Decision ‘08

The Lancet Study, Mach 2: Revenge Of The Bullspit

You may recall, if you’re the type to torture yourself with such things, a pretty widely debunked study that appeared in The Lancet that overstated the Iraqi civilian casualties after the first 18 months of the Iraq War by 300% or more, according to other widely respected and nonpartisan estimates, including those of human rights groups.

Now the same organization is back with a study that inflates the most widely held casualty count by over 1,000% - and they’re using the same ’scientific’ methodology:

A team of American and Iraqi epidemiologists estimates that 655,000 more people have died in Iraq since coalition forces arrived in March 2003 than would have died if the invasion had not occurred. The estimate, produced by interviewing residents during a random sampling of households throughout the country, is far higher than ones produced by other groups, including Iraq’s government. It is more than 20 times the estimate of 30,000 civilian deaths that President Bush gave in a speech in December. It is more than 10 times the estimate of roughly 50,000 civilian deaths made by the British-based Iraq Body Count research group. The surveyors said they found a steady increase in mortality since the invasion, with a steeper rise in the last year that appears to reflect a worsening of violence as reported by the U.S. military, the news media and civilian groups. In the year ending in June, the team calculated Iraq’s mortality rate to be roughly four times what it was the year before the war. Of the total 655,000 estimated “excess deaths,” 601,000 resulted from violence and the rest from disease and other causes, according to the study. This is about 500 unexpected violent deaths per day throughout the country. The survey was done by Iraqi physicians and overseen by epidemiologists at Johns Hopkins University’s Bloomberg School of Public Health. The findings are being published online today by the British medical journal the Lancet. The same group in 2004 published an estimate of roughly 100,000 deaths in the first 18 months after the invasion. That figure was much higher than expected, and was controversial. The new study estimates that about 500,000 more Iraqis, both civilian and military, have died since then — a finding likely to be equally controversial.

Yes, I suspect it will, at that. As a public service the Lancet team has examined some other events in history and provided up to the minute ’scientific’ estimates of losses:

Gettysburg: 4.78 million dead, 18.9 million wounded

Hiroshima: 1.93 billion dead, every living soul wounded

The recent Israeli/Hezbollah conflict: 450,000 dead, curiously, only 3 wounded….

65 Responses to “The Lancet Study, Mach 2: Revenge Of The Bullspit”

1 too many steves Says:
October 11th, 2006 at 4:52 am
In other news, Ted Williams, famed Red Sox slugger, hit a home run to left field in the final at bat of his career. The reported attendance at the game on September 29, 1960 was 10,454. The game was not televised, but in interviews with Red Sox fans over the past 46 years I have learned that more than 2,000,000 people actually witnessed Ted hitting the oltimate home run of his storied career.
2 HP Brandy Says:
October 11th, 2006 at 6:40 am
Except that, as the reports of the new study show, the methodology is sound and the Lancet figures have been corroborated several times. Nor was the original study ‘debunked’ -it was smeared and that’s an important difference.

The Iraq Body Count figure is a measure of media-reported dead not the actual figure which they state repeatedly is likely to be far higher.

I would suggest that before you rush to write about this, you actually educate yourself about the subject matter. The original Lancet report has still not been challenged in any peer-reviewed journal and for very good reason: it’s solid work.
3 Mark Says:
October 11th, 2006 at 6:47 am
Yes, HP, 500 people a day have died violent deaths - every single day - yet somehow no one’s noticed!

Go sell it somewhere else…
4 Undertoad Says:
October 11th, 2006 at 7:03 am
Well they’re learning. Last time they released the report a few days before the election, which meant it was lost in the heavy news cycle and didn’t have any impact. This time they’ve released it a month out, which will give it time to percolate.
5 Decision ‘08 » Blog Archive » And The Blind Shall Lead Them… Says:
October 11th, 2006 at 7:20 am
[…] 2008 : Oct 11th - 7:20am « The Lancet Study, Mach 2: Revenge Of The Bullspit […]
6 too many steves Says:
October 11th, 2006 at 7:22 am
This just in: tree falls in forest, no humans present, no sound recorded.

Reading this report I find that the only part of this that is demonstrably scientific is based on the claim that the 1,849 sampled were chosen at random. Good start. The rest gives no insight as to credibility and, thus, the ultimate value of this work. They looked at death certificates, fine, did they determine who of the dead were civilians, insurgents, and other mischief-making combatants? Seeing as it is being published in a medical journal perhaps there will be peer review this time. Check me on this, but is the absence of a peer review challenge considered to be definitive proof of accuracy? I must have been out playing frisbee rather than in class the day that was taught.
7 Blue Crab Boulevard » Blog Archive » Bull Says:
October 11th, 2006 at 7:29 am
[…] UPDATE: Others: Right Wing Nut House, OTB, Decision '08, Astute Bloggers, Gateway Pundit, […]
8 wilbur Says:
October 11th, 2006 at 8:29 am
HP Brandy; the original Lancet study was pretty thoroughly ruined, credibility wise, when the 2004 UNDP report (http://www.iq.undp.org/ILCS/overview.htm) came out (oddly enough, some people have even tried to twist the UNDP figures to support the Lancet’s published ones, but you’d have to bend them in to a virtual pretzel to do so).

