Detection of Crime, Resource Deployment, and Predictors of Success: A Multi-Level Analysis of CCTV in Newark, New Jersey, 2007-2011 (ICPSR 34619)

Version Date: Sep 24, 2019 View help for published

Principal Investigator(s): View help for Principal Investigator(s)
Eric Piza, John Jay College of Criminal Justice; Joel Caplan, Rutgers University. School of Criminal Justice; Leslie Kennedy, Rutgers University. School of Criminal Justice

https://doi.org/10.3886/ICPSR34619.v3

Version V3 ()

  • V3 [2019-09-24]
  • V2 [2019-06-27] unpublished
  • V1 [2019-06-27] unpublished

The Detection of Crime, Resource Deployment, and Predictors of Success: A Multi-Level Analysis of Closed-Circuit Television (CCTV) in Newark, NJ collection represents the findings of a multi-level analysis of the Newark, New Jersey Police Department's video surveillance system. This collection contains multiple quantitative data files (Datasets 1-14) as well as spatial data files (Dataset 15 and Dataset 16). The overall project was separated into three components:

  • Component 1 (Dataset 1, Individual CCTV Detections and Calls-For-Service Data and Dataset 2, Weekly CCTV Detections in Newark Data) evaluates CCTV's ability to increase the "certainty of punishment" in target areas;
  • Component 2 (Dataset 3, Overall Crime Incidents Data; Dataset 4, Auto Theft Incidents Data; Dataset 5, Property Crime Incidents Data; Dataset 6, Robbery Incidents Data; Dataset 7, Theft From Auto Incidents Data; Dataset 8, Violent Crime Incidents Data; Dataset 9, Attributes of CCTV Catchment Zones Data; Dataset 10, Attributes of CCTV Camera Viewsheds Data; and Dataset 15, Impact of Micro-Level Features Spatial Data) analyzes the context under which CCTV cameras best deter crime. Micro-level factors were grouped into five categories: environmental features, line-of-sight, camera design and enforcement activity (including both crime and arrests); and
  • Component 3 (Dataset 11, Calls-for-service Occurring Within CCTV Scheme Catchment Zones During the Experimental Period Data; Dataset 12, Calls-for-service Occurring Within CCTV Schemes During the Experimental Period Data; Dataset 13, Targeted Surveillances Conducted by the Experimental Operators Data; Dataset 14, Weekly Surveillance Activity Data; and Dataset 16, Randomized Controlled Trial Spatial Data) was a randomized, controlled trial measuring the effects of coupling proactive CCTV monitoring with directed patrol units.

Over 40 separate four-hour tours of duty, an additional camera operator was funded to monitor specific CCTV cameras in Newark. Two patrol units were dedicated solely to the operators and were tasked with exclusively responding to incidents of concern detected on the experimental cameras. Variables included throughout the datasets include police report and incident dates, crime type, disposition code, number of each type of incident that occurred in a viewshed precinct, number of CCTV detections that resulted in any police enforcement, and number of schools, retail stores, bars and public transit within the catchment zone.

Piza, Eric, Caplan, Joel, and Kennedy, Leslie. Detection of Crime, Resource Deployment, and Predictors of Success: A Multi-Level Analysis of CCTV in Newark, New Jersey, 2007-2011. Inter-university Consortium for Political and Social Research [distributor], 2019-09-24. https://doi.org/10.3886/ICPSR34619.v3

Export Citation:

  • RIS (generic format for RefWorks, EndNote, etc.)
  • EndNote
United States Department of Justice. Office of Justice Programs. National Institute of Justice (2010-IJ-CX-0026)

Closed-Circuit Television (CCTV) Camera Viewsheds

Access to these data is restricted. Users interested in obtaining these data must complete and sign a Restricted Data Use Agreement, describe the research project and data protection plan, and obtain IRB approval or notice of exemption for their research.

Inter-university Consortium for Political and Social Research
Hide
2007-11 -- 2011-04
2011-01-01 -- 2013-03-31
  1. The final report for this study (Piza, Caplan, and Kennedy, 2013) discusses a process evaluation and outcomes evaluation for Component 3. For dissemination purposes, the narrative documents from the process evaluation were merged into a quantitative dataset, Targeted Surveillances Data (Dataset 14).

  2. This collection contains 169 Geographic Information System (GIS) spatial data files. Component 2: Impact of Micro-Level Features Spatial Data (Dataset 15) contains 72 shapefiles (each consisting of approximately 7 files). Component 3: Randomized Controlled Trial Spatial Data (Dataset 16) contains 93 shapefiles (each consisting of approximately 7 files).

