Wednesday, October 11, 2006

Another October surprise

Remember the Lancet Study that claimed 100,000 Iraqi deaths back in 2004? Well now there is another study claiming 600,000. Rick Moran has more. This is so ridiculous. Where did the coalition hide all the bodies?

27 comments:

chuck said...

Tim Blair has a long thread on this here.

chuck said...

Here's a clickable link to the file Lancet.

Morgan said...

Here is a .pdf of the study.

Morgan said...

Beat me to it, Chuck.

chuck said...

I'm expecting the Lancet to start running ads in the back pages for penis enlargement, cialis, and colonic irrigation. Hey, everyone needs to make a living.

Rick Ballard said...

Chuck,

I think they'll go with phrenology. That or Lysenkoism. After all, they're scientists.

Syl said...

Heh. It's better than that.

655,000 excess deaths. Over and above what Saddam would have accomplished.

Doesn't this like DEFINE jumping the shark?

Syl said...

And, if the Left accepts the Lancet study of this time in 2004, that means that...voila...there had to have been 555,000 excess deaths..

IN JUST TWO YEARS.

That's um like 23,125 per month!

Anonymous said...

Rick:

Hasn't Lysenkoist genetics been accepted wholesale by the NEA already?

This Lancet stuff is all so boring and predictable.

Unknown said...

Rick Moran's got some lefties in comments who do not doubt this for one moment. Makes perfect sense to them. In fact we are gullible if we doubt it. Yep.

Like I said wehre are the bodies? I have seen pictures of Hirohima and Dresden and Iwo Jima and Antietam etc. These numbers would dwarf those casualties.

Morgan said...

Some quick analysis and asociated questions:

Estimated number of deaths due to airstrikes (based on Lancet data):

March '03-April '04: 12,720
May '04-May '05: 27,560
June '05-June '06: 42,400

Car bombs killed 63,600 people from June '05-June '06.

The questions:

Can coalition airstrikes really have been 3.5 times as deadly in '05-'06 as in the first year post-invasion, and have killed more than 40,000 people in the last year? I know it was reported late last year that the number of strikes had risen significantly, so the number of strikes must be available from the military. Alternatively, has there been a trend toward hitting more densely populated areas?

How many car bombs were detonated in Iraq from June '05-June '06? How many people would they have to have killed, on average, for the total to have been 63,600 people? I would think that the number of car bombings would generally be accurately reported. If they killed (on average) 10 people each (which seems high based on the most recent car bombings reported by Iraq Body Count), that would mean 6,360 car bombings in 13 months, or 16 every day. Is there any data consistent with that? If not, there is reason to believe that the sample is biased.

I think these are the real questions. I believe the methodology of the study from a statistical standpoint is relatively sound - sampling appears random (though apparently skewed toward urban dwellers) - but the collection of data may have been biased toward those with deaths in the family (for example, those who had experienced deaths may have been more likely to opt into the survey).

There could also have been deliberate malfeasance, of course. Les Roberts did run for Congress as a Democrat, and who knows who Riyadh Lafta hangs around with. But even if the results were skewed deliberately, the fingerprints should be seen in the data, so there's no need to impugn people's motives.

Luther said...

I noticed several particularly egregious sentences.

"constitutes a humanitarian emergency" (p6) and, wait for it, their solution... "an independent international body to monitor compliance with the Geneva Conventions and other humanitarian standards" (p8). I am sure the Baathist's and Quida will be extremely cooperative with that scenario.

"proportion of violent deaths attributable to coalition forces "could be" (my quotes) conservative estimates"(p6). You either know or you don't, "could be" they are also overestimated. Seems to me that "could be" does not belong in a scientific journal.

"Coalition forces have been reported as targeting all men of military age" (p7) Heck of an assumption, especially considering their listed references.

But, for me, this is perhaps the worst...

"From January 2002, until the invasion in 2003, virtually all deaths in Iraq were from non-violent causes".

The sheer audacity of that statement is beyond belief.

Morgan said...

One partial answer:

"In the first four months of this year, insurgents detonated 284 car bombs against civilian, police or military targets, according to U.S. military statistics. During the same period last year, insurgents used 349 car bombs."

The higher number yields about 1,134 over 13 months. How accurate is the count, I wonder? Could violent deaths really be overstated by a factor of 6 (or more)?

Rick Ballard said...

"Could violent deaths really be overstated by a factor of 6 (or more)?"

Could the BLS really have "honestly" missed 810,000 jobs?

How do I lie to thee?

Let me miscount the ways.

Morgan said...

Another partial answer. I compiled all 265 unambiguously identified "car bombing" reports with death tallies from iraqbodycount.net - not a complete list, maybe, but presumably representative (except that all of them must have at least one fatality).

