Cost of War - 655k Iraqi Dead - DPChallenge Forums

Canon EF 24-70mm f/2.8L USM

10/11/2006 01:29:37 PM · #26

Originally posted by hokie:

As long as we challenge the political and religous dogma of any group of people...their very existence is threatened and ..I believe..leads to violence.

I agree with you wholeheartedly that intolerance is the root of pretty much all of the issues. IMO, further complicating the issue is that tolerance is mistaken for appeasement. However, while you may be right, I would prefer not to think that it is inevitable. Idealism, perhaps, but without it there is not much hope.

violinist123

Canon EOS-6D

10/11/2006 01:33:24 PM · #27

Originally posted by LoudDog:

Come on man. You say this is just made up for spin purposes? 1 month before mid-term elections? You're talking crazy.

I'm curious how a death toll grows to 655,000 before someone makes a big deal out of it. You'd think 10,000 or 100,000 or 250,000 would be milestones worthy of a cnn headline. I guess it was just an oversight on someone's part....

Personally I don't care if it's 650 or 650,000. The Iraqis are no different than any of their neighbors in that they kill each other over foolish reasons with or without an American policy to point at as an excuse. Take the frigging oil fields, get my gas prices down, and start redeploying troops to Seoul.

hokie

Nikon D2X

Nikon AF-S DX Zoom-Nikkor 17-55mm f/2.8G IF-ED

10/11/2006 01:37:14 PM · #28

Originally posted by ddpNikon:

Is this just an acceptable tragedy of living in a country that has been taken over by terrorist forces? How many people, Iraqi's or others, would have died from the terrorist leaders if the US never went over there?

The mistake made by people who support the war in Iraq is that they talk about the here and now. That is also the mistake made by people who oppose the war.

The vision longterm is we face the expansion of Islamic extremism in the Mid-East before we face it in Europe and other parts of the world.

There has been an unprecedented amount of muslims being imported into Europe (Particularly France and Spain) to cope with aging populations and lack of low income workers.

As a matter of fact, The European experience seems to be mimicking the United States experience with immigration from Mexico. Cultural clashes and the changes that are occuring to established regional identities. Of course, the United States doesn't have to deal with religous extremists or petroleum funded terrorism from the Mexicans. If we did we probably wouldn't be over in the Mid-East but would have 300,000 soldiers in south Texas and along the southern borders!

There is the answer! Have Mexico become a religous extrmeist state...sorta like Texas or Oklahoma....and fund them with petroleum money. If only Osama really knew how to get the United States out of the Mid-East!

Message edited by author 2006-10-11 13:39:33.

10/11/2006 01:39:17 PM · #29

Originally posted by LoudDog:

I know statistics. It's a big part of my job and was a focus in my schooling. You can't talk to only 5% of the population, count 629 deaths and extrapolate out to 655,000 deaths, unless you read the book "how to lie with statistics" and are trying to come up with a bogus number to make people's head spin.

That is 500 deaths per day since the start of the war! Where are all those bodies going? What are the causes of these 500 deaths per day?

The number is BS.

The report is here:

//www.thelancet.com/webfiles/images/journals/lancet/s0140673606694919.pdf

It seems reasonably thorough to me. I read 18 months ago in the UK's The Independent paper about the crisis in Baghdad's morgues: they could not keep up with the consistent flow of bodies, 200-300% up from the pre-war period.

The statistics and cross checking indicates that the survey has given a result consistent with information from other sources. It also makes the point that there are almost no war situations in history where any more than 20% of deaths get reported.

10/11/2006 01:52:42 PM · #30

Originally posted by hokie:

There has been an unprecedented amount of muslims being imported into Europe (Particularly France and Spain) to cope with aging populations and lack of low income workers.

I am not sure that this is really true - we see a lot of Eastern European immigration to the West, which is probably largely Christian into largely atheistic communities. There are a lot of Turkish immigrants in Germany, and limited immigration from North Africa to Spain (although that is a difficult trip compared to other overland routes).