Anyway, it really is getting beyond the joke when someone tells us that the figure is now 655,000 excess deaths. I’d be surprised if you could prove that much carnage in amore violent place, (such as, say, Darfur), let alone Iraq. If the excess deaths are so high, why aren’t we seeing them? It’s not as though you could hide something so extensive.
9 Xanthippas Says:
October 11th, 2006 at 8:59 am
Who knew bloggers were such statistical experts? Oh…wait…you’re not. Oh, but you link to other statistical experts who “debunked” the original story. Oh wait…no, you didn’t.

The numbers are as solid as they can possibly be guys, given the state of things in Iraq. It’s wholly saddening to me to know that fellow Americans think that we can somehow erase the shame of having started this war by arguing down the numbers killed by it. Would this war still be worth it to you if the number was half that many? One-third? One-fourth? Are things better now than they were under Saddam, when we can confirm by the number dead in the morgue that 3,000 people die violently a month? Do you think Saddam killed 3,000 of his own people a month? Do you think the Iraqis are appreciate of the freedom we have bestowed upon them, when they can’t let their children go to school because someone will bomb them, or kidnap them, or force them to wear something?

There is a consistently disturbing trend among anyone who supported this stupid war to attempt to void reality by arguing it out of existence. It doesn’t work that way. If reality is not in conformity with your opinion of the war and it’s consequences, it’s your opinion that needs revision.
10 Mark Says:
October 11th, 2006 at 9:03 am
Xanthippas, you believe, then, that 500 people are killed daily in ‘excess’ of what would normally be expected…fascinating…and tell me, how did your conversation with Santa go today?…
11 Dmac Says:
October 11th, 2006 at 9:16 am
Where on earth are these idiot savants coming from these days, Mark? I suspect they’re all sock puppets of one truculent commenter…
12 Dennis Says:
October 11th, 2006 at 9:24 am
Either that or it’s Give a Troll a Cookie Day.
13 Mark Says:
October 11th, 2006 at 9:26 am
Nah, they’re legit…yesterday, they were coming from a Salon link; today, from a link from Greenwald…
14 Tom Says:
October 11th, 2006 at 9:27 am
“The numbers are as solid as they can possibly be guys, given the state of things in Iraq”

I’m afraid that’s the key right there - “given the state of things in Iraq”. It is probably simply impossible to do, at this point, what the Bloomberg School of Public Health is attempting to do. The techniques that they are using are designed for disease reporting, not for use in a war zone, given the confusion, selective memory, political agendas, and disinformation one finds there. I think that it is notable that most anti-war groups following the war in Iraq expressed skepticism of both Lancet studies, for that very reason. War is not disease, and the techniques for studying one are not necessarily applicable for studying the other. I’m no fan of the Iraq War, but I’d be skeptical of these numbers unless there was some pretty strong additional confirmation.
15 jimbobuddy Says:
October 11th, 2006 at 9:35 am
Mark: Cleverly sarcastic - apparently to you - putdowns do not answer X’s direct question to you. It’s avoidance pure and simple. Come back when you have the guts to face the fact that you , like me, have been duped into accepting a vicious war against expendable brown people. Or PLEASE,… explain to me the big picture of how this foreign policy catastrophy advances U.S. security. Perhaps you could review the crimes committed at Abu Graibe, Gitmo, and Haditha (among others), and then educate me as to how they draw millions of muslims to our side in our so-called ‘war on terror’ ?
16 david Says:
October 11th, 2006 at 9:35 am
Would the methodology be more “sound” and “scientific” were it to result in an estimated 35,000 Iraqi civilians dead?
17 Tom Says:
October 11th, 2006 at 9:54 am
“Would the methodology be more “sound” and “scientific” were it to result in an estimated 35,000 Iraqi civilians dead?”

No, it wouldn’t. But the fact that the number is so high helps make the flaws more apparent. You simply cannot extrapolate 537 deaths for a war-torn country of 25 million people, many portions of which are inaccessable. Of those 655,000 excess deaths. 601,000 are supposed to be violent. Is there ANY indication that fighting of that level is going on or has gone on, anywhere in the country? Right now, Baghdad is the most violent part of the country, and the violence there has worsened dramatically in the past three months - and still, there are nowhere NEAR 500 deaths per day, even in the most violent part of the country, during its most violent period. During the Battle of Falluja, the bloodiest episode in the whole war, there were not 500 deaths per day. I’ve talked to many U.S. soldiers who have served in Iraq, both pro and anti-war, and have never heard any indications of fighting at this level, anywhere in the country.

