Author: Rachael Gilbert <[email protected]>
License: Public domain
Source: https://github.com/rcg798/politix
I wanted to play with voter data in R.
This takes a bunch of voting data from csvs:
- an anonymized list of people with their age, race, gender, and partisan score i.e. likelihood of voting Democrat (candidate_covariates.csv)
- an anonymized list of people with their age, race, gender, who they plan to vote for when surveyed after messaging (outcome) and whether they voted in the last election (turnout) (candidate_outcomes.csv)
- an anonymized list of people with the messaging treatment they were assigned to (candidate_rand.csv)
I analyze the following:
- Did either of the candidate messaging strategies increase voter's likelihood to vote for the Democratic candidate?
- Did either/both of the messaging strategies persuade voters to support the Democratic candidate?