Where for art tho Lancet threads?

Email Me When A New Comment Is Made
Show Only Poster Name and Title Sort in Reverse Order Posted

Pages: PREV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 NEXT

eu4ea 10/19/2006 12:11:42 AM

I'll just give you one from Carl Sagan:

"extraordinary results require extraordinary proof" Nope, I'm not just refering to the Lancet's study. I'm questioning your motivation:

The fact is that you guys are holding up to a tremendous level of scrutiny the study that has produced the most ordinary results (this war closely resembles other wars), while happily giving a pass to the extraordinary results (this war is like no other war before) produced by sources as unreliable as political hacks in hotel rooms (such as Mr. Moore) and diletantes reading newspapers (such as Iraqbodycount.org).

Further: This is not the first time this happens. Just before the summer, the UN Assistance commision for Iraq came out with their results, which concluded that an average of 100 people were getting killed every day based on the records they collected from the Bagdhad morgue. This same fiery discussion happened back then, with the passionate denials (study not credible, iraqbodycount, assorted 'gotchas', etc.). The 100 dead/day figure is now common wisdom.

Further yet: The reason why you're doing this is an emotional attachment to the idea that the war is goign well. That's uninteligent. Wishing that the war was goign well (which I most certainly do) is one thing, blinding yourself to reality on the ground is quite another. Far from being patriotic, denying reality leads to defeat - and thus it is a tremendous disservice to the nation.

Now, truning back to Mr. Moore - here are some quotes from his previous 'research' before he authored the 'debunking' you quoted:

============================

I was in Iraq for nine months myself, and your analysis of a Westerner's life resonates. I was confined to my hotel for much of the time, and felt my freedoms erode as I became increasingly hunted.

However, I was in Iraq doing public opinion research. While I was increasingly unable to travel, I was consistently getting info on what Iraqis are thinking, and still get frequent polls.

51% of Iraqis think that Iraq is on the right track. Compare this, for instance, to Michigan where Governor Granholm inspires a 34% "right track" response on polling in her state.

More than 60% of Iraqis think that the Iraqi government is doing a good job.

Iraq is a tough place to be a Westerner, but the Iraqis are optimistic, statistically speaking.

20,000 to 30,000 insurgents, many from outside Iraq, are trying to oppress the 20 million or so Iraqis who say on polls that they want democracy, rather than a theocracy or a baathist style dictatorship.

After nine months doing a dozen or so polls and about seventy focus groups in 13 Iraq cities, I defer to those who
have the most information on Iraq's future - the Iraqi people. And they are optimistic
=============================================

Iraqis delighted with the occupation! The source of all trouble are 20-30,000 foreign fighters! No-one wants an Islamic government!

I'd say those are some pretty extraordinary results, wouldnt you?

In fact, it sounds like the work of a paid political hack holed up in his hotel room. Wait - Steven Moore actually *was* a paid political hack holed up in his Bagdhad hotel room - specifically a political hack paid by Paul Bremer, a famously incompetent political hire himself.

Next, let's go through Mr. Moore's claims:

"Survey results frequently have a margin of error of plus or minus 3% or 5% -- not 1200%."

Yes, Steven. Peacetime surveys in a developed nation following the same methodolgy often have a <5% error rate. However, did it ever occur that the "1200%" you claim is the difference between an acceptable methodology based on dangerous fieldwork and your own prefered approach?

That difference is nothing more than the difference between secondary-source aggregation studies (such as those carried out by you sitting in a Bagdhad hotel room, or by Iraqbodycount sitting in London reading the newspaper) and an on-the-ground population survey that involves constantly traveling into very dangerous areas and putting yourself at great personal risk.

"The UNDP's survey, in April and May 2004, estimated between 18,000 and 29,000 Iraqi civilian deaths due to the war. This survey was conducted four months prior to another, earlier study by the Johns Hopkins team, which used 33 cluster points and estimated between 69,000 and 155,000 civilian deaths."

Yes, Steven. The UN measured civilian deaths before the insurgency really got underway, while the Lancet study measured all deaths after the insurgency and sectarian violence got underway. Tell us again why this is surprising; you should know, at the time you were holed up in a Bagdhad hotel room doing 'research' whose outlandish results no doubt pleased yr employer, Paul Bremer.

And so on. The tone is nicely polished, but the content isnt credible and neither is Mr Moore.

Now, moving on to the clustering issue - that's certainly the main weakness in the Lancet's study that Mr. Moore has picked on. He's right that it is the weakest part of the survey. However:

1 - The Lancet survey was done by people who, unlike Mr. Moore, actually left their hotel rooms. Not only that, they put in thousands of miles traveling to every single province of Iraq and did so at a time when Iraq was radicaly more dangerous that it was during Mr Moore's hotel stay. While it certainly would be better if the Lancet team had physically gone through twice as many roadblocks and exposed themselves to many more roadside bombs in order to visit additional sample sites, it's disingenous to accuse them of being lax.

