Throughout my career at Socrata I often have the opportunity to find ways to get external data from APIs like the Census and Bureau of Labor & Statistics to help customers benchmark themselves against their peers. Here’s a presentation I gave about one such adventure using FME + python + the Census API for the City of New Orleans.
Lessons Learned
- Variables estimated by the Census may become more granular over time and it’s important to check categorical data for all years to either roll them if you want to track the same variable over time. For example, Commute Time (B08303)
< 10 minutes
is now broken into 2 segments,< 5 minutes
and5 - 9 minutes