Right now, no one really has any idea how many Iraqi civilians have died in the war. Seeing as how many researchers are still arguing about how many people died in World War 2, this is not surprising.
All that can be said at this point is that such a high number, coming from an avowed opponent of the war utilizing an inappropriate methodology is rightly subject to skepticism.
18 The Ugly American Says:
October 11th, 2006 at 10:04 am
There is no way to put it without being offensive. Anyone who believes this study is a fool. Not your average every day fool but the kind of fool who will believe anything.

You do not need to be a statistician, or scientist to understand this study is bogus. You need comon sense.

There are hundreds of independent newspapers in Iraq all reporting the news daily. Each with it’s own bias. Iraq body count uses these news reports to come up with their numbers.

To believe this study you have to believe that litterally 500 people a day are being killed and no one is reporting it. Do you believe that?

by the way the study makes it clear that 600,000 of these deaths are violent deaths. Again do you believe that Iraqi newspapers are failing to a few hundred car bombs, sniper attacks, suicide bombers, school bus shootings, beheadings, bank roberys, school and business invasions?

if you do you are a fool.

Tens of thousands of people have died. That is tragedy enough to cause serious misgivings for the most ardent war supporters. Blatantly inflating numbers like this dishonors every one of the people who did die in this war, in an attempt to sway opinion.

It is disgusting.
19 Tom Says:
October 11th, 2006 at 10:05 am
“Would the methodology be more “sound” and “scientific” were it to result in an estimated 35,000 Iraqi civilians dead?”

No, it wouldn’t, but it does help to point out the invalidity of the method. You simply cannot extrapolate 537 deaths to 655,000 for a war-torn country of 25 million people, many parts of which are inaccessable.

In truth, we have little indea how many Iraqi civilians have died since 2003. This is not surprising, seeing as how researchers are still arguing over the number of dead in World War 2. For now, however, I think that we can say that such a high number, coming from an avowed opponent of the war using an inappropriate methodology, is rightly subject to skepticism until more evidence comes forth. After all, if a fierce opponent of Cuba’s government came out with a study based on interviews saying that Castro had actually killed ten times more people than previously thought, wouldn’t most Leftists be skeptical, at least until more data was in? I know I would be…
20 too many steves Says:
October 11th, 2006 at 10:07 am
Of course not, the result does not legitimize the method, the method legitimizes the result. Cripes.

It is to define the world incredible to state that an average of 500 excess deaths have occured per day for the past three years in Iraq. It could turn out to be true, sure, but given objective (i.e.; factual) evidence and widespread public examination and scrutiny, that volume of death is unlikely to be proved and wholly improbable.

Refuting this study claim does not legitimize the war, but it appears believing this claim assuages some people’s guilt and allows them to have a good cry.
21 Mark Says:
October 11th, 2006 at 10:17 am
jimbobuddy, I don’t come back here, this is my blog. You came to me…

Look, this is not about the rightness or wrongness of the Iraq War, it’s about a study that is plainly contradicted by reality.

Simply put:

There is no way in hell that 500 ‘excess’ Iraqis die every day because of the U.S. invasion.

None…nada…zilch…

For this to happen every single day for 3 years and no one - Iraqi officials, opponents of the war, human rights groups, journalists - to notice is a flat impossibility…
22 Dmac Says:
October 11th, 2006 at 10:20 am
“…yesterday, they were coming from a Salon link; today, from a link from Greenwald…”

But as we all know from Greenwald’s past, his readers have become veritable experts in the use of sock puppetry to advance their spurious and unprovable “arguments.”

“It’s avoidance pure and simple.”

Sounds like another classic case of projection…

“…anyone who supported this stupid war…”

I like the grammatical flourish here - but it could stand some improvement, as it were…how about this…”anyone who supported this stoopid war is a moron, and I should know, I am edumacated!”

Let’s hope the Diebold machines suddenly start malfunctioning when this highly erudite audience goes to the polls…after they finish their one - hitters from their bongs….so they’ll probably forget all about it in the first place.
23 Sensible Mom Says:
October 11th, 2006 at 10:22 am
Iraqi Death Estimate

Burnham dontated $750 to his buddy, Les Roberts, in 2005 and 2006. You see, Les Roberts rain the the democratic primary. Birds of a feather, campaign together by releasing “studies” to get democrats elected.
24 Dominion Says:
October 11th, 2006 at 11:00 am
I wonder if anyone on this blog has even the smallest scrap of evidence that 500 excess deaths a day is impossible. I mean, I keep reading you guys saying that, but as yet, not one of you have backed it up. Arguments from incredulity are considered a logical fallicy. Just in case you were wondering.
25 Mark Says:
October 11th, 2006 at 11:05 am
Dominion, how could such a large number of deaths occur every day and no one notice? Iraq is full of people who are against the war: NGOs, human rights groups, Iraqi groups, journalists - no one has ever claimed such an incredible level of carnage.

