Nowhere to Go But Up

Polling Iraqis: "Although the percentage of Anbar residents who have a favorable view of local security has increased to 38 percent from zero in March, 62 percent still rate security negatively overall."

So, okay, I'll concede that this counts as progress. Unfortunately, on aspects of the situation that weren't already at the lowest-possible level, we're not doing as well: "Meanwhile, the level of satisfaction in other quality-of-life categories -- including the availability of jobs, supply of clean water and freedom of movement -- has decreased since March."

Comments

the poll must be inaccurate, matthew: i've been assured by no less than george bush that anbar is a sucess story. and bush wouldn't make stuff up, would he?

Posted by howard | September 10, 2007 11:44 AM

Ok, let's go over this once again: You're living in an area where a covert insurgency kills anybody who annoys them, and their families, too. And some dude you don't personally know starts asking you questions. Do you tell him what you really think, or what you think won't annoy the insurgents? Depends, I suppose, on the relative weight you put on speaking your mind, vs not having your family murdered...

It really, really annoys me when polsters pretend that they can get reliable numbers out of areas where people get murdered for speaking their minds. It's right up there with conducting "confidential" on the street interviews with a government supplied translator, for sheer bloody minded refusal to face reality.

Plus or minus 2.5%? They're freaking insane if they think that's really the margin of error.

Posted by Brett Bellmore | September 10, 2007 11:51 AM

Do you tell him what you really think, or what you think won't annoy the insurgents?

This is a fine point, as long as you acknowledge that it points to the conclusion that the progress in Anbar is actually overstated. After all, if you're afraid that you may get shot by an insurgent for answering a survey honestly, your answer is not likely to be "things are terrible here, I might get killed by an insurgent at any moment, please keep the US troops here for my protection." Instead, your answer would be "everything is fine, now go away please."

Posted by Steve | September 10, 2007 11:57 AM

I am unclear on the point Mr Bellmore is trying to make. Is it that people who are asked whether they feel secure will lie because they will get killed if they answer the question? Does Mr Bellmore think that the poll number is too high or too low? If this is the point, I don't think we need a poll.

Posted by Peter Driscoll | September 10, 2007 12:34 PM

Who gives a shit what the Iraqis think of us. What is this Sally Field's Oscar speech? We can't leave Iraq until they "like us, they really really like us"?

Kiss my ass Iraqis. I never cared about your country and have no clue why the republican party is bankrupting the country I do care about (the USA) for you asswipes.

Rebuild your own damn country.

I'd like my compatriots HOME. NOW.

Rather than toss my tax dollars onto the fire you call a country, I'd sooner give it to national healthcare or the people of New Orleans.

You know, AMERICANS.

Posted by Jim | September 10, 2007 12:41 PM

Anyone know how many Iraqis are actually left in al Anbar province?

The press keeps using pre-war population numbers, but judging by recent photos of cities like Ramadi(pre-war pop. 435,000), there's no way the current population is close to the pre-war numbers.

"To plunder, to slaughter, to steal, these things they misname empire; and where they make a wilderness, they call it peace."

--Tacitus

And all that...

Posted by alphie | September 10, 2007 12:48 PM

I think the plus or minus 2.5% error they cite is a joke, that's my point. Sample size error is frequently the smallest error contributor in a poll, and polsters rarely acknowlege that, and it pisses me off. Especially when, in cases like this, there are obvious questions about whether people would even feel safe speaking their mind to a stranger.

As I said, it's right up there with conducting "man in the street" interviews in a police state, using a government supplied translator, and thinking that you're getting honest responses.

And yet the media do this sort of thing all the time, and insist on pretending they're presenting reliable data. It's a joke, and a dark one.

Posted by Brett Bellmore | September 10, 2007 12:49 PM

This summary is actually too kind. You need to read the actual poll and not just the WaPo's summary.

The amount of Iraqis who support attacks on US troops has actually increased since March (from the same poll). It is now 57% and 93% in the Sunni community (50% in the Shi'ite community). The number of Iraqis who support an immediate withdrawal of US troops has increased 12% since March. The number of Iraqis who think the surge has WORSENED things is 70% (12% think it has improved things). And so on. (here a link to the full findings: http://news.bbc.co.uk/2/hi/middle_east/6983027.stm)

Brett, in places like Anbar, shouldn't this "fear" have dissapeared now? I think the numbers Matt cites are perfectly understandable if you know that in the city of Fallujah, the unemployment rate is 80% and driving has been made temporarily illegal.

Posted by Ben P | September 10, 2007 12:52 PM

Anyone know how many Iraqis are actually left in al Anbar province?

Rep. Tom Lantos says it's about five percent of Iraq's total population.

Posted by Matthew Yglesias | September 10, 2007 12:54 PM

Sorry the link above is broken. Try this: http://news.bbc.co.uk/2/hi/middle_east/6983027.stm

Posted by Ben P | September 10, 2007 12:55 PM

Brett, let's put aside your false comparison to a man on the street interview in a police state; let's put aside the fact that you're not alone in knowing some of the risks (the polling organizations know it too, and attempt to respond).

the point is: if you're afraid to tell a pollster what you think, then anbar isn't the magnificent success story of legend. steve challenged you on this; you ignored it so you could vent on some favorite hobby horse.

now, deal with it: what does it say that by your account, you're too scared in anbar to explain how secure you are to a pollster?

