qphotonyc wrote:
i've also heard that most gi's vote republican so what exactly is your point?

My point is that if you routinely oversample an unrepresentative group, you get a biased result.

Zogby has that reputation. 

Again, his brother is highly active in Democratic politics according to his bio on the Arab American Institue website.

This is hardly an unbiased result.

Clear?

Stephen Rawls wrote:
Zogby has been often criticized for over sampling Democrats ...

but by who exactly, i wonder?
rush limbaugh? bill o'reilly? ann coulter?
they certainly dont have an axe to grind, LOL.

fyi:

Zogby: Selected Client List


A brief sample of our clientele includes:

    * Microsoft Corporation, Cisco Systems, Inc., eToys Inc.

    * Bureau of the Census, USA

    * Philip Morris EEMA

    * MCI Telecommunications, Frontier Cellular

    * Chrysler Corporation

    * United Nations, Office of the General Secretary, (Global Poll, 22 nations), Data Base Marketing (South Korea), American Chamber of Commerce Egypt, Egypt's International Economic Forum

    * Reuters North America and Latin America, New York Post, USA Today, Fox Television Network, Gannett News Service, Campaigns & Elections, Bulletin News Network, Cincinnati Post, St. Louis Post Dispatch, Buffalo News, and daily newspapers in Syracuse, Albany, Rochester, Binghamton, Utica, Watertown, Ithaca, and Poughkeepsie,

    * National Bank and Trust Company, Savings Bank of Utica, Herkimer County Trust, First Source Federal Credit Union

    * Blue Cross and Blue Shield

    * St. Jude Children's Research Hospital, Scranton Medical Center, Cortland Memorial Hospital, St. Lukes Memorial Hospital, Slocum Dickson Medical Center, St. Elizabeth Medical Center, Bassett Healthcare, Northeast Medical Center

    * National Environmental Trust

    * ACCESS of Dearborn. MI,

    * Munson-Williams-Proctor Institute (museum)

    * New World Foundation, North Country Tracking Service

    * New York State Department of Labor

    * American Civil Rights Institute

    * ConMed Corporation, Welch-Allyn (medical manufacturers)

    * Oneida Indian Nation

    * National Association of Manufacturers

    * Labor Research Institute

Stephen Rawls wrote:

My point is that if you routinely oversample an unrepresentative group, you get a biased result.

Zogby has that reputation. 

Again, his brother is highly active in Democratic politics according to his bio on the Arab American Institue website.

This is hardly an unbiased result.

Clear?

No, that's a lie, his polls have no such reputation, that's simply the lie/spin/excuse Republican apologists came up with to explain away the exit polls which actually are part of the proof that the election was rigged. There is zero evidence that he oversampled Democrats or skewed his polls.

Stephen Rawls wrote:
Clear?

I'd have to say 58% of your point is clear, while 42% of it is either somewhat or very unclear, is not understandable at all, or is unclear.

You think they're a biased source, that much is clear.  How they could have miss-sampled in this survey leading to incorrect numbers is unclear.

Vita Brevis wrote:
I'm willing to bet if someone had polled the soldiers in the trenches of WWI or the marines at Iwo Jima or at any other long, bloody battle in history, most would have shown a clear preference for going home over staying in combat.

agreed. these results are not as surprising as some earlier ones showing higher support levels in uniform. but then things have been getting worse & some of these gi's are on their 2nd or 3rd tour.

qphotonyc wrote:
An overwhelming majority of 72% of American troops serving in Iraq think the U.S. should exit the country within the next year, and nearly one in four say the troops should leave immediately, a new Le Moyne College/Zogby International survey shows.

The poll, conducted in conjunction with Le Moyne College’s Center for Peace and Global Studies, showed that 29% of the respondents, serving in various branches of the armed forces, said the U.S. should leave Iraq “immediately,â€? while another 22% said they should leave in the next six months. Another 21% said troops should be out between six and 12 months, while 23% said they should stay “as long as they are needed.â€?

Different branches had quite different sentiments on the question, the poll shows. While 89% of reserves and 82% of those in the National Guard said the U.S. should leave Iraq within a year, 58% of Marines think so. Seven in ten of those in the regular Army thought the U.S. should leave Iraq in the next year. Moreover, about three-quarters of those in National Guard and Reserve units favor withdrawal within six months, just 15% of Marines felt that way. About half of those in the regular Army favored withdrawal from Iraq in the next six months.