The study is flawed…see Tom’s excellent comments above.
26 RightWinged.com Says:
October 11th, 2006 at 11:06 am
Bogus Lone Study Claims Iraq Death Count Much Higher

This is just getting ridiculous. What’s said is the unbalance in the NY Times article make the AP’s article seem fair, though the fact that it’s pumping the story so heavily and without many comments from contradicting surveys makes their…
27 The Mahablog » Adding Up the Commas Says:
October 11th, 2006 at 11:44 am
[…] Update: Glenn Greenwald checks out some rightie sites and notes (sarcasm alert) that “Bush followers have become overnight expert statisticians.” But as Glenn explains in an update, these and other righties who dismiss the study out of hand “do not actually understand what the study is examining.” They (and other of the above-linked Bush followers) seem to be laboring under the misunderstanding that the 650,000 death toll is the number of Iraqis who have died violent deaths since our invasion. That is not what the study is purporting to measure. The study is comparing the mortality rate of Iraqis during the time of our occupation (including deaths by any cause, such as disease, famine, or anything else) to the mortality rate prior to the occupation, and based on the post-invasion increased mortality rate (13.1 deaths per 1,000 persons post-invasion versus the pre-war 5.5 figure), calculates that more than 650,000 Iraqis have died during the occupation than would have died during the same time frame in the absence of the invasion. […]
28 Flopping Aces » Blog Archive » Another Bogus Study Says:
October 11th, 2006 at 11:59 am
[…] Decision ‘08 […]
29 Wake up American Says:
October 11th, 2006 at 12:35 pm
The Lancet Report

In January 2006, it was revealed that data had been fabricated in an article by the cancer researcher Jon Sudb� and 13 co-authors published in The Lancet in October 2005.
30 Texas Rainmaker » George Bush Has Personally Killed One Trillion Innocent Civilians Says:
October 11th, 2006 at 12:42 pm
[…] Others: Allah asks, “If 600,000 people have been killed, where are all the bodies?” LGF points to the timing of this, and the last, report from Lancet. Mark Coffey puts the Iraqi deaths in context of other wars. It is war, remember? 1 Comment » […]
31 CottShop Says:
October 11th, 2006 at 1:04 pm
What the yellow bellied terror coddling liberal whelps don’t understand is that people are going to die in mass numbers whether we are there to help or not. Iraq is in a fight for their lives, and they have no choice in the matter because the terrorists have made it perfectly clear that they will not tolorate freedom. The Iraqi civilians proved beyond doubt that they were willing to confront this affront to their freedoms when they went to the polls under threat of death, and colored their fingers purple. That was a bold statement that the terror apologizing left just can’t get through their thick skulls. They can’t comprehend the fact that a brutalized population has said ENOUGH!
32 only me Says:
October 11th, 2006 at 1:06 pm
Who cares? 23.3mm people to go
33 too many steves Says:
October 11th, 2006 at 1:25 pm
How in the world did us righties who dismiss the study findings arrive at “laboring under the misunderstanding that the 650,000 death toll is the number of Iraqis who have died violent deaths since our invasion.”?

Well, I for one was apparently thrown off by this in the Washington Post report cited above:

“A team of American and Iraqi epidemiologists estimates that 655,000 more people have died in Iraq since coalition forces arrived in March 2003 than would have died if the invasion had not occurred… This is about 500 unexpected violent deaths per day throughout the country.”.
34 Purple Avenger Says:
October 11th, 2006 at 1:55 pm
Who knew bloggers were such statistical experts? Oh…wait…you’re not.

I’ve taken graduate level stats courses. This “study” is pure undaulterated bullsh**.
35 Crust Says:
October 11th, 2006 at 2:09 pm
Purple Avenger writes:

I’ve taken graduate level stats courses. This “study” is pure undaulterated bullsh**.

You would be more convincing if you explained your reasoning. Have you looked at the study’s methodology and identified some flaws? (For what it’s worth, I have a PhD in pure math and have a decent understanding of statistics though I am by no means an expert.)
36 Allan Says:
October 11th, 2006 at 2:39 pm
Re: unexpected

So if Saddam were still in power, what would the expected casualty figures be?

Why didn’t the Lancet study “unexpected” deaths under the Hussein regime?

