To be able to order states by their level of gerrymandering, you first have to define how you measure level of gerrymandering.
|
Ideas |
Metrics |
One of my ideas that I couldn't figure out how to implement was a measurement for the demographics of a region. I also wanted a way to measure how urban or how rural a district was.
|
The metrics I used were as follows:
1. The difference between the state population's vote and the state district's vote. 2. The number of lines it takes to accurately represent the districts border. |
DataTexas 35th:
lines: 12 Population: 48.54% Trump 45.71% Clinton Total Population: 52.5% Trump 43.5% Clinton Difference: 6.17 Score: 18.17 Florida 5th: Lines: 12 Population :61% Clinton 37% Trump Total Population: 49% Trump 47.8% Clinton Difference: 25.2 Score: 37.2 Illinois 4th: lines: 23 Population: 17% Romney 81% Obama Total Population: 40.73% Romney 57.60% Obama Difference: 47.13 Score: 70.13 Michigan 14th: lines: 15 Population: 18% Trump 79% Clinton Total Population: 47.6% Trump 47.4% Clinton Difference: 61.2 Score: 76.2 North Carolina 12th: lines: 29 Population: 62.29% Trump 32.28% Clinton Total Population: 50.5% Trump 46.8% Clinton Difference: 26.31 Score: 55.31 Ohio 9th: lines: 9 Population: 45.36% Trump 49.88% Clinton Total Population: 51.8% Trump 43.7% Clinton Difference: 12.62 Score: 21.62 Pennsylvania 7th: lines: 46 Population: 47.0% Trump 49.3% Clinton Total Population: 48.6% Trump 47.9% Clinton Difference: 3 Score: 49 ProblemsOne of the problems I encountered was that it was difficult to find data for some districts, like Ohio, North Carolina, and Texas. My solution was to find which counties were in that district, then find the average of all their votes. I think that what I ended up with might not be as accurate because of it. I also struggled with whether you should measure gerrymandering by the shape of a state, or by the vote, so I ended up with a combination of the two.
|