By Howard B. Hall, Captain, Baltimore County, Maryland, Police Department; and Emily N. Varga, Statistical Analyst, Baltimore County, Maryland, Police Department
|Photographs courtesy of the Baltimore County, Maryland, Police Department|
aw enforcement agencies are facing unprecedented reductions in resources that have affected all aspects of policing. At the same time, demands for service are rising while the critical mission of preserving public safety remains constant. Now more than ever, police executives are forced to prioritize resources and ensure the maximum possible return on investment.
The return on investment for traffic safety is simple. A reduction in the deaths, injuries, and property damage that result from traffic crashes has a direct impact on the safety of communities. Recent evidence suggests that the investment is returning dividends. Fatalities nationwide are at the lowest level in decades. In fact, the National Highway Traffic Safety Administration (NHTSA) projects that the number of fatalities in 2010 will be 3 percent lower than in 2009, despite an estimated increase of 21 billion miles traveled by American motorists.1 These outcomes are a direct result of the efforts of frontline law enforcement officers to address traffic safety problems.
How, when faced with shrinking budgets and resources, should law enforcement executives ensure that traffic safety activities remain a priority? The answer to this question lies in the use of data-driven practices to deploy resources and conduct enforcement and other traffic-related activity. The availability of data and analytical processes is improving daily for both large and small agencies. Many of the same techniques that have been used successfully to support the CompStat model and target criminal activity can be applied to traffic safety problems. The four main principles of CompStat include accurate and timely intelligence (analysis products); effective tactics (a good plan); rapid deployment; and relentless follow-up and assessment.2 These techniques provide the operational supervisor the opportunity to deploy limited resources to the areas where they are needed the most and have the potential to do the most good, regardless of whether the problem relates to crime or traffic safety.
In recent years, the theory of place-based policing has taken hold. Place-based policing focuses “on places where crimes are concentrated, and it begins with the assumption that there is something about a place that leads to crimes.”3 Evidence suggests that place-based tactics provide the opportunity to reduce the number of incidents at identified locations, offer a stable geographic target for police intervention, and present a more efficient approach to deploying police resources.4 Could this theory also be applied to traffic safety problems, especially crashes? Data from the Baltimore County, Maryland, Police Department (BCOPD) and several others suggest that it can. Traffic crashes not only have been found to cluster in specific areas but have been found to overlap with crime hot spots, providing the opportunity to deploy resources in a manner that improves traffic safety and deters crime.5The BCOPD and an increasing number of agencies throughout the country are having success with a model of deployment known as Data-Driven Approaches to Crime and Traffic Safety (DDACTS). DDACTS is defined as “an operational model that uses the integration of location-based crime and traffic data to establish effective and efficient methods for deploying law enforcement and other resources.”6
The model is based on the research related to place-based policing and has been shown to reduce the incidence of crime and crashes in the areas where it has been deployed. It is based on seven guiding principles:
- Partner/stakeholder participation
- Data collection
- Data analysis
- Strategic operations
- Information sharing
- Monitoring, evaluating, adjusting
The remainder of this article will focus on several of these principles that are particularly important to data-driven traffic safety efforts. In particular, the needs for data collection, analysis, strategic operations, and evaluation will be discussed.
Types of Traffic Safety Data
Every agency has data that can be used for analysis. The most common type is crash data. Like crime data, crash data are usually available from an agency’s computeraided dispatch (CAD) system or records management system (RMS). Most states use a standard form for crash reports, which is submitted to a central repository in hard copy or electronic format. The crash report provides the most robust data since it includes crash types, causation, roadway factors, driver demographics, and other specific information that can be used to complete an analysis. Some states have excellent systems for this purpose. For instance, Louisiana uses a system known as LACRASH that provides timely data to participating agencies. Data from crash reports are collected and analyzed on a daily basis and are accessible via the Internet.7 There are still some states that cannot produce timely crash data; therefore, it is incumbent on agencies in those areas to make the best use of available resources. Fortunately, most CAD systems provide timely data, including the day, the time, the location, and the severity of the crash. While this is not ideal, it is enough to produce a spatial analysis of incidents. If an analysis of CAD data reveals a pattern of crashes, it is relatively easy for an analyst or an officer to obtain hard copies of the incidents in that specific area for a manual review of other factors. The ultimate goal is to obtain as much specific information as possible to be used in the design of countermeasures for crash reduction.
Identifying the types of data needed for data-driven traffic enforcement goes beyond examining crash reports. While crash data may be the easiest to obtain either from the reports themselves or from an RMS, there are other resources that can be valuable as well.
- A citation database is a valuable resource. Point locations for enforcement stops can be extracted from this data set along with violation types. These data sets allow field personnel to evaluate prior enforcement in an area. This will assist in designing new operations that build on historical activity. It will also help to ensure that enforcement activity is being conducted in the right places. In the absence of a citation database, the CAD system may offer an alternative. Traffic stops can be entered just like any other call for service, so the requisite data related to location, time, and even type of violation (if captured by the system) can be extracted.