Oh, that’s because we know Saddam killed hundreds of thousands of his own people and buried them in mass graves. That’s a fact. No need to statistically sample that and fiddle with the regression co-efficients.

Is Iraq violent? Yep, way too much for anyone’s liking. What the Lancet study tries to do with its ‘unexpected’ label is mythologize about an Iraq that is yet to be, because they certainly aren’t basing their expectations on the kindness of Saddam, Uday, and Qusay.
37 rightwingprof Says:
October 11th, 2006 at 2:50 pm
Except that, as the reports of the new study show, the methodology is sound and the Lancet figures have been corroborated several times.

Take a statistics class.
38 Tom Says:
October 11th, 2006 at 3:20 pm
Saying that there is no reason for Iraqi civilian deaths to be that high is not an “argument from incredulity” - the figure is FAR higher than anyone else, pro or anti war has come up with, and there is nothing going on in Iraq that would lead anyone to believe that it was correct. For this figure to be true, one of three things would have to be true: 1) There are massive ongoing military operations going on inside Iraq that no one knows about; 2) Iraq is not only in a civil war now, it has been for the last three years, with mass bloodshed in the countryside, unreported by American, Iraqi, or International media; or 3) Ordinary crimes of violence have skyrocketed by 600-700 per cent, once again without any reporting of the fact. All of these are possible, but none are likely. As Freeman Dyson once said, “apples might start to rise tomorrow, but that proposition does not deserve equal time in physics classrooms”.

A couple more facts. First of all Baghdad contains almost one-fourth of Iraq’s population. It is probably the most violent place in the country, and the most reported on. If almost 600 people were dying every day over the last 3 years, at least a quarter of them would be in Baghdad, probably more given the violent nature of the place. That means about 150 deaths per day. The media would certainly notice this, and report on it - have we seen this? No - in the last month or two we have seen a few days where it was close, but not before, and not consistantly.

Also, most of the people who accept this study also accept the 2004 study that put Iraqi civilian casualties at 100,000. Let’s assume that that report was true. If it was, that means that over 550,000 Iraqi civilians have died in the last 2 years - a rate of 760 per day. Does anyone think that the media, Human Rights Watch, etc, would miss this?

Look, I’m not criticizing any one’s motives here. But these numbers just do not seem credible, and one does not have to be a Bush supporter to say so.
39 TallDave Says:
October 11th, 2006 at 3:30 pm
There were four problems with the initial study that even the Lancet apologists (e.g. Tim Lambert) could not deny

1) The study did not say 100,000 excess deaths. The 95% confidence interval was 8,000 to 194,000. Thus, the only statement that could be made with 95% confidence (the general standard for making statistical claims) was that at least 8,000 more people died. If you didn’t already know that, you didn’t know there had been a war.

2) The study used the three (iirc) months before the war, which were not representative of the average 83,000 people per year who were killed DIRECTLY (not excess deaths) during Saddam’s reign. Using the entirety of Saddam’s reign as a benchmark, using the same methods one would probably find the war actually saved lives.

3) There were methodological problems. The study used bootstrapping methods, did not sample the peaceful pro-American Kurdish areas despite using them in the extrapolation, and the accuracy of death certificates issued in Iraq is obviously questionable, to say the least.

4) The study’s authors did not even pretend to be objective. They specifically intended the results to serve a political agenda, and said so. That is always the mark of very bad science.

The last three problems have apparently not been addressed. And I can’t WAIT to see the confidence interval on this one.
40 TallDave Says:
October 11th, 2006 at 3:36 pm
Oh, and keep in mind the violence in largely confined to Sunni areas. Sunnis are about 1/5 of Iraq’s population, or about 5 million. To believe a tenth of them have died violently is somewhat absurd. We would also expect to find five times as many wounded, or 2.5 million, 5:1 being the rough wounded-to-killed ratio.

In other words, half of all Sunnis in Iraq would have been injured by violence in the last three years. Oh, and no one noticed.
41 endorendil Says:
October 11th, 2006 at 3:54 pm
Funny thing, you call the study “widely debunked”, but point to no resources except your own statement of the same. Where is the counter-analysis? Is it the UN report? That report was done early on, before the civil war got going (most interviews from march through may of 2004). It didn’t focus on the mortality rate at all, whereas the Lancet study focussed exclusively on that. The UN report mentions the Lancet study, but doesn’t attempt to explain the difference. It doesn’t explain how it dealt with Fallujah. It admits to using wrong weights, based on very old population estimates, which for example didn’t take into account the repopulation of the marshes, and they ended up selecting villages that turned out to be empty. The Lancet study obtained up-to-date numbers from the governates.

