I need some data for my neighborhood - how hard can it be?
I’ve been thinking more and more about local data, and how to reconcile what we want to know with what’s available to us.
This was spurred on by Thoughtwell’s 2-1-1 Counts report released earlier this winter, which takes a closer look at variations around Columbus and Franklin County. On the one hand, we got to look at resident needs and demographics at a more granular level than Columbus as a whole, and with large enough populations that reasonable estimates could be formed. On the other hand, we measured data at a ZIP code level which, while recognizable, are subject to wild variation in the area covered and political boundaries. This is understandable, as their intended purpose is for effective mail delivery.
Are ZIPs acceptable for local data? We have alternatives, but your mileage may vary on effectiveness. The most commonly used subunit is a Census Tract, subunits within counties designed to have a population of no more than 8,000. How are those plotted in Columbus, where I’m writing?
As we can see, there are lots of subunits, and some of those even cross political boundaries (thanks, annexation policies). One Tract is not equal to one neighborhood however, with neighborhood boundaries coming in multiple shapes and sizes. How compatible are these two?
This map overlays Tracts with neighborhood boundaries from Columbus’s open data portal. Once again, we’re hard pressed to find reconciliation between boundaries where data are measured and a community’s own understanding of neighborhoods.
Other combinations of Census and local boundaries don’t fare much better. We could try Community Planning Areas:
Or Area Commissions: