This
1:42 minute video shows a risk terrain map for gun shootings as a
layer in Google Earth (The map shown here is actually
one of the final products from the RTM workshop webinar). Using the "Street View" tool, this video demonstrates how
differences between
lower-risk places and higher-risk places (as defined by the RTM) can be observed in
real terms within the environmental context of micro-level places
in Irvington.
RTM
is an approach to spatial risk
assessment that "paints a picture" of the landscape
where crimes are most likely to occur given the spatial influence
of certain environmental
risk factors. It can articulate perceptions of risk at places
beyond merely referencing past occurrences of reported crimes.
From this basic (and quite "cool") exploration of
street-level context and place-based risk, the process of
operationalizing spatial influences of crime correlates for use in
a risk terrain model seems to be a method with both face validity
and statistically significant predictive validity.
So,
here's an idea: Create and validate risk terrain maps based upon
environmental factors that are currently known to correlate with the specific type of crime (i.e., based on empirical research). Then export
the map as a KMZ file and open it in Google Earth. Using the 'Street
View' functionality
of Google Earth, "walk along" the streets in various high-risk
and low-risk places to identify new mitigating and aggravating risk
factors that might improve your risk terrain model. For example, while
doing this with the risk terrain map for shootings in Irvington, we
quickly noticed that the environmental context of lower-risk places often
included churches or had street intersections with traffic lights nearby.
Traffic lights might represent places with higher-than-normal volumes of
people/traffic (i.e., more guardianship). Churches in these areas might
have mechanisms that create similar deterrent effects on shooting
incidents. These mitigating features of the environment could be
operationalized to risk map layers, empirically validated, and included in
new risk terrain models. Google Earth can provide additional street-level
context to the "first iteration" of maps produced with RTM.
(This may be a fun and useful activity to try at CompStat
meetings!).
