Dealt from a stacked deck

Back in December I wrote about mathematical ways to show district maps are gerrymandered. In that previous post I wrote about the Pennsylvania case, which was before the state Supreme Court. It showed the map created by the GOP in 2011 was highly gerrymandered.

Now Voters Not Politicians, the campaign to end gerrymandering in Michigan that I’m a part of, has done a similar study for Michigan. Dr. Dan Magleby, professor political science at Binghamton University, used census data and a mapping program to randomly generate 10,000 district maps of Michigan. This is similar to the study that was done in PA. He then randomly selected 1,500 maps and did some statistical analysis on them.

The GOP likes to claim the maps elect more GOP members of Congress because of “self-sorting,” which means we tend to live in areas where the other people are similar to ourselves. Magleby used mean-median difference as a measure of bias towards one party or the other. Then he charted how many of those 1,500 maps showed each bias value. He got a typical statistical curve centered on -1.7. This is the amount of bias expected from self-sorting. The whole curve stretches from -0.2 to -3.5. And the current Congressional map has a bias of -6.7.

The campaign’s conclusions:

* The current map is highly biased. We’ve been dealt from a stacked deck. The current system is grossly unjust.

* The expertise and technology to draw fair maps does exist.

So visit the campaign website to donate or volunteer (or both!).