Friday, September 14, 2007

More Than One Million Iraqis Murdered Since 2003

A poll released today by O.R.B. -- a British polling company that has tracked Iraqi public opinion since 2005 -- places the number of Iraqis murdered since the 2003 invasion at over a million.

A representative sample of 1,461 adults, aged 18 years or older, answered the following question:

How many members of your household, if any, have died as a result of the conflict in Iraq since 2003 (i.e. as a result of violence rather than a natural death such as old age)? Please note that I mean those who were actually living under your roof.
Seventy-eight percent of the respondents answered "none". Sixteen percent answered "one". Five percent answered "two". One percent answered "three". And a fraction of one percent answered more than three.

"Given that from the 2005 census there are a total of 4,050,597 households this data suggests a total of 1,220,580 deaths since the invasion in 2003."

The poll also questioned the people surveyed on what caused the deaths of their loved ones. According to the results, 48% died from a gunshot wound, 20% from the impact of a car bomb, 9% from aerial bombardment, 6% as a result of an accident and 6% from another blast/ordnance.


Reference:

O.R.B. Poll

6 comments:

Anonymous said...

Wow, 16 % with one dead is one in 6. That is like looking down your street picking every 6 house and them having a death in the family due to violence. If you add up the %'s that is almost 23, or between 1/5 or 1/4. I wonder who has looked at that ORB poll in the UK's or USA's government and if or how they can factor it that data into feasible solutions to this conflict?

The one concern I have about the poll is how random the sample was. I am assuming they are only picking towns/cities with current US military occupation but I wonder how much those figures fluctuate from town to town or city to city?

Anonymous said...

jesus

Paul Sunstone said...

Hi Rob! Welcome to the blog! Thanks for putting those figures in perspective!

Amuirin, that was my first response to those numbers, too.

Webs said...

I wrote on this topic here:

The one concern I have about the poll is how random the sample was. I am assuming they are only picking towns/cities with current US military occupation but I wonder how much those figures fluctuate from town to town or city to city?
This survey and the one the Lancet performed awhile back are both well conducted surveys. They both used a random probability sampling where you list out and number each area you are going to survey. Then you have a random number generator give you x number of randomly generated numbers. Then you survey those areas.

In most studies, the larger your "n" (sample size) the better. But there is a cut off point. Eventually as you n gets larger you will get to the point where your sample gets closer and closer to the mean. This is bad cause it will lead you to believe your findings are significant when they are not.

When surveys are conducted as the ORB one with good demographics, a decent response rate, and a large sample size, you find the correlations are very accurate. Take political polls. Nearly all political polls are insanely accurate. But their n is very high. This is because they have conducted these surveys many times and know what areas to conduct them in to get generalizable information.

But this aside, this survey and the Lancet actually had areas with current conflict and past conflict in Iraq. So some areas where a hot zone and others, not so much. This created a problem because the hot zones where harder to get information back from, since there is a war going on. But the survey was set up so each area surveyed had equal opportunity, so we know the data is randomly collected.

Some areas certainly will have low response rates, but this, just like anything else is expected to look like a normal bell curve. If anything can be said about these studies it's their estimations of the numbers, which many argue is far off. But this is the third study to come out now, and all are pretty conclusive. PEOPLE ARE DYING...

Paul Sunstone said...

Excellent comments, Webs! Thank you for shedding so much light on this poll.

Webs said...

No prob!

I'm a grad student fresh out of stats, and I had a good teacher :)