Program Profile: Public Surveillance Cameras (Baltimore, Maryland)

This is a public surveillance system that includes a network of cameras and components for monitoring, recording, and transmitting video images to monitor crime in different areas of Baltimore, Md. The program is rated Promising. There was a statistically significant decrease in total monthly crime in the downtown, Greenmount, and tri-district areas of Baltimore between pretest and posttest. However, there was no statistically significant difference in the North Avenue area.

A Promising rating implies that implementing the program may result in the intended outcome(s).

A Promising rating implies that implementing the program may result in the intended outcome(s).

Program Goals/Program Components

Public surveillance systems include a network of cameras and components for monitoring, recording, and transmitting video images. These systems are typically equipped with night vision, color recording, and have the ability to pan, tilt, and zoom. Most cameras are pre-programmed to scan an area for a predetermined time and pattern, yet can also be operated remotely by security personnel or an automated computer system to focus on a particular area of interest.

In 2000, officials in the city of Baltimore, Md., were committed to using technology and data to positively impact the crime rate in their city. City officials envisioned a similar system in Baltimore that was already widespread in other major metropolitans (such as London, England and Chicago, Ill.) and were confident that such a system could help reduce the historically high levels of crime, specifically violent crime.

Prior to the implementation of public surveillance cameras, Baltimore experienced 11,183 violent crimes and 48,653 total crimes, resulting in the seventh-highest violent crime rate and 28th overall highest crime rate in the United States (La Vigne et al. 2011). Public surveillance cameras offered the opportunity to use technology that could aid in the prevention, detection, and investigation of these crimes.

Program Theory

Public surveillance systems are grounded in the theory that people will be less inclined to commit criminal activity in the presence of cameras that are recording activity, as this would place them at greater risk for apprehension. Furthermore, proponents of public surveillance systems also believe that such systems have the ability to increase perceptions of safety among citizens, as well as encourage citizens to use public spaces that are now guarded by surveillance (La Vigne et al. 2011). By increasing the number of citizens using public spaces, more individuals can potentially serve as witnesses to crimes, presenting the possibility of greater crime reduction. Additionally, video footage of a crime may also help in investigations and prosecutions.

Target Sites

With the ultimate goal of crime reduction, cameras were placed throughout the city of Baltimore in areas with the highest crime rates. Installation areas included the 50-block downtown area and other selected neighborhoods. Crime rates were determined by incident and arrest reports, as well as input from district commanders. Overall these areas had a disproportionate number of shootings, murders and assaults, and were known for drug use. The selected areas were assessed to determine the feasibility of camera installation within the existing environment, which resulted in the installation of light poles to support the cameras, as well as the trimming of trees to ensure the camera’s range was not obstructed.

Furthermore, necessary precautions were taken to ensure that the cameras did not infringe on the civil liberties of community members. An individualized range of motion for each camera was set to capture the desired images. Cameras were designed to follow their individualized pattern for 24 hours, moving from right to left and zooming in on desired locations. A variety of Baltimore Police officers, including retired officers, also have the ability to monitor the cameras, selecting the specific areas they want to focus on within the camera’s range. Given their experience, these officers were aware of crime hot spots, persistent threats, and specific body language to suggest the occurrence of a crime.

Civilians can also function as monitors with the proper training. All monitors must pass a criminal background check and a drug test. After passing these tests all monitors are given a manual which outlines the specifics of monitoring, such as activities that can and cannot be monitored.

Study

As part of a cross-site evaluation of public surveillance cameras, Baltimore, Md., along with two other cities, was evaluated by La Vigne and colleagues (2011) to investigate the impact of public surveillance cameras on crime control within their respective implementation area. Both qualitative and quantitative data were used in the evaluation. Qualitative data was used in the process evaluation of the cameras, which included interviews with various city officials, as well as a review of city policies, budgets, and various other documents. Quantitative data was used in the impact analysis of the cameras, which was gathered from reported crimes, demographics, land use, camera installation, locations and types, and costs data. Finally, the evaluation also included a cost-benefit analysis, exploring the relation of the costs associated with the implementation of public surveillance technology to any reductions in crime and increased efficiencies in the criminal justice system that can be attributed to the implementation of the cameras.

