Search results

The purpose of this paper is to describe and test a quantitative harm-focused approach to offender selection for investigation and surveillance. The approach incorporates a measure of crime harm as well as a time-decay function that adjusts the score downward for offenders who desist from crime.

Across 10 of 21 high-crime police districts in the city of Philadelphia, the authors compare the mean harm scores of 60 prolific offenders selected by district analysts, 60 prolific offenders selected citywide by detectives assigned to the Gun Violence Reduction Task Force and the top 60 prolific offenders chosen from a harm-score generated list of known offenders in the ten high-crime districts.

The offenders on the harm-focused list have significantly greater mean harm scores than the offenders identified by the crime analysts or task force personnel. They have a significantly greater mean number of gun crime episodes in their offending history as well.

The harm-focused approach uses arrest data that may not accurately reflect convictions and which miss undetected criminal activity. A leader of a criminal organization who orchestrates criminal activities but does not engage directly may have a low harm score. Arrest data may also suffer from some inherent bias. The approach also requires the creation of a crime harm index. Determining the operational impact on overall crime reduction by focusing on offenders with higher harm scores will require further research.

Clinical methods of target selection based on officer intuition, opinion and experience may have limitations in terms of effectiveness and accuracy. They also lack transparency and may incorporate bias, a critical consideration given the current crisis in police-community trust and legitimacy. The actuarial method of weighing the harm of past offending with a crime harm score may be more acceptable and defendable to the community. It also identifies offenders with a higher frequency of involvement in gun crimes. Until methodological limitations are better understood, a compromise may be to start with the harm-score method (data-driven) and supplement this initial list through intelligence and investigative information.

The paper expands crime harm indices to quantify offender triage lists. The authors also empirically demonstrate through a case study that the approach is more effective at identifying harmful offenders than methods that solely rely on the experience or intuition of either crime analysts or detectives.

The purpose of this paper is to evaluate the violence-reduction effects following an FBI-led gang takedown in South Central Los Angeles.

The time series impact of the intervention was estimated using a Bayesian diffusion-regression state-space model designed to infer a causal effect of an intervention using data from a similar (non-targeted) gang area as a control.

A statistically significant 22 percent reduction in violent crime was observed, a reduction that lasted at least nine months after the interdiction.

The research method does make assumptions about the equivalency of the control area, though statistical checks are employed to confirm the control area crime rate trended similarly to the target area prior to the intervention.

The paper demonstrates a minimum nine-month benefit to a gang takedown in the target area, suggesting that relatively long-term benefits from focused law enforcement activity are possible.

Longer-term crime reduction beyond just the day of the intervention can aid communities struggling with high crime and rampant gang activity.

Few FBI-led gang task force interventions have been studied for their crime reduction benefit at the neighborhood level. This study adds to that limited literature. It also introduces a methodology that can incorporate crime rates from a control area into the analysis, and overcome some limitations imposed by ARIMA modeling.

The Kensington transit corridor runs between Huntingdon and Allegheny stations in the Kensington area of Philadelphia, Pennsylvania, and is one of the largest illicit drug areas in the country. The authors report qualitative findings from ride-alongs with transit police officers assigned to a vehicle patrol dedicated to reducing the response time to opioid overdoses in and around the transit system (trains and buses) in this large open-air drug market. This study's focus was on management and mitigation of the criminogenic harms associated with the illicit drug environment.

For ten months, transit officers patrolled the Kensington transit corridor in a dedicated vehicle (callsign “Oscar One”). Oscar One operated during either an early (8 a.m. to 4 p.m.) or late (4 p.m. to midnight) shift, between September 2020 and June 2021. 269 shifts were randomly selected for Oscar One from 574 possible shifts. Researchers accompanied Oscar One for 51 observations (19%), 45 of which were completed by the authors. Semi-structured interviews occurred during these shifts, as well as ethnographic field observations.

Four main themes emerged from the study. These centered on the role of law enforcement in a large drug market, the politics of enforcement within the city of Philadelphia, the policing world around risk and proactive engagement post–George Floyd, and the sense of police being overwhelmed on the front-line of community safety.

Police officers have a community safety as well as a law enforcement mandate, and this study explores the community safety and harm mitigation role from their perspective. The article draws on their words, based on approximately 400 h of field observation.