2- The Lancet itself identified clustering as the primary source of uncertainty in this survey. Aside from number of clusters, every other aspect of the methodology (the weighted geographical distribution, qualifications of the survey team, number of households surveyed, number of individuals surveyed, etc) are uniformly excellent.

3- Reduced cluster numbers have an effect on the confidence range of the results (which the Lancet team reflected) not on the global results. It is just as likely that the casualty rate was under-reported as it is that it was over-reported.

4- The survey's overall result (2.5% excess mortality over 3+ years of vigorous armed conflict) is entirely consistent with what we've seen in comparable wars in the past.

5- The age and sex distribution of the mortality (men more affected than women, violent deaths heaviest amongst military age males, excess non-violent excess heaviest amongst the elderly/very young, etc.) is entirely consistent with what we have seen in previous wars.

6- The geographical distribution of the mortality (heaviest in Bagdhad and Al-anbar, lowest in the Kurd areas and the calmer souther nprovices, etc.) are fully consistent with our experience of the fighting there.

7- The trends evident in the survey data (the present year being the most violent, an acceleration of the increase in mortality, an increase in the number of casulaties caused by intra-iraqi violence, etc.) is fully consistent with our experience of the fighting there.

Etc.

I can easily go on, but the point is clear: this is a political hack with a history of doing his 'research' while hanging out in a hotel rooms doing a political hatchet job. The ony thing that's surprising is that the WSL would publish this bunk.

Heart,

eu4ea

Quote Reply

eu4ea 10/19/2006 2:22:49 PM

I'm still waiting, WhiteB. Got any more Carl Sagan handy?

Quote Reply

Pseudonym 10/19/2006 6:20:00 PM

"Now, moving on to the clustering issue - that's certainly the main weakness in the Lancet's study that Mr. Moore has picked on. He's right that it is the weakest part of the survey. "

Weak point?

LOL.

It's true, a Liberal will not accept the truth when it hits them in the face.

"The fact is that you guys are holding up to a tremendous level of scrutiny the study that has produced the most ordinary results"

LOL. Sure, it's Sagan whose doing the political hack job, not Lancet. >LOL.

Go do some basic research on cluster sampling, then come back with more than attacking the source, when you cannot counter THE COLD HARD FACTS.

Quote Reply

Plutarch 10/19/2006 7:06:48 PM

In Baghdad, thousands of bodies have been pulled from the Tigris, but the deaths aren't reported. How bad is the violence?

Still, even these figures don't tell the whole story. For that, a visit to Medical City is in order. The Ministry of Health has instituted a strict policy for journalists, requiring them to seek permission before visiting the facility. Those allowed in get only a truly sanitized tour; more often than not reporters are barred from entering. But at the gate, guards who have worked at the facility tell a chilling tale. "Last year, I saw maybe 1,000 bodies a month coming into the morgue," says one man who, fearing for his life, requested his name not be published. "Now we're getting nearly 1,000 a week."

All, he says, are victims of sectarian violence, both Sunni and Shia, but the officials at the morgue inside Medical City will not tell you that. "The officials don't want us talking to the media," says another guard, also requesting anonymity. "I've heard them telling people that most of the deaths are because of terrorists, but I've also seen the bodies myself and I can tell you that most of them were executed by death squads."

While he describes the bodies, a dump truck pulls out of the facility. The guards open the gate, holding back a rush of people from all over central Iraq hoping to get in to look for loved ones. As the truck passes, the smell of decomposing flesh fills the air. "That's just the clothes from bodies pulled from the Tigris over the past few days," says the first guard. "The trucks come and go regularly." The stench is overwhelming. People cover their noses and mouths with cloth. Women wail.

link

IBC isn't even close to the number of actual deaths.

Quote Reply

swhitebull Interview with the Lancet study Author on Methodology and results 10/20/2006 5:31:39 PM

www.pajamasmedia.com/2006/10/joisting_with_the_lancet_the_p.php

�

swhitebull� - you be the judge

Quote Reply

PlatypusMaximus 10/25/2006 8:36:19 AM

4yr. X 100deaths/day = 146,000

Want leap year? You got it, buddy!...146,100

As of May this year, it was reported that 50 deaths/day was a sign of new, increased violence.

146,000 might well be acurate, assuming Arab governments do not exaggerate. 146,000 might well equal 650,000, but I'm not betting my hard-earned GED on it.

Quote Reply

eu4ea 10/29/2006 4:58:43 PM

Sadly, I have to say that I'm more amused than amazed by the on-going Lancet controversy. Some folks just love to go on - *anything* other than admitting a mistake or (heaven help us!) abandoning failed policies.

