not including the running mates. When Biden and Palin are added to the mix, Obama’s Democratic support remains at 86%, while McCain’s increases to 92%.
After the McCain “Veep” announcement on Friday, Palin was almost immediately hailed as a strong conservative, and those voters have rallied to the GOP ticket, the survey shows. Republicans gather in St. Paul, Minnesota this week to officially nominate McCain and Palin as their presidential ticket.
Overall, 52% said the selection of Palin as the GOP vice presidential nominee helps the Republican ticket, compared to 29% who said it hurt. Another 10% said it made no difference, while 10% were unsure. Among independent voters, 52% said it helps, while 26% said it would hurt. Among women, 48% said it would help, while 29% said it would hurt the GOP ticket. Among Republicans, the choice was a big hit - as 87% said it would help, and just 3% said it would hurt.
This is inconsistent, of course, with the Gallup and Rasmussen tracking polls which continue to show Obama/Biden in the lead.
So, what’s the story ?
Well, here’s a note about the methodology of the Zogby poll that I think calls it’s findings into question:
Zogby International conducted an online survey of [2020 Likely Voters].
A sampling of Zogby International’s online panel, which is representative of the adult population of the US, was invited to participate. Slight weights were added region, party, age, race, religion, gender, education to more accurately reflect the population. The margin of error is +/- 2.2 percentage points. Margins of error are higher in sub-groups.
And, anyone can sign up to be on the interactive panel that the poll is based on.
Given that, it doesn’t seem that the Zogby Poll is truly a random poll meant to be a representative sample of the voting public. Therefore, I wouldn’t trust it’s conclusions without further evidence.
August 31st, 2008 at 8:39 pm
Zogby’s online polling has always been a bit questionable.
August 31st, 2008 at 8:54 pm
Read it again, Doug: “A sampling of Zogby International’s online panel.” In other words, they didn’t take the first 2020 people who signed on. They used the demographics to determine the sample.
Zogby’s more unpredictable because he doesn’t control for party ID (he’s of the theory that it is a dependent variable, and thus flows with the campaign). There’s a raging debate within pollsters on the wisdom of this. Zogby’s way looked great in 2000; far less so in 2004 . . .
August 31st, 2008 at 11:31 pm
It’s still a largely self-selected poll, which makes its results questionable at best.
I’d also note that, during the primaries, the Democratic primary in particular, Zogby’s polling done from the interactive panel was always an outlier and never consistent with the actual results.
September 1st, 2008 at 12:03 am
Im on Zogby’s list.
The online polling is questionable. I posted about this a few years ago for a paper in a statistics class I was taking.
Zogby also does telephone polling and it is more accurate.
The big story is that according to Rasmussen - Obama was three up before the DNC and was three up now. No bounce to speak of.
September 1st, 2008 at 1:25 pm
At fivethirtyeight.com, a pretty sophisticated ranking system is used to rate the reliability of a couple of dozen polling organizations. The Zogby interactive polls are way down near the bottom of the list (and the Zogby telephone polls aren’t a whole lot better).
There’s also a lot of discussion there about post-convention bounces.
September 1st, 2008 at 4:40 pm
This is a change. Usually its the Gallup poll which is heavily biased towards republicans. If the GOP is losing in Gallup, they’re in big trouble, folks.
September 3rd, 2008 at 2:35 pm
Reading thru this(comments as well) all I see is “unpredictable” “sampling of” “questionable” “never consistent”…i would like to know who DOES take this poll seriously? Fox news , maybe….