Rural Geocoding: Why is it so gosh darn difficult?

Maps have been around for a long time. Going back thousands of years into the history of civilization, humans have charted the landscape and identified points of interest to serve as a guide and tool.

Despite that, my GPS still can’t seem to find that rural address and ends up leading me on a wild goose chase in my Subaru. Why, after all this time, can’t so many of us get easily and accurately from Point A to B?

The answer is, of course, extremely complicated. Anyone who works with geocoding addresses quickly finds that mapping errors have been around just as long as maps themselves, and that even with satellite images and GIS mapping vehicles driving around the country, there are many challenges to linking an address to a point on a map, and that difficulty increases the more rural you get.

While I touched on this challenge in an earlier post, I’d like to delve deeper into these issues, as it’s one that becomes extremely valuable when we talk about providing polling place directions to voters and a citizen’s ability to vote is at stake.

When rural routes were first established, there was very little need to know exact locations or relative distances. As a result, sequential numbering of houses began as inaccurate, if it existed at all, and many maps included mere address ranges on roads, rather than specific points assigned to each address.

With the advent of Enhanced “911” (E911), many rural addresses have been assigned sequential addresses with road names (rather than rural route or fire lane numbers) and pinpointed better to help fire and rescue services responding in an emergency. E911 is nearly complete across the nation, but some areas are still transitioning.

Even when you have an address, many software applications still lack the exact geo-coordinates of an address and will deliver a “best guess” by approximating the location within the given range. When you’re out on the South Dakota plains where one piece of property may last for miles, it can become extremely inaccurate for a software application to “guess” that address number 500 is halfway between numbers 1 and 1,000. This margin of error is the cause of many adventurers taking the “scenic route.”

When applying this to Election Day information, geocoding gets even more complicated when you consider there are two points to map: the voter’s address and the polling place address. If both are rural addresses, then any mapping service will have an extremely difficult time providing accurate directions from a voter’s home to her polling place. These challenges exist beyond elections, too. Performing health & environmental impact studies in rural areas is challenging for the same reason: how can you know if someone got sick from a nearby water source if the location of their house has a 20-mile margin of error?

To solve this challenge, we’re working closely with our partners to avoid voter confusion. The solution we implemented in 2010, for example, was to display, when the mapping service was unsure, only the text of the polling place address. Users, if unsure of where that location was, could request a map and directions to the poll location – and the mapping service’s usual caveats for rural addressing would also appear. While this is an imperfect solution, it reduced confusion and led to a smoother process.

You must be logged in to post a comment.