I would like to bring to the attention of the Ai-geostats
community the availability of an improved version of the CrimeStat spatial
statistics program. The program (version 2.0) is free. It was developed by
Ned Levine & Associates of Houston, Tx for the Mapping and Analysis for
Public Safety program of the National Institute of Justice (NIJ) under a
research grant. The program is Windows-based and interfaces with most
desktop GIS programs. The program provides statistical tools for law
enforcement agencies and criminal justice researchers in their crime
mapping efforts. However, it can also be used for a variety of other
applications for which researchers would be interested. It is being used by
many analysts and researchers in the United States and throughout the
world, including those in geophysics.
CrimeStat II has a collection of statistical tools for the
analysis of point/incident locations and includes a range of diagnostic and
modeling spatial statistics, including statistics for measuring spatial
distribution, for examining distances between incident locations, for
detecting hot spots, for interpolating one-variable and two-variable
density surfaces, and for analyzing space-time interactions.
The program inputs incident locations (e.g., the residence
location of a patient) in 'dbf', 'shp', ASCII, or ODBC formats using either
spherical or projected coordinates. It calculates various spatial
statistics and writes graphical objects to ArcView®, ArcGIS®, MapInfo®,
Atlas*GIS, Surfer® for Windows, and ArcView Spatial Analyst©. Version 2.0
includes many improvements to the functionality of the program including
ODBC access, the ability to save parameters and alternative reference
files, the output of Monte Carlo simulation data, and an improved help menu
that is linked to the manual.
New statistics in version 2.0 include the mode, the fuzzy mode,
the Spatial and Temporal Analysis of Crime module (STAC), a risk-adjusted
nearest neighbor clustering routine, the Knox index, the Mantel index, and
a Correlated Walk Analysis module. Many of the existing routines from
version 1.1 have been improved. Six of the statistical routines also have a
Monte Carlo simulation to approximate confidence intervals around the
Among the many applications for which CrimeStat would be useful is
in detecting spatial changes in the location of facilities by time period
(e.g., shifts in the locational pattern of manufacturing firms over time),
in identifying concentrations of incidents within small areas (e.g., the
location of motor vehicle crashes), in assessing where disease incidents
cluster relative to a baseline population (e.g., tuberculosis cases
relative to population), in identifying time sensitive hot spots (e.g., a
rash of auto thefts), in estimating the likely origin of a serial offender,
and in predicting the likely time and location for a next crime committed
by a serial offender (or the next time and location of an emerging
CrimeStat is accompanied by three sample data sets and a manual
that gives the background behind the statistics and examples. The current
manual includes examples on the use of CrimeStat by other researchers,
including some in planning. The program and sample data sets are in
Windows-based zipped files that can be downloaded. The manual is a set of
individual files in 'pdf' format that can also be downloaded.
The program, the manual, and sample data sets can be downloaded
free from the NIJ archivist at:
Ned Levine, PhD
Ned Levine & Associates
* To post a message to the list, send it to ai-geostats@...
* As a general service to the users, please remember to post a summary of any useful responses to your questions.
* To unsubscribe, send an email to majordomo@...
with no subject and "unsubscribe ai-geostats" followed by "end" on the next line in the message body. DO NOT SEND Subscribe/Unsubscribe requests to the list
* Support to the list is provided at http://www.ai-geostats.org