We really enjoyed Sarah’s map. Remarkably, Sarah's metrics led the adult division in all three of the categories that she focused on: compactness, competitiveness, and limiting county splits. We were also impressed by the way she mapped Pittsburgh, using known communities of interest and the junction of rivers. Sarah’s level of outreach with her colleagues were impressive. Her work is a model for exactly the kind of conversations that need to be had across the Commonwealth in 2021 as new election maps get drawn for real.
My goal was to minimize splitting counties and communities of interest, while also attempting to prioritize compactness and competitiveness. During the process, I discussed the priorities I identified with fellow members of our Food Policy Council, where I work.
Gerrymandering limits the candidate slate, and affects election outcomes across the state. These elected officials' decisions then impact big topics, such as food insecurity. There was a general consensus that using existing boundaries, like counties, could help reduce confusion for voters who are also assigned to school board, city council, legislative, and congressional districts. Through my personal experience as a volunteer canvasser and enthusiastic voter, I can attest that the confusion is real.
I ended up with this map, in which 53 of Pennsylvania's 67 counties remain intact, a majority of districts are competitive, and most districts are minimally spread or distorted.
The larger urban cores of Pennsylvania posed the largest challenge to this approach. As a Pittsburgh resident, I have learned the importance of the city's unique geography to neighborhoods and municipalities and therefore identity and politics. To the extent it was possible, I wanted to use the rivers to guide the division of Allegheny County. Barring one unfortunately-shaped census tract, it seemed to work fairly well. I did feel limited in my ability to consider community borders on the east side of the state and therefore leaned heavily on District Builder's transparency toggle and clusters of smaller census tracts to guide my decision-making when splitting a county was necessary. This was easier with larger cities in rural areas, such as Reading and Scranton, and much more difficult closer to Philadelphia.
Maintaining county lines made it more difficult to distribute the population evenly among districts. On more than one occasion a district was almost devoid of county splits, only to be over- or under-populated by a frustratingly small margin. It also often felt at odds with competitiveness, given that party loyalties tended to be concentrated within certain counties. For example, I considered a larger split somewhere in District 8 to alter the 1.1 percentage point difference between registered Democrats and Republicans, but favored preserving county lines as much as possible. The district still felt sufficiently competitive.
Overall this approach allowed me to create a map that promotes a clear voting process and responsive candidates, both of which are necessary to the democratic process.
Maps are fun! Thanks for hosting this competition!