72% of U.S. Troops want out of Iraq within 1 year

[Balance]

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

 

A Zogby poll released today showed that an overwhelming 72% of our troops in Iraq believe that we should completely withdraw from Iraq within one year.

 

I guess they want to "cut and run" like Kerry and Murtha, eh?

 

I'm just waiting for Rove's spin on this one.

[juddling]

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

 

A Zogby poll released today showed that an overwhelming 72% of our troops in Iraq believe that we should completely withdraw from Iraq within one year.

 

I guess they want to "cut and run" like Kerry and Murtha, eh?

 

I'm just waiting for Rove's spin on this one.

 

 

Sample size: 944

 

Troop level: 140,000 (about that)

 

percentage of soilders asked : .006

 

i guess that 680 soliders in the sample size speaks for the rest. Yet another worthless poll.

[southsider2k5]

Yeah I am sure the troops all want to be in Iraq fighting a war. Sounds like a great way to spend your time if you ask me... of course they want to be out of Iraq in a year. Hell I wonder how they found 29% of people who wanted to stay

 

Also interesting...

 

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 weekenders want out quicker as opposed to the guys who are "in".

Edited February 28, 2006 by southsider2k5

[Balance]

i guess that 680 soliders in the sample size speaks for the rest.  Yet another worthless poll.

Well, that's what it means to conduct a scientific poll. So, yes, that sample is statistically representative of the whole.

[knightni]

HEY! I NEVER SIGNED UP FOR THE ARMY TO FACE COMBAT!

 

:rolly

[Balance]

Yeah I am sure the troops all want to be in Iraq fighting a war.  Sounds like a great way to spend your time if you ask me... of course they want to be out of Iraq in a year.  Hell I wonder how they found 29% of people who wanted to stay

 

Also interesting...

The weekenders want out quicker as opposed to the guys who are "in".

So, you're questioning the troops' courage and commitment, then?

[samclemens]

HEY! I NEVER SIGNED UP FOR THE ARMY TO FACE COMBAT!

 

:rolly

 

seriously, no kidding. when you are a soldier, you do whatever the f*ck they tell you to do, trusting in your superior's judgment and excluding a war crime or something like that.

 

i believe that zogby polls were discounted on this site for having "suspect methods" by posters on both sides of the eisle here at soxtalk. i dont remember what thread it was, but i think i remember bigsqwert going into a bit more detail about zogby. (correct me if i'm wrong, i dont have the time or effort to go back through old posts and find it)

[southsider2k5]

Damn this stuff is way more interesting than the snippet that was posted...

 

Those are strangle numbers for people who are repeating tours in Iraq.

 

Three quarters of the troops had served multiple tours and had a longer exposure to the conflict: 26% were on their first tour of duty, 45% were on their second tour, and 29% were in Iraq for a third time or more.

 

The troops think are are well armed and protected. They also think they are improving the daily lives over there.

 

Just 30% of troops said they think the Department of Defense has failed to provide adequate troop protections, such as body armor, munitions, and armor plating for vehicles like HumVees. Only 35% said basic civil infrastructure in Iraq, including roads, electricity, water service, and health care, has not improved over the past year.

 

A majority thinks that the insurgency is coming from abroad, but at the sametime they don't think the insurgency would end if the borders could be completely sealed. At the sametime 80% still hold a favorable view of the Iraqi people. Interesting.

 

The continuing insurgent attacks have not turned U.S. troops against the Iraqi population, the survey shows. More than 80% said they did not hold a negative view of Iraqis because of those attacks. About two in five see the insurgency as being comprised of discontented Sunnis with very few non-Iraqi helpers. “There appears to be confusion on this,” Zogby said. But, he noted, less than a third think that if non-Iraqi terrorists could be prevented from crossing the border into Iraq, the insurgency would end.

[southsider2k5]

So, you're questioning the troops' courage and commitment, then?

 

Yeah that's what I am doing... GMAFB.

 

I am saying it is pretty obvious to me that the troops would rather be home than having their lives in danger in a foreign country. If you need to read between the lines and twist that into something else, have at it.

[whitesoxfan101]

Hahaha the good old "less than 1 percent of the troops sample size want to leave now, so we better do it" argument. We need to get out of Iraq, but this poll is about as meaningful as Guatemala's opinion on U.S. music.

[juddling]

Well, that's what it means to conduct a scientific poll.  So, yes, that sample is statistically representative of the whole.