Tension has arisen over the last few years out of second or third generation immigrants. The UK received a lot of immigrants from Pakistan and India following the break up of the commonwealth and those territories. France received a lot of immigrants following the break up of its empire. They and their descendants are now part of our multicultural societies (largely, reasonably well integrated in the UK).

The tension is not brought by radicals arriving in this country, but by local people who are feeling increasingly disenfranchised by local and international issues.

10/11/2006 01:57:26 PM · #31

Originally posted by routerguy666:

You'd think 10,000 or 100,000 or 250,000 would be milestones worthy of a cnn headline. I guess it was just an oversight on someone's part....

They reported on 100,000:

//edition.cnn.com/2004/WORLD/meast/10/29/iraq.deaths/

Tokina AF 12-24mm f/4.0 AT-X Pro DX

10/11/2006 02:02:19 PM · #32

Originally posted by theSaj:

Originally posted by "legalbeagle":
In my opinion, the spread of advanced weaponry among less advanced nations would be best controlled by a strong demonstration of the application of the rule of law and respect for international law.

Hey, I agree...now if you can just get them to agree. So far, most of our attempts to do so have failed miserably.

Which part of starting an war in Iraq in contravention of international law demonstrates the strong application of the rule of law?

With his individual moralistic approach to international politics, GWB has the exact opposite policies to those that I promote.

milo655321

Nikon D70

10/11/2006 02:10:37 PM · #33

Originally posted by theSaj:

As I've been following this issue for nearly 20 years. And since 9-11 was only 5 yrs ago. I answer this with a RESOUNDING "No!"

Bali? Madrid? London? Jordan? Isreal?

Spork99

OM System OM-1

Canon EF 100mm f/2.8 USM Macro

10/11/2006 02:38:14 PM · #34

It's all England's fault, Winston Churchill specifically, for creating Iraq after WWI. Putting together groups of people (Kurds, Arabs and Turkomen) who share a great animosity for each other under a common governance is not a recipe for success.

Most people think Iraq is a creation of the people that lived there years ago and that it has existed for centuries. They don't realize that it's a construct of the post-WWI British government.

Judith Polakoff

Nikon D7000

Nikon AF-S DX NIKKOR 18-200mm f/3.5-5.6G ED VR II

10/11/2006 02:57:38 PM · #35

Originally posted by theSaj:

I am open to solutions and advice, thoughts on better tactics and policies.

Reform of our energy policy. An obvious possible solution, but we seldom talk about it. I don't know how we can claim to be fighting a "war on terror" while we're actively financing the war machine of the "enemy."

Here's an interesting article on one possible approach:

Toward a Global Energy Transition

As an aside, it was reported yesterday that over 1 million Iraqis have left the country entirely due to the increasing sectarian violence. So much for exporting democracy...

10/11/2006 04:25:33 PM · #36

Originally posted by legalbeagle:

Originally posted by LoudDog:
I know statistics. It's a big part of my job and was a focus in my schooling. You can't talk to only 5% of the population, count 629 deaths and extrapolate out to 655,000 deaths, unless you read the book "how to lie with statistics" and are trying to come up with a bogus number to make people's head spin.

That is 500 deaths per day since the start of the war! Where are all those bodies going? What are the causes of these 500 deaths per day?

The number is BS.

The report is here:

//www.thelancet.com/webfiles/images/journals/lancet/s0140673606694919.pdf

It seems reasonably thorough to me. I read 18 months ago in the UK's The Independent paper about the crisis in Baghdad's morgues: they could not keep up with the consistent flow of bodies, 200-300% up from the pre-war period.

The statistics and cross checking indicates that the survey has given a result consistent with information from other sources. It also makes the point that there are almost no war situations in history where any more than 20% of deaths get reported.

I read the report.

