Scholarly Works, Center for Geospatial Information Technology (CGIT)
Permanent URI for this collection
Browse
Recent Submissions
- Asymmetries in Potential for Partisan GerrymanderingGoedert, Nicholas; Hildebrand, Robert; Travis, Laurel; Pierson, Matthew (2024)This paper investigates the effectiveness of potential partisan gerrymandering of the U.S. House of Representatives across a range of states. We use a heuristic algorithm to generate district maps that optimize for multiple objectives, including compactness, partisan benefit, and competitiveness. While partisan gerrymandering is highly effective for both sides, we find that the majority of states are moderately biased toward Republicans when optimized for either compactness or partisan benefit, meaning that Republican gerrymanders have the potential to be more effective. However, we also find that more densely populated and more heavily Hispanic states show less Republican bias or even Democratic bias. Additionally, we find that in almost all cases we can generate reasonably compact maps with very little sacrifice to partisan objectives through a mixed objective function. This suggests that there is a strong potential for stealth partisan gerrymanders that are both compact and beneficial to one party. Nationwide, partisan gerrymandering is capable of swinging over one hundred seats in the U.S. House, even when compact districts are simultaneously sought.
- Attribute Standardization of Car Crashes and Its Potential UsesMitchell, Allison; Hamilton, Lonnie III; Newman, Joseph (Virginia Tech, 2019-04-26)The primary focus of our research endeavor centers around standardizing the spatial attributes of police-reported crash records in the Commonwealth of Virginia. The Center for Geospatial Information Technology at Virginia Tech (CGIT) is working in support of the Virginia Department of Motor Vehicles Highway Safety Office’s mission to improve public safety. This data will be used by highway safety officials to identify particularly dangerous intersections and road segments across the commonwealth. We evaluated the crash factors and characteristics present in the dataset to better understand the potential that geospatial techniques can provide to the highway safety community. We elected to analyze crashes involving the black bear (Ursus Americanus) to see what observation could be made. To start, it was necessary to define the criteria to identify crashes involving bears. This was initially done manually by using an SQL request to obtain all records from 2018 crash data where the word ‘bear’ is referenced in the officer’s narrative. From there, we conducted a manual sort of the remaining data to help craft future, more efficient SQL requests for other years Once all records involving bears have been found, the data will be rendered in ArcGIS. Some exploratory analyses we plan on conducting involve identifying routes with high incidences of bear-related crashes, overlaying the crashes with known Wildlife Urban Interfaces (zones where housing density >6.17 housing units/km2 and vegetation cover >50%), and overlaying the crashes with the known habitats of the black bear in the commonwealth to observe if and how they may differ.