“They resorted to mathematics”

Lindsay Beyerstein of Majikthise rounds up some disputation of the latest Lancet study [.pdf] of excess Iraqi deaths since the 2003 invasion from some of our comrades on the right, who will let nothing deter them from their commitment to bravely defending the powerful. My favorite quotation was this:

When the statistics announced by hospitals and military here, or even by the UN, did not satisfy their lust for more deaths, they resorted to mathematics to get a fake number that satisfies their sadistic urges.

Couldn’t leave well enough alone, could they, the calculator-using scoundrels?


13 thoughts on ““They resorted to mathematics”

  1. Ananth

    Well I am glad that you have decided to blog on something in current events. Maybe something NK next. Anyway, as I am sure that the numbers of Iraqi deaths is more than the 50,000 that the Iraqi and US governments give, this statistic is really bullshit. They take a statistical survey, and try to verify what the surveyents say by looking for death certiifcates. Apparently if they can find 80% as verified then the assume that the reports are accurate and can use them to extrapoliate, so it leads to this question.

    Secondly, their pre-invasion are questionable at best. Are we really to assume that during the era of Saddam, the Lancet group were able to accurately poll areas of Iraq? Do you think they really got an accurate count of how many people Saddam had killed? Also, some of their results at a glance seem to defy logic. They suggest that the non voilent mortality rate is almost 30% higher now than it was post Saddam invasion. There is no way that availibility of health care, drugs, and quality of health care has gone DOWN since ending the embargo and having Saddam use the oil profits to line the pockets of French officials and build palaces.

    Finally, they apply a range of total excessive deaths and try to provide a 95% confidence level, which again may be an accepted practice for calculating mortality rates, but doesn’t mean its accurate.

    I can’t speak for the rest of the right wingers, but I have no problem with math. The problem I have is something we in the programming business like to refer to as Garbage in Garbage Out.

  2. Steve

    We all knew this was coming, but it’s good to have our own specimen in our little greenhouse. This is how it works for postmodern conservatism: you don’t actually have to know what you’re talking about; in fact, it’s better if you don’t. You must assume that all history and human learning up to this point is worthless, and then you just throw questions at every detail you can think of. How do you account for this? This can’t be right. That goes against common sense. Peer-reviewed medical journals have a well-known liberal bias. And so on. For instance, the very first sentence of the article Ananth links to is wrong. The Lancet study does not say that there have been as many as 655,000 excess deaths. They say that there’s a 95% certainty that the number of excess deaths is between 426,369 and 793,663. There have been as many as 793,663. As few as 426,396. The 655,000 is the number that’s most likely. The question raised in the article, about why no one has 500,000 death certificates, is also answered in the Lancet article, as you might expect, unless you find it necessary to make excuses for people who like to casually ruin countries in unprovoked wars. But who wants to pay attention? There are atrocities to ignore.

    Ananth, your entire comment amounts to, “I have no idea how this study works. I disbelieve!”

    Lindsay links to her own post on the study in the above-referenced post, and that post, in turn, links to this explanation. You’re not being skeptical, you’re being contrarian.

  3. Ananth

    First of all, you don’t know what you are talking about any less than I don’t know what I am talking about, so don’t get all high faluting like you
    1) Have a PHD in Statistics
    2) Carefully reviewed and addressed every thing in the Survey

    Secondly, before I even wrote a word down, I did do research. I looked at other estimates that were linked at iraqibodycount.org that showed comparasions to their admittedly low estitmate to other credible estimates, which find that the Lancet survey is an outlyer, with an incredibly large data range for a 95% confidence level. This survey that is being compared to Lancet was done by or for UN or it is being linked to by the UN, so don’t flame me for not being technically correct, and it has a much larger data set and much tighter range for the 95% CI. Having an 87% range between your low estimate and your high estimate is fairly large range. Does it completely invalidate the survey? No. But is this survey the end all proof of just how atrocious Iraq has been? No.

