How Reliable Is the Zogby-Journal Poll?

“The poll’s Web site describes the process as ‘interactive’ — that is, it’s an Internet poll, based on a database of individuals who have signed up to participate,” writes Burka. “It is not a random sample; the polling organization solicits responses by e-mail. In addition, the poll takers make about 20 to 50 phone calls in the state where a race is taking place. The poll does not mention a screen for likely voters.”


“As I have written repeatedly, the poll that I think is the least credible is Zogby/Wall Street Journal/Battleground States,” adds Burka. “I can’t believe the Journal allows its name to be attached to this so-called poll.”

http://www.cjr.org/politics/how_reliable_is_the_zogbyjourn.php

The Columbia Journalism Review is the gold standard for journalism...


Please discuss...

in my email but haven't had any in ages. I don't know why I got deleted from their list. Maybe I answered "poor" too many times when asked about Bush's performance. :shrug:

Mz Pip
:dem:

My copy of their "take this survey" email shows up there regularly.

Keeping it handy as we will surely need it again,,,, and again and again.

If the pre-teens are clued in, why are voting age adults still clueless?

See here, for example:

http://www.democraticunderground.com/discuss/duboard.php?az=show_mesg&forum=132&topic_id=3758205&mesg_id=3758205

but you're complaining because someone else has pointed out what's wrong with the Zogby poll? Okay, I see how it goes.

because one poll is subsequent to the other poll and there is a downward trend.

that's absurd. She's the front runner. I don't particularly like it, but it's true. Yes, there's been some slide in her poll numbers lately, but outside of Iowa, she's 15+ points up in every state, SC where she's 10 points up. And the point about interactive polling is one whe should all remember.

Respectfully, have you ever taken any social science research courses or statistics classes?


It has nothing to do with what I like or don't like... If I am slavishly devoted to anything that anything is science...That's why I'm not a Fundy...

Also, the net is richer and more educated than the rest of America... Plus it has much less people of color and older folks using it on a regular basis....

like Rasmussen and SurveyUSA. It's irrelevant as a point of distinction because both internet based polls and telephone based polls have a known selection bias (internet based polls have a sampling error because the universe of people who participate is not random or perfectly representative and phone base polls have sampling errors related to the fact that many demographics don't have land lines and many demographics are considerably less willing to participate in phone polls, e.g., men are less than half as willing as woman in some racial sub-demographic groupings and minorities are generally less willing to participate in such polls as non-Hispanic whites). Both poll types require sampling correction and normalization. Both internet-based polling and robo-call polls (and manual phone call polls, too, which also introduce additional poll questioner bias issues) do not random sample but correct for the sample bias.

here at the "D"U...it got the "true" progressives foaming...

Double-assed standard around here sometimes. I can't wait until we have the damn nominee already. It might make it a bit more polite around here again.

anyway you want to twist it to fit what you need. polls are crap.

Watching poll-deniers twist themselves into pretzels trying to defend this one as accurate...

Zogby interactive polls may be informative, but they're not scientific, regardless of who the candidate is.

over the years, I have to say, internet polls mean nothing to me. :)

It's not an open poll where it can be "DUed."

That makes it much better, but I think it's still likely to create biased results.

:shrug:

I think it's interesting to see the various results, but I don't live my life by them, you know what I mean? People go up, people go down, and really the only poll that matters is on election day. :)

:)

Here's the link:

http://www.realclearpolitics.com/polls/

In SurveyUSA Election November poll of Iowans sponsored by KAAL-TV Rochester-Mason City-Austin, Barack Obama will fare significantly better than
Hillary Clinton in the general election. Obama easily beats all of his Republican opponents (with 50% or more against each of them), while Clinton's
margin is significantly slimmer against Giuliani, Romney, and Huckabee, and she even loses to McCain.

Link to poll: http://www.surveyusa.com/client/PollReportEmail.aspx?g=2d211d2a-ee13-40ab-8455-5a70b9b55bfe

So this Zogby poll just reinforces the Survey USA's results in a further downward trending. So much for Clinton's electability argument. I can only hope
Iowans make decisions based on electability AND change.

And since you're so fond of Survey USA:


http://www.surveyusa.com/client/PollReportEmail.aspx?g=8d04025f-3b99-41fd-b488-a470924222b1


http://www.surveyusa.com/client/PollReportEmail.aspx?g=09c3361e-ab99-45ed-b520-0b2c6c8942b3


http://www.surveyusa.com/client/PollReportEmail.aspx?g=8684e5c8-ef95-410f-b1d2-e3f623b34ed2


This is a test of (your) intellectual honesty... If you believe that Survey USA's Iowa results are accurate then you must of necessity believe their results from Kentucky, New Mexico, and Ohio are accurate as well... If you don't then I don't have an honest debating partner and any further discussion is futile and will only descend into the trading of personal insults...

It also tells me that comparing national head-to-head match ups with state polls is comparing apples with oranges.
However, trending is an important rough indicator of voter sentiment, especially in the early primary states, which is
why the latest ABC/Washington Post poll of likely Iowa caucus goers (which shows Obama leading Clinton as a first among
equals) and the recent CNN/WMUR NH poll (where Clinton's lead dropped 10 points in a month) is so interesting.

