I am not a statistician. The statistics class I took in college was easily in my top two least favorite courses^[a].

Still, the graphs in this article about the stimulus spending don’t seem to show what the article says they show:

The stimulus bill “includes help for those hardest hit by our economic crisis,” President Obama promised when he signed the bill into law on Feb. 17. “As a whole, this plan will help poor and working Americans.”

But FOXNews.com has analyzed data tracking how the stimulus money is being given out across the 50 states and the District of Columbia, and it has found a perverse pattern: the states hardest hit by the recession received the least money. States with higher bankruptcy, foreclosure and unemployment rates got less money. And higher income states received more.

The transfers to the states having the least problems are large. Even after accounting for other factors, each $1,000 in a state’s per capita income means that the state got $21 more per capita in stimulus funds. With a spread of almost $38,000 in per-person income between the top and bottom states, this has a sizable impact. High-income states get considerably more stimulus money.

Admittedly, I’m just looking at the graphs here, not running any actual numbers, but it seems to me like there’s a pretty low correlation on any of the scales shown. I’ll grant that it seems true that the stimulus isn’t being distributed in a way to help the hardest hit the most, but it seems unclear that the opposite is true, as the headline would have us to believe.

Does anyone with more statistics education than me have an explanation that would support either the Administration’s line or that of this article? I’d also be interested in seeing if there was a graph that showed the correlation (or lack of same) between stimulus outlays and political contributions to Democrats (particularly those in the administration).

1. The only course even close on the “least favorite” list was taught by a professor who seemed to have a tendency to grade somewhat erratically – when comparing one returned test with a few classmates’ copies of the same test, we discovered that the same mistake that cost one person 2 points might have cost another anywhere from 10 to 15 points, with no real consistency across the tests (that is, he didn’t seem to be favoring any particular student, exactly; he just seemed to be scoring things randomly), which was an… eccentricity… that I particularly despised. [↩]