Spring 2022
Rice affiliates Mason Reece and Bob Stein worked as part of a multi-university team identifying where (ie the physical polling location) that people cast their ballots, if they voted in-person. They were particularly interested in studying the use of professional sports stadiums as polling locations, since this was a mostly new thing in 2020 and at least anecdotally they were very popular and well-run. Among the projects larger goals and data collection efforts, they used administrative voter files containing information on where each voter in a county cast their ballot in 2020.
They geocoded each of the voters in those county datasets and created some simple exploratory maps for the team to peruse, and which will most likely be part of their final report shared with the stakeholders invested in the project. Maps attached are for Washington DC and for Mecklenburg County, NC. There are markers for each of the stadiums used as polling places in the county, and blue voters are people who voted at the stadium, whereas red voters voted at some other type of polling location. They’ve used what is effectively kernel density in ArcGIS to achieve this, thus the varying opacities. In addition there is an alternative map for DC, which uses shades of blue to identify overall voter turnout in that precinct, and then uses a dot of varying size to show how many voters in that precinct cast their ballot at a stadium.
One of the main uses of ArcGIS was using the Network Analyst feature to calculate drive-times and geodesic distances between voters and each of the stadiums. There really isn’t a great way to visualize this but they have been making great use of the feature.
Density Map: Washington, DC
Density Map: Mecklenburg County, NC
Voter Turnout Map: DC Precinct Level
* information provided by Mason Reece