acs2pgsql

This repository imports American Community Survey (ACS) data into PostgreSQL. The hard work has already been done by Census Reporter. We add the following features:

• Make the state (e.g. Illinois, California) a variable
• Fix issues with the 2009 ACS 5-year data
• Run the workflow using docker or drake

Usage

1. Clone the repository and create the file acs2pgsql/default_profile by specifying PostgreSQL credentials and the desired U.S. State in the following template:

PGHOST=
PGDATABASE=
PGUSER=
PGPASSWORD=
STATE=Illinois
STATE_ABBREV=il

If the state name has multiple words, remove the space: e.g. for North Carolina use STATE=NorthCarolina.

2. Change to the acs2pgsql directory. To use docker, run docker build --rm -t "image_tag" .. Docker will create an image and then extract, transform, and load the ACS data into your PostgreSQL database. To use drake, run drake.

3. Consult the census-postgres documentation for full technical details. See also the census documentation for each dataset, especially the:

• Appendix (e.g. 2015) for information about census tables, and the
• Sequence Number and Table Number Lookup (e.g. 2015) for information about variables.

TODO

• Expand to Decennial and ACS 1yr and 3yr
• Rename folders per state