    So the guy I linked to had the first sentence wrong. I understand what a 95% confidence level is, even if he doesn’t. But the question he has is still valid. I don’t see any where they address the issue of the actual number of death certificates issued. They discuss how they feel that getting 92% confirmation rate was acceptable based on mitigating reasons as to death certifcates were not issued. If there is something in the PDF. please point it out to me.

    I never said I have no idea how the study works. I understand it completely. Its not hard. Its sampling, and thats all. I just find some the numbers a bit hard to believe, and these numbers taken in context and extrapoliated out what they translate to (IE there are 438 deaths by voilence per day since 2004, so the amount is being under reported is only 5% percent, and I know they said the highest accuracy besided bosnia was 20% and as lows 5% in guatamala, but I think that their is much more reporting on the ground in Iraq than any other war in history) Prewar Iraq had a lower rate of voilent deaths than Sweden? There are lots of things like that.

    Also two other points, which you may or may not pay attention too, Omar who is quoted above, actually lives in Iraq, and was pissed off at what he percieved to be a propaganda action to descredit the worthiness of what is going on in His country, and that Lindsay so nicely labeled him a coward for it.

  4. Tom

    Ananth, the guy you link to pulls some statement completely out of his ass saying that 80% of deaths should be able to be verified via government documents. Wha?

    I’d like to see him point me to the statisticians who came up with that figure. And tell me if I could’ve found 4.8 million death certificates down at the county office in Berlin circa 1945 (a quick googling shows that records exist for about 5.3% of deaths in Auschwitz).

    Also, the non-violent death rate included cases when the cause of death could not be verified, i.e., they disappeared. So if its higher than one might expect it was because the Lancet was erring on the side of the rightwing nutjobs, and assuming that ZERO of the people who died of unknown causes were killed as a result of coalition presence.

    And it’s very easy to get Saddam-era figures, by asking questions like, “before 2003, how many people in your family died per year?”.

  5. Ananth

    Actually, his 80% percent number was off, but if you read the report, the Lancet people got documentation for 92% of the 547 or deaths they had reported to them. That is how the decided whether or not the people reporting the deaths were telling th truth or not.

    So if you are going to use sampling and then projections to determine how many deaths from voilence there were shouldn’t then all the things that you used in your projection be projected out. So even going by their 95% CI should’t there be between something like 300k to 630k death certificates on file that state voilence as the cause of death? How would it be only the people they surveyed actually had death certificates but then the projections wouldn’t?

  6. Ananth

    One minor correction, his 80% number is not off now that I think about it. He said if 80% were verifible by death certificates, which is about the percentage that was verified via death certificates (87% were asked for death certificates, of which 92% were verified with certificates, which gives 80.04% of the deaths as being verified with certificates).

  7. Eric

    It’s all but impossible to get an accurate count in a war-torn country. The nit-picking (ie- getting at the truth) is important, but I’d like to fill the role of the person who steps back for a second. Without conceding anything either way – is the war somehow ok if there have ONLY been 100,000 unnecessary deaths??? Did the holocaust kill 10 million or 20 million? It matters, but when you’ve already reached fucking awful shouldn’t everyone at least be able to admit that much.

  8. Ananth

    The number abosolutely doesn’t mean a thing to me, and it doesn’t change my opinion on the war. If the cause is worth it, then no matter how high the number gets, it should still be worth it. If the cause is not worth it, then anything over 0 is unacceptable. The only time actual numbers makes a difference in whether an endeavor was worth it or not is if the endevour was specifically designed to stop a certain number of killings, in which case mortality deltas are the only thing that matters.

    Also, I don’t accept your characterization as unnecessary . Unfortunate, tragic, heartbreaking sure, but unnecessary implies that they could have been avoided and the desired results still could be attained. I am not sure if the desired outcome will be attained or not, but I am not willing to say the whole thing is lost yet. See me again in 10 years, and I think well have a better idea.

