100,000 Iraqi Civilians have died in current war

[Schrodinger's Cat]

I thought the following article was very interesting, especially since it appears in the highly prestigous Chronicle of Higher Education:
http://chronicle.com/temp/email.php?...5m3h7noo5ikert

The stance of Bush and the Pentagon seems to be "if we don't know about it, we don't have to care." As the article itself notes, look at the widespread outpouring of sympathy and assistance for the victims of the Tsunami's versus the public's almost total indifference to the news of the deaths estimated by this study.

[Torrere]

At least we liberated them from the hardships of living in war-torn Iraq.

[xoxoxoBruce]

look at the widespread outpouring of sympathy and assistance for the victims of the Tsunami's versus the public's almost total indifference to the news of the deaths estimated by this study

Sympathy and assistance for the living.

[Griff]

On the eve of a contentious presidential election -- fought in part over U.S. policy on Iraq -- many American newspapers and television news programs ignored the study or buried reports about it far from the top headlines.

I don't recall the major party candidates differing substantially on the war. A better Dem candidate could have argued US policy but with his voting record Kerry couldn't.

[Undertoad]

Well Madame Albright took credit for 500,000 deaths during sanctions so perhaps we are seeing an improvement here.

The paper that they published carried some caveats. For instance, the researchers admitted that many of the dead might have been combatants. They also acknowledged that the true number of deaths could fall anywhere within a range of 8,000 to 194,000, a function of the researchers' having extrapolated their survey to a country of 25 million.

Why don't they just make the number up? It will have the same impact.

[Schrodinger's Cat]

Originally Posted by Undertoad

Well Madame Albright took credit for 500,000 deaths during sanctions so perhaps we are seeing an improvement here.



Why don't they just make the number up? It will have the same impact.

You probably didn't read the entire article. It went on to describe the sampling techniques used, and how they were well within the criterea for sociological/scientific studies. The article also noted that this same researcher was the one who made the estimates of the numbers killed in the Rwandan Civil War and that these estimates were widely accepted as being accurate. The 100,000 figure is within a 95% confidence level of being the correct one.

[Undertoad]

I read the entire article and was unimpressed. No I do not believe the number. Partly because despite the fact that they found somebody to say the guy's methodology was sound, to me it seems utterly ludicrous. But mostly because without a massive coverup, it would be impossible to kill that many people without having bodies littering the landscape that somebody would notice. Even the non-rebuilt hospitals would be full.

[Beestie]

Why would someone use "sampling techniques" instead of an actual body count? Besides, statistics should only be applied to recurring events as they tend to zero in on the liklihood of an outcome over a number of observations.

I think, for example, there is more than a 99% confidence interval that the winning lottery ticket was not, in fact, a winner. Over time, the "99% of the time the ticket will not be a winner" conclusion will be proven correct. But you can't take an average and apply it to a single observation. At best, its meaningless and at worst, its very misleading.

Its been my experience that when stats meets politics, hold your nose.

[Troubleshooter]

"Lies, damn lies, and statistics."

[flippant]

Yeah- It makes more sense not to have a count at all- especially when the numbers aren't well liked.

Flippant

[Schrodinger's Cat]

Originally Posted by Beestie

Why would someone use "sampling techniques" instead of an actual body count? Besides, statistics should only be applied to recurring events as they tend to zero in on the liklihood of an outcome over a number of observations.

I think, for example, there is more than a 99% confidence interval that the winning lottery ticket was not, in fact, a winner. Over time, the "99% of the time the ticket will not be a winner" conclusion will be proven correct. But you can't take an average and apply it to a single observation. At best, its meaningless and at worst, its very misleading.

Its been my experience that when stats meets politics, hold your nose.

The pentagon refuses to do a civilian body count. A body count which comes from an Iraqi source would be automatically suspect. Thus, it was left to a highly respected researcher from an American university to investigate the true number of civilian deaths in the war.

In case you haven't noticed, private Americans are not exactly welcome these days in Iraq. Thus, westerners cannot just show up at Iraqi funeral homes and ask them how many war dead they are burying today. Even the Iraqi's who helped gather the data were frightened if it got out that they were working for an American researcher. Death certificates were requested (and supplied 63% of the time) of those households which answered positively to having a war inflicted death among its members in the past year.

I think possibly you are misunderstanding the principles of statistics. I"ll buy that lottery ticket which has a 95% chance of being a winner since you don't want it.

As for where are all the dead bodies? They buried them.

[Undertoad]

Under normal combat situations there will be two injured for every one dead. Under bombing situations the ratio could be higher. Where are the 200,000 injured? Where are the hospitals full of crying patients with missing limbs? Where are the photo ops for the insurgents? How could such a thing be covered up?

[Happy Monkey]

Originally Posted by Beestie

I think, for example, there is more than a 99% confidence interval that the winning lottery ticket was not, in fact, a winner. Over time, the "99% of the time the ticket will not be a winner" conclusion will be proven correct. But you can't take an average and apply it to a single observation. At best, its meaningless and at worst, its very misleading.

Interesting. So you think this survey is probably the "lottery winner" survey that falls outside the 95% confidence area, and moreover the real number is below, not above, that number.

I guess that's the attitude of most people buying lottery tickets. Though the average lottery ticket has a much lower cost.

[Happy Monkey]

Originally Posted by Undertoad

Under normal combat situations there will be two injured for every one dead. Under bombing situations the ratio could be higher.

Under "no medical facilities" situations the ratio could be lower.

[russotto]

To reiterate what UT said earlier, the actual figure, taking into account only sampling error, was 101,000 plus or minus 93,000. When your error bars are of the same magnitude as your data points, you don't have data; you have junk.