Code to extract Franklin County, OH voter file information and analyze patterns in voting behavior. There are two versions of the code, one working with data stored in a flat JSON and one working with a SQLite DB.
To create flat JSON for full voter file and sub-samples split every 100k voters, go to makeSamples/ and run
$ python makeFlatVoterFileSplitSamplesJSON.py
Sub-samples to be used for calculating correlations and training voter likelihood algorithm. To test general content of subsample, run
$ python testSubSamples.py
Code running analysis is accessed by going to analysis/ and running
$ python simpleAnalysis.py
The same general output is created in a SQLite DB by going to sql/ and running
$ python makeVoterTable.py
Voters are split into sub-samples identified in SUBSAMPLE column. To test general content, run
$ python testTable.py
Code analyzing SQLite DB run by going to analysis/ and running
$ python runAnalysisSQL.py