openelections-data-wi

Pre-processed election results for Wisconsin elections

Using the data

The election results files are located in folders by year. The file names are standardized as described here: http://docs.openelections.net/archive-standardization/

office_table.xlsx is a table showing the presence or absence of data for each office in each election.

Some data may be missing. This and other known problems are documented at https://github.com/openelections/openelections-data-wi/issues

Warning: Some elections were recounted. Some of the data here are pre-recount results. This occurs in cases when the recounted results are available only in a different form. See Issue 34 for details.

We have included the following offices in our results:

• President
• Senate
• House
• Governor
• Lieutenant Governor
• Attorney General
• State Treasurer
• Secretary Of State
• State Superintendent Of Public Instruction
• State Senate
• State Assembly
• Supreme Court
• Court Of Appeals
• Circuit Court
• District Attorney

Using the code

parser.py processes files based on cached metadata in local_data_cache/elections_metadata.json
To update this metadata from the OpenElections API, run python fetch.py wi -m
(fetch.py fetches data files based on the cached metadata)

To re-parse all data files:

python parser.py

To parse data files for specific elections, append one or more election ids:

python parser.py 426

Elections will be processed in the order they appear in the metadata.

A folder local_data_cache keeps a local copy of input data files. To update it:

python fetch.py wi

To fetch input files for specific elections, append one or more election ids:

python fetch.py wi 1577 404

Spot tests to check a few records from each results file are in tests/wi-elections.tests.csv. To run, use:

python run_spot_tests.py [<tests_filepath>]

Tests to validate the CSV output using csv-test can be run (2014 example):

npm install
node_modules/csv-test/bin/csv-test tests/csv-test-config.yml '2014/*' tests/csv-test-validators.yml

Process for adding new election data

• Update metadata at openelections.net, note new id #
• python fetch.py wi -m to update local metadata file
• python fetch.py wi <id> to download input data for new election
• Pick random records from downloaded data files, add them to tests (at least one record per file)
• python parser.py <id> to parse input data, write results file
• Add line(s) from office_table.csv to office_table.xlsx
• python run_spot_tests.py
• Fix any problems found
• Push to repo
• Make PR to update main repo