The number includes and is probably comprised mostly of terrorist, military members and police officers as the about 70% of that 655,000 is adult males that died a violent death. Plus, the sample was far from random, it was only taken in densly populated areas, they used the cluster method, their observation period for the pre-war was too short (why???), they missed 2 of the 18 areas...

Also, per their report, violent deaths only accounted for 2% of deaths in Iraq pre war! (55% post war which is where most of the 655K comes from).

10/11/2006 04:51:42 PM · #37

Also,

Per the report the pre war crude death rate in Iraq was 5.5. Per this chart, the US is in the low 8's and the 5.5 puts Iraq as one of the safest places to live!

//www.infoplease.com/ipa/A0004395.html

10/12/2006 05:33:13 AM · #38

Originally posted by LoudDog:

Look here:

//www.bartleby.com/151/fields/25.html

Lots more 2003 death rates - Bahrain 3.99, Egypt is at 5.35, Jordan at 2.62, Kuwait 2.62, Iran 5.54, Iraq 5.84, Libya 3.49, Morocco 5.78, Oman 3.97, Qatar 4.43, Saudi 5.79, Syria 5.04, Tunisia 5.02, West Bank 4.62.

There are a lot of reasons why mortality rates may be high or low - you should know, if you deal with statistics, that your "gut feeling" is easily mislead. A low mortality rate does not equal safety.

PS the Iraq statistic of 5.84 corresponds with the early 2003 stat of 5.8 given at the bottom of page 4 of the Lancet 2006 report.

Message edited by author 2006-10-12 05:47:51.

10/12/2006 05:42:55 AM · #39

Originally posted by LoudDog:

The number includes and is probably comprised mostly of terrorist, military members and police officers as the about 70% of that 655,000 is adult males that died a violent death.

The study concludes that a large proportion are civilians. over 30% (though declining) due to coalition troops. The remainder for a variety of reasons, but still deaths that exceed the pre-war background levels. I am not sure that it really matters whether people die at the muzzle of a coalition soldier, or as a consequence of coalition failure to institute peace. They are still deaths consequential to coalition invasion.

Originally posted by LoudDog:

Plus, the sample was far from random, it was only taken in densly populated areas, they used the cluster method, their observation period for the pre-war was too short (why???), they missed 2 of the 18 areas...

Why was the pre-war analysis period too short? It was sufficient to give an accurate reflection of the pre-war death rate.

The study indicates that it was random and does not indicate that it was only carried out in densely populated areas. Populated, yes, but the survey was conducted of clusters of people proportionate to population density.

They explain that mistakes by a survey group (not unreasonable mistakes given the difficult circumstances) led to data being collected in the wrong region, and therefore being excluded (in order to protect the objectivity of the data). This is an exampe of rigour, not failure.

RonB

Canon EOS-300D Rebel

10/12/2006 01:29:05 PM · #40

Originally posted by legalbeagle:

Originally posted by LoudDog:
Plus, the sample was far from random, it was only taken in densly populated areas, they used the cluster method, their observation period for the pre-war was too short (why???), they missed 2 of the 18 areas...

Why was the pre-war analysis period too short? It was sufficient to give an accurate reflection of the pre-war death rate.

I hardly think that ANY pre-war statistics are valid, since many people just "disappeared" with no record of whether they were "relocated" or just dumped into mass graves without "death certificates" being filled out.

Originally posted by legalbeagle:

The study indicates that it was random and does not indicate that it was only carried out in densely populated areas. Populated, yes, but the survey was conducted of clusters of people proportionate to population density.

They explain that mistakes by a survey group (not unreasonable mistakes given the difficult circumstances) led to data being collected in the wrong region, and therefore being excluded (in order to protect the objectivity of the data). This is an exampe of rigour, not failure.

