Predicting Burglary Hotspots

advertisement
Predicting Burglary Hotspots
Shane Johnson, Kate Bowers, Ken Pease
Repeat Victimization
• Prior victimisation is an excellent predictor of future risk
(Burglary, DV, CIT, hotel theft……)
Number of repeat burglaries per interval
• Repeat burglary victimization occurs swiftly (e.g. Polvi et al.,
1991) 250
200
150
100
50
0
1
6 11 16 21 26 31 36 41 46 51 56 61 66 71 76 81 86 91 96 01
1
Weeks between events
Conference’s name here
00.00.00
Theories of rv
• A sports team loses the first two matches of
the season. Why did it lose the second one?
Was it because the first result reflected the
fact that it was a poor team, and it was still
a poor team at the time of the second
match? This is a flag account. Alternatively,
did the first result destroy its confidence so
that it played tentatively in the second
match? This is a boost account
Conference’s name here
00.00.00
Explaining Repeat Victimisation
Risk heterogeneity/Flag hypothesis
Some households are always at more risk than others
– Flag accounts alone encounters problems explaining the timecourse of repeat victimisation (Johnson, 2008)
Johnson, S.D. (2008). Repeat burglary victimisation: A Tale of Two Theories. J Exp Criminol, 4: 215-240.
Conference’s name here
00.00.00
Explaining Repeat Victimisation
Boost Account
•
Repeat victimisation is the work of a returning offender
•
Optimal foraging Theory - maximising benefit, minimising risk and keeping
search time to a minimum– repeat victimisation as an example of this
– burglaries on the same street in short spaces of time would also be an
example of this
•
Consider what happens in the wake of a burglary
– To what extent is risk to victim and nearby homes shaped by an initial
event?
Conference’s name here
00.00.00
Ashton Brown and Senior
• “The house would be targeted again ‘a few
weeks later’ when the stuff had been
replaced and because the first time had
been easy...”“It was a chance to get things
which you had seen the first time and now
had a buyer for”.“Once you have been into
a place it is easier to burgle because you
are then familiar with the layout, and you
can get out much quicker”
Conference’s name here
00.00.00
Gill and Pease (and Everson)
• repeat robbers of the same target were
more determined, more likely to carry a
loaded gun, and more likely to have
committed a robbery where someone had
been injured. They had longer criminal
records, were more likely to have been in
prison before, and for a sentence upwards
of five years. They planned their robberies
more, and were more likely to have worn a
disguise.
Conference’s name here
00.00.00
Repeat Victimisation Makes Time
Central
• Surprising how often
time is neglected in
police mapping.
• Repeat victimisation is
a special case of risk
communication
Conference’s name here
00.00.00
Fortnightly variation
Burglary
Concentration
High
Low
Conference’s name here
00.00.00
Morning shift
Conference’s name here
00.00.00
Afternoon shift
Conference’s name here
00.00.00
Overnight shift
Conference’s name here
00.00.00
An analogy with Disease Communicability
• Communicability - inferred from closeness in space and time of
manifestations of the disease in different people.
++ +
++
+
Conference’s name here
++ +
+ +++ +
+
area
+
burglarie
s
00.00.00
Neighbour effects for all housing: Near repeats
2500
Not all on same day
Number of burglaries
2000
Burgled home
1500
1-week
2-week
1000
500
0
5
4
3
2
1
0
1
2
3
4
5
Doors apart
Bowers, K.J., and Johnson, S.D. (2005). Domestic burglary repeats and space-time clusters: the dimensions of
risk. European Journal of Criminology, 2(1), 67-92.
Conference’s name here
00.00.00
4500
num ber of burglaries
4000
3500
same side equivalent
opposite side
3000
2500
2000
1
2
3
4
5
6
7
8
9
10
house numbers apart (opposite side of street)
Conference’s name here
00.00.00
Communicability of Risk
8
6-8
6
4-6
4
2-4
2
0-2
residual
0
-2-0
-4--2
-2
-6--4
-4
300
400
500
600
700
800
900
1000
200
1 2
3 4
5 6
repeats
time(months)
100
-6
distance (m)
Johnson, S.D., and Bowers, K.J. (2004). The burglary as clue to the future: the beginnings of prospective hot-Spotting. European
Conference’s
here
00.00.00
Journal of
Criminology, name
1(2), 237-255.
International comparison
(burglary)
• Using a modified technique to test for disease
contagion
– Demonstrated pattern is statistically reliable in five
different countries:
• USA, UK, Netherlands, Australia, New Zealand
Johnson, S.D. et al. (2007). Space-time patterns of risk: A cross national assessment of residential burglary
victimization. J Quant Criminol 23: 201-219.
