Some tools to model the election and make maps on the Congressional level. You should be familiar with Python and R.
fetch.py Download the US Census American Community Survey 1-Year
data (2005-2019) into a single file by congressional district. The
districts are coded in a text field which is column 1. It is best
if you get a developer ID and add it to the end of the string in this
file on your copy but it is not necessary to get this data.
You can find out more here: How to download all congressional districts from the census API
join.py An example of how to create a standard format of congressional
district names that can be used by the Pandas join function.
house.R Extract house results from R data set by congressional
district for any year and state. If you have the politicaldata and
tidyverse and optparse libraries, this file will allow
extracts of a congressional election for one year either for a state
or for all states.
presidential.R Extract presidential results from R data set by
congressional district for any year and state. Same as the House of Representatives
example above.
congressionalmap.Rmd R notebook to look at maps on the congressional
district level.
get_house_results.py The 2020 results were not available yet on the politicaldata
library that we used for house and presidential races above. This file gets those results
from the Politico website, which had the most up-to-date results as of this writing.
house_changes.py Combine the district and determine a winner for the
house.R files and the get_house_results.py tools above. This is
so that we can look at districts over the 8 years of 4 congressional
cycles from 2012 to 2020.
Maps are available by individual or groups of states