Did you or did you not link an Iowa poll done by Survey USA to indicate Obama does better than Clinton against potential Republican opponents?


And if you did what do you have to say about the three polls I linked that were done by Survey USA during the same time period that show Clinton doing better than Obama in the more Electoral College Vote rich states of New Mexico, Ohio, and Kentucky?

Although I do acknowledge national trending and those in NH in a lesser but still significant light. My opinion - and I could be wrong -
is based on what happened in 2004. I firmly believe that the winner in Iowa will win it all. Time will tell if I am right.

You cited a poll done by Survey USA showing Barack Obama doing better in Iowa than Hillary Clinton against the Republican field...

I then cited three polls by (Survey USA) of roughly the same vintage that showed Hillary Clinton doing better than Barack Obama against the Republican field in the Electoral College Vote rich states of Ohio, Kentucky, and New Mexico...


Either we agree that Survey USA, the pollster (you) chose to cite, is reliable or it isn't...

It is what it is....

Now that I understand your question, the answer is yes with a qualifier -- I don't think the polls anywhere except Iowa matter because if, say, Obama
or Edwards win Iowa - then those polls you cite will be meaningless. I also think the Zogby poll is valid as well, by the way. I do not think any Rasmussen
polls are valid due to the pollster's partnership with Hillary's Karl Rove's close friend and former partner. But you knew that.

Since Edwards is polling at three percent or so among African Americans who make up about three of every ten Democratic primary voters Clinton will beat him fairly easily... Especially when you couple that with Hillary Clinton's huge advantage with women voters...

Now, if Obama wins IA he finishes Edwards and him and Hillary Clinton slug it out through Super Tuesday 2/5 where half the delegates will be chosen on one day...

Even if all of Zogby's techniques couldn't be taught in a university course, that critique seemed irrelevant after he called the election correctly. By February 1998, Zogby had been asked by Rush Limbaugh to do a poll on the Monica Lewinsky scandal, one that, by emphasizing moral questions, would differ from surveys showing widespread support for the president in the face of Kenneth Starr's inquiries. Zogby obliged. The five-question poll opened with the following: "Suppose you are ready to hire a candidate who is well-qualified for the job, but then you find out that they like to have consensual sex with subordinates. Still hire them?"

http://www.prospect.org/cs/articles?article=john_zogbys_creative_polls

Kerry at one point early in 2003 had been the leader in New Hampshire where he was a virtual favorite son due to the extensive media coverage New Hampshire voters had always received of Kerry's career beamed into their tiny state from neighboring Massachusetts. Kerry bled support in New Hampshire when voters there lost faith in his ability to beat Dean for the nomination, and the person who gained most of Kerry's support was Clark.

Iowa was critical in 2004 because Kerry reestablished himself as a winner there after he bet virtually all his remaining chips on that state. Clark did not compete in Iowa so Kerry did not have to split votes with Clark there. That pumped new life into Kerry's campaign and it pumped new confidence in Kerry into many New Hampshire voters who had been predisposed to back Kerry all along until his campaign went off the tracks. Since Kerry won New Hampshire after winning Iowa it can not be logically challenged that Iowa set the tone for the entire national race.

But what if Gephardt had won in Iowa with Kerry a close second in front of Dean? Or what would have happened if Edwards squeaked out a narrow victory in Iowa with Kerry coming in second over Dean? Would Gephardt or Edwards have gone on to poll higher in New Hampshire than Kerry? I think that is not at all certain. Clearly Kerry would not have run away with New Hampshire in the same way, but he still might have won there. And had that happened, the results in Iowa though obviously still significant would not have had the ring of final truth to it that we now look back and bestow on Iowa 2004.

Kerry may or may not have won the nomination if he lost Iowa but won New Hampshire, but had he still won the nomination after losing Iowa and winning New Hampshire, we might today be talking about the ultimate importance of New Hampshire, not Iowa. Unlike in 2004 when both Howard Dean and John Kerry had strong regional advantages in New Hampshire to work with, this time there is no regional favorite son running in New Hampshire. If Obama or Edwards win in Iowa, they do not have a special relationship with New Hampshire's voters to count on the following week. The winner of Iowa may not win New Hampshire. And if that comes to pass, the momentum for the Iowa winner coming out of Iowa will not continue to build unchecked.

The idea that Hillary Clinton can't come back from a loss in Iowa and even NH and SC is fanciful...Here's the primary schedule:


http://en.wikipedia.org/wiki/Democratic_Party_(United_States)_presidential_primaries,_2008


Even in the worst case scenario she goes into Super Tuesday with wins in Nevada, Florida , and Michigan... And even if the delegates from MI and FL aren't seated the media will still cover those primaries and perception still matters, giving her momentum going into Super Tuesday...


Now, if she loses all those races prior to Super Tuesday it's over!!!

I get one like almost every week or so. It's great and I think they are scientific.

Google it...

It's a non-probability sample...It has no basis in science....