Module 1 - Crime Analysis

  The topic of this course's first module was hotspot analysis. We were tasked to make three maps of 2017 Chicago homicides using local clustering. The three methods used were grid-based mapping, kernel density and local Moran's I.

   From left to right: grid-based, kernel density and local Moran's I.

   The steps I took to create these maps were as follows:

Grid-based

     First, I used the Spatial Join Tool to join the overlaid grid with the 2017 homicides feature layers. Then, from the grids that contained greater than 0 homicides I exported the top quintile into a new feature layer. I used the Dissolve Tool to finalize the grid-based map.

 Kernel Density

    I started by using the Kernel Density Tool on the 2017 homicides point feature layer. I changed the output feature to consist of 2 classes and used the Reclassify Tool. Lastly I used Select By Attributes to export the gridcode 2 features to use as the final map.

Local Moran's I

    The first step was to use the Spatial Join Tool to join the census tracts and 2017 homicide features. After, I used the Cluster and Outlier Analysis (Local Moran's I) Tool on the spatially joined features using a calculated Crime Rate field. To finalize the map I used the Dissolve Tool to dissolve previously exported HH features.