I found an interesting dataset on political contributions in the state of Illinois.. The downloadable .zip contains multiple tab delimited database files which contain the relationships between Donations, Committees, and Candidates.
Out of curiosity in seeing Comcast’s political influence in Illinois over time, I parsed the 650,775mb file called Receipts.txt. Below is a bar chart of yearly recorded donation totals from 2000 through 2015-08.
Not being very politically oriented, I wanted to somehow relate the donations to candidates. But in the form of the available data, it appears Donors make contributions to Committees, and Committees support a Candidate. But I do not know if Candidates and Committees are a One to One relationship at the time of typing this.
Parsing the text file called CmteCandidateLinks.txt, I related the Committee Id with the candidate Id. Parsing the text file called Candidates.txt I relate the Candidate Id to the Candidate name.
Lots of candidate duplicates per donation entry. Majority of Committees represent the same candidate under different ids, while some committees represent multiple candidates. Example here:
So I decided to distribute each donation amount between a potential multitude of candidates. I did this by dividing each donation by the number of candidates which belong to the committee recipient. From that, I got this list of Comcast’s Top Illinois Candidates.
