655,000 Iraqis die because of war - CNN headline (Page 2)

Wrong.  There is most definitely valid science behind it.  The model used in this report has been developed based on recorded / known death rates of past conflicts.  This isn't a new method.  The interviewees in the majority of the cases produced death certificates or other proof of the reported deaths.

Most of the other/lower mortality counts are based on morgue reports, which in past conflicts have shown to be only a fraction of the actual deaths.  See Congo, East Timor, Bosnia, Rwanda, and other conflicts. 

Please, read the study before you dismiss it.  Most of your "objections" are covered in the analysis  and explanation of the data.

"Let's do a little sanity check on this number." 

Here's the sanity check on this number.

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

Regardless of your opinion of CNN's intent, read the actual report and the methods used to derive their estimate.  I'm anything, but an expert on these matters and realize that all studies have their limitations, but it would seem to be a credible study, and if so, its certainly news worthy.  So if reporting the news sways the public opinion, so be it.

If the Right Wing can use a vote to ban gay marriage to keep Bush in office, then surely its fair to use scientific studies and facts to mobilize those voters on the other side of the fence, if that's even what's going on in this case.

The "science" is in how the data sets are collected and analyzed.  Furthermore, the study states that death certificates were used to verify the accounts of the interviews.

Nobody below a Master/PhD level in Stats can answer that question.  This kind of sample is not suited to a plug and chug solution(as if there is such a thing in Stats).  The situation as it stands in Iraq makes it difficult/impossible to get accurate data.

How you feel about Iraq drives you interpretation of the data set.  Strong opinions abound on both sides without really knowing anything about statistical methods.

Unfortunately, anything that is put out now is considered "political".  Granted, they are probably right. 

If it's only takes a Headline to sway your opinion and it doesn't cause you to read further, than you are playing into the Leming (sp) trap that is today's media (both conservative and liberal).

continuing until the election:

You're going to have the right saying, "If we didn't go into Iraq, then we would have had a nuclear bomb go off in everyone's living room and it's only up to us to save the world from itself"  Without us knowing what to do then you'd all be dead.

You're going to have the left saying, "See...they can't do it.  We can and the war was wrong, GW sucks and we can change things.  We will help everyone just let us have a chance"

Unfortuntely, those extreme views, IMHO, are of the 1% on each side that don't represent the majority of us.

What I said was wrong was your assertion that it was 'just a bunch of interviews' and 'embellishments'. If 92% of the reported deaths were backed up with death certificates, I don't think that's quite people trying to blow things out of proportion. Statistical sampling and modeling is accepted as a method to produce reasonable conclusions about given subject where 100% observation is not possible. Based on the size of the sample size, you can set a reasonable confidence level ( expressed as the +- X % that you see ). The families/households polled in the study were randomly selected, with a certain number of 'households' being selected from a given region based on that regions population density.

That hardly seems to me to be made up data and analysis of the data.

From everything in this thread, news reports, and links I have gathered a couple of thoughts I would like to share:

1. The group that released these statistics is bias and trying to push an agenda.  Whether this means they purposely invalidated their research to prove a point or push their agenda remains to be seen.  IMO anyone who does not think that they have this agenda is just being niave.

2. As Mark Twain so eloquently put it "There are lies, damn lies, and then there are statistics".  Whether this study said there have been 20 deaths, 20,000 deaths or 600,000 deaths it should be taken with a grain of salt.  The method of data extrapolation is an accurate one as long as you do it right.  The group could have gone to the most violent parts of the country or looked for other signs as to who they were interviewing for these stats.  Also there could be an error in the extrapolation of the data.  For example they could have used a logrithmic factor instead of a linear or vise versa.  The point is, if you look hard enough you can "prove" or  "disprove"  anything  with statistics.

3. Why would the liberal media try to bring down the republicans before this election, we seem to be doing a pretty good job of it ourselves.  It would be much easier and fun for them if they just sat back and watched the republican party further alienate the gay community, as well the hispanics, anti-pedophilia, and rational thinking communities.

You're simply making an assertion (the research is invalid), for which you offer no proof or logical reasoning, and then state that anyone who doesn't agree with your assertion is naive. Your logical construct (argument) is invalid.

What proof do you offer that the research is invalid?

I did not mean to imply that the research was invalid because of some agenda, just that they probably had one.  Looking back point one was not a very good one due to the fact that I have no facts or assertions to their motavation fpr conducting/releasing the study.  I still believe the group has an agenda but what I believe was not really relavent to what I was trying to say.  Hope that clears it up.

What prood do you offer that the research is valid? 

From the CNN site on this report:

"The private British-based Iraq Body Count research group puts the number of civilian deaths at between 43,850 and 48,693. Those figures are based on online media counts and eyewitness accounts."

So then, we have analytical extrapolations vs eye witness accounts and media reports.  Hmmm, think I'm more inclined to believe the latter.  It would be the acme of incompetance for the media to overlook or misplace 600,000 dead bodies.

~Mike

Mike, I'm not making assertions either way. Your conclusion that the media is biased and trying to rig the election is based upon a premise that the study is invalid. So, I'm just asking you to offer some validation for your premise before even getting to whether you've reached a logical conclusion.

I don't believe either, but I'm curious - wasn't the official body count in the initial invasion 25k +?

Lets split the difference... hell, lets take your number: Yippi! we only killed 50, 000 iraqis in retribution for the 2500 THEY DID NOT KILL. Joy.

That opinion - there is bias in the media - seems to be universal.

Asserting an unsupported premise (e.g. report is unsound) to support his conclusion (media is biased) is an unsupportable logical construct.  That's my point.

You have a good point on your head Renee.

:-)

~Mike

The media isn't everywhere in Iraq, and doesn't report every death.  The Iraqi Body Count even state that they are only counting civilian deaths that are verfied/published reports by the media.  If the AP doesn't report it, does that make them less dead, even if Iraqi Body Count doesn't add it their tally? 

Yeah, so maybe they miss 10% of the civilian deaths...so maybe the number is 45,000 instead of 40,000.  But 655,000 is off the frickin' charts man!  How do you miss-count 600,000 dead people?!  Imagine about 9 major professional football stadiums filled with people....that's what we're talking about.....over half a million people.

~Mike