The quantitative portion of the evaluation, or impact analysis, gathered crime data from four specific areas within the city of Baltimore—the downtown, Greenmount, North Avenue, and Tri-District areas—between January 2003 and April 2008. Each selected area had approximately 30 or more cameras. Seven crime categories were investigated: all crime, violent crime, inside larceny, outside larceny, motor vehicle theft, burglary, and robbery. Inside larceny was defined as crimes occurring inside businesses, whereas outside larceny was defined as thefts occurring in open spaces. To analyze the impact of public surveillance cameras on crime control, crime rates in the camera’s target area—200 feet—were compared to the crime rates in matched comparison areas for three of the four selected areas, using two-tailed t-tests in the analysis. A matched comparison area was not possible for the fourth selected area, the downtown area. Therefore for the downtown area, a time series analysis was conducted to identify statistically significant changes in average monthly crime counts within the camera’s target area of 200 feet during the investigation period. By using a time series analysis, changes associated with the implementation of the public surveillance cameras can be investigated.

For the Greenmount, North Avenue, and Tri-District areas, comparison groups were used. The comparison areas were selected based on similarities in land use, crime rates, and socioeconomic measures. The Greenmount matched comparison area was located slightly north of the camera area, which (similar to the treatment area) is mostly residential, included a few retail businesses, a large high school, and a recreational park. The North Avenue matched comparison area was located south of the target area, which (similar to the treatment area) is predominately residential, includes a handful of churches and schools, and is bordered by thoroughfares. The Tri-District area matched comparison group was located on the far eastern side of the city, beyond two other cameras that were not used in this analysis. This matched comparison area was similar to the treatment area in that it is predominately residential, with a similar style of row houses, and included restaurants and retail stores scattered throughout the area.

Although the study design was adequate to analyze the impact of public surveillance cameras on crime, the study authors did note the difficulty of finding true matched comparison areas. For example, due to the unique makeup of the downtown area, as well as the high crime rates that occur there, a matched comparison area was not feasible. Additionally, the authors noted the limited options of comparison areas for the Tri-District area. As a result, the Tri-District’s matched comparison area included an area that was just beyond two other camera areas that were not included in this study. The Tri-District area and its corresponding comparison area were matched on the same similarities as the Greenmount and North Avenue areas and their corresponding comparison groups; however, the Tri-District area and its comparison area were not as similar to one another as the other areas and their respective comparison groups. Subgroup analyses were not conducted.

La Vigne and colleagues (2011) also conducted a cost-benefit analysis using a cost-collection instrument. Overall, the evaluation found that the benefits that result from reduced crime are greater than the costs associated with the implementation of the public surveillance cameras. Although a delay in benefits was found, due to the initial start-up costs, the benefits exceeded the costs of camera implementation around the 29th month after the public surveillance cameras had been installed.

The use of public surveillance cameras requires careful implementation. The placement of cameras is crucial to maximize their effectiveness; therefore, geographic information systems (GIS)—which examine crime patterns, input from police regarding crime hot spot areas, and stakeholders involved in the implementation process—are typically used to guide the camera placement. For example, in the study by La Vigne and colleagues (2011), the qualitative portion included research questions on various public surveillance topics, such as: reasons for investing in the cameras, the decision process for implementing cameras, involvement of the community in the decision-making process, advantages and disadvantages of their implementation, types of cameras used, monitoring of cameras, and use of cameras in arrests and investigations. Responses were gathered through three sources:

1. in-person or telephone interviews with stakeholders involved in the implementation process, as well as those affected by its installation;
2. collection of documents related to the acquisition and installation process—such as contracts, training materials, and policies; and
3. on-site observation of the camera monitoring process to document how the cameras are monitored, how alerts are issued, and how data is gathered for investigative purposes.

Interviewees were identified initially by the police department, and then those interviewed were asked to provide the names of others involved and affected by the installment of public surveillance cameras.

Furthermore, a decision must be made to use either a passive system (which relies on the retrieval of previously recorded images that are viewed after the fact) or an active system (which is monitored in real time usually by police officers, retired police officers, or volunteers). With both systems, those viewing the footage must be trained on the type of activity to watch for, which is why those with law enforcement experience are typically thought to be better suited in these roles. However, regardless of previous experience, all personnel viewing camera footage are given a training manual to guide them through the process.

Additional information about the implementation process at the Baltimore, Md., site can be found in the 2011 study by La Vigne and colleagues.

These sources were used in the development of the program profile:

Following are CrimeSolutions-rated programs that are related to this practice:

Closed Circuit Television (CCTV) Surveillance

Public surveillance systems include a network of cameras and components for monitoring, recording, and transmitting video images. Public surveillance cameras are designed to reduce both property and personal crime. This practice is rated Promising for reducing overall crime, property crime, and vehicle crime, and rated Ineffective for impacting violent crime.

Evidence Ratings for Outcomes

	Crime & Delinquency - Multiple crime/offense types
	Crime & Delinquency - Property offenses
	Crime & Delinquency - Vehicle crime
	Crime & Delinquency - Violent offenses

Why might a practice's outcome ratings differ from the ratings of specific programs encompassed by that practice?