Predictable Policing: measuring the Crime Control Benefits of Hotspots Policing at Bus Stops
Objectives: A fairly robust body of evidence suggests that hotspots policing is an effective crime prevention strategy. In this paper, we present contradictory evidence of a backfiring effect. Methods: In a randomized controlled trial, aimed at reducing crime and disorder, London's 'hottes...
| Authors: | ; |
|---|---|
| Format: | Electronic Article |
| Language: | English |
| Published: |
2017
|
| In: |
Journal of quantitative criminology
Year: 2017, Volume: 33, Issue: 4, Pages: 809-833 |
| Online Access: |
Presumably Free Access Volltext (Resolving-System) |
| Journals Online & Print: | Drawer...
|
| Check availability: | HBZ Gateway |
| Keywords: |
| Summary: | Objectives: A fairly robust body of evidence suggests that hotspots policing is an effective crime prevention strategy. In this paper, we present contradictory evidence of a backfiring effect. Methods: In a randomized controlled trial, aimed at reducing crime and disorder, London's 'hottest' 102 bus-stops were targeted. Double patrol teams of Metropolitan Police Service uniformed officers visited the stops three times per shift (12:00-20:00), 5-times per week, for a duration of 15 min, over a 6 month period. Crucially, officers arrived and departed the bus stop on a bus, with significantly less time spent outside the bus stop setting. Outcomes were measured in terms of victim-generated crimes reported to the police and bus driver incident reports (DIRs), within targeted and catchment areas. We used adjusted Poisson-regression models to compare differences in pre- and post-treatment measures of outcomes and estimated-marginal-means to illustrate the treatment effect. Results: DIRs went down significantly by 37 % ( p = 0.07) in the near vicinity of the bus stops (50 m), by 40 % in the 100 m catchment area ( p = 0.04) and marginally and non-significantly in the farthest catchment (10 %; p = 0.66), compared to control conditions. However, victim-generated crimes-the primary outcome measured in previous experiments-increased by 25 % ( p = 0.10) in the near vicinity, by 23 % ( p = 0.08) and 11 % ( p ≤ 0.001) within the 100-150 m catchment areas, respectively. Conclusions: These findings illustrate the role of bounded-rationality in everyday policing: reductions in crime are predicated on an elevated perceived risk-of-apprehension. Previous studies focused on clusters of addresses or public facilities, with police moving freely and unpredictably within the boundaries of the hotspot, but the patrol areas of officers in this experiment were limited to bus stops so offenders could anticipate their movements. Hotspots policing therefore backfires when offenders can systematically and accurately predict the temporal and spatial pattern of long-term targeting at a single location. |
|---|---|
| ISSN: | 1573-7799 |
| DOI: | 10.1007/s10940-016-9312-y |