The importance of crime mapping
for developing intelligence
Spencer Chainey
Director of Geographical Information Science
UCL Jill Dando Institute of Crime Science
Introduction
• Developing intelligence
– In general terms
– Understanding scanning and analysis
• Developing intelligence using crime mapping
– Two examples
• Scanning by non-analysts
• Developing explanatory intelligence
Developing intelligence
• Is not,
– merely about creating pretty maps, identifying patterns and
comparing trends
– Or just confirming the pattern you are asked to analyse!
• It is about examining the underlying conditions of both the
simple and complex problems
– Explaining rather than just presenting and describing the crime
problem
Developing intelligence
• Its about applying SARA: Scanning:Analysis:Response:Assessment
– ‘Scanning’ is about identifying, defining, classifying and exploring
the nature of the problem
• Setting up testable ideas – hypotheses
• It also involves thinking about what data to use
• ‘Analysis’ interprets the problem and generates
intelligence that helps to understand and explain the
problem
– It requires the specifics of the crime problems to be explored
• Analysis is vital for an intelligence-led and problemoriented approach
– And geographical analysis is a vital component of analysis
Developing intelligence
3i Model
Ratcliffe (2004)
Applying SARA against this model
Developing intelligence
• So we need to be better at scanning
– Involve non-analysts
• They are probably best placed to pose testable ideas
• And better with our analysis
– Generate explanatory rather than just descriptive intelligence
products
– Intelligence products that are fit for the purpose of helping to direct
how to bring about a positive impact on the criminal environment
• Two examples
– Scanning tool for non-analysts
– Being crime specific and testing hypotheses
Involving non-analysts in scanning
• Often interested in exploring beyond performance statistics
• If we want them to pose testable ideas,
– We need to develop information systems/technology that suit
this audience
• Easy to use, reward the user with the information they require
within three mouse clicks, available on one page, encourages
them to explore
Producing explanatory analysis
“We think it relates mainly to local residents having their cars
stolen at night” (The Police)
T h e ft o f v e h ic le s b y tim e o f d a y (a o r is tic tre n d )
Theft of vehicles by time of the day
350
300
250
200
150
100
50
2200+
2000+
1800+
1600+
1400+
12.00
1200+
1000+
0800+
0600+
00.00
0400+
0
0200+
•
Analysis involves breaking
the problem apart and
exploring the specifics of the
crime problem
We have a series of questions
that we can turn into hypotheses
0000+
•
23.59
Victims
Mar 01 – Feb 04
Theft of vehicles
Camden victims (58%)
Non-Camden victims (42%)
Type of vehicles stolen
Vehicle type description
Offences
%
Hatchback
1258
21.7%
Saloon
1433
24.7%
Estate
220
3.8%
People carrier
45
0.8%
Convertible
120
2.1%
Sports
42
0.7%
4 X 4's
4
0.1%
Moped or scooter
1494
25.8%
Motor cycle
755
13.0%
Motor caravan
11
0.2%
Van
274
4.7%
Other
50
0.9%
Not known
23
0.4%
Cars
Motorbikes
Scooters and mopeds
Recovery status of vehicles
Recovery status of stolen vehicles in Camden
60.0%
% all vehicles
% cars
50.0%
% motorbikes
40.0%
% scooters and mopeds
30.0%
20.0%
Unknown
Other
Stolen/Taken
Recovery status
Rec
Part/Dam
Rec Part
Rec Intact
Rec
Damaged
Rec DM/Prop
Damaged
No Interest
0.0%
Not Stolen
10.0%
T h e f t o f v e h ic le s b y t im e o f d a y ( a o r is t ic t r e n d )
Theft of vehicles by time of the day
350
300
250
Place: space
and time
200
150
100
50
2200+
2000+
1800+
1600+
12.00
1400+
1200+
1000+
0800+
0600+
0400+
00.00
0200+
0000+
0
23.59
Vehicle type
Camden
Clerkenwell (n)
Clerkenwell(%)
Car
51%
41
18%
Sports or convertible
3%
5
2%
Scooter or moped
26%
95
42%
Motor cycle
13%
70
31%
Van
5%
3
1%
Other
2.0%
10
4%
Not known
0.5%
0
0%
So it’s not all to do with residents having their cars stolen at night …
Summary
To be better at developing intelligence
• We need to be better at scanning
– We first of all really need to understand what is meant by scanning and
what it involves
– Identifying, defining, classifying and exploring the nature of the
problem
– Setting up testable ideas – hypotheses
– Involve non-analysts
• They are probably best placed to pose testable ideas
• But provide them with the tools they can use
•
We need to be better at analysis
– Generate explanatory rather than just descriptive intelligence products
– Intelligence products that help to deliver impact on the criminal
environment
•
Crime mapping can help us
– Recognising the inherent geographical quality of crime
– Not just in terms of presenting information
– But understanding why crime occurs