The Washington Post’s August cavalcade of incredibly poorly-done articles continues apace with a page-three story this morning headlined Suburbia USA: Fat of the Land?

The story concerns a report from the University of Maryland that shows that there is a correlation between the ‘sprawl index’ of where you live and how fat you are, on average. Presumably this is because people who live in certain places never walk anywhere, because there’s nothing to walk to.

The ‘sprawl index’ is a number derived from doing calculations on a number of figures describing an area, including gross population density, percentage of the population living at less than 11500 people per square mile, percentage living at more than 12,500 people per square mile, average block size in square miles, and the percentage of ‘small’ blocks — under .01 square mile.

I have a big problem with this ‘sprawl index’. The report — see it here — doesn’t mention precisely how all these figures go together to give the index, but the numbers don’t seem to make sense to me.

To begin with, the sprawl index looks like some kind of weighted density rating; a higher number means less sprawl. Manhattan’s sprawl index, the highest in the country, is 352.07; Westchester County’s sprawl index is 128; Cook County, IL, is 150. Exactly what these numbers mean, except that Manhattan is 3.52 times as dense as the national average, and that Chicago is apparently twice as ‘sprawling’ as Manhattan, it’s hard to say.

On top of that, the sprawl index makes absolutely no distinction between a dysfunctional, sprawling suburb, and open countryside. Loudoun County, VA, has a sprawl index of 94.57; Goochland County, VA an index of 67.59. Both of these must be awful places, right? Wrong. Both counties are largely multi-million-dollar horse farms, with small towns dotted here and there. The eastern edge of Loudoun County has its share of grotty suburbs, but in general the county is composed of the kinds of settlements that one would think anti-sprawl advocates would favor.

It’s clear that if your goal was to build some pretty awful suburbs, but to get a low sprawl score, it’d be possible: just make the houses fairly dense (which they are already, actually), make the roads connect, and so on. It won’t mean anything for real walkability if your residential zoning ‘pods’ are ten miles across. The problem isn’t that people live at too low a density, exactly; it’s that they live in residential ghettos with nothing but other tract houses for miles around.

All of that said, the study’s conclusions — that the suburbs, as they’re built now are a public health threat — are probably true. This is the first time I’ve seen this clearly stated from a non-crackpot source. Since you can justify almost anything these days by calling it a solution to a public-health problem, maybe this is the thin end of the wedge that will result in the human habitat being more, well, habitable in the future.