Hide

This project is a multi-level analysis of the Newark Police Department's Closed-Circuit Television (CCTV) system. The project aims to contextualize the places where CCTV cameras deter crime better than other places, evaluate the process by which the Newark Police Department currently responds to criminal incidents detected by the CCTV system, and measure the effects that dedicated patrol units assigned to and dispatched by CCTV operators has on deterrence and clearance rates for crimes that occur in areas of Newark that are visible to CCTV cameras.

This study includes three distinct analyses, referred to as "Component 1," "Component 2," and "Component 3."

Component 1 (Parts 1 and 2) focused on the period from November 2007 through the end of 2010. Data were compiled from the Newark Police Department's weekly Video Surveillance Unit (VSU) activity reports, which list all incidents that occurred in Closed-Circuit Television (CCTV)-areas of the city (both CCTV detections and 9-1-1 calls-for-service). Each incident appearing on the VSU weekly activity reports was individually referenced in the CAD system to collect additional data.

Component 2 (Parts 3-10, and 15) analyzed the context under which CCTV cameras best deter crime. Units of analysis were the viewsheds of the individual CCTV cameras, which accounted for two particular obstructions: 1) immovable objects (e.g. traffic signs and telephone poles), and 2) foliage (e.g. leaves from trees and bushes). It examined the influence of several micro-level factors (environmental features, line-of-sight, camera design and enforcement activity) on changing crime levels in CCTV viewsheds.

Component 3 (Parts 11-14, and 16) was a randomized experiment designed for the purpose of overcoming specific "surveillance barriers" common to CCTV, namely a high camera-to-operator ratio and the lack of an immediate response to CCTV-detected incidents due to the differential-response nature of police dispatch. An additional camera operator was funded to monitor a subset of Newark's CCTV cameras over 40 separate four-hour tours of duty in the Summer of 2011. Two patrol units were assigned to the operators, and were tasked with exclusively responding to incidents of concern detected on the experimental cameras. Units of analysis began as camera viewsheds, which were constructed during the early stages of component 2. Viewsheds that overlapped or lay directly adjacent (less than one-city block) to each other were grouped together into singular "schemes" for the purpose of the experiment. A randomized block design was utilized in the assignment of schemes to the experimental and control groups. In addition to conducting an outcome evaluation, the study conducted a process evaluation in an attempt to contextualize program effects and any policy implications generated from the findings. The process analysis primarily focused on the activity of the experimental operators and patrol officers. Data were collected for the process analysis through qualitative methods. Researchers observed the activity of the CCTV operators from within the control room, as well as the response of the responding officers via the CCTV camera feeds, and used a pre-constructed form to record specific aspects of their observations.

Component 1 and Component 2 utilized Closed-Circuit Television (CCTV) detections and CCTV viewsheds as units of analysis, respectively. These datasets include the entire universe of detections and viewsheds in the study area; no sampling was conducted. Component 3 includes all CCTV viewsheds within a standard deviation ellipse of the study area. The ellipse represents one standard deviation around the mean center of the group of points, outlining an area both concise in size and inclusive of a large proportion of the overall places of interest. Overall, 38 schemes fell within the boundary of the standard deviation ellipse and were chosen for inclusion in the experiment.

For Component 1, Part 1 (Individual CCTV Detections and Calls-for-service Data) includes data on a total of 8,115 incidents from the weekly Video Surveillance Unit (VSU) reports for the period November 2007 through 2010. This includes 1,385 CCTV detections and 6,730 calls-for-service (CFS). The VSU reports include all incidents that occurred in CCTV-areas of the city (both CCTV detections and 9-1-1 calls-for-service). A total of 13,368 incidents were included in the weekly VSU reports for that period, but 5,253 cases were dropped based on criteria determined by the researchers. Part 2 (Weekly CCTV Detections in Newark Data) includes weekly summary data for the 165 weeks (spanning from Sunday through Saturday) from November 2007 through December 2010.

Component 2 looked at the 117 viewsheds (which excludes overlapping viewsheds) corresponding to 141 of the system's 146 cameras. These 117 viewshesds exclude overlapping viewsheds, 5 cameras that were out of service for about a year, and 13 viewsheds where the police department did not have precise information regarding the installation date. Parts 3-8 contained data on subsamples of those viewsheds, while Parts 9 and 10 contained data for all 117 viewsheds.