It went from Saddam is the devil and WMD *will* be found to "well, he was err _planning_ on making WMDs"
Next it went from "the world-wide coalition" to the amazing shringking coalition.
Next was "a few diehard insurgents" to "insurgents weakening!" to "insurgents really weakening this time" to hum, we still have 140,000 troops, but that's not because there's a growing insurgency
Then there was "capturing bagdad is the turning point!" to "capturing saddam is the turning point!" to "leveling fallujah is the turning point!" to "elections are the turning point!" to "killing zarquawi is the turning point!". And there is no turning point - it has all steadily progressed: into a larger and nastier civil war.
Next it was "minimal iraqui casulaties" to "maybe 30,000 (bush, december)" to "50-60,000" to "well, maybe lots but we cant count 'em"
Next it was "the UN is nuts to claim 100 dead/day in bagdad alone" to "there are only 100 dead/day in bagdhad - things are goign well"
Etc. etc.

Each one of those was argued and denied with the same vitriol that is now directed at the Lancet's study.

The Lancet controversy is just the last one on this long and sorry line. I predict estimates of 500k to 1M dead will be commonplace within 6-9 months. But by then we'd move on to argue about something else. Like whether it was our plan all along to divide Iraq in 3. Or whether it's really Iraq's government's fault that this whole thing is a bloody failure, not ours.

Heart,

eu4ea

Quote Reply

shek 10/29/2006 8:01:11 PM

A string of red herrings. Roberts et al fails on its own accord. Also, you've still never addressed my benchmarking questions, the first of which deals with their underestimation of pre-war deaths leading to overestimating the "excess" death rate, and the second of which deals with their misleading attempts to benchmark to other modern conflicts despite the lack of evidence that they serve as valid benchmarks..

Quote Reply

eu4ea 10/29/2006 10:33:55 PM

Hardly "a string of red herrings".

More like a string of well-documented instances where the argumentations of the Bush faithfull (not 'Republicans' - those are two very different species) rang on and on, before being overtaken by reality.

And it is my assertion that this specific case (questioning the finer points of Lancet's clustering and benchmarking methodology) will soon come to precisely the same fate.

Regarding your points about clustering and benchmarking, I have already addressed them above. I'll be happy to cut and paste as required, and in the meantime please explain why you find other surveys more credible. I'd be particularly interested to learn why I should put faith in iraqbodycount.org with their laughable 'sit in london and read the newspaper' methodology, or the comical "30,000 or so" figure put forth by Bush in December.

If you want to go one better, you can go to the UN comission for Iraq, and their 100 dead/day in Bagdad estimate based on morgue records. I'd be happy to ignore the fact that we had the very same discussion (their methodology is suspect, they are biased, international press reports adds up to less deaths, etc.) when that report came out five months ago.

Heart,

eu4ea

Quote Reply

shek 10/29/2006 11:01:08 PM

If you say you've addressed the benchmarking issue, then I'd love to see you cut and paste what you think addresses it, as all I've seen are assertions that a paragraph from Roberts et al should be taken as gospel, akin to your claim that it was a "gold standard" survey. Sorry, but the gold plating isn't sticking to it so well. In fact, it's been pretty well peeled off.

Also, just like you want to have a double standard regarding the CIA numbers (count the before numbers, but not the after numbers), it seems like you want a double standard on the UNDP numbers. You're quoting them above, and yet you dismissed their survey results cited by Mr. Moore (remember your double standard on ad hominem attacks - where you attacked Mr. Moore as a political hack, but Roberts is a pure as the driven snow) and linked to by me.

Your line of reasoning in this thread is sure built upon a house of double standards.

BTW, I've never argued a lick about IBC, so why do you insist on dredging up that tired strawman?

Anyways, I'm standing by for your attempt to argue why two 25 year conflicts, one of them with an inflated death count by your "gold standard" researchers citing a secondary source, and a conflict where 500K have died in a year after the peace from diseases not found in any significant numbers in Iraq serve as appropriate benchmarks.

Quote Reply

Arbalest 10/30/2006 2:00:46 AM

It seems that there are still questions about the Lancet's study.

A short Google effort to check on 2006 Iraqi population density reveals the following:

Population of Iraq: 22-25 million

Population distribution: about 75% live in the Tigris-Euphrates corridor (approximately Mosul to Basra), and there is a large peak around Baghdad.

About 20% of the population are Kurds, and their areas in the north have been essentially free of killings since April 2003. The south and the west seem to have had some killings, but the vast majority of killings seem to be in the Baghdad area and center of the country.

Just using news reports and population locations, it looks like the populations in the Kurdish areas (20%), the southern Shia areas (another 20%) and the western districts (another 25%) should not have been included in the Lancet’s calculations, as the day-to-day reality in those areas is considerably different.

