3 ms·
"But unlike postal codes in other countries which represent small areas suitable for in-depth geographic inquiry (Google “N6C 2B5” for a Canadian example), ZIP
by DrScump 10y ago
"But unlike postal codes in other countries which represent small areas suitable for in-depth geographic inquiry (Google “N6C 2B5” for a Canadian example), ZIP codes are woefully unqualified as units of analysis."
First of all, don't Canadian zip codes have the exact same limitation? Postal codes in general are designed for the convenience of mail delivery and nothing else. (For that matter, mailing address often misidentifies the actual city of residence if one lives near a boundary and always states a phony city if one is in unincorporated territory.)
There are two methods I can think of immediately to drill down into better detail. For one, zip codes are nine digits now, and using all nine digits (Zip+4) yields extremely narrow detail (generally no larger than the destination's building; often, homes have their own Zip+4. But that still leaves the problem of matching address to political boundary.
For that, there are many datasets that match street address to the underlying parcel, and knowing the parcel provides you with actual city (if in incorporated territory), state political district(s), city political district, Congressional district, etc. It's used by mailhouses for sending political mailers to the correct jurisdiction all the time.
- maxerickson 10y agoI don't know what the Canadian system is, but the example points to it being street/block level: https://www.google.com/search?num=20&q=postal+code+N6C+2B5 https://www.google.com/search?num=20&q=postal+code+N6C+2B5 As far as I understand zip+4 is for a given sorting route, so it won't necessarily have nice spatial properties. The USPS anyway considers zip codes to be collections of addresses, so it's sketchy to use them as areas. The problem with the parcel data sets is getting access to them in a coherent format without spending money. I actually came across this link on a the twitter feed of someone behind Openaddresses (https://openaddresses.io/ https://openaddresses.io/ ). They often get addresses from parcel data, https://usopendata.org/2016/06/07/parcels/ https://usopendata.org/2016/06/07/parcels/ but don't restrict themselves to it.