"Predictive Policing" - Crime Mapping Software that Predicts When and Where Crime Will Happen
Posted on Sep 30, 2013 1:15pm PDT
A new tool used by some police departments uses software that computes
algorithms of past crime information to “predict” when and
where future crimes will take place. As the recent ABA Journal article
reports, “the idea is to compile past crime details, run them through
algorithms and identify future hot spots of specific crimes, such as burglary,
down to individual blocks or even smaller areas. “ See ABA Journal,
Predictive policing ma help bag burglars – but it may also be a
constitutional problem, by Leslie Gordaon (Sept. 1, 2013)
(seettp://www.abajournal.com/magazine/article/predictive_policing_may_help_bag_burglars–but_it_may_also_be_a_constitutio/).
In 2011, predictive policing was named one of the top new inventions by
Time Magazine. See id. It allows police departments to patrol certain
areas at certain times and take other preventative measures in an effort
to reduce crime.
Does it Work?
The FBI’s website posted an article reporting that Santa Cruz and
Los Angeles both developed studies to test the effectiveness of the predictive
policing software. See Preditive Policing: Using Technology to Reduce
Crime, By Zach Friend, M.P.P. (4/9/2013) (see
http://www.fbi.gov/stats-services/publications/law-enforcement-bulletin/2013/April/predictive-policing-using-technology-to-reduce-crime). In Santa Cruz, based on its study, the police department believes the
program reduced burglaries by 19 percent over a 6-month period. The LAPD’s
controlled experiment also established the model’s effectiveness.
In Los Angeles, they used the model in the Foothill Division of the city.
The department distributed maps of “hot spots” to officers
at the beginning of their shifts. On some days they used the predictive
policing software algorithms to establish the hot spots and on other days
they produced the maps using traditional LAPD hot spot predictors. The
officers had no idea whether the map they received was one produced by
the algorithms or traditional measures. The result of the study established
that the algorithm method provided twice the accuracy of the traditional
maps. The LAPD has now expanded the program to other divisions serving
a population of approximately 1.5 million people.
See id.
What’s the Problem?
If the predictive policing software is just used to deploy police officers
to certain areas, there are probably no constitutional issues. If merely
an increased police presence is all it takes to reduce crime, the predictive
policing method is working as intended. However, if the officers begin
to use the predictive policing hot spot information as a reason to suspect
people of criminal activity based on them being in a certain area at a
certain time, there are true Fourth Amendment concerns. Simply because
a person is at the “hot spot” area should not subject the
person to greater risk of a search compared to a person displaying the
same activity in a non-hot-spot location. The software data cannot take
the place of reasonable suspicion based on articulable facts observed
by police officers – and those articulable facts cannot include
that the map printout from predictive policing software indicated crime
was likely at that spot. The use of this technology provides a real concern
for “hot spot profiling,” that is placing a person under extreme
scrutiny based solely on being present in a hot spot instead of actually
displaying suspicious behavior. As this software becomes more popular,
there will inevitably be court cases raising these Fourth Amendment issues.