Reducing the Lancet’s initial figure of 655,000 by 65% yields 229,250.

I suspect that with a little more news analysis, it will be easily possible to exclude another 5-10% of the Iraqi population from the calculations, as they, too, do not live in areas with the same conditions as those of the Lancet’s survey. This puts the death toll between 164,000 and 197,000

It looks like the Lancet’s statistics can be correctly applied only to the 5 to 8 million people who live in Baghdad and certain other nearby areas.

650,000 deaths represents 8-10% of the population in the area. This might represent 15% (almost 1 of 6) of the adult male population, as women and late middle-aged and elderly men are unlikely to be targets.

Surely such a change in day-to-day life should be noticed.

Surely corroborating evidence, such as an increase in claims for public assistance by widows and orphans, is available. If entire families were involved, a decrease in school attendance, consumption of groceries, electricity, and an increase in vacant apartments, should be evident.

Where is the corroborating data?

If the Lancet’s calculations are really only valid for Baghdad and environs, and not Iraq as a whole, then 170,000-190,000 deaths seems to make much more sense, and would represent 5% (1 of 20) of the young adult male population (the ones most likely to be involved in fighting).

My experience with political hacks has been that they frequently try to spin only one or two things (change "Baghdad-and-environs-only" to "all of Iraq") to get the result that they want. There are then fewer "things" to keep track of, a detailed look might bury the investigators, and when discovered, plausible deniability becomes easier.

Quote Reply

shek 10/30/2006 7:18:37 AM

questioning the finer points of Lancet's clustering and benchmarking methodology

Finer points? Um, I'd call it the heart of their study. If it fails, then they fail. Well, their benchmarking fails against other established and credible studies of Iraq (but they never mention these numbers, why do you think that's so eu4ea?). It also fails when you try to motivate their "other conflict" benchmarking strategy. Anyways, since it's just a cut and paste job for you, you should have a detailed point by point reply up in no time. However, given that my points, the same points that have been out there for two weeks or so now went unanswered before, I suspect it won't just be a cut and paste job.

Quote Reply

shek 10/30/2006 7:19:13 AM

questioning the finer points of Lancet's clustering and benchmarking methodology

Quote Reply

PlatypusMaximus 10/30/2006 7:48:46 AM

Un study, my arse. the UN simply regurgitated what the Iraqi deputy information minister told them. Which you became YOUR latest battlecry. Show us....on what day since that figure appeared have more than 100 iraqis died? 100 is the average, so maybe you could find 2 days in which more than 100 Iraqis died...Pick a date, any date...still waiting; been waiting since you spewed that Iraqi garbage.

All along, it has been Iraq, Iran and North Korea.

and their enablers.

Quote Reply

Plutarch Arbalest Reply 10/30/2006 12:13:02 PM

It seems that there are still questions about the Lancet's study.

A short Google effort to check on 2006 Iraqi population density reveals the following:

Population of Iraq: 22-25 million

Population distribution: about 75% live in the Tigris-Euphrates corridor (approximately Mosul to Basra), and there is a large peak around Baghdad.

Reducing the Lancet’s initial figure of 655,000 by 65% yields 229,250.

It looks like the Lancet’s statistics can be correctly applied only to the 5 to 8 million people who live in Baghdad and certain other nearby areas.

Surely such a change in day-to-day life should be noticed.

Where is the corroborating data?

"Just using news reports and population locations, it looks like the populations in the Kurdish areas (20%), the southern Shia areas (another 20%) and the western districts (another 25%) should not have been included in the Lancet’s calculations, as the day-to-day reality in those areas is considerably different."

That isn't quite correct; the Shia districts in the South are almost as violent as Baghdad. Basra sees two or three poltical assassinations per hour every day, and the Mahdi Army has taken over Amarah in the East

Also the western districts are Al Anbar, Najaf, and Ninawa, hardly immune to violence.

Fighting in most wars in history is localized. There wasn't a lot of killing going on in Maine during the American Civil War but Maine was affected by the war nonetheless (economically, socially, etc.)

That being said I think you are closer to the death toll (160-200,000) which is still pretty significant

Quote Reply

Pages: PREV 1 2 3 4 5 6 7 8 9 10 11 12 13 14 NEXT

	News As History - November 8, 2009
	Home News How to Make War Wars Around the World Austin Bay's On Point Dirty Little Secrets Features Al Nofi's CIC Prediction Market Wargames Measure of Respect On War and Warfare Videos Photos Jokes Community Military Discussion Boards Military Jokes Military Photos Military Book Reviews Military Movie Reviews Military Links Wargame Store StrategyPage Subscribe Login Feedback About Us Search Account Manager Advertise With Us