 

 

Besides it does matter what position the soldiers were in when asked. If you ask a guy stuck in a foxhole with bullets flying over his head every day then hell yeah he's gonna want to leave. Now ask a bunch of supply people back at the bases who maybe see the good going on and they will tell you something different . again....you can take a poll over there and have it skewed to any side you want.

[whitesoxfan101]

An accurate stat: 100 percent of small sample size polls are full of s*** and biased in some way. This is why we have elections, referendums, etc...

[southsider2k5]

Besides it does matter what position the soldiers were in when asked.  If you ask a guy stuck in a foxhole with bullets flying over his head every day then hell yeah he's gonna want to leave.  Now ask a bunch of supply people back at the bases who maybe see the good going on and they will tell you something different .  again....you can take a poll over there and have it skewed to any side you want.

 

And that really bears out when you read the section about what the percentages were by branch. The everyday people had much hirer levels than did the weekend warriors.

[Balta1701]

Sample size: 944

 

Troop level: 140,000  (about that)

 

percentage of soilders asked : .006

 

i guess that 680 soliders in the sample size speaks for the rest.  Yet another worthless poll.

Yes, it actually does give you valuable information about the opinions/attitudes over there. Because once you get to a certain sample size, if you've truly done random sampling, then it's almost impossible for the overall survey results to be off by a significant percentage. In this case, how you should read this poll is thus:

 

"If I were to take all 140,000 U.S. troops in Iraq, break them into groups of 944 at random, and ask each of these groups the same question, 95% of those groups would have responses within plus or minus 3.3% of the overall result of this poll."

 

That does, however, assume that you have gotten a truly random sample. That's sometimes the one problem. For example, the CBS poll we saw last night has a major problem in that they don't have enough Republicans being sampled for whatever reason, so they're normalizing their poll results by assuming that a higher portion of the population are Republicans than they're actually seeing.

 

Interestingly, unlike the CBS poll, this poll actually has data which can be used to check whether or not their sample is truly random. From the poll:

 

Three of every four were male respondents, with 63% under the age of 30.

Because this poll took demographic information, it actually has something you can compare it to. If those numbers are plus or minus about 3% away from the actual numbers that the DOD has in Iraq, then it strongly suggests that they did manage to get a random sample. If, on the other hand, like 60% of the soldiers in Iraq were actually over 30, or 50% were female or something like that, and the sample set was made up as described, then the sample would almost certainly not be random. Unless those numbers are inaccurate, there's no reason to believe that this poll is not representative of the entire army in Iraq within a couple of percentage points.

[Balta1701]

And that really bears out when you read the section about what the percentages were by branch.  The everyday people had much hirer levels than did the weekend warriors.

Keep in mind that since you're talking about select branches...you're talking about a subset of the poll, and therefore, you're talking about groups with much higher margins of error. Probably on the order of plus or minus ten to fifteen percent over each group.

[Balta1701]

i believe that zogby polls were discounted on this site for having "suspect methods" by posters on both sides of the eisle here at soxtalk. i dont remember what thread it was, but i think i remember bigsqwert going into a bit more detail about zogby. (correct me if i'm wrong, i dont have the time or effort to go back through old posts and find it)

Zogby and Gallup are 2 opposite ends of a spectrum in the general elections. In the General, Zogby assumes that the party registration of whatever sample he gets may be biased, and he takes the step of normalizing his results by the party affiliation of the previous election. In other words, if his sample records 33% Republicans, Democrats, and Independents, but the last election was 40-30-30, he'll assume he missed some Republicans and recalculate.

 

Gallup doesn't do that...Gallup assumes that Party ID is a variable...that people can leave and enter parties depending on how they feel at the time. While this is true, it leaves the possibility that Gallup's sample may be biased by something like having a larger portion of Democratic voters either at work when Gallup calls or using only cell phones, which cannot be polled.

 

Which method works? Well, in the last election, Gallup was closer to correct; there was a shift towards the Republicans due to additional votes for Bush among the elderly and among those who are in the top 10% of wage earners. Zogby missed this, because he assumed that party affiliation was holding constant. However, it's also possible that next election, Gallup could easily be the one who is off, because they could miss some bias in their sample due to things like people not being home or choosing to not respond when called (If more Republicans choose not to answer the poll than Democrats, suddenly they have a biased sample.)

 

In this poll, it looks like none of that is any concern at all. This Zogby poll appears to have taken several steps to appear random, they have demographic information which can be cross-checked and compared to army statistics (which I don't have but i'm betting others do), and there's no adjustment based on any standard as far as I can tell. So in this case, the only reason to disbelieve Zogby's method is if he didn't get a representative sample...which as I said above, could be checked by comparing his demographic information to the makeup of the army as a whole.

