Problem:
The goal of this project is to evaluate 2 different spatial pattern analysis techniques to determine the clustering of data.
Analysis Procedure:
ArcGIS was used for this analysis. Tools included Cluster and Outlier Analysis (Anselin Local Morans I) and Hot Spot Analysis (Getis-Ord Gi*). Data used was provided by NCSU’s GIS instructor. It was from GIS Tutorial 2 – Chapter 9.
The assignment consisted of two different exercises to determine clustering of data. The first involved exploring clustering of Fire House incident calls from January 2015. The tool used for this was Cluster and Outlier Analysis (Anselin Local Morans I). The distance was set as inverse distance and the tool was run. For this the clustering, z score and census data were displayed on the map so it was necessary to manipulate the symbology to appropriately represent the ranges of data observed. This process was then repeated for a fixed distance. The second exercise was exploring clustering of income within Dallas County. For this exercise the Hot Spot Analysis (Getis-Ord Gi*) tool was used with a fixed distance. In this case the map already displayed the wanted results.
Exercise 6.1a 
Exercise 6.1b 
Exercise 6.2
Application and Reflection:
The assignment has taught me two additional ways to determine if clustering is occurring in my spatial data. Being able to manipulate the symbology to display ranges for the results is crucial for layers like income because there are so many different possible values. The hot spot analysis tool would be useful when evaluating clustering of incomes compared to flood maps. The data needed to do this would be a state wide census for North Carolina as well as the FEMA flood map for North Carolina. This would allow me to specify a fixed distance to evaluate if there is a clustering of income across the state of North Carolina using the census and hot spot analysis. From this the flood maps could be overlayed and explored to see if there is an overlap between income and flooding. A hot spot analysis could also be performed on the census data and overlayed with the flood maps. This is important because people who are in danger of flooding situations could be financially impacted by such an event. And historically, lower income communities have been developed in this area. It would allow FEMA to better allocate money to help people out of the flood prone areas.
Spatial Problem Solving Portfolio 