Lies, Damn Lies and Statistics

I work in a research unit, and long-time readers know how I feel about innumeracy, so I am more attuned than most to the subtle ways people who want to lie to you can manipulate the numbers to make their arguments sound valid.

But there's nothing subtle about this one. This came up and smacked me between the eyes.

At the VA website there is a page that shows the number of wounded in Iraq. For several months this figure was up in the 50,000 range, and then on January 10 it plummeted, without explanation, to 21,649.

So what's up with that?

Well, it turns out that the lower figure is the number of strictly combat-related injuries. The rest of them are due to illness, vehicle crashes and other accidents, etc., a number that the VA insists isn't relevant. See the full story over at VA Watchdog

Never mind that it helps to hide the true cost of war (for example, those injuries or illnesses likely would not have happened if the individual hadn't been in Iraq to begin with). And the cost, to be borne for several generations, of caring for the veterans who have been disabled in those incidents. Remember, if it occurred while you were in the service, it is by definition a service-connected disability.

It wouldn't surprise me a bit if the VA came up with a new definition of service-connected disability. Look at the money it would save if disabled veterans were helped only for combat wounds and nothing else.

Jesus, I'd better shut up. I might be giving the evil fuckers in the BFEE some ideas and lose my own VA disability payments...