Many times we all see people using GIS in a way to make a point. We all know maps lie so that is why special interest groups take to maps more than any other format. I happened to see a link on Ogle Earth showing the worst gerrymandered districts in the U.S. Congress. I’m not saying that this information is not useful, but when you add a statement like this on your readme it shows either you are uninformed or you have an ulterior motive.
“We leave it open as a challenge to the GIS community to come up with some algorithm for creating districts that is both “fair” and which cannot be misused by politicians.”
OK, I can’t speak to other states, but Arizona has an Independent Redistricting Commission made up of five members: two Democrats, two Republicans, and one Independent. They look at many factors and try and eliminate gerrymandered districts. OK so why did they end up with Arizona Congressional District 2? Well there was a huge feud between the Navajo and Hopi tribes about representation and the Hopi Tribe was worried about being canceled out by the larger Navajo tribe in Washington (there is much history here between the two tribes so you might want to read up a little on that to see what this is such a big deal here in Arizona). So that is why you have the second district running all the way around the mollogon rim to get to the Hopi Reservation (take a look at this map on the Hopi visitor page. Now you see what that district includes). Beyond that issue, you also have to deal with a very urban state. Most of the population of Arizona is in the large cities of Phoenix and Tucson so you have to work around including millions of acres that have almost no population with those compact highly populated counties of Maricopa and Pima.
The Arizona Independent Redistricting Commission is tasked with creating competitive districts so in a state like Arizona, you’ll end up with districts that will have a little gerrymandering in them to try and capture equal votes. Sure that doesn’t explain district 2, but as I said the Hopi/Navajo feud is a very big deal. Oh and guess what? The Arizona Independent Redistricting Commission uses GIS to distribute the district boundaries in a fair method and provides Shapefiles and even Maptitude files for citizens to review.
You can bash the legislative districts all you want and I’ll agree there are some really bad ones out there, but to blindly include Arizona’s Second District and claim that the citizens of Arizona don’t understand the issue is an insult to everything we have done here. The citizens of Arizona voted for Proposition 106 in the 2000 General Election for the purpose of creating competitive districts. I have to question the motives of the creator of this Google Earth KML when the call out a district in a state that has voted for an independent redistricting commission. Either they don’t have a clue about what they are talking about or they have an ulterior motive beyond fair districting.
Lets take one quick look at the Arizona Congressional District Map.
Knowing what you now know about the Hopi/Navajo issues, I think you’d say that the Arizona Independent Redistricting Commission did a very good job, don’t you?
Update - Thanks to b99 in the comments of this post for finding the actual 2006 Georgia redistricting plan which as you can see is very different from the one presented in the Google Earth KML. Yet another nail in the coffin of this KML exercise.
15 responses so far ↓
1
Gretch
// Jul 23, 2006 at 11:46 am
That’s pretty interesting James. I live in a state where we don’t have such a commision and all our districts are pretty much all one sided.
I think the GE person should have asked for ideas. He put forth his conclusion when it wasn’t completely informed. He should have said here is what I found, please let me know if you have any comments or suggestions.
The readme file at least to me seemed like he wrote it up before he did his research. Kinda like I have my conclusion, now let me get some “facts” to back it up.
2
Stefan
// Jul 23, 2006 at 1:49 pm
As a student of political systems, I can’t help but point out that gerrymandering a first-past-the post electoral system is indeed fair as long as the aim or constraint is that a two-party duopoly needs to be maintained. Once that assumption is out the window, then it would make far more sense to distribute representatives according to statewide results. If a Hopi party collects 10% of the votes and the Navajo party 15%, and the libertarians 10% and the social conservatives 15% and the fiscal conservatives 15% and the greens 8% and the welfare statists 13% and the immigrant party 14%, then the real fun can begin, with governing coalitions formed that potentially are far more responsive to voter intentions, and far less likely to remain incumbent if those intentions are not met.
3
James Fee
// Jul 23, 2006 at 3:16 pm
Its not so much as the Hopi party vs the Navajo party (as I don’t believe there is such a thing) as it is perceived representation. The Navajo Nation has about 175,000 on its reservation, while the Hopi Reservation has a population of about 9,000. It is a little hard to balance those populations out. I don’t believe the Hopi’s really care if it is a Dem or Rep that represents them, only that they get a fair shake compared to the Navajo. It is hard to do that with such a disparity of population sizes. Also due to the fact that the Navajo Nation surrounds the Hopi Reservation, it is really hard to spit them off any other way than how it was done.
I’m saying Arizona is far from the problem and in fact is the solution. An independent redistricting commission both protects populations that need protection and balances out the districts. You can’t look at the map of Arizona districts and no tell me Arizona has a very fair and balanced system.
If it wasn’t for the Hopi tribe getting the raw end of the stick for so many years in Arizona, they might not have felt that they need this kind of protection. The Arizona district is a very special case and just dropping it into Google Earth with no explanation as to why is very disingenuous.
4
Anon
// Jul 23, 2006 at 3:37 pm
Looks like someone hasn’t done his homework too James. Do you even know who the author of this project is?
