Plurality Now Favor Leaving Iraq

[MattyT]

I'm saying that it's IMPOSSIBLE to establish a solid "THERE ARE EXACTLY 600,000 PEOPLE KILLED IN THIS CONFLICT" when issues such as politics are involved. It's in the best interest of all the involved parties to inflate casaulties, especially to the foreign press. It makes their cause more likeable and makes the other guy look more diabolical.

802647[/snapback]

From an AP article....

 

The researchers, reflecting the inherent uncertainties in such extrapolations, said they were 95 percent certain that the real number lay somewhere between 392,979 and 942,636 deaths.

 

If there is any real news from this study it may be in the sentence that followed in the article.

 

Even the smaller figure is almost eight times the estimate some others have derived.

 

General Casey thinks that the numbers in the study are way too high and that the highest number he has seen is 50,000.

 

So basically the study is probably way too high while the gov'ts numbers are probably way too low and the truth is somewhere in between.

[chicot]

Which isn't what they did.  Had they, I'd give it more credence.  They went around and polled people and families ("So...know anyone who's dead?"), and statistically extrapolated that to an overall death toll for the country. 

 

I've seen that method used before.  When they eventually compare it to actual body counts, I've never seen it end up accurate.

802659[/snapback]

 

Talking to people who actually handled the body will undersample. A great many bodies are not taken to the morgue and are buried soon after the person died. And they didn't just say "So ... know anyone who's dead?", they asked for the death certificate which they got in 92% of the cases.

[TheMadCap]

Even if your figure for casulties is accurate (I haven't read the article, nor do I intend to do so) how can one determine who killed who? How can you differentiate between a victim of insurgent violence and Allied kills?

[UConn James]

Even if your figure for casulties is accurate (I haven't read the article, nor do I intend to do so) how can one determine who killed who? How can you differentiate between a victim of insurgent violence and Allied kills?

803142[/snapback]

 

I don't think they are differentiating, just estimating the # of people who've died as a result of the war. And insurgent violence is part of the war.

[Bungee Jumper]

Even if your figure for casulties is accurate (I haven't read the article, nor do I intend to do so) how can one determine who killed who? How can you differentiate between a victim of insurgent violence and Allied kills?

803142[/snapback]

 

They don't. It's strictly a statistical measure: "The pre-war Iraqi death rate was 5.5 per 1000, now it's 30 per 1000, therefore 600000 people were killed because of the invasion and occupation." That's due to all causes...breakdowns in sanitation and health care, violence, car accidents, suicides.

 

And having read the article this afternoon (yeah, I picked up a copy of the Lancet and read it. I figured at least one person in this thread should know what the !@#$ they were talking about. ) and done the math...I still disagree with the methodology. Looks to me like, based on their data, the number should be closer to 120k or so. Like I said earlier, I've never seen statistical estimates match up with the actual body count; it usually greatly overestimates actual deaths (the classic "One million Iraqi children die each year because of sanctions is a great example.)

[TheMadCap]

They don't.  It's strictly a statistical measure: "The pre-war Iraqi death rate was 5.5 per 1000, now it's 30 per 1000, therefore 600000 people were killed because of the invasion and occupation."  That's due to all causes...breakdowns in sanitation and health care, violence, car accidents, suicides.

 

And having read the article this afternoon (yeah, I picked up a copy of the Lancet and read it.  I figured at least one person in this thread should know what the !@#$ they were talking about.  ) and done the math...I still disagree with the methodology.  Looks to me like, based on their data, the number should be closer to 120k or so.  Like I said earlier, I've never seen statistical estimates match up with the actual body count; it usually greatly overestimates actual deaths (the classic "One million Iraqi children die each year because of sanctions is a great example.)

803228[/snapback]

 

 

Well then that study is misleading. It would be like saying 1.2 million Americans crap thier drawers everyday, then not mentioning that most of them are infants and old folks...

[Bungee Jumper]

Well then that study is misleading. It would be like saying 1.2 million Americans crap thier drawers everyday, then not mentioning that most of them are infants and old folks...

803427[/snapback]

 

Except that they were able to say, statistically, that a certain number of deaths were due to gunfire. Of course, the standard deviation of their study is so wide, you can use their numbers to say that 400k people have been killed in Iraq, 500k of them by gunfire...

 

But that's not even one of my complaints. My biggest complaint is that the demographic makeup of Iraq - calculated using the same methodology - don't bear out their findings. Death by violence is generally age-dependent to some degree; 20-40 year old males are more likely to be shot than a three month old baby or an 80-year old grandmother, for example. Doing some quick back-of-the-envelope calculations using the study's numbers and some conservative estimating...the 20-40 year old male demographic in Iraq should have seen a 20% contraction in the past three years. But the demographic estimates - using the same methodology as the Lancet study - say otherwise. That makes me wonder if the methodology is applicable to begin with.

 

I also think it's a rather silly point to base policy on. Pro- or anti-war, either group should have learned from Vietnam that measuring success or failure by body counts is preternaturally stupid and useless.