From the actual document you reference:

Originally posted by TheLancetDocument:

"The third stage consisted of random selection of a main street within the administrative unit from a list of all main streets. A residential street was then randomly selected from a list of residential streets crossing the main street. On the residential street, houses were numbered and a start household was randomly selected. From this start household, the team proceeded to the adjacent residence until 40 households were surveyed. For this study, a household was defined as a unit that ate together, and had a separate entrance from the street or a separate apartment entrance."

So, far from being "random" the "selected" households were CLUSTERED ( as they indicate in the summary ), but ONLY on residential streets that CROSSED A MAIN STREET. And, I'm curious, were single person households considered, since a single man/single woman cannot logically "eat together" by himself/herself?

Now, if I was a domestic terrorist ( aka "insurgent", aka "freedom fighter" ) where do you think I would detonate my explosive device to achieve maximum effect? An area on a quiet residential street away from a main street? or actually on or near a main street?

Is it really any wonder that the number of fatalities due to violence is greater near a main street?

WHY, for the sake of statistics, did they not just randomly select ANY residential street in the target areas? I believe that the answer is quite clear - at least to me it is.

Obviously, when you extrapolate the higher fatality rate that occurs near major streets to an entire country you get a higher number than you would get if you extrapolated the lower fatality rate that would be found in a mix of areas both near and not near major streets.

Interesting how the media overlook the bias built into the method.

Message edited by author 2006-10-12 13:33:03.

10/12/2006 02:51:20 PM · #41

Originally posted by legalbeagle:

Why was the pre-war analysis period too short? It was sufficient to give an accurate reflection of the pre-war death rate.

Lets say I was comparing my DPC scores before and after I converted 100% to DSLR and lets say I've submitted to 100 challenges and only my last 9 photos are post change. Also, for the sake of argument lets say I have not improved a lick since my first challenge entry. If I took the average of the 9 post change photos and only compared them to the average of the last 3 pre change photos will it give me a fair comparison? Maybe. However, if any of those last three photos were above or below my normal score it would throw my comparisin way off.

When doing a comparison of current to past, the past is your baseline. It is important that your baseline be accurate for any comparison to be accurate so the baseline period looked at should actually be longer then then what it is being compared to. Here the "was" time period was about 1/3 the length of the "is" time period. Thus, the margin for error on the "was" number is much higher then the margin of error on the "is" number because it was a much smaller sample.

I find it very odd that they cut off their observation period at 2002? If 2002 was an unusually safe or unusually unsafe year the results would be totally skewed. I'm too busy to look it up today, but I'd be curious as to what the crude death rates were in Iraq in each year prior to 2002. If they are consistent with 2002 then the survey may be pretty accurate. If they are higher then 2002 then I'd have to guess they picked 2002 as a cut off to make their war death toll seem larger.

10/13/2006 05:43:39 AM · #42

Originally posted by RonB:

I don't think that pre-war Iraq was teeming with Stalinist disappearances in the way that you suggest: it happened, and people did disappear, but hardly in the massive scale that would be necessary to impact significantly on the crude death rate, and this would be picked up by the survey in any case. The mass killings tended towards specific incidents, such as Fallujah.

Originally posted by RonB:

From the actual document you reference:
Originally posted by TheLancetDocument:
"The third stage consisted of random selection of a main street within the administrative unit from a list of all main streets. A residential street was then randomly selected from a list of residential streets crossing the main street. On the residential street, houses were numbered and a start household was randomly selected. From this start household, the team proceeded to the adjacent residence until 40 households were surveyed. For this study, a household was defined as a unit that ate together, and had a separate entrance from the street or a separate apartment entrance."

So, far from being "random" the "selected" households were CLUSTERED ( as they indicate in the summary ), but ONLY on residential streets that CROSSED A MAIN STREET. And, I'm curious, were single person households considered, since a single man/single woman cannot logically "eat together" by himself/herself?

In my experience, the middle east is full of many thousands of villages, consisting of a few streets crossing a main road. Just because there is a main road, it does not mean that there is a major population centre: it can be a very small village.

