What do you mean the closest thing we get to an agreed upon statistic is 600K? The JH study disagreed with all the other studies done, IT was the anomoly. Just because you prefer the higher number it doesn't make it correct. See, you are saying ALL other studies are wrong.

huh? Tell us this. what would be these samples? odd or otherwise. How as a "number cruncher"(is that non PC?) would you go about getting a approximate number without having the absolute number counted?

__________________
Cussing out low class inbreds isnt uninteligent, its honest

Good typing is not inteligent its dexiteritous.

Everything you just said is total bullshit

Patriotism is the last refuge of a scoundrel.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

Have you read the study? You are using the words experts without even reading what the experts wrote. You and I debated before and my argument was the Lancet was the most reliable study at that point in time with the range of deaths given. You liked the number 15,00-30,000. All death counts are an estimate and again this study while sound with solid numbers is clear that it is an estimate and most likely and underestimate. I think Bush believed it was around 15,000 and other studies only used solid body counts. There is no circling of wagons-do you still believe the Bush numbers are the closest in accuracy? This study only serves to back up the Lancet study of the range of estimate.

The reality the abstract states it's limitations and the reasons it is likely an underestimate.

Take the time to read the methodology-they were the same-cluster sampling-except the second study did not count homes that were completely destroyed and non family member reports of a person being killed and the remaining family fleeing. the JH would have counted that and the Iraq Ministry of health did not.

I read the methodology of the JH study.

When did I ever say that Bush's numbers (a couple of years ago) were accurate? Are you trying to put an argument in my mouth that was never there? Why would you do that?

Sam, the WHO took part in this report as well, you have left that out - a couple of times now.

The JH study stood alone, all other studies/surveys are more in line with this one.

Yes you are correct The Lancet range was around 150,000- 950,000. Both studies used the cluster sampling method-this most study did not include deaths by two witnesses reporting a death among a family fleeing the city and the JH/MIT study did report deaths where there was a bombed out house and two witnesses reporting a death or deaths . In addition the JH study included deaths (such as a man dying from an MI when his child was killed or a 12 year old boy dying of an appendicitis when previously there was a hospital that was now bombed out). The JH study included deaths from infrastructure destroyed in the war and included witnessed deaths by a non relative. The Iraq Ministry of Health study di not include those and admits most likely an underestimate so the range by JH with the different criteria puts it right in range.

The dfference was the polling technique and the recent study was quick to admit it was a likely underestimate.

NEJM -- Violence-Related Mortality in Iraq from 2002 to 2006

Considering the different polling used and the new study not giving a range but suggesting the likelihood of an underestimate it falls in the low range of the JH study.

I am not sure of any other methodology that could be used during this type of conflict. I think it shows the ridiculous statements put out by Bush and others of 15,000 with a high of 30,000 were people quickly accepting numbers they liked.

Either way both studies were concluded before the period in 2006 of the highest numbers of bloodshed and deaths. Same methodology -different more inclusive criteria in the JH study and evidence of an underestimate in the recent study accounts for the differences.

Okay I read that cluster sampling dealy. Okay so Norm. In your opinion is that a best technique..and why or why not?

in mine it is not, based on the limited explaination of the tehnique from where I looked it up and the example used to further explain it.

__________________
Cussing out low class inbreds isnt uninteligent, its honest

Good typing is not inteligent its dexiteritous.

Everything you just said is total bullshit

Patriotism is the last refuge of a scoundrel.


To view links or images in signatures your post count must be 10 or greater. You currently have 0 posts.

In our debate my stance was that the JH study was the closest with its range of any other study. You disagreed -the other studies at the time ranged from 15,000-30,000. Anyway I think this second study was forthcoming in its limitations and and the different criteria only serves to back up the range reported by the JH study.

The second study did not include infrastructure deaths . both studies concluded before the bloodiest time period-clearly the death toll is in the hundreds of thousands and I always believed and within the range of the JH study. I knew the 15,000-30,000 put out by the WH was so off base.

It is limited-and only an estimate. Epidemiologists seem to believe it is the best during conflicts like this-but again it is limited. The two studies are not completely comparable because the JH study included non violent deaths related to destroyed infrastructure.

Those examples are all well and good Sam, but the JH study said somewhere around 90% were dying from deaths related to things such as gunfire, and it was still quoted at over 600K. If the deaths you are referring to above were added to this study, it would still be 400K less. The upper range for this more extensive survey is still only a little over a third of that total.

What was the percentage of non-violent deaths in that study. I seem to remember it being around 10%, so the estimate was still much greater than this study.

From the JH study:

The death toll in Iraq after the U.S.-led invasion in March 2003 has been the subject of much discussion, with numbers differing by as much as a factor of 10. A national survey of a sample of population clusters that was conducted in mid-2006 estimated that an additional 654,965 persons had died during the 40 months since the U.S.-led invasion, as compared with prewar numbers. This number included 601,027 excess deaths due to violence

It is still four times greater than this new study, when only deaths directly related to violence is factored in.