Posted by howard | September 10, 2007 1:00 PM

I'm pretty sure someone from Bush's camp will be able to explain why being afraid to talk to a pollster is a sign of progress but Brett doesn't get the job done. howard's got it right. Anbar is worse than we think and the polls show it. Iraqi behavior shows it. But we're not leaving no matter what, so it is of no moment anyway.

Posted by Peter Driscoll | September 10, 2007 1:22 PM

I am saying nothing more than what I'm saying: That it pisses me off when the media present unreliable polls as hard data. They know they can't even get good results from exit polling, and still they pay posters to go in war zones and collect numbers that are about as reliable as somebody's magic 8 ball, and then represent them to their readers as solid information!

It's a pet peeve of mine.

And, sure, the polsters probably told the AP just how dubious the numbers were; The AP chose to misrepresent to readers their reliablity, they weren't duped.

And, by the way, given that the rewards of talking straight with a polster are minimal, and the potential downside is heartrending, just how dangerous do you think a place has to be before polls like this are worthless? I'd say, not very dangerous at all.

Posted by Brett Bellmore | September 10, 2007 1:31 PM

Unfortunately, Brett sticks with the line that the poll is "worthless," as though it could be off in either direction.

You see, maybe Anbar is dangerous, but people are saying it's safe because they're afraid of the insurgency. Or, alternatively, it's equally likely that Anbar is safe, but people are saying it's dangerous because they're afraid of... what, I'm not exactly sure, if it's safe!

It's entirely possible that this poll understates the level of danger in Anbar, yes, but I doubt that's the point Brett was aiming for.

Posted by Steve | September 10, 2007 1:46 PM

The insurgency clearly wants the American people to think the surge isn't working. Why wouldn't a resident of Anbar reason that the insurgency wants him to tell a polster that?

Anyway, my beef is with media outlets which deliberately mislead their audience about how reliable polls are. Which way they're unreliable has no bearing on that complaint.

I guess we're not allowed to care about whether the media is lying to us, independent of the merits of the cause that lie might be advancing...

Posted by Brett Bellmore | September 10, 2007 2:11 PM

The insurgency clearly wants the American people to think the surge isn't working. Why wouldn't a resident of Anbar reason that the insurgency wants him to tell a polster that?

Let's not obscure the point with sophistry. The poll asked people about security. It didn't ask them whether "the surge is working."

Clearly, what the insurgency wants is for American troops to leave. Telling pollsters that it remains dangerous in Anbar is likely to lead to an extended troop presence, rather than the opposite.

The idea that it's safe to walk the streets in Anbar, but people are telling pollsters that it's not safe in order to please the insurgency, is one of the craziest things I've heard in a long time. Tell me, if people truly have a "favorable view of local security," why would they be so afraid of the insurgency that they'd have to lie about it?

The complaint that the media is lying is similarly laughable. Look, aside from crazy Republicans who think that Iraq is like Indiana in the summertime, everyone gets that Iraq is a chaotic and war-torn place. People are free to draw whatever conclusions from that fact that they like; as you've demonstrated, people are even free to draw crazy conclusions like "maybe it's really safe and they're lying when they say it's unsafe."

If the poll reported optimistic attitudes among Iraqis and the media included a disclaimer that said "we might not be able to trust the opinions these people are offering," there would be screaming to high heaven about the media inserting its own biases into the news. You're looking for a reason to discount this poll because you don't like the message it conveys, Brett, but your argument is truly hopeless. "People might be lying about feeling unsafe!"

Posted by Steve | September 10, 2007 2:36 PM

"Telling pollsters that it remains dangerous in Anbar is likely to lead to an extended troop presence, rather than the opposite."

I think you have a remarkably naive view of the political impact of news from Iraq, certainly much more naive than the insurgency's.

"Tell me, if people truly have a "favorable view of local security," why would they be so afraid of the insurgency that they'd have to lie about it?"

Because things don't have to be very dangerous at all, for the lie to be the sensible choice. Cost benefit, and there's no appreciable benefit from telling a polster the truth. So the cost dominates, unless the probablity of incuring it is vanishingly small.

And the lie is that the margin of error is 2.5%, plus or minus, when they know damned well there are other sources of error that doesn't include. Calculate it yourself, that's just the error based on the size of the sample, it doesn't even take into account the possiblity that people who refused to respond aren't randomly distributed. That's not the margin of error, it's the lower limit for the margin of error, if everything else was perfect.

I've been pissed off for decades about the way the media mishandle polls, and they're only getting worse.

Posted by Brett Bellmore | September 10, 2007 3:10 PM

I guess we're not allowed to care about whether the media is lying to us, independent of the merits of the cause that lie might be advancing... Brett is trying to point out that the poll may be better or worse than the Post says, so we should ignore it. In fact, the reason that he gives for its unreliability demonstrates pretty convincingly that the people in Anbar are not secure. I have no idea what it says about the poll. Someone should ask the 3 million Iraqis in Jordan and Iran how safe they feel. They might be lying, but they left Iraq.

Of course the WaPo could have cured this with a headline that says, "Many Iraqis Feel Safe" with a Hiatt-inserted lede that says: Iraqis are proving to feel more secure after the surge by opening their doors to strangers.

Posted by Peter Driscoll | September 10, 2007 3:19 PM

"In fact, the reason that he gives for its unreliability demonstrates pretty convincingly that the people in Anbar are not secure."

Well, duh, it's a war zone. I never said they were secure, I just said that we don't know their opinions on the matter to plus or minus 2.5%. And that the AP shouldn't claim we do.

Posted by Brett Bellmore | September 10, 2007 3:22 PM