I have no idea whether single person households were considered, but again, in my experience, families traditionally tend to be very close and stay living in the same building with each other even after marrying etc. This is moreso in villages than cities, but still very common. I am not sure why you think that including or not including single person households might skew the survey results.

Originally posted by RonB:

Now, if I was a domestic terrorist ( aka "insurgent", aka "freedom fighter" ) where do you think I would detonate my explosive device to achieve maximum effect? An area on a quiet residential street away from a main street? or actually on or near a main street? Is it really any wonder that the number of fatalities due to violence is greater near a main street?

This presupposes that people are dying because of explosive devices - people are being shot, most often. I don't think that shootings are limited to main streets.

Originally posted by RonB:

Obviously, when you extrapolate the higher fatality rate that occurs near major streets to an entire country you get a higher number than you would get if you extrapolated the lower fatality rate that would be found in a mix of areas both near and not near major streets. Interesting how the media overlook the bias built into the method.

I think that this is a pretty desperate criticism of the survey. The analysis has been peer reviewed, which will have been much more rigourous than your sniping at the methodology.

If you were to investigate remote smallholdings living away from main streets (which must represent a small proportion of the population) and find (as I am sure you would) that they are less affected, and if this has not been taken into account in the Lancet analysis (it is not clear exactly what factors have been taken into account to balance the figures already - but it is the kind of factor that could already have been taken into account) then you might manage to reduce the figure by a few percent.

The figure might be unpalatable, and you can reject it, but in the absence of any formal studies (which the Lancet report calls for) it is the best information available as to the likely consequences.

I am not sure how, if the figure could be brought down to, say, 500,000, that would make any significant difference to the issue at hand: the coalition invasion has resulted in many hundreds of thousands of deaths (and god only knows how many more serious injuries). Given the absence of substantive justification, pressure must be maintained to hold the executive to account.

Canon EF 24-70mm f/2.8L USM

10/13/2006 05:50:48 AM · #43

Originally posted by LoudDog:

I find it very odd that they cut off their observation period at 2002? If 2002 was an unusually safe or unusually unsafe year the results would be totally skewed. I'm too busy to look it up today, but I'd be curious as to what the crude death rates were in Iraq in each year prior to 2002. If they are consistent with 2002 then the survey may be pretty accurate. If they are higher then 2002 then I'd have to guess they picked 2002 as a cut off to make their war death toll seem larger.

I accept that this would change the results. I note from the sources linked already that death tolls do not tend to vary wildly, but gradually change over time (absent war/famine etc - not the case here). So it would be surprising if 2002 was very different to other years.

It would also be surprising if the survey was trying to maximise the number: there do not appear to be any political objectives that it is trying to achieve, other than establishing the need for a formal study. Obvious manipulation of the stats would risk being observed and held up as a reason for rubbishing the study; those responsible would have a lot more to lose by manipulation than they would have to gain (unlike the detractors from the study, who have an extremely strong political interest in keeping death tolls as low as possible).

violinist123

Canon EOS-6D

10/13/2006 08:59:04 AM · #44

Everyone outside of the guys who put the study together dispute the numbers, including people on both sides of the political fence. The guy in charge of the study admits to getting it fast tracked through the publication process. Why the hurry? Again, with elections in a month it does raise an eyebrow.

Get the oil flowing, redploy to Seoul. Body counts are irrelevant.

RonB

Canon EOS-300D Rebel

10/13/2006 01:39:14 PM · #45

Originally posted by legalbeagle:

Originally posted by RonB:
I hardly think that ANY pre-war statistics are valid, since many people just "disappeared" with no record of whether they were "relocated" or just dumped into mass graves without "death certificates" being filled out.

I don't think that pre-war Iraq was teeming with Stalinist disappearances in the way that you suggest: it happened, and people did disappear, but hardly in the massive scale that would be necessary to impact significantly on the crude death rate, and this would be picked up by the survey in any case. The mass killings tended towards specific incidents, such as Fallujah.

