Evidence-Based Policing Matrix
Neighborhood – Rydberg et al. (2018)
Study Reference:
Location in the Matrix; Methodological Rigor; Outcome:
Neighborhood; General; Proactive; Rigorous; Mixed findings
What police practice or strategy was examined?
This study evaluated the implementation of a directed patrol intervention to reduce violent crime in Flint (MI) from January 2012- March 2014. Originally based on the data-driven approaches to crime and traffic safety (DDACTS) model, this intervention involved targeted increases in vehicle patrol presence within seven hot spots of violent crime throughout the city. Hot spot areas were identified based on spatial mapping of violent crime data and ranged in size from 13-364 blocks, often comparable to a police precinct. Increased patrol was achieved via collaboration between state and local police, with upwards of 33 state troopers and five sergeants involved and 14 hours per day dedicated to enforcement activities. Program activities primarily included proactive traffic stops focusing on locating fugitives and contraband.
How was the intervention evaluated?
Researchers compared trends in violent crime incidents (all violence, homicides, aggravated assaults, and robberies) before and after program implementation between the targeted areas and two sets of matched comparison areas that did not receive the program. Comparison areas were selected both within the targeted city and in a neighboring city (Detroit). These comparison areas were matched to the target areas on characteristics such as size, pre-intervention crime trends, population, racial characteristics, and measures of disadvantage (and weighted via synthetic control methods to best approximate pre-intervention crime levels and trends in the target areas). Impacts were estimated separately for each target area and all target areas combined (via meta-analysis).
What were the key findings?
The evaluation provided mixed results. Compared to the control areas within the targeted city, program hot spots experienced a significant 18% increase in total violence and a 33% increase in aggravated assaults, with no significant changes in homicide or robbery. Compared to the control areas within the neighboring city, program hot spots experienced a significant 26% increase in aggravated assaults and a significant 24% decrease in robberies, with no significant changes in homicide or total violence.
What were the implications for law enforcement?
It is unclear why the intervention produced evidence of both desirable and undesirable results. The authors suggest these results may be related to the DDACTS model not being fully implemented. Ideally, the model would involve problem-oriented policing tactics employed in areas where crime and traffic accidents overlap. The current intervention relied simply on increased patrol presence in areas with violent crime problems. The authors also note some evidence of a spike in domestic assaults in Flint that did not occur in Detroit. Domestic assaults are unlikely to have been affected by the intervention but perhaps accounted for the increase in aggravated assaults observed in treatment locations relative to the Detroit control locations.
Where can I find more information about this intervention, similar types of intervention, or related studies?