This is war, and hundreds of thousands of civilians have been killed in it. That was to be expected, of course, by everyone except the Bush team. And for what? To make Al-Qaeda’s recruiting work better? To prove Osama right? To make sure that Iran and North Korea realised that they really, really needed nukes if they wanted to survive? What has the US gained from the war in Iraq?
42 Mark Says:
October 11th, 2006 at 4:00 pm
Nice try, endorendil, but this thread is long enough without going off on a wide-ranging ‘whither the war’ tangent…TallDave summarizes the objections to the original study above, but really, I assume you have heard of this thing called ‘Google’ - you can’t miss it…here’s one to get you started…
43 crust Says:
October 11th, 2006 at 4:08 pm
TallDave, thanks for getting into specifics. I haven’t read the first study, so let me reply to your specifics with regards to the current study:

1) I agree with you that 8,000 from the old study is a low number; that basically says you couldn’t say anything useful with 95% confidence. The new study has a larger sample and therefore greater statistical power (95% confidence interval on excess fatalities is 390K-940K).

2) The current study considers deaths deaths that occurred between January 1, 2002, and the invasion of March 18, 2003 for the Saddam period. So it reflects excess fatalities relative to the rate in that period. As you say, that may be very different than relative to the entirety of Saddam’s reign (e.g. if you include the Iran-Iraq War and/or Saddam’s purges after Gulf War I it may well look very different).

3) The current study does include clusters from Kurdish areas (see Figure 3). Clusters were first “selected systematically by Governorate with a population proportional to size approach” (p.2). The study is not based on death certificates; rather it is based on household interviews.

4) My sense is at a minimum, the authors are pretending to be objective in their analysis. I agree the timing of publication (twice) points to a political leaning, but that doesn’t in itself show that the methodology is flawed.
44 endorendil Says:
October 11th, 2006 at 4:12 pm
TallDave, read up on confidence intervals. The only thing you can say with 95% certainty is that the number of deaths is between 8,000 and 194,000. The median was 98,000, which is the “most likely” estimate.

Also, the Lancet study used 14.6 months before the invasion, not 3 months. Please give some credable source for your 83,000 number Saddam killed per year. Even during the war years, estimates of excess deaths were less than 100,000 per year.

As to the death certificates, are you saying that people fabricated them (LOL)? They used them to cross-check reported deaths and death dates, as a verification on their result (something the UN study didn’t do).
45 crust Says:
October 11th, 2006 at 4:32 pm
If you look at Figure 3, you will notice there are two governates they didn’t sample. Here is the discussion about this (p.8):

The miscommunication that resulted in no clusters being interviewed in Duhuk and Muthanna resulted in our assuming that no excess deaths occurred in those provinces (with 5% of the population), which probably resulted in an underestimate of total deaths.

There was one other point I didn’t reply to from TallDave: he mentioned “bootstrapping”. I assume, he doesn’t mean bootstrapping in the technical statistical sense, which isn’t relevant here. I think he is just referring to the use of random sampling, but that’s inevitable (you can’t exactly take a census in the midst of a civil war or whatever you want to call the conflict) and of course we rely on random sampling in other contexts all the time (e.g. opinion polls). Apologies if I misunderstood what you’re saying here and have replied to a straw man.
46 Tom Says:
October 11th, 2006 at 5:51 pm
Anyone who has ever participated in a statistical study of this nature knows how difficult it is to obtain unambiguous data in peacetime, in a calm environment, with relatively literate, unafraid, unbiased subjects. To believe that this can be done with accuracy in the maelstrom of wartime Iraq is quite a leap of faith. If your initial data is inaccurate, it doesn’t matter how much validity (or lack thereof) your statistical manipulations have. As I noted above, this study MIGHT be accurate, but there are certain things that we would be seeing in Iraq if it were. On the whole, we are not seeing these things, certainly not over the entire three-year period covered. Most people who are familiar with Iraq, both pro and anti war, have put forward civilian casualty figures much lower than this. This report is an outlier, and unless more substantiating information is forthcoming, I am inclined to doubt.

By the way, the Primary Investigator in this study has described the Iraq War as a “fiasco”. That it may be, but those are hardly the words of a dispassionate scientist.
47 Darth Executor Says:
October 11th, 2006 at 7:12 pm
I have neither the time nor the patience to analyse the study in detail but one thing in particular caught my eye:

From the study: (available on CNN)

“Deaths were recorded only if the person dying had lived in the household continuously for three months before the event. In cases of death, additional questions were asked in order to establish the cause and circumstances of deaths (while considering family sensitivities). At the conclusion of the interview in a household where a death was reported, the interviewers were to ask for a copy of the death certificate. In 92% of instances when this was asked, a death certificate was present.”

