As I mentioned in Join me at PyData Seattle 2023, yesterday I presented a 90 minute tutorial on the censusdis package for querying and working with U.S. Census data in Python. The material I covered is now available online in the censusdis-tutorial repository on GitHub. This contains the notebooks I presented during the tutorial. There are no slides. The entire presentation was a live demo using these notebooks.
If you would like to run the tutorial notebooks live on a virtual machine in the cloud at mybinder.org, follow the links in the README.md file.
The first notebook, Tutorial.ipynb, dives right in to censusdis with a “Hello, World,” query and then builds on that example to add variables, geographies and maps. The notebook then proceeds to introduce the censusdis metadata API, a powerful tool for discovering the nuggets of data you want that are hidden in amongst the tens of thousands of variables the U.S. Census makes available.
The second half of the tutorial uses the notebook Newark.ipynb to demonstrate how we can go quickly from a research question like, “where are there concentrations of children living in poverty in Newark, NJ?” to discovering the census variables that cover this topic and the relevant geographies to downloading the data and producing a geographic visualization suitable for sharing with stakeholders.
A video of the tutorial should be available in the near future. I will post an update when it is.