Human Rights Watch reported in November, 2004 ( ref: here):

"During the past thirty years, the government of Saddam Hussein engaged in three wars and numerous campaigns to repress the Kurdish, Shi`a, and Marsh Arab populations, resulting in the disappearanceâ��and, most certainly, the deaths--of between 250,000 and 290,000 people. By February 2004, the Combined Forensic Team of the Coalition Provisional Authority (CPA) had collected information on 259 mass graves in Iraq. Of these, U.S. military criminal investigation teams had conducted preliminary assessments of fifty-five sites by February 2004."

I consider that estimate to be significant. And that's just the estimate of those who "disappeared", not those whose deaths were carried out "in the open". As for the survey, I cannot believe that only 2 people died a violent death in all of Iraq during the 14-month pre-war period covered by the survey, yet that is the conclusion drawn from the survey. The survey says that NO pre-war deaths were the result of gunshot. Zero pre-war violent deaths due to gunshot in the households surveyed, extrapolated to include the entire country, equals...well, zero. Do you really believe that not a single person in Iraq died as a result of a gunshot from January, 2002 until March, 2003? I sure don't.

Originally posted by legalbeagle:

Originally posted by RonB:
From the actual document you reference:
Originally posted by TheLancetDocument:
"The third stage consisted of random selection of a main street within the administrative unit from a list of all main streets. A residential street was then randomly selected from a list of residential streets crossing the main street. On the residential street, houses were numbered and a start household was randomly selected. From this start household, the team proceeded to the adjacent residence until 40 households were surveyed. For this study, a household was defined as a unit that ate together, and had a separate entrance from the street or a separate apartment entrance."

So, far from being "random" the "selected" households were CLUSTERED ( as they indicate in the summary ), but ONLY on residential streets that CROSSED A MAIN STREET. And, I'm curious, were single person households considered, since a single man/single woman cannot logically "eat together" by himself/herself?

In my experience, the middle east is full of many thousands of villages, consisting of a few streets crossing a main road. Just because there is a main road, it does not mean that there is a major population centre: it can be a very small village.

12 out of the 47 clusters ( 0ver 25% ) were in Baghdad - hardly a village of a just a few streets. No detail is given as to which cities/villages were selected in large regions with populations heavily concentrated in just a few major cities.

Originally posted by legalbeagle:

I have no idea whether single person households were considered, but again, in my experience, families traditionally tend to be very close and stay living in the same building with each other even after marrying etc. This is moreso in villages than cities, but still very common. I am not sure why you think that including or not including single person households might skew the survey results.

Because the death rate was computed based on the number of people in the households at the start of the "recall period" + the number of births - the number of deaths - the number remaining. If, in a given household there were 2 to start with, 3 births, no deaths, and 4 moved away, leaving one, those 3 births wouldn't be counted in computing the death rate.
Oh, and fwiw, according to the Lancet, separation of combatant from non-combatant deaths during interviews was not attempted.

Originally posted by legalbeagle:

Originally posted by RonB:
Now, if I was a domestic terrorist ( aka "insurgent", aka "freedom fighter" ) where do you think I would detonate my explosive device to achieve maximum effect? An area on a quiet residential street away from a main street? or actually on or near a main street? Is it really any wonder that the number of fatalities due to violence is greater near a main street?

This presupposes that people are dying because of explosive devices - people are being shot, most often. I don't think that shootings are limited to main streets.

OK. Let me rephrase the query: Of all those who die under violent circumstances, where would you suppose most of them lived? on a quiet residential street away from a main street or near a main street? In a rural village? or in a major city ( like Baghdad )?

Originally posted by legalbeagle:

Originally posted by RonB:
Obviously, when you extrapolate the higher fatality rate that occurs near major streets to an entire country you get a higher number than you would get if you extrapolated the lower fatality rate that would be found in a mix of areas both near and not near major streets. Interesting how the media overlook the bias built into the method.