If the methodology was as good as these gentlemen suggest, then one should expect roughly 550k death certificates issued by the government. Except the government says the study is bunk and nobody knew about these 550k confirmed deaths including all the left wing vampires looking for stupid crap to whine about.

So what happened? I can think of a couple of reasons who make the people in charge of the study look like either dumb liars or really stupid morons:

A) The people who made the study lied. I find this the most likely because most liberals I run into are stupid enough to add extra baggage they think will help that ends up contradicting the rest of their report.
B) The Iraqis surveyed lied about who died and showed the poll teams death certificates of their long dead grand grand uncle’s grandmother.
48 Dominion Says:
October 11th, 2006 at 7:55 pm
Wow, I really have to compliment you people on your big bulgy brains. I ask a simple question. Does anyone around here have a smidgen of evidence that 500 additional deaths a day is excessive. Just a tad, a thirsty man does not demand a gallon of water…a glass will do.

Instead I get the same argument that was going on before. 500 extra deaths a day is excessive because *I* can’t believe there are that many deaths. I mean, have you heard about that many deaths? Why there would be funerals from sun up to sundown. Where’s all the news about these deaths?

Sheesh! How silly can you get? Oh I forgot, this is a right wing blog, it can get very silly around here. Much MUCH sillier than this I know.

I got news for you clowns, and I have a feeling your not going to like it. Just because you have failed to hear that the death toll in Iraq is huge is only proof that you like to run around the room with your finger in your ears screaming “La La LA LA LA! I CAN’T HEAR YOU”. It is in no way, shape, or form evidence of anything.

Well other than your own stupidity. But heck it ain’t hard at all to find evidence of that.
49 Mark Says:
October 11th, 2006 at 8:12 pm
Dominion, quite persuasive…tell me, who’s more stupid? The dumb blogger like me or the guy who wastes his valuable time talking about how dumb the blogger is on his dumb blog?…
50 Never Yet Melted » That Lancet Study Says:
October 11th, 2006 at 10:30 pm
[…] Mark Coffey […]
51 Brock Says:
October 11th, 2006 at 11:51 pm
I can’t help but wonder how skeptical you all were when Bush was pushing the war, and said that Saddam had killed half a million or so of his own people. Did you demand to know his sources? Did you debate the credentials of the authors of the estimate? Did you roll your eyes and ask where all the bodies were? By the way, where are they? The only “mass graves” they found after the invasion had a few hundred people in them.

Somebody said that anybody who believes this scientifically conducted, peer-reviewed survey is a fool. What does that make the people who still support Bush, after he has proven to be wrong time after time?
52 only me Says:
October 12th, 2006 at 8:16 am
I say again, WHO CARES?
53 TallDave Says:
October 12th, 2006 at 8:29 am
Endo, apparently you don’t understand why confidence intervals are done or you wouldn’t make that statement. A wide confidence interval has consequences. A 95% confidence of interval of 8,000 to 194,000 is not the same as one from 94,000 to 106,000. In the latter case, we might say 100,000 had some statistical weight, in the former we cannot.

As to the death certificates, are you saying that people fabricated them (LOL)?

Yes! This happens all the time, even here in the U.S., for all kinds of reasons. You think insurgents who saw people’s heads off are above trying to skew a study?
54 TallDave Says:
October 12th, 2006 at 8:34 am
I assume, he doesn’t mean bootstrapping in the technical statistical sense

Yes, in fact I do. IIRC from the discussion at the time, the technique used is not uncommon, but inherently makes the sample somewhat less reliable.
55 TallDave Says:
October 12th, 2006 at 8:41 am
It was widely reported by our troops during the siege of Fallujah, for instance, that insurgents were issuing all sorts of documents claiming atrocities committed by Americans from the hospitals and municipal offices they had taken over. The deaths claimed actually exceeded the local population.

Also, remember Iraqis are often compensated for accidental injury/death or damage to property. There is a financial incentive to claim higher losses than actually happened.

And of course AQ as well as the Saddamists are well-versed in the media battle. Part of their mission, as stated in internal memos. is to discourage public support for the effort to democratize Iraq.
56 TallDave Says:
October 12th, 2006 at 8:49 am
Please give some credable source for your 83,000 number Saddam killed per year

You can check wikipedia or any other source. Saddam is conservatively estimated to have killed — directly, not by “excess deaths” — some 2 million people (who knows how much higher the excess deaths number would be using the Lancet method — 10 million? 20 million?) over the course of his 24-year reign in the Iran-Iraq war, the invasion of Kuwait, and various civil wars. UN estimates would actually be higher.