Purpose – This paper aims to provide an overview of organizational changes in the New Jersey State Police (NJSP) Investigations Branch and how the Branch has adapted to the paradigm of intelligence‐led policing. The paper also reports on interviews conducted to assess the impact on key staff affected by the change, through the medium of a drug‐gang investigation, Operation Nine Connect. Design/methodology/approach – Both semi‐structured and less formal interviews were conducted with 20 detectives, intelligence analysts and managers. Extensive quotes are employed to demonstrate key points. Furthermore, content analysis of documents related to organizational change in the NJSP and to a large drug‐gang intelligence operation was conducted. Findings – The paper identifies the key tenets of intelligence‐led policing, and describes progress made both organizationally and culturally to move the Investigations Branch to an intelligence‐led policing mindset. Organizational successes were reported, but more limited success was achieved in relation to changing police culture. Practical implications – The paper clarifies the meaning of intelligence‐led policing, and identifies potential road‐blocks to implementation for police departments wishing to move to intelligence‐led policing. Originality/value – The paper identifies the key tenets of intelligence‐led policing, outlines how these were used to determine greater geographic focus in the organizational structure of the New Jersey State Police Investigations Branch, and is a rare examination of the internal workings of a state police investigations branch in relation to a drug‐gang investigation. The paper will be of interest to police executives and managers, and intelligence professionals.

The purpose of this paper is to describe how the Philadelphia Police Department instituted a large‐scale randomized controlled trial of foot patrol as a policing strategy and experienced 23 percent fewer violent crimes during the treatment period. The authors examine whether activities patrol officers were conducting might have produced the crime reduction. The activities of foot and car patrol officers research takes a closer look at what types are examined separately and differences between car patrol activities pre‐intervention and during the intervention are explored. Activities of foot versus car patrol officers during the study period are compared across treatment and control areas.

Official data on police officer activity are used to compare activities conducted by foot patrol officers with those by car patrol officers in 60 treatment (foot beat) and 60 control areas consisting of violent crime hot spots. Activities of car patrol officers are described pre‐intervention and during the intervention. Foot patrol officers’ activities are described within treatment and control areas during the treatment phase of the experiment. Car patrol officers’ activities are reported separately. The statistical significance of changes in car patrol activity pre and during intervention is evaluated using a series of mixed model ANOVAs.

There were noticeable differences in the activities conducted by foot and car patrol. Foot patrol officers spent most of their time initiating pedestrian stops and addressing disorder incidents, while car patrol officers handled the vast majority of reported crime incidents. Car patrol activity declined in both treatment and control areas during the intervention but there was no statistically significant difference between the treatment and the control areas.

The major limitation of this study is the restricted set of data describing officer activity that is captured by official records. Future studies should include a more robust ethnographic component to better understand the broad spectrum of police activity in order to more effectively gauge the ways in which foot patrol and car‐based officers’ activities interact to address community safety. This understanding can help extend the literature on “co‐production” by highlighting the safety partnerships that may develop organically across individual units within a police organization.

The study provides evidence that individual policing strategies undertaken by agencies impact one another. When implementing and evaluating new programs, it would be beneficial for police managers and researchers to consider the impact on activities of the dominant patrol style, as necessary, to understand how a specific intervention might have achieved its goal or why it might have failed to show an effect.

The research contributes to the understanding of the separate and joint effects of foot and car patrol on crime. In addition, it provides police managers with a clearer picture of the ways in which foot patrol police and car‐based officers work to co‐produce community safety in violent inner‐city areas.

Expert assessment of organized crime (OC) group capabilities is often the basis for national threat assessments; it is rare, however, for variations in collective expert opinions of OC success factors to be systematically evaluated. The purpose of this paper is to examine the differences in how 150 criminal intelligence experts from a variety of national and organizational backgrounds sort and organize perceived attributes for OC group success.

The paper uses the Royal Canadian Mounted Police (RCMP) Sleipnir framework as a foundation for a Q-sort survey regarding the characteristics of OC group success. The survey was delivered to over 150 criminal intelligence specialists at a national conference in 2011. Descriptive statistics, seemingly unrelated regression, and biplots reveal different aspects of survey responses.

Results show that perceptions of the ingredients for OC group success both vary by nationality and by analysts’ level within the hierarchy of the law enforcement structure (local, state, national). These differences are marked; particular characteristics are viewed as differentially important for the perceived success of OC groups. Furthermore, the results suggest that there are shared and structured differences in perceptions of OC group success characteristics.

