PA 2018 Standardization and Typo Cleaning

[peter-redistricting-data-hub]

Hi, this is a pretty big update, so let me know how I can be the most helpful in sharing what I did.

I've upload a zip of the Jupyter notebook I used to generate the changes.
OE_Cleaning.ipynb.zip

The only change I imagine you might not want, is for two candidates marked DEM / REP, I update their party to match what is on ballotpedia. As I understand it DEM / REP is used when only one two-party candidate is on the ballot. I go into it a bit more in the Notebook.

In addition, I've upload a .txt file pa_2018_diffs.txt with, after these changes are performed, the differences between the official county results and those in this file both by county and at the state level.

Three big notes:

1. There aren't write-in votes in the official county results so those aren't compared
2. For county '087' which has by far the biggest differences, what we have seems to match the canvass report, but just differs with the official county results.
3. Aside from county '087' the biggest difference at the county total level is 14 votes. I think we'll be able to fix these, but I just wanted to push this out to prevent duplicative work of people finding or correcting the issues I've changed.

[peter-redistricting-data-hub]

rootsmusic helpfully pointed me to this link http://docs.openelections.net/guide/, which my code did not follow. I can try to rewrite my code / functions to fit with that workflow, though it will likely be ~2 weeks before I will be able to do so

[dwillis]

@peter-redistricting-data-hub thank you! I've been making fixes to 2018 general files piecemeal, and will reconcile them with your findings.