Component 3 included the CCTV viewshed schemes included in the experimental component of the study. Part 11 (Calls-for-service Occurring Within CCTV Scheme Catchment Zones During the Experimental Period Data) has data for the 36 Scheme Catchment Zones, and Part 12 (Calls-for-service Occurring Within CCTV Schemes During the Experimental Period Data) has data for the 38 schemes. Part 13 (Targeted Surveillances Conducted by the Experimental Operators Data) has data on 237 targeted surveillances conducted during the experiment. Part 14 ( Weekly Surveillance Activity Data) has weekly aggregated data for 99 weeks, the 55 week pre-experiment period, the 11 week experiment period, and the 33 week post experiment period.

Longitudinal
Crime incidents, Weeks, Closed-Circuit Television (CCTV) Camera Viewsheds

Geographic Information System (GIS) crime and officer activity data were provided by the Newark Police Department's CompStat Unit. CompStat similarly provided GIS files denoting the locations of several crime generators and attractors throughout the city. Additional data was obtained from InfoGroup.

Closed-Circuit Television (CCTV) cameras were utilized by researchers in order to create viewsheds of each camera site. Researchers also utilized the cameras to monitor the camera feeds during a targeted intervention in order to record CCTV operator monitoring activity and the subsequent response by patrol officers.

Researchers also compiled data from the Video Surveillance Unit's (VSU) weekly activity reports to allow for an analysis of the process by which the Newark Police Department responds to incidents detected by Closed-Circuit Television (CCTV).

The data for component 1 consist of 2 quantitative datasets. Part 1 (Individual Closed-Circuit Television (CCTV) Detections and Calls-For-Service Data) has 33 variables describing incidents included in the Newark Police Department's weekly Video Surveillance Units (VSU) activity reports, including date and time of incident, type of incident, case disposition, and camera site(s) that viewed the incident. Part 2 (Weekly CCTV Detections in Newark Data) has 22 variables describing the CCTV detections and calls-for-service incidents that occurred each week. The variables also include weather during each week, and when the week occurred relative to certain events.

The data for Component 2 consist of 8 quantitative datasets, and 72 Geographic Information System (GIS) shapefiles. Parts 3-8 (crime incidents data) each have 45 variables describing the physical features of the camera viewsheds, the number of crime incidents occurring in the viewsheds, and location quotients for the viewsheds. Part 9 (Attributes of CCTV Catchment Zones Data) and Part 10 (Attributes of CCTV Camera Viewsheds Data) have 80 and 82 variables respectively describing the physical features of the catchment zones or viewsheds, the number of crime incidents occurring in the catchment zones or viewsheds, and location quotients for the catchment zones or viewsheds.

The data for Component 3 consist of 4 quantitative datasets, and 93 GIS shapefiles. Part 11 (Calls-for-service Occurring Within CCTV Scheme Catchment Zones During the Experimental Period Data) has 94 variables, including calls-for-service counts for a variety of offense types, for three time periods: the actual experiment tours of duty, the entire days (e.g. 24 hour period) the experimental tours took place, and the entire 11-week experimental period, as well as for corresponding pre-experiment periods. Part 12 (Calls-for-service Occurring Within CCTV Schemes During the Experimental Period Data) has 168 variables, including calls-for-service counts (disagreggated by crime type) within CCTV schemes over a variety of time periods, as well as changes in calls-for-service between the pre-intervention and post-intervention time periods. Location quotients were also calculated. The variables also include counts of targeted surveillances and experiment-generated enforcement activities. Part 13 (Targeted Surveillances Conducted by the Experimental Operators Data) has 53 variables from the observation of activity of CCTV operators, including coded descriptions of the observed incidents and the outcome. Part 13 also includes the descriptive narratives that were provided for each targeted surveillance. Part 14 has 17 variables giving the number of CCTV detections, by crime type, that occurred each week.

N/A

N/A

Hide

2019-06-27

2019-09-24 Geographic Information System (GIS) spatial data files and associated documentation were added to the data collection.

2019-06-27 ICPSR data undergo a confidentiality review and are altered when necessary to limit the risk of disclosure. ICPSR also routinely creates ready-to-go data files along with setups in the major statistical software formats as well as standard codebooks to accompany the data. In addition to these procedures, ICPSR performed the following processing steps for this data collection:

  • Created variable labels and/or value labels.
  • Checked for undocumented or out-of-range codes.
Hide

This collection does not contain any weight variables.

Hide

Notes

  • The public-use data files in this collection are available for access by the general public. Access does not require affiliation with an ICPSR member institution.

  • One or more files in this data collection have special restrictions. Restricted data files are not available for direct download from the website; click on the Restricted Data button to learn more.