Average deaths from a car bombing = 5.38 (bootstrapped 95% CI 4.57-6.27) using the "reported maximum" figures, so this is the high end.

We need 11,819 car bombings in which at least one person died to account for 63,600 deaths. 11,819 bombings in 13 months versus 1,134 reported car bombings computed by extrapolating reports from an especially bad few months.

Could deaths from car bombs be overstated by a factor of 10? Or did they sample a neighborhood in which 30 people died in one or two events?

Similarly, average deaths from 54 air raids, air strikes, air to surface missiles and associated gunfire = 18.44 (bootstrapped 95% CI (9.70-26.17). We need 2,300 deadly air strikes (177 per month).

Blegging, now - anyone know of a military source tracking number of airstrikes?

Unknown said...

Morgan:

According to this 91,700 people have died in car bombs since the war began.

When and where?

As for airstiikes, we are talking about a full third of total casualties. That would be over 200,000 people. Think of the fire bombing of Dresden, it was a fraction of this number.

I don't give a damn what methodology this guy supposedly used it is insane on its face. And if it is true, where are they? They must be buried in Iraq, I assume they had names and families and communities.

They can find mass graves from a decade ago, surely they can find these people?

Unknown said...

Morgan:

I saw a comment at Time Blair and the guy made a point taht from 1940-45 the Allies flew something like 300,000 sorties against Germany and killed 569,000 German civilians. I don't remember exactly, but it took years and years and a lot of deliberate targetting of cities to come up with those numbers, and people noticed. The press noticed, it was hard to miss.

Rick Ballard said...

Morgan,

Total sorties flown are over 422,000 but that has no relation to the number of strikes. I've looked for an hour and I just don't find anything usable.

Personally, I'd like to see a study of the graveyard registers from the cemeteries in Baghdad. It's not like they don't exist.

Morgan said...

Rick,

Thanks for your efforts. I've seen media reports for certain months stating that the numbers came from the military, but nothing comprehensive.

In some sense, I feel like I'm trying to prove that certain aspects of a Dali painting are out of proportion to its other elements, in order to show that it isn't really a photograph.

Except that, in this case, a whole bunch of people do think it's a photograph.

terrye:

It's absurd on its face, isn't it.

Unknown said...

Morgan:

yes, it is. It reminds me of the ten thousand dead in New Orleans and people eating their young in desperation during Katrina. When you are dealing with people who can and will say anything it is hard to know how to respond.

Unknown said...

knucklehead:

How is it that Human Rights Watch sees no reason to question the findings?

Let them prove it, damn it, they made the claim.

Rick Ballard said...

What is weird is that all that it would take to disprove this idiocy is for someone to check the cemetery registers - or just go out and count burials for two weeks around Baghdad. Hell, find out if they've doubled the number of grave diggers on staff. Muslims don't go for cremation and they have strict rules on burial - they don't ship bodies around much at all (except for Green Helmet).

Someone at CENTCOM should pull their head out and send out some counters. This ain't brain surgery - if they're only planting 100 per day in Baghdad then the Hopkins guys are liars, if they're planting 250 a day then they're telling the truth.

Unknown said...

Rick:

I don't think it would matter, they would just call the Centcom liars.

But why didn't the people who did the study do that? Why didn't they go out and check cemeteries? We have no way of knowing that these people ever lived, much less that they died.

BTW, even 250 a day in Baghdad would not do it, unless there were several hundred dying somewhere else too.

Rick Ballard said...

Terrye,

250 is the right number for Baghdad and if it's disproven for Baghdad, it's disproven. In thinking about it, we probably have imagery on a daily basis that could be used as proof. I read that they're shipping unclaimed bodies down to Najaf for burial so an adjustment would have to be made but it still ain't brain surgery.

Unknown said...

Rick:

Even if they proved the numbers for Baghdad it does not make the rest of the numbers right. And I do not think they can prove the numbers for Baghdad.

Syl said...

From the New York Times:

The most recent United Nations figure, 3,009 Iraqis killed in violence across the country in August, was compiled by statistics from Baghdad’s central morgue, and from hospitals and morgues countrywide. It assumes a daily rate of about 97.

The figure is not exhaustive. A police official at Yarmouk Hospital in Baghdad who spoke on the condition of anonymity said he had seen nationwide counts provided to the hospital that indicated as many as 200 people a day were dying.


THe UN's figure is from morgues and hospitals. And even rumored nationwide counts (by whom?) puts the max at 200/day countrywide.

As I said, this report jumped the shark.

Luther said...

Numbers are easy. Many are confused by them. As I ineptly attempted to illustrate above, go for the narrative, the meat that tries to massage the sausage. The political agenda is quite obvious. The authors served their purpose. They got people to talk about numbers, and not intent.