Monday, February 26, 2007

See Blog Title for Article Description

We are taking a break from our blog silence to bring you Statistics, and what may or may not be called "lies." (Am I enjoying this too much? possibly)

At Gateway Pundit (via ASC, via ChrisB) today you will find the quote below, asserting that now after 6 years of Bush's Presidency:
US losses in Iraq and Afghanistan today (3525) are approaching the half way mark
(3750) of the military losses during the Clinton years.
Please note boys and girls the careful wording of this statement. It is good that the wording is SO careful, because it prevents the Gateway Pundit from actually outright lieing. Interesting no? Note how he is very specific about US losses In Iraq and Afghanistan, versus all soldiers lost in the Clinton administration. It only takes a moment or two to figure out that this is not a straight comparison.

The graph he shows is here.
He gets his Iraq statistics from this link, which I think includes only casualties that occured in Iraq or Afghanistan (Or possibly leaves out the first two years of Bush's presidency? ) He also links here, which is where he gets his statistics for Clinton.
Using that PDF the total number of military deaths under just the first four years of the Bush Administration was 5187. And for the purposes of comparison, the first four years of Clinton's Administration 4621.

And just for kicks, I made my own graph, and this graph shows the total number of deaths due to "Hostile Action" during All 8 years of the clinton administration, and the first FOUR years of the Bush Administration. PLEASE NOTE, this graph is HEAVILY skewed in favor of Bush!!

Clinton on the left, bush on the Right. The Blue bar is deaths from Hostile action, (Clinton had 1, Bush 1102, ) the Purple Bar shows total deaths.

Now I'm not saying that Gateway Pundit was exactly LIEing. I am saying that his comparison of Military deaths under Bush versus under Clinton is meaningless. Beyond the fact that the statistics he uses are from two completely different sources which means they are counted differently and therefore not comparable, the two Presidents are not comparable either. Bush's presidency is and will always be colored by 9/11, and our struggle against Terrorism, and Clinton's is colored by oral sex. Which do you think kills more soldiers?

Now I can understand wanting to be able to say that Bush killed fewer soldiers than Clinton. I mean wont that really get the left's panties in a twist. Quite frankly though, it just isn't true. It's not a straight statistical comparison that he is trying to make, and it ends up sounding a lot like a lie.

UPDATE: I had a discussion with Lance in the comments at ASC that I recommend everyone read. He says he really thinks the intention of Gateway's post was to point out that the same number (or more) soldiers can be lost in peacetime, and not to really make a comparison to Clinton's military. I still think the post could be clearer about what exactly those statistics that he is comparing are measuring. For while I guess it is technically stated, it isn't very explicit and easily misinerpreted. But I'm glad the intention was not to mislead.

I hope people will forgive the somewhat harsh tone of my post. I'd edit it but then apparently I'd be harming the intergity of the blogosphere or some such nonsense. I may have gotten a tad fired up for no reason. Maybe I should go out for a job in radio.


Murdoc said...

As far as I can tell, the general tone of most of this has been "you'd be surprised at how many people died in peace time compared to the war". And, yes, that's quite surprising.

The total DEATH RATE (which factors in differing sizes of militaries in different years) is 36% higher under Bush than it was under Clinton. Or, rather, Clinton had peace and still lost only 26% fewer troops than Bush who has conquered two countries and stuck around to fight two determined insurgencies. These numbers use some estimates and extrapolations, though I believe that they're probably very close. See how I got them here.

Now, about cherry picking which numbers to count and which numbers to ignore, why is it that you only count the 1 "hostile action" death in your chart? Why not include "terrorist attacks"? Are you suggesting that the soldiers killed in Mogadishu weren't killed by hostile action? If you're going to claim others are picking and choosing which stats to count and which not to, you should make sure you aren't doing so when proving them wrong.

Shinobi said...

As I said in my comments to this post I may have somewhat overreacted to the comparison. However I still think the chart linked to in this article is misleading because it is not clear. But again, I was mistaken.

I don't dispute that this war has been low on US casualties, compared to every other war that we have fought. (I see the idea of comparing it to peacetime deaths as well, it isn't how I would chose to make my point.)

As far as your criticism of my chart, I was using the definitions that were given on the .pdf from your site. "Hosile Action" is a separate catagory from "Terrorist Attack" in that datasheet. You are correct I could have, and possibly should have used "Deaths due to Hostile Action And Terrorist Attacks" which would bring the totals to 76 (for 8 years of clinton) and 1157 for 4 years of Bush. Do you think that would have changed the overall conclusion drawn from that chart? Possibly it would have.

Murdoc said...

I simply think that being off by 7600% on the number of deaths to enemy attacks under Clinton isn't a choice one should make when knocking others for cherry picking numbers.

For the record, and you'll see it's backed up by what I've written about these very numbers for the past 11 months, I'm not a big fan of playing too many games with these numbers other than pointing out that the death totals aren't as bad as one would probably think.

Shinobi said...

If you look through my post at no point will you see me represent the number of deaths as portrayed in my graph as being the number of deaths from "Enemy Attacks." I very specifically used the term "hostile action" because that is the term used on the .pdf. I did not represent that number as being the number of soldiers killed due to enemy attacks. My intention was to be true to the definitions as portrayed on the .pdf, so that anyone who wanted to could easily look and see where those numbers are coming from.