    Now, that being said, I know their are people out there that look at the numbers and to them that justifies things. This survey provided a large number so that people like our host can point to this large number and say, look at how much of a mistake this has been. If the number was smaller and more accurate, they lose some of their shock appeal and ability to sway more people to their side. So absolutely, I am going to nitpick on the accuracy of something like this when people are holding it up as if were handed down from the mount.

    Also, I really take issue with Steve’s earlier characterization of my opinion being
    “I have no idea how this study works. I disbelieve!” You are acting like a person of strong faith, who when questioned about certain aspects of it fails to address the question itself, and simply says that you need to have faith. I would really like at some point for you to actually address something I ask, with you know, an answer, other than there is another dumb conservative, who just doesn’t get science and math, and if only they were more educated on the issue, then they would see the world exactly as it is , which is how I see it.

  9. Steve

    The number abosolutely doesn’t mean a thing to me, and it doesn’t change my opinion on the war. If the cause is worth it, then no matter how high the number gets, it should still be worth it. If the cause is not worth it, then anything over 0 is unacceptable.

    Ananth, that’s monstrous.

  10. Ananth

    Yeah, monstrous. You are taking my comment about whether the actual number makes any difference to me as to if this is was worth it or not and you are going to construe it as if I am advocating not caring about civilian casualities or making every attempt to keep them to a minimum. I mean, if the civilian casuality count was 5000, would you have been ok with the Iraq war? If the civilian body count in Afghanistan were 200K as opposed to whatever it is, would it make it less legitmate to you?

    Look, I don’t expect, or even want everyone to agree with me. But just because I have a different view on things doesn’t make me any less of a thoughtful or understanding person than you. I mean any kind of disagreement with an issue reduces the discussion to how souless and uncaring conservatives are.I guess its too bad that I am brown, since it removes the not caring about lives if they are brown arrow from your quiver of remarks of indignation.

  11. Steve

    But just because I have a different view on things doesn’t make me any less of a thoughtful or understanding person than you.

    That is correct. It is not just because you have a different view on things that you are thoughtless and obtuse. It is because your different view causes you to characterize an epidemiological study, using the accpected methods for getting an accurate number in war and disaster situations like this, published in maybe the most prestigious medical journal in the world after having been reviewed by more than the usual number of experts (four; my understanding is that it’s usually two or three) as “handed down from the mount” as if it was received in an LSD vision or some shit, while having no problem seriously considering every ignorant quibble pulled out of the ass of a basement warrior with a jizz-stained copy of An Army of Davids and a bloody red-white-and-blue axe to grind. The quibbles of the ignorant are, after all, more convenient to your prejudices.

    Fuck being brown, you’re rich and American; the thousands of dead are not American, poor, and no one you know or ever have to personally worry about. So yeah, pretty easy to decide that we should kill as many of them as we need to in the name of our good idea. (And in case there’s any possible way that it would be unclear, I used the words “good idea” with bitter, bitter irony.)

  12. Ananth

    So every quibble is ignorant? Then why don’t you address them point by point, as the bastion of knowledge you seem to be , because I can’t find any real counterpoints aside from, this is a peer reviewed scientific survey, don’t question the projection. Also did I question the methods of study? Did I question the scientific methods? But in the end, when you get a result that does not jive with the numbers and reports that exist you have to ask why the discrepency? Either the reporting is off or the projection. This is not off by a percentage point here or there. This is something of orders of magnitude. Is iraq body count questioning the Lancet study because it has a bloody red-white-and-blue axe to grind? Don’t prescribe prejuidice as the reason why I question the number that the Lancet report generates. Yes, I found it high, and I took a deeper look as to what the numbers would actually mean. All these inconsistincies means that there was an error or bias in the sample data, or that people are under reporting deaths, everyone is failing to keep proper records by orders of magnitude, the number of wounded, dead, etc. Why don’t you explain to me how that all these things are lined up to make the projection correct?

    If anyone is prejudiced in this, it would be you. You so badly want to show everyone how much of a fiasco the Iraq invasion has been, that anyone who dares questions your assertions to prove this must be idealogically driven, obtuse, thoughtless, coward who must bask in the death of civilians.

Comments are closed.