I think that this is a pretty desperate criticism of the survey. The analysis has been peer reviewed, which will have been much more rigourous than your sniping at the methodology.

The paper by the Korean scientist who claimed to have created embryonic human stem cells was also peer reviewed before publication. Peer review doesn't mean a thing when the peers doing the reviewing don't have all of the FACTS. For example, the Lancet report does not break down the survey by city or town or village - only by region. Perhaps the "random" selection of cluster inadvertantly resulted in more cities being selected than villages. Just because the city/villages were selected at RANDOM does not guarantee that they were representative of the entire region in which they exist.

Originally posted by legalbeagle:

If you were to investigate remote smallholdings living away from main streets (which must represent a small proportion of the population) and find (as I am sure you would) that they are less affected, and if this has not been taken into account in the Lancet analysis (it is not clear exactly what factors have been taken into account to balance the figures already - but it is the kind of factor that could already have been taken into account) then you might manage to reduce the figure by a few percent.

Or by a large percent

Originally posted by legalbeagle:

The figure might be unpalatable, and you can reject it, but in the absence of any formal studies (which the Lancet report calls for) it is the best information available as to the likely consequences.

It is information, and it may be the best information that is AVAILABLE, but availability doesn't guarantee quality.

Originally posted by legalbeagle:

I am not sure how, if the figure could be brought down to, say, 500,000, that would make any significant difference to the issue at hand: the coalition invasion has resulted in many hundreds of thousands of deaths (and god only knows how many more serious injuries). Given the absence of substantive justification, pressure must be maintained to hold the executive to account.

Your arguments, as those of the Lancet, seem to be based on a position that there is a cause / effect relationship between the invasion and the deaths - yet you have provided no proof that such a relationship exists. There is no evidence, apart from speculation, that without the cause ( the invasion ) the effect ( the deaths ) would NOT have occurred just the same. Deaths by the thousands were occurring before the war ( though the survey data do not support what everyone knows to be true ) - what makes you believe that they wouldn't have continued or even escalated? Obviously, I'm not speaking of the combat deaths, as those ARE related to the invasion - but that number is in the tens of thousands, not the hundreds of thousands.

FWIW, here are some interesting extrapolations that I could make based on the same data published by the Lancet ( assuming that there really IS a cause/effect relationship ):

The war has resulted in a significant drop in deaths due to cancer. The post-war rate is only 75% of the pre-war rate.
The war has resulted in a significant drop in deaths due to chronic illness. The post-war rate is only 64% of the pre-war rate.
The war has resulted in a significant drop in deaths due to old age. The post-war rate is only 81% of the pre-war rate.
The war has resulted in a significant drop in men's deaths due to non-violent accidents. The post-war rate is only 62% of the pre-war rate.

In closing, be it known that I am NOT trying to minimize the loss of human life in this war. I'm merely questioning the validity of the survey.

10/13/2006 02:24:21 PM · #46

Originally posted by RonB:

The war has resulted in a significant drop in deaths due to cancer. The post-war rate is only 75% of the pre-war rate.

I noticed that as well, but I didn't mention it because it could be argued that the 7 more people that should have died of cancer in the sample, based on the baseline (which is flawed because the time period is too short...) could have been amongst the additional people that died a violent death. But it does show how easy it is to lie with statistics.

Their math and logic is correct, but their sampling method and sample periods are flawed. Thus it's difficult to believe their estimate without any actual proof.

Nikon AF Micro-Nikkor 60mm f2.8D

10/15/2006 02:25:53 PM · #47

Originally posted by RonB:

"During the past thirty years, the government of Saddam Hussein engaged in three wars and numerous campaigns to repress the Kurdish, Shi`a, and Marsh Arab populations, resulting in the disappearanceâ��and, most certainly, the deaths--of between 250,000 and 290,000 people. By February 2004, the Combined Forensic Team of the Coalition Provisional Authority (CPA) had collected information on 259 mass graves in Iraq. Of these, U.S. military criminal investigation teams had conducted preliminary assessments of fifty-five sites by February 2004."

