Recently I dumped all of the data out of one of our geographic tables in our database, and plotted a red dot for every record. This was the result:

Click the link for a larger version (1200×800).

This is an interesting graph. It appears to follow population, but there are also some patterns that follow state outlines: there are some curious voids in North Carolina and Minnesota, for example. In the west, we can almost make out highways; there’s a visible north/south track through Denver along I-25, and there’s a strong clustering down I-15 through Utah with a big blotch over Las Vegas.

Kinda neat, huh?

What you are looking at is every registered sex offender in the United States. My job is to collect and analyze that data, along with crime information.

Still neat? Sigh.

There are still some interesting patterns, however. Population centers follow the freeways in the western states. As the population density rises to the east, we lose track of individual roads; sex offenders are only absent where people cannot live. Physical barriers like coastlines and bays are clearly marked, of course; but you can also clearly see places where people merely aren’t allowed to live: the Okefenokee National Wildlife Refuge in Florida, the Allegheny and Susquehannock National Forests in Pennsylvania, and the cluster of tiny national parks all throughout upstate New York.

I am most interested, however, in North Carolina and Minnesota. Why are they so obviously underrepresented here? Are people just less likely to offend sexually in those states? Possibly; I am too cynical to believe that people in those states behave differently than in the other 48, but it could be that NC and MN convict sex offenders less often, or they could simply have less strict sex offender laws.

Ah, but see how trusting you are of my little dots? It could be that the NC and MN sex offender registry systems are just harder to get data from. While I haven’t worked with those states directly, I have worked on the code that aggregates this data, and I can tell you that no two states provide data the same way. Some states are sensible and reliable, while others are such a headbashing nightmare—Alaska, I am looking at you here—that you wonder if their IT department had to make do with wild monkeys because the trained ones were too expensive.

Here’s a really good possibility: Some states track sex offenders at different levels, and sex offenders under a certain level are tracked inside the state but not published to national registries. It could be that those two states have just as many RSO’s per capita, but they do not report misdemeanor offenses outside the state.

Or, who knows. It really could be that in North Carolina people are too nice—and in Minnesota it’s just too cold—for that sort of thing.

Note: some of the dots lie outside the US. This is not always a mistake. Sex offenders are required to keep their address registered even if they move outside the USA. There are a few thousand US sex offenders scattered across the globe—including one at the South Pole.