Avencia released a whitepaper that measured the gerrymandering of local and federal electoral districts across the U.S., using some software that they sell. I read it and it is pretty professional, but it brings me back to my point about not . As I’ve said before, there are special cases that create some of these districts and to claim that one is worse over another, just because it is spatially significant. Now don’t get me wrong, I think GIS is the perfect tool for such research, but it shouldn’t be the only tool.
**Update – **Robert Cheetham @ Avencia was kind enough to email me and let me know that I overlooked an important part of the whitepaper in that it does note that there are “legitimate reasons why districts end up with some convoluted shapes”. Also at the end of the report, Avencia notes (emphasis is mine):
The gerrymandering index described in this white paper attempts to quantify the extent to which a local or federal district may be gerrymandered, based on its level of compactness and that of its city or state. Because of the combined impacts of political boundaries and physical geography, other factors may be taken into consideration when looking into a particular district, such as shape, contiguity and respect for political subdivisions. Nonetheless, compactness measures are a reliable indicator that gerrymandering is likely and point the way to districts worthy of higher scrutiny.
As long as people go that extra step to analyze why districts end up like this, then we are on to something.
Update 2 –Avencia made a small update to the whitepaper clarifying the Arizona 2nd District. I’ve put the revised PDF here. I don’t ever recall seeing Avencia before (my bad), but they seem to do some pretty interesting work.