Master’s Defense for Kayley Smiley

Wednesday December 14th, 2022, 10 am – 12 pm in Student Commons Building 4113

Title: Border Analysis for Spatial Scan Methods

Committee: Joshua French (advisor), Adam Spiegler, Stephen Hartke

Abstract: Spatial scan methods are commonly used to identify and analyze geographic disease patterns. Most often, they are used to identify clusters or areas of high disease incidence. When testing for clusters, there are many proposed ways to construct the list of locations to test i.e., the set of candidate zones. Two of the most popular methods are the circular scan test and the elliptic scan test, which achieve their popularity from their accessibility, efficiency, and accuracy. One limitation of the scan methods is the inability to determine the exact border of a cluster. To combat this border uncertainty, Oliveira et al. proposed the F function, which calculates a value between [0, 1] for each region in the study area, where higher values indicate more evidence that the region belongs to the true cluster. In this paper, we make the F function for the circular scan test and the elliptic scan test available in R and explore the F function’s performance when used with both tests.