Slashdot Mirror
Mathematicians Study Effects of Gerrymandering On 2012 Election
HughPickens.com writes Gerrymandering is the practice of establishing a political advantage for a particular party by manipulating district boundaries to concentrate all your opponents' votes in a few districts while keeping your party's supporters as a majority in the remaining districts. For example, in North Carolina in 2012 Republicans ended up winning nine out of 13 congressional seats even though more North Carolinians voted for Democrats than Republicans statewide. Now Jessica Jones reports that researchers at Duke are studying the mathematical explanation for the discrepancy. Mathematicians Jonathan Mattingly and Christy Vaughn created a series of district maps using the same vote totals from 2012, but with different borders. Their work was governed by two principles of redistricting: a federal rule requires each district have roughly the same population and a state rule requires congressional districts to be compact. Using those principles as a guide, they created a mathematical algorithm to randomly redraw the boundaries of the state's 13 congressional districts. "We just used the actual vote counts from 2012 and just retabulated them under the different districtings," says Vaughn. "If someone voted for a particular candidate in the 2012 election and one of our redrawn maps assigned where they live to a new congressional district, we assumed that they would still vote for the same political party."
The results were startling. After re-running the election 100 times with a randomly drawn nonpartisan map each time, the average simulated election result was 7 or 8 U.S. House seats for the Democrats and 5 or 6 for Republicans. The maximum number of Republican seats that emerged from any of the simulations was eight. The actual outcome of the election — four Democratic representatives and nine Republicans – did not occur in any of the simulations. "If we really want our elections to reflect the will of the people, then I think we have to put in safeguards to protect our democracy so redistrictings don't end up so biased that they essentially fix the elections before they get started," says Mattingly. But North Carolina State Senator Bob Rucho is unimpressed. "I'm saying these maps aren't gerrymandered," says Rucho. "It was a matter of what the candidates actually was able to tell the voters and if the voters agreed with them. Why would you call that uncompetitive?"
The results were startling. After re-running the election 100 times with a randomly drawn nonpartisan map each time, the average simulated election result was 7 or 8 U.S. House seats for the Democrats and 5 or 6 for Republicans. The maximum number of Republican seats that emerged from any of the simulations was eight. The actual outcome of the election — four Democratic representatives and nine Republicans – did not occur in any of the simulations. "If we really want our elections to reflect the will of the people, then I think we have to put in safeguards to protect our democracy so redistrictings don't end up so biased that they essentially fix the elections before they get started," says Mattingly. But North Carolina State Senator Bob Rucho is unimpressed. "I'm saying these maps aren't gerrymandered," says Rucho. "It was a matter of what the candidates actually was able to tell the voters and if the voters agreed with them. Why would you call that uncompetitive?"
It would also be the end of the two party systems.
"The likes of Facebook and WhatsApp are free to those whose privacy is of zero value."
Not according to a comment by Justin Levitt, a law professor at Loyola University, in one of the linked articles and acknowledged by Mattingly - he did say that incorporating it might be an option for further research though. This was apparently a "proof of concept" with deliberately simple rules, but given the interest and positive feedback it seems to be generating I'd like to see what happens if this could be adapated to include a more complete set of rules, not to mention be adapted for other countries where this is a problem.
UNIX? They're not even circumcised! Savages!
Whilst the Republicans have played this game well in recent years, it's not that long ago that the Democrats were at it equally successfully, and in many states they still do it. Which is not to suggest that it's a good thing - but let's not get partisan about it.
Actually if anything they're more serious than the external ones. In an ordinary district the candidates from each major party have to compete for a majority of voters. In a gerrymandered "safe" district the other side is never going to win in the first place so the real contest isn't between each side but rather during the primary to see who's can be more extreme.
A bullet may have your name on it but splash damage is addressed "To whom it may concern."
Besides the effect on lawmaking (or failure to pass laws under gridlock), gerrymandering gives people on both sides of issues a sense of majority. "I won in a landslide, I must be right", combined with polarized news programming, has been demonstrated to make people dumber. Harvard Business Review has an interesting article this week on opinion reinforcement and groupthink this week [ https://hbr.org/2014/11/making... ], which compares focus groups from liberal Boulder CO USA and conservative Colorado Springs USA. The researchers documented the negative effects of grouping like-minded people in political discussions. I think gerrymandering has the same effect on political intelligence. Their own conservatism or liberalism appears validated by landslide elections in their districts.
Gently reply
The great virtue of 'first past the post' is that it forces parties to appeal to a wider group than their obvious supporters...
I'm not sure that's necessarily true, but what FPTP does do is push everything towards a two-party state. This is why you get, effectively, extremists on both sides. Case in point: UK and USA. Minor parties are pushed out, moderate viewpoints are ignored. FPTP directly leads to "Us v. Them" contests.
In fact, thinking more about your first point: I don't think it's quite true. FPTP encourages parties to talk negatively about their opponents rather than push their own positive points.