5
James Fee
// Jul 23, 2006 at 3:39 pm
@Anon, I know who Dimitri is. He’s got a history of writing stuff before thinking about it.
6
b99
// Jul 23, 2006 at 6:40 pm
I agree with James that independent redistricting commissions are more likely to produce fair districts. But there was an extraordinary amount of litigation in Arizona following the release of the 2000 census, so we are only talking about marginal improvements in the process.
http://www.senate.leg.state.mn.us/departments/scr/redist/redsum2000/redsum2000.htm
The Georgia congressional map referenced by Dmitri is not the congressional plan in place for the 2006 elections.
Here is the 2006 Georgia plan overlaying a voter registration map.
http://tinyurl.com/egdmk
It looks better and was upheld by the Supreme Court, but there are many (including myself) who would maintain it violates the Voting Rights Act.
I’ve developed thousands of voting plans over the years. There is no way that you can come up with a GIS algorithm for creating “fair districts”.
7
KG
// Jul 23, 2006 at 7:39 pm
I challenge anyone to come up with anything that “cannot be misused by politicians”. In my state we are working on a system to assist with redistricting or redlining. Changes in population and polling places necessite this process. Yet when you link census boundaries with political boundaries you can’t overlook the attributes associated with census blocks - racially and politcally delineated population statistics all freely avaliable but never before ‘tied’ so neatly up in a bundle as GIS can do.
The independent commisions should be mandated by the feds. But when did anyone ever concede any form of power in this world? They vote in their own pay raises!
8
James Fee
// Jul 23, 2006 at 8:15 pm
b99, thanks for that info on Georgia. I’ll be honest in that I haven’t really paid attention to other states redistricting plans.
As for Arizona, yea there were those court challenges, but given that this is a huge change from before you’d have to admit that it was a given. The end result is pretty good, maybe more creativity could have been come up with on the Hopi issue and we’ll have to see if the next go round is any better. I’m sure there will be litigation, but I feel much better about my state when I look at my congressional map.
Your point though is very well taken and better put than how I was trying to put it. GIS is only one tool of many that one can use in redistricting. Just because GIS says it should be one way doesn’t make it correct anymore than 1000 monkeys at computers trying to accomplish the same thing.
9
James Fee
// Jul 23, 2006 at 8:17 pm
KG:
“But when did anyone ever concede any form of power in this world?”
Arizona took that power from them. The citizens of this state stood up and said no more. I’ve never been more happy to have a proposition on the ballot than when I saw that one come up.
10
b99
// Jul 23, 2006 at 9:29 pm
Yes, James, anything that opens up the redistricting process to public review is a positive development.
Montana has an independent redistricting commission and, in 2003, they finally adopted a plan that gave the Blackfeet and Flathead tribes a shot at electing a state senator. By contrast, there is no redistricting commission in South Dakota and soon the state will have to redraw a portion of their legislative plan in Indian Country. A federal court has required them to do so and I think it is unlikely that the Supreme Court will overturn that ruling.
Tennessee 7 is the perfect example of a political gerrymander – redistricting is behind closed doors in Nashville. Tennessee 7 has nothing to do with the Voting Rights Act or communities of interest.
http://216.55.182.132/FairData/Congressional/map.asp?command=findall&vis=&field=Cong&map_option=loc_and_site&data=TN7
Caliper (makers of Maptitude — the preferred software for most people in the redistricting world) actually sold something called “Auto-District” for $8,000 in 2001. (I don’t see it for sale now.) The software would supposedly draw districts based on a set of parameters – maximum number of split counties, minority percentages, “compactness scores”, etc. I’ll bet nobody bought it because there are way too many factors to effectively automate the process.
11
Stefan
// Jul 24, 2006 at 1:30 am
“Its not so much as the Hopi party vs the Navajo party (as I don’t believe there is such a thing) “
I’m not saying that there is such a thing, I’m saying that such a thing might evolve as a natural consequence of switching to a system of proportional representation. I readily admit that’s very unlikely to happen.
And of course I have no idea who Dimitri Rotow is. But you have a fair point that you can make GIS, just like statistics, prove nearly anything.
12
Dimitri
// Nov 3, 2006 at 2:27 pm
James, et al,
I don’t read your blog on a regular basis, so I didn’t notice your misleading comments until today. You completely misrepresent my statements.
Anyone can see my comments from
http://www.manifold.net/downloads/gerrymanders.zip
…unzip the file, see the text commentary and view the KML for themselves using the always-stylish features of Google Earth.
As anyone can read for themselves, I did *not* “claim that the citizens of Arizona don’t understand the issue.” In point of fact, I didn’t mention Arizona at all.
If your comment was intended to argue with my point that the average citizen has no clue what Gerrymandering is, you picked a deceptively roundabout way to phrase your disagreement. You are either being deliberately deceptive, or showing that you don’t have very much experience discussing Gerrymandering with the average citizen.