By sampling a tiny, relatively peaceful portion of Saddam’s reign, they skew the comparison and give a misleading result. It would be as though they sampled only the most peaceful month of the occupation.
57 Tom Says:
October 12th, 2006 at 8:55 am
I’d like to point out that calling people stupid or evil is not an argument. The simple fact of the matter is, these numbers are FAR higher than anyone who has followed the war in Iraq considers plausible. Iraq Body Count, very much an anti-war organization, has tabulated 48,693 deaths in Iraq - and they REALLY look for them. IBC syas that they believe the numbers are actually higher than this, but even they expressed surprise at the Lancet numbers. It simply beggars belief that a committed organization like Iraq Body Count would have somehow missed 93% of all deaths in Iraq, along with the International media and everyone else. As I noted above, I do not have a closed mind on this issue - the study MIGHT be true - but if it is, there is something going on in Iraq that all major pro and anti war organizations in Iraq have missed. Applying Occam’s Razor, it seems much more reasonable to believe that the data used was taintd, especially given the situation in Iraq at the time.
58 TallDave Says:
October 12th, 2006 at 8:58 am
To illustrate the importance of the interval, consider that you have just built a bridge. You tell everyone “I’ve done a study, and this bridge can support 100,000 tons.” A few days later, the bridge collapses under the weight of only 50,000 tons. You scratch your head and say “Well, I don’t understand. The 95% confidence interval was 8,000 to 194,000, so 100,000 was the most likely number.”

Your engineer slaps you and says “You fool! That meant there was a 95% probability it wouldn’t collapse at less than 8,000 tons! There was no statistical weight to your 100,000 number. ”

100,000 was basically a guess.
59 On the new Lancet Study at politburo diktat 2.0 Says:
October 12th, 2006 at 1:40 pm
[…] Most of the blogosphere has weighed in on the new report, which estimates that there have been 655,000 excess deaths in Iraq since April, 2003. Commentators, left and right, have weighed in predictably, uncritically accepting it and sternly dismissing it, respectively. […]
60 Chaos Says:
October 12th, 2006 at 3:43 pm
Whoever that idiot whining for proof that 500 people aren’t dying a day…

Ever heard of burden of proof? Apparently not. It is the responsibility of those who conducted and wrote the study to account for the apparently unnoticed 500 deaths a day. It is not the responsibility of the people who express skepticism to prove a negative: that 500 people aren’t dying, unnoticed, a day in Iraq.

Whatever happened to simple logic?
61 DomesticatedDog Says:
October 13th, 2006 at 12:10 am
The information you are reading on this web page is false. The public has been misinformed as to what the data shows. Please visit the following link for more accurate information.

http://thestruggleforpower.blogspot.com/
62 Edward Says:
October 13th, 2006 at 5:54 am
News just in:(no not really)
Bush’s War Blamed for 6 Billion Deaths

A new study by Joes’ Pizza and Deli(open 7AM-9PM est/4AM-6PM pst) Mon-Sat, recieved by The Lancet proves 6 billion people will die within the next 250 years from varied causes. Bush’s Iraq war blamed for detracting from solving the rising death from old age pandemic. The exhaustive study was conducted by asking 2 of the 22,000 people alive now in his town if they expected to be alive in 250 years. To be fair, 18 dead people were also asked if they were going to still be dead in 250 years. The results were extrapilated to show that the entire population of Earth is going to die.
No film at 11.
63 Tom Says:
October 14th, 2006 at 2:00 am
#25 said: Dominion, how could such a large number of deaths occur every day and no one notice? Iraq is full of people who are against the war: NGOs, human rights groups, Iraqi groups, journalists - no one has ever claimed such an incredible level of carnage…

Actually, journalists report the carnage all the time but right-wingers consistently dismiss the reports as the work of a “liberal media that never reports the good news.” Reporter Jane Arraf recently wrote: “I don’t know a single family here that hasn’t had a relative, neighbor or friend die violently. In places where there’s been all-out fighting going on, I’ve interviewed parents who buried their dead child in the yard because it was too dangerous to go to the morgue.”

#46 said: By the way, the Primary Investigator in this study has described the Iraq War as a “fiasco”. That it may be, but those are hardly the words of a dispassionate scientist…

Ummm, I think they are the words of any human being who has been paying attention to what’s happening in this war. Just because someone is a scientist does not mean they give up their right to an opinion, especially one that is so patently obvious.

When are you right-wingers going to start holding your leaders accountable? Or is that “personal responsibility” credo you espouse just meant for people other than yourselves?
64 Mark Says:
October 14th, 2006 at 10:50 am
Yes, and when will you men of straw learn to accept the wisdom of the progressives?
65 RINO Sightings: Crime and Punishment Edition » Politechnical Says:
October 16th, 2006 at 7:41 am
[…] Mark at Decision ‘08 also examines the Lancet study, and uses history as a guide to the Lancet’s authenticity. Hilarity ensues. […]