For our final lab of the semester, we examined the effects of scale and resolution on the properties of spatial data and explored the concept of Modifiable Area Unit Problem (MAUP) as it relates to districting and gerrymandering.
In vector data, scale affects how spatial patterns are represented and interpreted. When data are aggregated to larger spatial units such as like moving from census block groups to county level, there is statistical variation causing a bias in redistricting.
In raster datasets, resolution determines the level of spatial detail captured in the grid cells. High-resolution rasters, with smaller cell sizes, preserve finer spatial variation, while coarse-resolution rasters, with larger cell sizes, generalize the landscape.
The way district boundaries are drawn can affect how populations are aggregated, sometimes diluting certain groups while concentrating others. This can lead to one political group being packed into a few districts or spread thinly across many, altering election outcomes. This concept is called Gerrymandering. For this exercise, I calculated one of the identifiers of Gerrymandering, which is compactness of the shape of a district boundary, using the Polsby-Popper score formula.
Below is a screenshot of Maryland’s 8th Congressional District, which had the lowest Polsby–Popper compactness score. The district is split into two separate polygons without any clear geographic justification, indicating an unnecessary division likely caused by political boundary manipulation rather than natural features.
Comments
Post a Comment