Sign In

Communications of the ACM

ACM TechNews

Mathematicians Are Deploying Algorithms to Stop Gerrymandering


View as: Print Mobile App Share: Send by email Share on reddit Share on StumbleUpon Share on Hacker News Share on Tweeter Share on Facebook
puzzle pieces form U.S. map, illustration

Mathematicians are implementing algorithms designed to halt gerrymandering — the partisan manipulation of U.S. Census data to redraw electoral maps for one political party's gain. These programs could dramatically affect redistricting, and offer an objective and practical test for flagging gerrymandered maps.

The different tools that various scientists have designed deliver an extreme-outlier test, comparing a map suspected of being gerrymandered with a large collection of bias-free maps, and generating a random sample that reflects the probability that any drawn map will meet various political criteria. The ensemble maps account for various district-drawing principles, incorporating their interaction with a state's geopolitical geometry.

Duke University researchers will load 2020 U.S. Census data into an algorithm and produce an array of typical, nonpartisan district plans for North Carolina; this should yield protective benchmarks, and in September the team will release the maps and hope state legislators consider them.

From MIT Technology Review
View Full Article

 

Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA


 

No entries found