[NorthSideSox72]

Zogby and Gallup are 2 opposite ends of a spectrum in the general elections.  In the General, Zogby assumes that the party registration of whatever sample he gets may be biased, and he takes the step of normalizing his results by the party affiliation of the previous election.  In other words, if his sample records 33% Republicans, Democrats, and Independents, but the last election was 40-30-30, he'll assume he missed some Republicans and recalculate.

 

Gallup doesn't do that...Gallup assumes that Party ID is a variable...that people can leave and enter parties depending on how they feel at the time.  While this is true, it leaves the possibility that Gallup's sample may be biased by something like having a larger portion of Democratic voters either at work when Gallup calls or using only cell phones, which cannot be polled.

 

Which method works?  Well, in the last election, Gallup was closer to correct; there was a shift towards the Republicans due to additional votes for Bush among the elderly and among those who are in the top 10% of wage earners.  Zogby missed this, because he assumed that party affiliation was holding constant.  However, it's also possible that next election, Gallup could easily be the one who is off, because they could miss some bias in their sample due to things like people not being home or choosing to not respond when called (If more Republicans choose not to answer the poll than Democrats, suddenly they have a biased sample.)

 

In this poll, it looks like none of that is any concern at all.  This Zogby poll appears to have taken several steps to appear random, they have demographic information which can be cross-checked and compared to army statistics (which I don't have but i'm betting others do), and there's no adjustment based on any standard as far as I can tell.  So in this case, the only reason to disbelieve Zogby's method is if he didn't get a representative sample...which as I said above, could be checked by comparing his demographic information to the makeup of the army as a whole.

 

For the record, I was the person samclemens was referring to. I took Squirt to task for the Zogby poll.

 

The reason I find Zogby polls to be nearly meaningless isn't about the numbers at all - it is about the method of harvesting respondants. Not sure if you all know this, but the way Zogby conducts his polls (the ones I am familiar with anyway) is to recruit volunteers over the web, and email them poll questions on various topics. This is entirely different than a random phone-based poll for many reasons.

 

Now, what he does with the raw result set aside, this method alone makes the data worthless IMHO. You already have a highly biased set who are willing to be part of the Zogby "panel", and further, you have built in biases for internet usage and savvy, as well as factors like people who have time to fiddle with that crap in their email. So no matter what he does to massage the results, the raw data is out of whack, so the results are not meaningful.

 

Therefore, I would not rely on this poll for much of anything.

[Balta1701]

Well, this poll appears to have been conducted using a different method: face to face sampling in Iraq.

 

The survey included 944 military respondents interviewed at several undisclosed locations throughout Iraq. The names of the specific locations and specific personnel who conducted the survey are being withheld for security purposes. Surveys were conducted face-to-face using random sampling techniques. The margin of error for the survey, conducted Jan. 18 through Feb. 14, 2006, is +/- 3.3 percentage points.

They certainly appear to have had no connection to being found over a computer. Which is probably why "Le Moyne college" was involved. Probably gave him manpower and access to the bases that he might not have had if it were just 1 person doing the sampling.

[NorthSideSox72]

Well, this poll appears to have been conducted using a different method: face to face sampling in Iraq. 

 

They certainly appear to have had no connection to being found over a computer.  Which is probably why "Le Moyne college" was involved.  Probably gave him manpower and access to the bases that he might not have had if it were just 1 person doing the sampling.

 

Well that does make a big difference. The Zogby polls I have seen did not use face-to-face methods. You are probably right about the partnering, that gave him the manpower he needed. Probably part of some Le Moyne college kids' Masters thesis or something.

[EvilMonkey]

http://www.prospect.org/print/V14/1/mooney-c.html

 

some additional info on Zogby.

[Balta1701]

Here, Here, Here, and here are posts from a blog raising questions about the methodology of this particular poll, for those who are interested.

[NUKE_CLEVELAND]

Ok here's a little more informed commentary on this deal.

 

 

There are very few soldiers who "want" to go to war, Id say about 10% or less. Those who "want" to go to war are either green troops who never faced the business end of a gun before or adventurers who most think are s*** nuts. We go to war because its our job not because we want to be there.

 

 

Most soldiers, as I do, want to see the Iraqi's get their act together so we can get out of there. While they dont want us to be there any longer than necessary, we also believe that if we just arbitrarily withdraw before our work is done then all we've done in the last 3 years there will be in vain.