"If you think the problem is bad now, just wait until we've solved it." --- Arthur Kasspe
Each state should create a competitive contract and to build a re-districting computer program. The requirements for the program should include
* The only data fed to the program is geographic markers the will provide convenient district borders (railroad lines, roads, rivers, county and city borders, etc.) and the number of people within each section. No other demographic data (age, race, previous voting patterns, income, etc.) will be input into the program
* the program will be completed 2 years before the redistricting and be open source so that anyone can inspect it and run it and get the same result
* the program will take a random seed as input and will generate different results based on that seed.
The geographic data will also be made public 2 years in advance of the redistricting
When the census data comes out it will be published as well.
On the big day they'll hold a lotto-type drawing to select the random seed. At that point anyone - researchers, journalists, some kid in his basement - can run the program and know the result before it is even published by the government. If the result isn't what everyone else expects we'll know there was funny business.
The program will be fair because the kind of data that allows gerrymandering simply won't be permitted as input. Any sneaky attempts to use something like population density as a proxy will be something anyone can find and complain about in the open source code. Neither party will be able look at the results ahead of time, see that by chance it gives a slight advantage to their opponents, and scuttle the process because the outcome won't be available until the random seed is drawn.
I often don't like the choices people make, but I like the fact that people make choices. That's why I'm a conservative.
Each state should create a competitive contract and to build a re-districting computer program. The requirements for the program should include
* The only data fed to the program is geographic markers the will provide convenient district borders (railroad lines, roads, rivers, county and city borders, etc.) and the number of people within each section. No other demographic data (age, race, previous voting patterns, income, etc.) will be input into the program
* the program will be completed 2 years before the redistricting and be open source so that anyone can inspect it and run it and get the same result
* the program will take a random seed as input and will generate different results based on that seed.
The requirements also include obvious stuff like how spread out or compact districts must be, how many can be disconnected, etc.
The geographic data will also be made public 2 years in advance of the redistricting
When the census data comes out it will be published as well.
On the big day they'll hold a lotto-type drawing to select the random seed. At that point anyone - researchers, journalists, some kid in his basement - can run the program and know the result before it is even published by the government. If the result isn't what everyone else expects we'll know there was funny business.
The program will be fair because the kind of data that allows gerrymandering simply won't be permitted as input. Any sneaky attempts to use something like population density as a proxy will be something anyone can find and complain about in the open source code. Neither party will be able look at the results ahead of time, see that by chance it gives a slight advantage to their opponents, and scuttle the process because the outcome won't be available until the random seed is drawn.
I often don't like the choices people make, but I like the fact that people make choices. That's why I'm a conservative.
And which side do you fear? The Republicrats or the Demoblicans?
Both, if you're smart.
Anecdotal evidence? Germany, Sweden, Finland, Norway, Denmark, Belgium, the Netherlands and more use proportional representation systems to some extent. Hardly unstable countries.
If I could design a voting process, I would use the condorcet method and proportional representation.
Just change to a proportional election system instead. Let the percentage of votes decide how many seats a party will get.
It will of course invite other parties to the election party as well.
If builders built buildings the way programmers wrote programs, then the first woodpecker would destroy civilization.
Personally, I find it all to be a bunch of bullcrap. Have you seen those voting districts that are along, squiggly lines that wander all over the place?
Yeah, and you know what? One of the most famous ones is in North Carolina, the site of this study.
And guess who created it and why? Democrats did, in order to secure a minority voting block big enough to elect a black person to Congress. Ever since, it's been one of the most litigated districts in the U.S.
I'm always shocked at how many people don't realize that this is one of the primary LEGAL rationales for gerrymandering -- back in the 1980s and 1990s you even saw unholy alliances between minority leaders and conservative Republicans conspiring to create awkward districts in some states that would give each group what they wanted: the minorities got enough people together in a district to elect a minority to Congress, and the Republicans got to excise many of those annoying mostly Democratic minority voters from their districts.
We are still living with that legacy in many states, and I frankly have found news coverage in recent years of gerrymandering to be lacking in discussion of this issue. It's not all just Republicans who have taken control of state legislatures -- we've also had a committed effort for quite a few decades to segregate voter districts in such a way that would allow more minorities in Congress.
But of course that creates a problem, because it ends up disenfranching non-minority Democrats who get stuck in all the surrounding districts that can no longer elect a Democrat because a large portion of Democrats were deliberately removed from swing districts to create the minority-majority district.
So the Democrats end up in a Catch-22. If they want to promote Congressional "diversity," they can create districts where minorities get elected, but they can end up screwing themselves over in the process because then all the surrounding districts become more Republican and make it more difficult for Democrats to actually achieve an overall Congressional majority.
It's certainly not the only issue that has led to Republican majorities in Congress -- but it's one that's not often talked about, and it has had some significant effects.
North Carolina, in fact, has a very famously gerrymandered district for this reason ( http://en.wikipedia.org/wiki/North_Carolina%27s_12th_congressional_district ); and I'm sure the simulation ignored that detail.
If a state has ten districts, and 10% of the population is black, under current federal law, they have to do their best to give blacks a majority vote in one district. But that generally means that nine of that states representatives can completely ignore the black vote. It seems to me that the world would be a better place if black voters made up 10% of each district's population: they could swing the election in any of those districts, and each representative would have a very strong reason to listen to their concerns.