I'm with Dana on the skepticism about the new Brookings Institution study of walkability. While I'm pleased to see a trend toward a greater number of walkable areas in cities, any study that leaves out the less-affluent neighborhoods can't give a very good read of an entire city. Walkability should be based on the ease with which all the city's residents can go about their daily life without needing a car, and a lot of the parts of DC not included in the Brookings study that would score considerably lower on walkability, since affluent neighborhoods tend to have better planning, greater proximity to amenities like grocery stores, and safer, cleaner pathways to traverse.
An easy way to investigate this a bit is to check out Walk Score, which I find to be a pretty useful tool for gaging the walkability of a specific address. While the Dupont Circle area scores a perfect 100, my own considerably less-affluent Petworth neighborhood scores much lower. But even then, the places they consider "grocery stores" in my neighborhood are more like corner stores where you might be able to get a six-pack or a hot dog, but not groceries, which is true of many poorer neighborhoods. One's ability to walk comfortably to a real grocery store is considerably lower, which is likely true for all the places the Brookings study left out.
Which is why I also have to question Dana's read on Manhattan as more walkable because you can get your groceries delivered. You'd only be able to use that service if you could afford to pay someone else to lug your food around, meaning it has little bearing on poor neighborhoods. Walkability is about whether you can do things like buy groceries on foot without a lot of hassle or discomfort, and in the parts of DC they leave out, you can't.
--Kate Sheppard