Blog: Geography

From National Geographic via Strange Maps, here's a map of which metropolitan areas have more single men than women, or vice versa:

More data behind the map would be very helpful; for example, this map's key is based on the absolute number of single men versus women, but since New York and LA are so huge, having the most disparity in absolute terms doesn't mean that they're the hardest places to find mates. And as several commenters point out on Strange Maps, many of the single women could be elderly widows, for example.

From one commenter, here's a census map comparing relative gender percentages by county, but not specifically for singles:

Colorado and Seattle are in trouble on both maps.

Here are some more great demographic maps, including religous membership and language use by county. For a good overview, see the maps on leading religion and primary non-English language.

Also fascinating are the maps of average family size, median income, and percent of adults lacking a high school diploma.

A few years ago I saw a map someone had made of the Bay Area with each town color-coded by race - green for towns that were majority white, gray for majority black, red for Latino, and purple for Asian, with the darkness of each color representing how large the majority was. I've never been able to find that map again, but this site has created a Google Maps mashup that lets you see the percentage of each race by census tract across the whole country.

The Bay Area is pretty segregated - Cupertino, Fremont, and a few other South Bay towns are majority Asian, Oakland majority black, parts of San Jose majority Latino, and most other areas majority white. You can play with it here. But an even more stunning depiction of segregation are this map versus this one of Washington, DC. It's as if someone drew a line straight down the middle of the District and surrounding counties and said, white people on this side, black people on that side.

What's America's most segregated city? Atlanta? Chicago? Miami? Detroit?

When I was at Tellme, I built the Restaurants channel for 1-800-555-TELL, the Tellme Voice Portal. Restaurants allowed the user to pick a city and state, and hear a list of nearby restaurants. Hearing a list of sushi places in Los Angeles isn't so useful, so I also added a way for users to pick individual neighborhoods for larger cities.

Unlike with city boundaries which are clearly defined, there's no definitive source for neighborhood boundaries, and actually there's really no unofficial source either. Individual cities' Web sites, and travel books, make some attempt, but they all disagree. I solved this at Tellme by just buying a bunch of books and picking out a center for each neighborhood as best I could, then finding restaurants by proximity to that point.

The Neighborhood Project uses Craigslist data to map neighborhoods based on what their residents call each area. They have it for San Francisco only right now, but it's fascinating. (My one complaint about the site is that the "my browser sucks" link should really be "my browser sucks, or else I have a big enough screen that I can look at the whole map and don't need your fancy magnification widget thingie".)

Some neighborhoods are clearly defined, while in other areas nobody seems to agree on what to call them (what's the area around Dolores Park? I've always wondered myself). Using housing listings is of course biased; nobody says they are in the Tenderloin. And some neighborhoods, like possibly Hayes Valley, are significantly larger than they "should" be because people want to claim they are in that neighborhood (are those really Noe Valley dots north of Market?)

It would be really interesting to see this map for NYC, and also to track it longitudinally over time. Some neighborhoods get larger as people want to claim they are in that neighborhood, but then sometimes people rebel against being swallowed up by an area and embrace a gritty name associated with a neighborhood's former character (like "Hell's Kitchen").