The troops have drawn different conclusions about fellow citizens back home. Asked why they think some Americans favor rapid U.S. troop withdrawal from Iraq, 37% of troops serving there said those Americans are unpatriotic, while 20% believe people back home don’t believe a continued occupation will work. Another 16% said they believe those favoring a quick withdrawal do so because they oppose the use of the military in a pre-emptive war, while 15% said they do not believe those Americans understand the need for the U.S. troops in Iraq.

The wide-ranging poll also shows that 58% of those serving in country say the U.S. mission in Iraq is clear in their minds, while 42% said it is either somewhat or very unclear to them, that they have no understanding of it at all, or are unsure. While 85% said the U.S. mission is mainly “to retaliate for Saddam’s role in the 9-11 attacks,â€? 77% said they also believe the main or a major reason for the war was “to stop Saddam from protecting al Qaeda in Iraq.â€?

http://zogby.com/news/ReadNews.dbm?ID=1075

News Flash:  "Soldiers Want to Go Home!"  News at 11. 

Now that I have been a smart ass!  This is actually a very interesting article if for no other reason than looking at the disparity between Marines and Soldiers. 

PS...if you want to see typical POLL bias (not just Zogby) just look at how they group numbers together when they report it.

Example: "The wide-ranging poll also shows that 58% of those serving in country say the U.S. mission in Iraq is clear in their minds, while 42% said it is either somewhat or very unclear to them, that they have no understanding of it at all, or are unsure."

How many did they have to group together to get to 42%?  4?  Paired against one question showing 58%.  Not to mention the problems from asking one pro question against 4 con questions.  You will get uneven results just from test/poll fatigue.  Polls are very rigged.  It is not difficult to manipulate results or to manipulate answers via the poll/test.

EDIT:  The only pollsters I trust are the guys at the Pew Reasearch Center.

qphotonyc wrote:
An overwhelming majority of 72% of American troops serving in Iraq think the U.S. should exit the country within the next year, and nearly one in four say the troops should leave immediately, a new Le Moyne College/Zogby International survey shows.

The poll, conducted in conjunction with Le Moyne College’s Center for Peace and Global Studies, showed that 29% of the respondents, serving in various branches of the armed forces, said the U.S. should leave Iraq “immediately,â€? while another 22% said they should leave in the next six months. Another 21% said troops should be out between six and 12 months, while 23% said they should stay “as long as they are needed.â€?

Different branches had quite different sentiments on the question, the poll shows. While 89% of reserves and 82% of those in the National Guard said the U.S. should leave Iraq within a year, 58% of Marines think so. Seven in ten of those in the regular Army thought the U.S. should leave Iraq in the next year. Moreover, about three-quarters of those in National Guard and Reserve units favor withdrawal within six months, just 15% of Marines felt that way. About half of those in the regular Army favored withdrawal from Iraq in the next six months.

The troops have drawn different conclusions about fellow citizens back home. Asked why they think some Americans favor rapid U.S. troop withdrawal from Iraq, 37% of troops serving there said those Americans are unpatriotic, while 20% believe people back home don’t believe a continued occupation will work. Another 16% said they believe those favoring a quick withdrawal do so because they oppose the use of the military in a pre-emptive war, while 15% said they do not believe those Americans understand the need for the U.S. troops in Iraq.

The wide-ranging poll also shows that 58% of those serving in country say the U.S. mission in Iraq is clear in their minds, while 42% said it is either somewhat or very unclear to them, that they have no understanding of it at all, or are unsure. While 85% said the U.S. mission is mainly “to retaliate for Saddam’s role in the 9-11 attacks,â€? 77% said they also believe the main or a major reason for the war was “to stop Saddam from protecting al Qaeda in Iraq.â€?

http://zogby.com/news/ReadNews.dbm?ID=1075

It they're using Random Sampling, that would control for the over or under representation of either parties.  The key is to get a representative sample of the soldiers in Iraq.  If there are more Republicans than Democrats, for example, you would want more Republications in your survey (Random Sampling attempts to adjust your sample so that the sample closing representations all of the population you're attempting to study).

Seems like they're doing something right. Pretty impressive list of clients:

-Microsoft Corporation, Cisco Systems, Inc., eToys Inc.

