Duke study shows steep tilt in redistricting maps

Published December 5, 2014

Editorial by News and Observer, December 4, 2014.

Researchers at Duke University recently established what North Carolina Democrats already know too well: The GOP-drawn redistricting maps are hopelessly weighted in favor of Republicans.

Still, it’s interesting to see just how weighted they are. A recent North Carolina Public Radio report focused on  a mathematical assessment led by Duke math professor Jonathan C. Mattingly and one of his students, Christy Vaughn. Mattingly was intrigued by the results of the 2012 election in which Republicans won nine out of 13 congressional districts despite there being more votes statewide for Democrats. (This fall the GOP won 10 out of 13.)

“Right away I was very interested in this project because it’s just such a stark result that so few seats were (awarded to Democrats) when the popular vote was so different,” Vaughn told WUNC.

Mattingly and Vaughn developed a mathematical algorithm that could redraw the state’s congressional district boundaries into 100 different maps. All of the maps met two basics requirements for a congressional district: Each district had to have roughly the same population, and each district had to be compact.

When the pair ran the 2012 vote counts through the 100 map versions, this was the result: Between six and nine Democrats were elected 95 percent of the time.

No wonder Republicans have cooled the enthusiasm they had while in the minority for districts drawn by nonpartisan legislative staff.

The Duke maps did not account for racial-balance requirements that Republicans say forced them to pack African-American voters into the few districts Democrats won. And there’s no law against gerrymandering for political gain.

Still, Mattingly hopes the results will give pause to Republican lawmakers who, no matter their feeling about Democrats, still have some affection for democracy.

“It wasn’t representative of the will of the people, and you ask yourself, is this democratic?” the professor said. “I mean, if we really want to be in a democracy, we should put in safeguards like we do for other things. We should put in safeguards to protect against gerrymandering in either direction.”

December 5, 2014 at 11:52 am
Richard Bunce says:

There is an easy answer to this issue...

"Mattingly was intrigued by the results of the 2012 election in which Republicans won nine out of 13 congressional districts despite there being more votes statewide for Democrats."

The members of the US House are elected by House of Representative Districts within a State, not at large by the entire State population. It may be intriguing but it is also irrelevant.

December 5, 2014 at 11:59 am
Richard Bunce says:

... and I favor all election districts be created by algorithm with only number of districts, area to be divided, distribution of population in the area, and perhaps County boundaries. No population demographic data such as age, sex, political party affiliation, religion, ethnicity, etc. other than address or perhaps better yet Census block would be allowed as input to the algorithm.

Any direct human intervention, even by a supposedly nonpartisan panel, will be subject to gaming the system. Interesting that the left has only recently become interested in this topic... after decades of abusing redistricting, including using the US DoJ to redraw State maps based on Census "race" data which is nothing more than the respondents self identification that cannot be validated and can change each time asked.