We are conducting a research study of environmental risks and trying to estimate the number of residential structures exposed to a certain risk. Much of the spatial footprint data we could find is limited and has a lot of missingness, so we thought we would use addresses instead. If we could get the addresses, we could geocode them, but… need the addresses first. I saw there is address data in the TIGER data but it seems to just be streets and lines…
Thank you,
Claire
Dear Claire,
The TIGER / Line data have address ranges associated with streets - you will find the ranges in the “All Lines” shapefile (the range attributes are LFROMADD, LTOADD, RFROMADD, RTOADD. The address range essentially provides you with the address ranges on each side of a given street. This, of course, isn’t the same as the actual number of structures / units on the given side of a street, but it’s the best you have from Census. Also, there is missingness in the address range data in the TIGER data because it’s hard to keep up to date.
You could combine the address range information with the number of housing units on a given census blocks (from the 2020 decennial census) to estimate the number of housing units on a given road. Then, you would probably just linearly interpolate the housing unit locations along the road segment.
You could also look at OpenAddress.io or see whether your states of interest have an address point dataset that’s publicly available.
Sincerely,
Dave Van Riper
IPUMS
1 Like