I consider that estimate to be significant. And that's just the estimate of those who "disappeared", not those whose deaths were carried out "in the open". As for the survey, I cannot believe that only 2 people died a violent death in all of Iraq during the 14-month pre-war period covered by the survey, yet that is the conclusion drawn from the survey. The survey says that NO pre-war deaths were the result of gunshot. Zero pre-war violent deaths due to gunshot in the households surveyed, extrapolated to include the entire country, equals...well, zero. Do you really believe that not a single person in Iraq died as a result of a gunshot from January, 2002 until March, 2003? I sure don't.

This is exactly my point: 250k or so over 30 years. The majority of that was before the Gulf war. Very little in the last few years (statistically) - even on a straight line calculation, you would only expect to see 20 or so people on the survey.

The survey comes with a margin of error. While I am sure that people died from gunshot wounds in Iraq in 2002, it is not unbelievable that relatively few people did so as to be statistically insignificant and within that margin of error.

I know that some people have a view of the countries in the area as uncivilised and teeming with the very worst ravages of dictatorship. From my experience, this just is not the case. The area is/was fundamentally safe (sometimes as a consequence of over-policing).

Originally posted by RonB:

The clusters are located within *provinces* in proportion to the population count. Baghdad city is in Baghdad province, but not all of Baghdad province is composed of city. It is a small populous province - more study is required to calculate stats for the province because there are more people there and the risk of a disproportionate survey is therefore greater. Again, you mistake rigour for fault (taking more surveys in Baghdad to get a better picture, rather than accidentally doing just one survey and it happenning to be in an area that has suffered well or badly).

Originally posted by RonB:

Not at all - the death rate consists of how many people have died. It has nothing to do with birth rate.

The Lancet explains that separation of causes of death would be dangerous and unlikely to be recordable with any degree of accuracy - it could not be done. Why criticise it for this?

Originally posted by RonB:

No - the whole science of statistics is that it is guaranteed to be inaccurate - but within a predictable margin of error. Here, c.95% confidence of success, but with a number of noted factors that could affect the validity of the results. 12,000 is a big survey.

Originally posted by RonB:

Yes - it is an assumption, but a reasonable one, that the deaths are related to the invasion and subsequent collapse of law and order. There could be other reasons, but they are very well hidden.

Originally posted by RonB:

FWIW, here are some interesting extrapolations that I could make based on the same data published by the Lancet ( assuming that there really IS a cause/effect relationship ):

The war has resulted in a significant drop in deaths due to cancer. The post-war rate is only 75% of the pre-war rate.
The war has resulted in a significant drop in deaths due to chronic illness. The post-war rate is only 64% of the pre-war rate.
The war has resulted in a significant drop in deaths due to old age. The post-war rate is only 81% of the pre-war rate.
The war has resulted in a significant drop in men's deaths due to non-violent accidents. The post-war rate is only 62% of the pre-war rate.

In closing, be it known that I am NOT trying to minimize the loss of human life in this war. I'm merely questioning the validity of the survey.

There are possible reasons for this (lower chances of diagnosis, sick people being moved, people at risk dying of other causes etc). I don't pretend that the survey has all of the answers (nor does it), but that the cost (which is often overlooked by those focussing only on iraqbodycount) should be acknowledged. Even if you halve the figures, more people died following the coalition invasion than Saddam Hussein had managed to kill in 30 years.

Message edited by author 2006-10-15 14:28:48.

David Ey

Nikon D80

10/15/2006 03:43:18 PM · #48

They actually prefer to chop off the head.

dacrazyrn

Nikon D300

Nikon AF-S DX NIKKOR 18-200mm f/3.5-5.6G ED VR II

10/15/2006 06:25:30 PM · #49

Studies margin of error
+/- 655,000

yanko