-Bureau of the Census, USA

-Philip Morris EEMA

-MCI Telecommunications, Frontier Cellular

-Chrysler Corporation

-United Nations, Office of the General Secretary, (Global Poll, 22 nations), Data Base Marketing (South Korea), American Chamber of Commerce Egypt, Egypt's International Economic Forum

-Reuters North America and Latin America, New York Post, USA Today, Fox Television Network, Gannett News Service, Campaigns & Elections, Bulletin News Network, Cincinnati Post, St. Louis Post Dispatch, Buffalo News, and daily newspapers in Syracuse, Albany, Rochester, Binghamton, Utica, Watertown, Ithaca, and Poughkeepsie,

-National Bank and Trust Company, Savings Bank of Utica, Herkimer County Trust, First Source Federal Credit Union

-Blue Cross and Blue Shield

-St. Jude Children's Research Hospital, Scranton Medical Center, Cortland Memorial Hospital, St. Lukes Memorial Hospital, Slocum Dickson Medical Center, St. Elizabeth Medical Center, Bassett Healthcare, Northeast Medical Center

-National Environmental Trust

-ACCESS of Dearborn. MI,

-Munson-Williams-Proctor Institute (museum)

-New World Foundation, North Country Tracking Service

-New York State Department of Labor

-American Civil Rights Institute

-ConMed Corporation, Welch-Allyn (medical manufacturers)

-Oneida Indian Nation

-National Association of Manufacturers

-Labor Research Institute

Stephen Rawls wrote:

My point is that if you routinely oversample an unrepresentative group, you get a biased result.

Zogby has that reputation. 

Again, his brother is highly active in Democratic politics according to his bio on the Arab American Institue website.

This is hardly an unbiased result.

Clear?

Then according to your logic, if I routinely sample republican GIs, then the numbers quotes in that article would be even worse for Bush.  As it is, if they are over sampling Republicans, then Democrats are being underrepresented.

Who care who his brother is.  My brother has no influence over me in any way, shape, or form.

/tim

Aaron_H wrote:

No, that's a lie, his polls have no such reputation, that's simply the lie/spin/excuse Republican apologists came up with to explain away the exit polls which actually are part of the proof that the election was rigged. There is zero evidence that he oversampled Democrats or skewed his polls.

Here's a quote from an article John Zogby wrote, "Myth 1 -- Polls predict the winners and losers. Actually, a poll is only a snapshot of a moment in time. It can point to trends, but things can change on election day, when a lot of undecided voters make up their minds. We do try to ask "projective questions" – i.e. to see how people will react to situations and messages, but a poll can only measure a fixed moment in time.
Myth 2 -- Polls determine the outcome of an election (therefore, why bother to vote?). Polls generally only confirm what professional observers (and many voters themselves) already know – whether a race is close or not. Early polls can have an effect on a candidate’s ability to raise funds, but they do not shape how an election will turn.

Source: http://www.zogby.com/about/article.cfm

Tim Baker wrote:

Then according to your logic, if I routinely sample republican GIs, then the numbers quotes in that article would be even worse for Bush.  As it is, if they are over sampling Republicans, then Democrats are being underrepresented.

Who care who his brother is.  My brother has no influence over me in any way, shape, or form.

/tim

I am with Tim.

You can attack polls and pollsters without worrying who their relatives are.

Aaron_H wrote:
No, that's a lie, his polls have no such reputation, that's simply the lie/spin/excuse Republican apologists came up with to explain away the exit polls which actually are part of the proof that the election was rigged. There is zero evidence that he oversampled Democrats or skewed his polls.

Actually, referencing Zogby's poll credibility, this very topic was the subject of today's Day by Day cartoon.

CLEARLY, Zogby DOES have 'such reputation'.

http://www.daybydaycartoon.com/?CartoonDate=03-02-2006

Stephen Rawls wrote:

Actually, referencing Zogby's poll credibility, this very topic was the subject of today's Day by Day cartoon.

CLEARLY, Zogby DOES have 'such reputation'.

http://www.daybydaycartoon.com/?CartoonDate=03-02-2006

LOL.
clearly you're joking... i hope.

BCG wrote:
i would add that 10 out of 7 pollsters can fuck with numbers enough to show you what they want to tell you.

Unless the pollster comes to a conclusion that agrees with your opinions ... then it's gospil.

Stephen Rawls wrote:
Actually, I'm not. Zogby's polls have been and are currently a joke. It shouldn't come as a surprise that they are parodied.

Quite revealing was Zogby's admission on the Hugh Hewitt radio show that the poll was commisioned by an anti-war patron:

HH: Good. John, your poll has created quite a lot of controversy out of Iraq. Who paid for it?

JZ: It was a gentleman who is a very wealthy individual who is anti-war, but he had absolutely no influence on any of the questions at all.

In fact the entire interview was quite revealing, until Zogby got defensive and hung up. You can read the transcript or listen at

http://www.radioblogger.com

A more complete Fisking of Zogby's polling methods in this particular poll can be found here:

http://www.mysterypollster.com/main/200 … pol_2.html

Interesting here that Zogby admits "The survey did not involve a "random probability" sample of all American troops serving in Iraq."

