[quote=Prime95;98564]If I understand your question, no.[/quote]Then to what possible categories of cause of excess deaths, if not disease, illness, or accident (note my amendment), [I]do[/I] you attribute the 1-4x difference?

[quote]Where the UN report only counts violent deaths and the Lancet report counts excess deaths from all causes including violence I would not find a Lancet number of 2-5x surprising.[/quote]To what possible causes of excess deaths other than violence, disease, illness, or accident do you attribute the 1-4x difference, such that you wouldn't find it surprising?

Another try:

To what possible categories of causes would you attribute the difference between (A) the UN report count of only violent deaths and (B) the Lancet report count of excess deaths from all causes including violence?

Or:

What sorts of causes would you imagine to be included in "all causes", but not "violence"?

[QUOTE=cheesehead;98566]To what possible causes other than violence, disease, illness, or accident do you attribute the 1-4x difference, such that you wouldn't find it surprising?[/QUOTE]

I believed your original question asked what amount of undercounting would seem plausible? My answer is up to 80% undercount would not seem unreasonable.

I don't understand what your current line of questioning is getting at.

I found the comparison of Iraq death rates to the U.S. (as opposed to other Iraq studies) interesting. Here are my findings after a short amount of online research and a quick attempt to normalize the numbers so that we can compare apples to apples.

U.S. murder rate for 2005: 5.6 per 100,000
New Orleans murder rate: 57 per 100,000
Iraqbodycount.org is 62,000 over about 4 years. That's a yearly rate of 60 per 100,000.
U.N 2006 violent death estimate for Iraq 34,000. Iraq population is about 26 million. This equals 131 deaths per 100,000.
The Lancet survey estimates 655,000 excess deaths since 2003. Of those 92% were due to violence. Thus, Lancet finding is an [I]increase[/I] of 650000 * 0.92 / 3 years / 26 million population or 773 per 100,000. Assuming a pre-2003 violent death rate of about 25, Lancet is saying the Iraq yearly death rate has been about 800 per 100,000.

I've been revising my post #133 while you were posting #134-135, George. Maybe the last line in it now makes sense?

- - -

[quote=Prime95]I believed your original question asked what amount of undercounting would seem plausible? My answer is up to 80% undercount would not seem unreasonable.[/quote]Then I asked, or intended to ask, whether you attributed the up-to-80% (1-4x difference) to disease and illness (and accident, as I added), since the UN count was for only violent deaths.

Or were you saying that the UN count was undercounted by 50-80%, compared to a (hypothetical) more accurate Lancet figure which was only 2-5x the UN figure (rather than 20x), for violent deaths only?

[QUOTE=cheesehead;98571]Or were you saying that the UN count was undercounted by 50-80%, compared to a (hypothetical) more accurate Lancet figure which was only 2-5x the UN figure (rather than 20x), for violent deaths only?[/QUOTE]

Yes. The difference introduced by Lancet also counting excess non-violent deaths is small.

In other words, I'd like to believe the U.N. wouldn't undertake a study that captures less than 20% of the real number - less than that would be grossly incompetent.

[QUOTE=Prime95;98570]U.S. murder rate for 2005: 5.6 per 100,000... Lancet is saying the Iraq yearly death rate has been about 800 per 100,000.[/QUOTE]

I'd like to refine the above to be a tad more accurate for comparing apples to apples. The Lancet study was over 3.2 years - not 3.0. The Lancet and IBC study are over the entire war time frame. Since violence is escalating one would expect a higher death rate during 2006. To better compare these two studies to the U.N. 2006-only study, I've arbitrarily assumed that the IBC count for 2006 was 30% of the 4 year total rather than 25%. And the Lancet study 2006 count is 35% of the 3.2 year total rather than 31.25%. This has only a minor effect on the numbers I originally posted.

U.S. murder rate for 2005: 5.6 per 100,000
New Orleans murder rate: 57 per 100,000
Iraqbodycount.org is 62,000 over 4 years. Estimated 2006 death rate is 72 per 100,000.
U.N 2006 violent death estimate for Iraq 34,000. This equals 131 deaths per 100,000.
The Lancet survey 2006 death rate is 650000 * 0.92 / 26 million * 0.35 + 25 = 830 per 100,000.

If my math and assumptions are right, this makes the Lancet rate 6.33 times the U.N. rate and 11.5 times the IBC rate. Note that the Lancet study does not include data from the second half of 2006 which was pretty nasty.

An aside!
 

:wink:

Well whatever the numbers I claim that George W made the same mistake his father made-- He did not pull out soon enough ! :grin:

Mally :coffee:

Maybe that's called "faith-based contraception" :grin:

Dead Right!
 

:smile:

Well Anything but a concentration camp- Like Gautanamo ?

Mally :coffee:

Getting back on track ...
 

[quote=masser;98563]Lastly, a kind request: how does one put a quote within a quote on this forum?[/quote]The tags used to denote a quote are (without the internal spaces that I insert here to keep the parser from treating the following as actual markup commands)

[ quote ]

(quoted text)

[ /quote ],

which produces, when I remove the spaces between brackets and commands:

[quote]

(quoted text)

[/quote]

- - - - -

To put a quote within a quote, just nest a set of [ quote ] [ /quote ] within another set, like:

[ quote ]

first part of outer quote's text

[ quote ]
inner quote's text
[ /quote ]

second part of outer quote's text

[ /quote ],

which produces, when I remove the spaces between brackets and commands:

[quote]

first part of outer quote's text

[quote]
inner quote's text
[/quote]

second part of outer quote's text

[/quote]

and so on ...

Getting back on track ...
 

masser,

My comments in post #132 were neither pedantic nor intended to shut you up. Instead they were intended to persuade you to move out of rhetorical-device mode and into straightforward-discussion mode.

I myself often used rhetorical devices in the past, but right now I'm more interested (at least in serious discussions like this one) in facilitating communication than in one-upping or in other games (in the transactional analysis sense). So when I see rhetoric interfering with sincere communication, I say so.

I genuinely want to see your answers to questions I asked you.

- - -

But I'm not perfect, so when someone else sees me using rhetoric to interfere with sincere communication in a serious discussion, he/she is invited to point it out.