Analysis for an election audit of the Colorado Board of Elections for a recent local congressional election, to determine the winning candidate and the largest voter turnout based on county. The analysis required the follwing data to be presented:
- The total number of votes cast.
- The list of all candidates in the election.
- The total number of votes received per candidate.
- The total number of votes per county.
- The percentage of votes for each candidate.
- The winner of the election.
- The county with the largest voter turnout.
-
After conducting the analysis, a total of 369,711 votes were cast during this congressional election from the counties in the precinct, those being Denver, Jefferson, and Arapahoe. As can be referenced on the image below, the breakdown of the total votes per county lists Denver, the county with the largest number of votes, with an 82.8% of the popular vote achieving a total of 306,055 votes. The county of Jefferson received a 10.5% of the popular vote with 38,855 votes while Arapahoe received 6.7% of the popular vote with a total of 24,801 votes.
-
The running candidates and the results per candidate can are listed below and can also be referenced on the screenshot below:
- Diana DeGette received 73.8% of the vote which is 272,892 votes, was the winner of the election.
- Charles Casper Stockham received 23.0% of the vote which are 85,213 votes.
- Raymon Anthony Doane received 3.1% of the vote which are 11,606 votes.
The code developed to run this analysis is a great and efficient tool for the election commission to analyze data sets of big scales, like this audit, the code is extremely customizable and can be easily run to analyze any data set by making small adjustments. The script can be applied to any electoral tabular dataset simply by changing the source of the data on the code itself, as can be seen in the screenshot below.
As can be seen in the following image, it's very easy to manipulate the script to be run for any sort of election, be it a congressional or even a federal election, as it was designed to pull the candidates from the dataset itself. On the second screenshot, it can also be referenced how the input for the county could be switched to a city, or even a state and the code can be easily modified to gather this data on its own as opposed to being manually entered shows how flexible and usable this code can be for the commission.
Candidate Field easily modifiable
County can be switched to modify the target regions
Data Source: election_results.csv
Software: Python 3.6.1, Visual Studio Code, 1.38.1