# One Person, One Vote Analysis

## Evaluating the effective power of individual votes in United States elections

### <Tyler Woebkenberg, Github: https://github.com/tw1574, Net ID: tw1574>

Problem Description: What is the effective power of a single vote for a citizen of the United States?  This answer depends on many factors - geography, demographics, political landscape, etc.  Can we narrow down to the specific factors which affect the power of a vote?
Data: There are numerous datasets that exist that can help to support this analysis
• US Census data - A rich catalogue which includes population, eligible voting population, demographics, some voting data.  This data exists at various levels, so I would expect some necessary transformation in order to merge with other data.
• Library of Congress data - This resource guide compiles a list of online and print resources that contain U.S. election statistics for both federal and state elections.  This data exists at various levels, so I would expect some necessary transformation in order to merge into with other data.
Analysis: From a tool perspective, the analysis will be completed mostly in Python, with Google Docs and Tableau supporting data preparation, exploration, and profiling.  For the analytical piece, I intend to use cartography in Python to visualize the data, but also I will use clustering to understand which factors have similar and strongest effects on votes.  If there is time, I will continue the work to perform an autocorrelation on the Congressional Districts to understand the relationships within States, and even across State boundaries regarding voting habits.
References:
• Some work for this particular project has already begun as part of the CUSP 2016 Fall Hackathon.  It can be referenced here
• The continuation of this project was also inspired by a researcher at U of Michigan who produced similar graphs for 2016, 2012, 2008, and 2004.  These were based on population, though, not eligible voters on which I intend to focus.

Deliverable: As an output of this analysis, I would expect to produce a statistical conclusion sufficient enough to be able to begin to infer which factors contribute to the "power" of an individual vote in the United States.  This will be a combination of plots, and output of statistical analysis in a Jupyter notebook.