I think most people have sufficient common sense and worldliness not to need to peform the actual experiment, but for those who doubt my word, go out to your local shopping mall and choose people at random and ask them to explain Gerrymandering to you. You’ll encounter darn few who know what it is at all and perhaps only one or two in a hundred who can accurately explain to you the mathematics of it.
Although I’ve never tried that experiment in Arizona, I think there would be little doubt in any sensible person’s mind that you’d get about the same results in a shopping mall in Phoenix as you would in New York. So I think it’s fair to believe my assertion that most people don’t understand Gerrymandering applies in Arizona as well.
If you try that experiment you may want to test my other assertions that a) it is difficult to explain to average people what Gerrymandering really is because most citizens don’t have the math skills to understand it (trivial though the math may be), and b) on those occasions when you are able to explain it you’ll find that average folks really are appalled by the practise.
It doesn’t matter whether they are liberal or conservative, Dems or Repubs, white or black, rich or poor. A very high percentage of average people who learn what Gerrymandering is will judge the practise to be anti-democratic. Go ahead and try it if you doubt my word.
To your general criticisms: I didn’t set out to “prove” anything other than to show how GIS can be used to make computations to find Gerrymandered districts. If you think you can cobble up a better way, have at it. That’s part of my challenge to the GIS community.
There are a thousand stories about how those districts ended up Gerrymandered and there is no end of stories why in a particular case a given Gerrymandered district is demonstrably for the best. That’s how politics works: people who do such things are never stupid about how they sell their handiwork.
One of the implications of the technical work I reported is that you can’t have a debate about whether or not a particular Gerrymander makes sense unless you can detect the Gerrymander in the first place. Since I trust that no one reading this blog is so lame as to not have at least *some* pride in their GIS technical skills, let’s see how anyone who disagrees with the formula I used would do the job better. And, let’s not have any girly-man, nitwit, unmathematical distractions - let’s see your math, your spatial SQL if you have it.
As I said, once you see a Gerrymander you can argue about whether it is good or bad. My own view is that the practise is ultimately anti-democratic. I believe that if the average voter understood the role of Gerrymandering we’d have less of it. But I take it no further than that: show the Gerrymandering that exists and let people make up their own minds what they think about it.
And, unlike you, I don’t slam efforts to engage the GIS community in educating other citizens. By providing rich and powerful GIS that any person can afford I make it easier for average people to use GIS to examine issues like Gerrymandering for themselves, to make up their own minds and to educate themselves about the complexities involved.
If you believe in democracy, as I do, you believe in giving people the tools and the data so they can have a fighting chance of seeing the lay of the land and coming to their own conclusions.
In that context, my quote you took out of context to start this thread and so contemptuously dismissed shows my faith in the good will and sensibility of the GIS community, faith which you appear not to have. The full quote would be:
“We leave it open as a challenge to the GIS community to come up with some algorithm for creating districts that is both “fair” and which cannot be misused by politicians. It is easy to be outraged by Gerrymandering, but not so easy to find a solution that ensures every vote counts the same as every other vote. Not so easy at all!”
That you showered contempt on the above does not speak well either of your intelligence or your good faith.
To find an electoral procedure that does indeed assure every vote counts the same as every other vote is a noble objective, a challenge worthy of our GIS community. It’s a shame you are such a lightweight that you would criticize that objective.
By the way, considering your comment “@Anon, I know who Dimitri is. He’s got a history of writing stuff before thinking about it. ” [link omitted] …Pausing just a moment to thank you for increasing the search engine scores of my writings by linking to them, if you really believe I didn’t think through my analysis of why Microsoft has won the desktop over open source, why is it you are writing a blog about Microsoft-only ESRI stuff and aren’t using GRASS or some other such open source package? The reason, of course, is that what I wrote is indeed well thought through and is true.
Regards to all,
Dimitri
13
ManifoldUser
// Nov 3, 2006 at 2:45 pm
Hey Dimitri, when will we be seeing a Manifold Blog? Many of us hang out here because there isn’t any other place to go. It is kind of funny that if you google “manifold blog” you get this one. That’s how I found it.
14
Dimitri
// Nov 3, 2006 at 4:45 pm
That is funny! It was not easy to train James to mention Manifold in his posts but we seem to have succeeded.
It’s not really part of this thread, but if you hang out here on a regular basis I suppose it’s OK to follow your lead with a reply.
The Manifold forum is at http://forum.manifold.net - it’s an unmoderated, free-wheeling discussion.
Many Manifold users have their own blogs which, like this one, often cover more than just the immediate GIS product at hand. I was pointed back into the spatiallyadjusted blog by Simon Greener, who has a blog at
http://www.spatialdbadvisor.com/blog
Simon is well known at Oracle and ESRI for his talents with spatial DBMS and spatial SQL, at one time being one of ESRI’s strongest pitchmen for ArcSDE.
Simon has long argued for certain technical improvements in the next generation of spatial DBMS. He is so expert his ideas are getting a lot of traction with major players such as Oracle, Microsoft and IBM.
15
Mike
// Nov 6, 2006 at 2:32 pm
It is funny. More blogs people!
Leave a Comment