1 of 27

CrimeStat IV

Susan C. Smith

Christopher W. Bruce

Revised by: Thomas Mueller

2 of 27

About CrimeStat PowerPoint

  • This PowerPoint was revised from the Susan C. Smith and Christopher Bruce’s work. The Original can be found at:
  • http://www.icpsr.umich.edu/CrimeStat/workbook.html

​

  • Their PowerPoint and Manual was developed for CrimeStat 3.0. This PowerPoints was revised for CrimeStat 4.0
  • The GeoTech Center wants to thank both of them for allowing us to use this tremendous resource.

​

​

3 of 27

Chapter Five��Hot Spot Analysis�Part 1

4 of 27

In Chapter Five...

  • Summary of different hot spot routines
  • Mode and fuzzy mode
  • Nearest neighbor hierarchical spatial clustering
  • Spatial and Temporal Analysis of Crime

5 of 27

Introduction

  • Identifying hot spots
    • A spatial concentration of crime, or
    • A geographic area representing a small percentage of the study area which contains a high percentage of the studied phenomenon
    • Can be on a variety of scales
      • A hot address
      • A hot office building
      • A hot block
      • A hot area

6 of 27

Hot Spot Routines

  • Mode
    • Identifies the geographic coordinates with the highest number of incidents
  • Fuzzy Mode
    • Identifies the geographic coordinates, plus a user-specified surrounding radius, with the highest number of incidents
  • Nearest-Neighbor Hierarchical Spatial Clustering
    • Builds on NNA by identifying clusters of incidents

​

7 of 27

Hot Spot Routines

  • Spatial & Temporal Analysis of Crime (STAC)
    • Alternate means of identifying clusters by “scanning” the point and overlaying circles on the map until the density concentrations are identified
  • K-Means Clustering
    • User specifies the number of clusters and CrimeStat positions them based on the density of incidents

8 of 27

Hot Spot Routines

  • Aneslin’s Local Moran statistic
    • Compares geographic zones to their larger neighborhoods and identifies those that are unusually high or low
  • Kernel Density Interpolation
    • A spatial modeling technique

9 of 27

Mode – 1

  • Just counts the number of incidents at one spot
    • Note: same address vs. X & Y coordinates
      • Which is your records management or CAD system receiving?
      • How would this effect the mode?

10 of 27

Mode – 2

  • Would you think this would be rare?
  • Crimes in a parking lot will usually have the address of the business even if the parking lot surrounds the building.
  • Same address if a crime happens in the north or south parking lot. Thus same coordinates

11 of 27

Mode

  • Set up a new CrimeStat session
  • In Data Setup add the shapefile: theftsfromautos
  • Set Projected and Feet
  • Add the rest of the information similar to previous labs

12 of 27

Mode

13 of 27

Mode

  • Check Mode on Spatial description / Hot Spot Analysis I tab
  • Click Save result to: theftsauto1
  • Click compute

​

14 of 27

Mode

  • Top 45 locations, ordered by frequency

​

15 of 27

Mode

  • Save result to (.dbf)
  • (You could then import to GIS)

16 of 27

Mode

  • Add theftfromautos, streets, and CityExt Shapefiles
  • Then add Modetheftfromautos1.dbf

​

17 of 27

Mode

  • Right Click on DBF – Display X Y data

​

18 of 27

Mode

  • Save result to (.dbf)
  • (You could then import to GIS)

19 of 27

Mode

  • Input Table – Modetheftsfromautos1
  • Output – Keep the standard name
  • X = X, Y = Y,
  • Coordinate System – theftsfromautos (it will change it to Lincoln
  • Click Ok

20 of 27

Mode – Map Result

21 of 27

Fuzzy Mode

  • User can specify a search radius around each point
    • Hence, it will include all of the points within that radius in the count
  • For agencies with GPS data, may be only way to find hot spots
    • Unlikely two incidents will have identical X & Y coordinates

22 of 27

Figures 5-3 and 5-4: Accidents at several intersections. The agency has been ultra-accurate in its geocoding, assigning the accidents to the specific points at the intersections where they occur. The mode method (left) would therefore count each point only once, whereas the fuzzy mode method (right) aggregates them based on user-specified radiuses

23 of 27

Fuzzy Mode

  • Return to CrimeStat screen
  • Uncheck Mode / Check Fuzzy Mode
  • Search radius of 500 feet

​

24 of 27

Fuzzy Mode

  • Save result to DBF file to c:\CrimeStat\Data\Lincoln\HS.dbf

​

25 of 27

Fuzzy Mode

  • Compute
  • Note different results from Mode

​

26 of 27

Fuzzy Mode

  • Now complete the similar steps in ArcGIS Pro from the Mode option

​

27 of 27

Fuzzy Mode