If you have "no place to go," come here!

Troop Levels: Let's Change the Frame

chicago dyke's picture

Like the actual number of Iraqi dead that "nobody" knows (I believe in the Lancet numbers of 300,000-600,000), it's hard to get a good idea of the actual numbers of people serving in Iraq. It's important for us to remind people of the total commitment: nearly 400,000.

So, let's add in the additional 100,000 troops in the region and the 100,000 contractors of whom 25,000 are estimated to be security personnel.

183,000 troops
100,000 additional troops in the region
100,000 contractors
383,000 total

How would Americans react if they knew that we had 383,000 personnel in the region?

Exactly. Let's all start speaking of our forces in numbers that reflect the truth, and not Rummy's fantasy forces of light.

No votes yet


hobson's picture
Submitted by hobson on

Maybe I missed someone posting it already or you're referring to it but the BBC has an article that says the Iraq study was "robust" and the methodology can't be faulted. More here

This is all a NY deli sandwich that is just too big to eat unless you're willing to risk a heart attack or severe indigestion, at least.

Submitted by [Please enter a... (not verified) on

I don't have the full text handy tonight and that WaPo article left the c.i.s out, but the Lancet study had ~650K as the ML estimate, with 95% confidence interval from ~300K to >900K. The methodology is solid, and the shortcomings of so few independent clusters are reflected in the large confidence interval. But there is a substantial chance that the real number is LARGER than 650K.

sorry, but I just taught cluster sampling to my graduate class...

chicago dyke's picture
Submitted by chicago dyke on

it would've proven that i'm an unserious, conspiracy theory, rabid lamb type of blogger who doesn't understand "the facts."

you'll notice how few blogs use that latter number. even left leaning ones. the "accepted" range barely extends beyond 300K, in most places.