Conference’s name here
00.00.00
Near Repeats: Patterns in detection
data?
For pairs of crimes:
– Those that occur within 100m and 14 days of each other
76% are cleared to the same offender
– Those that occur within 100m and 112 days or more of
each other
only 2% are cleared to the same offender
Johnson, Summers & Pease (2009) Offender as Forager: A Test of the Boost Account of Victimization,
Journal of Quantitative Criminology, in press.
Conference’s name here
00.00.00
Forecasting - ProMap
Risk
High
Low
Bowers, K.J., Johnson, S.D., and Pease, K. (2004). Prospective Hot-spotting: The Future of Crime Mapping? The British J. of Criminology, 44, 641658.
Conference’s name here
00.00.00
Forecast Accuracy – Next 7 days
Accuracy (%)
Thematic
30
KDE
50
ProMap
63
ProMap (Backcloth)
71
Johnson, S.D., Bowers, K.J., Birks, D. and Pease, K. (2008). Predictive Mapping of Crime by ProMap:
Accuracy, Units of Analysis and the Environmental Backcloth, Weisburd, D. , W. Bernasco and G.
Bruinsma (Eds) Putting Crime in its Place. New York: Springer.
Conference’s name here
00.00.00
ProMap – Next 7 days (Merseyside, UK)
100
90
Promap
70
Simulation 95th percentile
Simulation mean
60
50
40
30
20
10
Johnson et al. (2008)
95
10
0
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
0
5
10
Percentage of burgary identified
80
Percentage of cells searched
Conference’s name here
00.00.00
ProMap*Backcloth – Next 7 days
(Merseyside, UK)
100
90
Percentage of burgary identified
80
Promap*Rds*Houses
70
Simulation 95th percentile
Simulation mean
60
50
40
30
20
10
Johnson et al. (2008)
0
5 10 15 20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 00
1
Percentage of cells searched
Conference’s name here
00.00.00
Retrospective KDE – Next 7 days
(Merseyside, UK)
100
90
Percentage of burgary identified
80
Retrospective KDE
70
Simulation 95th percentile
Simulation mean
60
50
40
30
20
10
Johnson et al. (2008)
95
10
0
90
85
80
75
70
65
60
55
50
45
40
35
30
25
20
15
5
10
0
Percentage of cells searched
Conference’s name here
00.00.00
Thematic map – Next 7 days (Merseyside, UK)
100
90
Percentage of burglary identified
80
Beats by rate per HH
Simulation 95th Percentile
Simulation Mean (N=99)
70
60
50
40
30
20
10
Johnson et al. (2008)
0
-
5
10
15
20
25
30
35
40
45
50
55
60
65
70
75
80
85
90
95 100
Percentage of area
Conference’s name here
00.00.00
Confidence versus Accuracy
High Accuracy
Pearson’s correlation (r) =
non significant
Spearman’s rho (rs)
= non significant
Low Accuracy
Confidence
does
not predict
accuracy
Low Confidence
Conference’s name here
High Confidence
00.00.00
Towards an Operational Tool
1. Ensure police geocoding accuracy
2. Check spatio-temporal patterns for each offence
type, initially and periodically thereafter
3. Obtain senior police prevention preferences by
crime type and crime mix
4. Optimise patrol routes
5. Identify stable and emergent spates and hotspots
for bespoke action
Conference’s name here
00.00.00
Presumptive routing
Conference’s name here
00.00.00
Publications
Johnson, S.D., & Bowers, K.J. (2004). The Burglary as Clue to
the Future: The beginnings of Prospective Hot-Spotting.
Bowers, K.J., & Johnson, S.D. (2005). Domestic Burglary
Repeats and Space-time Clusters: the Dimensions of Risk.
Johnson, S.D., & Bowers, K.J. (2004). The Stability of Spacetime Clusters of Burglary.
Bowers, K.J., Johnson, S.D., & Pease, K. (2004). Prospective
Hot-spotting: The Future of Crime Mapping?
Bowers, K.J., & Johnson, S.D. (2004). A Test of the Boost
explanation of Near Repeats. Western Criminology Review.
Johnson, S.D., Bowers, K.J., Pease, K. (2005). Predicting the
Future or Summarising the Past? Crime Mapping as Anticipation.
Launching Crime Science.
Conference’s name here
00.00.00
Conference’s name here
00.00.00
Thank you
Download