The survey has identified distinct differences between the characteristics for OC group's success perceived by analysts in the USA, Canada, and beyond. Furthermore, the organizational level of the analyst (local, state, national) shapes the perceptions of success factors. It is possible variations identified merely reflect differentials in training and experience, i.e. different organizational perceptions of the same problem. That aside, the patterning of results seem likely to be based to some degree on external factors linked to OC group operations, and not just on individual characteristics of the surveyed intelligence professionals.

The current research raises a number of questions regarding the confidence that should be placed in OC group assessments. The research has highlighted areas of professional dissonance that were not apparent from the RCMP Sleipnir research alone. Causes of the dissonance in assessments, and connections of these variations to both intelligence analysts’ experience, training, and organizational ethos; and to OC group capabilities, seem deserving of additional attention.

Expert intelligence analyst interpretation of OC group capability is central to most national risk and threat assessments, yet the assessment processes themselves are rarely examined. This is a unique survey of over 150 intelligence personnel that highlights significant differences in perceptions of OC groups, differences that raise questions about how the authors evaluate the OC threat.

The purpose of this paper is to generate information about the contours of police responsiveness, focussing on how quickly and precisely police make firearm arrests after a shooting incident.

Using a modified version of the Knox close pair method, a spatio-temporal clustering technique, over 11,000 shooting incidents and firearm arrests between 2004 and 2007 in Philadelphia, PA were analyzed.

Police are responding quickly and in a geographically targeted fashion to shootings. Across Philadelphia elevated patterns of firearm arrests were approximately two and a half times greater than would be expected if shootings and firearm arrests lacked a spatio-temporal association. Greater than expected patterns of firearm arrests persisted for roughly one-fourth of a mile and for about one week from the shooting incident but the strength of these associations waned over space and time. The pattern of police response varied slightly across different police divisions.

The current method uncovered spatio-temporal patterning and determined when these patterns were significantly different from what would be expected if the events were completely independent. Specific events and processes surrounding each event are not known.

Findings can help inform the knowledge about police behavior in terms of how police produce arrests.

The patterns observed here provide more micro-level detail than has been revealed in previous studies regarding police responsiveness to firearm violence while also introducing a more integrated spatially and temporally specific framework.

To examine the incidence and prevalence rate of drug‐related complaints against police in the New South Wales Police Service (Australia) and compare these rates to officer demographics.

Seven years of complaints data (1993‐2000) are examined. The data showed that of nearly 40,000 complaints amounting to over 80,000 allegations, less than 2 percent relate to drug‐related allegations. These allegations were isolated and the pattern of officer demographics from these incidents were compared to the police service as a whole, with the aim of exploring if particular groups (such as length of service, age, gender etc.) were particularly susceptible to attracting drug‐related allegations.

The most common drug‐related allegation was for supplying drugs. The distribution of drug‐related complaints follows the general demographic pattern of officers in the police service, though female officers attracted fewer drug‐related allegations. Adverse findings, while rare, are most likely to be recorded against lower ranking police officers who have served less time in the police service.

The paper shows that demographics alone are not sufficient to identify officers at risk of being on the receiving end of a drug‐related complaint. The age, service and rank analysis conducted in this paper has not revealed any particular groups that are more susceptible to allegations of drugs misconduct. This paper therefore supports the idea that a more thorough early warning system tailored to individual officers may be necessary for an effective strategic complaints system.

With a pool of nearly 80,000 allegations to draw upon, the research employs one of the largest data sets ever examined. The findings are therefore sufficient to provide robust statistical comparisons, and are of interest to police practitioners, law enforcement managers, and researchers.

This paper seeks to empirically describe the role of research and planning units within contemporary, local police organizations in the US.

Data were collected from a national survey of police organizations, municipal police agencies and sheriffs' offices in the US and analyzed using univariate statistics.

The findings demonstrate that the task scope of research and planning units (RPU) within local law enforcement organizations is heterogeneous. RPUs perform a range of tasks and these tasks differ from one agency to another. When separate tasks are aggregated into broader categories, the data reveal that, overall, RPUs focus primarily on administrative tasks.

These findings suggest that RPUs primarily focus on administrative support tasks rather than research and planning projects. Thus, RPUs may be underutilized by law enforcement organizations. These findings suggest that administrators consider how the task scope of RPUs can be refocused to help law enforcement agencies achieve their goals.

This paper empirically updates the understanding of the tasks and functions of contemporary police research and planning units.