Now since you seem to have difficulty in understanding my points, let me list them.

1) This is a pollster who has been wildly inaccurate in the past... Kerry in a sweep, for example in 2004.

2) This is a pollster who's brother has frequently been a Democratic and Palestinian acitivist, according to his own bio, therefore creating questions of bias on this subject.

3) Zogby has a past reputation for oversampling liberal Democrats and passing the information off as a representive sample. Enough so that he's parodied in comics.

4) The poll was paid for by an anti-war patron.

5) Zogby is defensive enough about the methodolgy of the poll that he refused to answer any questions about it and then hung up on a radio interviewer.

Again, my advice to you is before you post check your sources.

so far you're the only one trying to debunk the poll and you've dragged in his brother , the person who commissioned the poll, a cartoon and a conservative talk show host. should we likewise discard any polls conducted by the rightwing wall street journal and invoke doonesbury comic strip references to make a leftist point? the closest you've come to making a serious point is the mystery pollster who states:

John Zogby insists it is enough that those of us who have heard more about his survey's methodology conclude that it was "honestly and objectively done."  I think he misses an important point. Consumers of Zogby's Iraq troop poll data also need to understand where it fits on the continuum between strict probability-based sampling and non-random convenience sampling.  Zogby certainly believes that "security concerns" prevent further disclosure, that we do not "need to know" more.  Perhaps.  But without knowing more, it is hard to decide whether to trust the results.

so for him "it is hard to decide whether to trust the results."
big deal. that hardly makes the poll worthless or unworthy of posting here.

Stephen Rawls wrote:
Actually, I'm not. Zogby's polls have been and are currently a joke. It shouldn't come as a surprise that they are parodied.

Quite revealing was Zogby's admission on the Hugh Hewitt radio show that the poll was commisioned by an anti-war patron:

HH: Good. John, your poll has created quite a lot of controversy out of Iraq. Who paid for it?

JZ: It was a gentleman who is a very wealthy individual who is anti-war, but he had absolutely no influence on any of the questions at all.

In fact the entire interview was quite revealing, until Zogby got defensive and hung up. You can read the transcript or listen at

http://www.radioblogger.com

A more complete Fisking of Zogby's polling methods in this particular poll can be found here:

http://www.mysterypollster.com/main/200 … pol_2.html

Interesting here that Zogby admits "The survey did not involve a "random probability" sample of all American troops serving in Iraq."

Now since you seem to have difficulty in understanding my points, let me list them.

1) This is a pollster who has been wildly inaccurate in the past... Kerry in a sweep, for example in 2004.

2) This is a pollster who's brother has frequently been a Democratic and Palestinian acitivist, according to his own bio, therefore creating questions of bias on this subject.

3) Zogby has a past reputation for oversampling liberal Democrats and passing the information off as a representive sample. Enough so that he's parodied in comics.

4) The poll was paid for by an anti-war patron.

5) Zogby is defensive enough about the methodolgy of the poll that he refused to answer any questions about it and then hung up on a radio interviewer.

Again, my advice to you is before you post check your sources.

new fox poll:
http://www.foxnews.com/story/0,2933,186634,00.html

using your logic, w's numbers are really much worse since fox is notoriously pro bush.

Terry Breedlove wrote:
Aaron I think your having paranoid delusions again. Did you forget your medication.

No, the guy I responded to is having them though and so are you if you agree with him. I wish there was a medication to protect people like you from propaganda and brainwashing though, so sad and disgusting how easily it's done to get people to support people and policies diametrically opposed to their own best interests and opposed to simply being human.

Tim Baker wrote:

Here's a quote from an article John Zogby wrote, "Myth 1 -- Polls predict the winners and losers. Actually, a poll is only a snapshot of a moment in time. It can point to trends, but things can change on election day, when a lot of undecided voters make up their minds. We do try to ask "projective questions" – i.e. to see how people will react to situations and messages, but a poll can only measure a fixed moment in time.
Myth 2 -- Polls determine the outcome of an election (therefore, why bother to vote?). Polls generally only confirm what professional observers (and many voters themselves) already know – whether a race is close or not. Early polls can have an effect on a candidate’s ability to raise funds, but they do not shape how an election will turn.

Source: http://www.zogby.com/about/article.cfm

For one thing I was talking about exit polls and your quotes are addressing pre-election polls. But his last sentence doesn't make much sense, campaign funding is generally one of, if not the, biggest factor in an election. So if polls affect ability to raise money then they are in fact helping shape elections. But all that is beside the point here.