A fascinating tool that mathematically analyzes congressional district boundaries:
So the gerry index can be calculated using this equation:
I=(L/A)/(X/A)=L/X
Where I is the gerry index, A is the area of the district (e.g. square miles), L is the length of its boundary, and X is the length of the boundary of a circle of the same area A.
It ranges anywhere from 1, or perfectly smooth, on upwards. By experimentation, it seems a rating less than 2 is okay. Between 2 and 2.5 is suspect. Anything above 2.5 runs a high chance of being gerrymandered...
Iowa 2 ranks pretty low (1.82), but we have that weird, pseudo-no-partisan thing. Illinois 7, the "squashed headphones" Hispanic-majority district, rates 5.32 and the editorial comment "That Junk's Gerrymandered up the Ying-Yang!" Same with North Carolina 12, the I-85 district (5.35).
The tool is purely mathematical and doesn't look at Voting Rights Act requirements to create majority-minority districts.
Have fun playing.
UPDATE: I played some more; some of the wackiest districts (I looked at Florida and Illinois) don't process properly. It looks like, if the district meets at a point, the program only analyzes a piece of the district. Or sometimes (Louisiana 2) it doens't show up at all. I also can't get it to work with the states with one at large seat (I was guessing nice big square Wyoming would have an extemely low score.) Districts at the edges of states seem to score low, because state boundaries tend to be straigher than gerrymander lines.
Still, a fun toy that I hope will get fine tuned.
No comments:
Post a Comment