- Citizen complaints are another resource. Citizens often report traffic problems to police. These complaints, combined with other databases (such as an RMS or citations), can facilitate a more robust analysis by verifying information from other sources or by helping to identify areas where additional analysis is needed.
- Survey information should be used to verify the types of violations and their severity. The most common type of survey is for speed. A speed survey should be conducted to verify that speeding is a problem and help determine the times that it occurs. Several types of electronic speed survey devices are available that will collect data for an extended period of time. Agencies that do not have this equipment should contact engineering authorities, who often conduct this type of survey for road design purposes. Surveys related to occupant protection usage are frequently conducted to assess the rate of compliance. While there is not an electronic system for conducting occupant protection usage surveys, agencies may find that volunteers or auxiliary personnel are available for this purpose. Obviously, there are any number of additional surveys that could be conducted based on an identified or suspected problem. It is up to the analyst or the officer conducting the study to identify these data needs.
A number of other government agencies may have access to traffic safety data. Each state has a highway safety office that has a role in traffic records. These offices are likely to support local efforts to obtain this type of data. For instance, the Maryland Highway Safety Office works with the National Study Center at the University of Maryland to provide data to local agencies. Engineering authorities at the local and state levels are also a good source of information. These agencies are concerned with traffic safety and roadway design and sometimes maintain their own databases of information. The idea is to examine what resources are available and determine how they can be used to complete a thorough analysis of traffic safety problems.
Once the types of traffic data are identified, the next step is to evaluate the quality of the data that are available. The quality of data depends of several factors, including accessibility, accuracy, completeness, timeliness, and cleanliness.8
Accessibility. As previously mentioned, analysts may not have access to all types of data that may be ideal for data-driven analysis. There are many resources available to supplement agency data, and these should be fully explored.
Accuracy. There are numerous ways that inaccurate data can be produced. Incident locations may be misspelled, erroneously reported, or even omitted. How will the exact location be pinpointed if the incident address was omitted? The severity of incidents may be incorrect. If a victim dies after being initially reported as injured, will the incident be updated? These are concerns that need to be addressed when considering what data to use.
Completeness. This is related to accuracy. Reports from the field should not be accepted without all of the required information. After a report is generated, it ultimately finds its way to the desk of a supervisor for review. The better the quality control is for reports, the better the output and the available data for analysis.
Timeliness. With regard to timeliness, the more current the data, the more current the analysis. As noted above, some states have very timely crash data systems while others lag behind, sometimes for a year or more. Delays of this length render the data nearly useless. Agencies should do everything possible to overcome this obstacle. As mentioned previously, utilizing CAD data may be an option. Some agencies have elected to complete their own data entry to eliminate this problem.
Cleanliness. If data are used to drive traffic enforcement, or any type of analysis for that matter, the identified target areas will be only as accurate as the data used to identify them. Officers, dispatchers, report reviewers, and command staffs alike are responsible for the successful completion of this task. Simply put, inaccurate data provide inaccurate results and will not lead to the effective use of enforcement resources.
The goal of data analysis with respect to traffic enforcement is to gain a deeper understanding of the traffic concerns in a particular area. This will provide valuable insight into where traffic enforcement and other countermeasures are most likely to be effective. There are countless ways that traffic data can be analyzed, and there is no universal technique that will work for all data sets. The type of analysis depends on the data set. It is important to use the appropriate data set in the analysis. In Baltimore County, there are more than 32,000 crashes each year. An analysis of all traffic crashes would yield very different results than an examination of personal injury crashes only. If traffic fatalities are a concern, analyzing those crashes alone would give insight into the traffic conditions and causations that are attributed to fatalities. For these reasons, analysts must carefully consider the data set to be used.Hot Spots. A common type of analysis is creating hot spots. Hot spots are often created to illustrate geographic areas with a high concentration of crashes. With a large data set, this technique can be very useful. There are a number of software packages that can perform this type of analysis, including CrimeStat, which is a product of the National Institute of Justice and can be interfaced with a geographic information system to provide mapping products.9 There are also several Internet services that can be used for this purpose. With a smaller data set, hot spot analysis may not be the best choice. Using pin mapping techniques and plotting each crash or crash type on a map may work better. It is possible to create maps of this type with computer mapping software or even from Internet mapping sites.
Causation. Another type of analysis considers crash causation. Being able to identify the cause of crashes (that is, failing to keep right of center, left-angle turning, or a blind curve) allows for better directed traffic enforcement. For example, if crashes at an intersection are caused by left-turning vehicles that fail to yield right of way, the use of speed enforcement may do little to discourage those incidents. Countermeasures should target the specific cause of crashes to be the most effective.
Temporal analysis. Knowing where the hot spot or traffic problem is located is only the first part of the analysis. A temporal analysis is also important. This includes the time (either exact or often a range of time), the day, the month, and so on, that the incidents most frequently occur. Having an idea of when the crashes or traffic problems are occurring allows for a more exact deployment of resources. If analysis reveals that crashes are most highly concentrated during the evening rush hour at a particular location, deploying resources during the overnight hours might not be effective. It is all about being in the right place at the right time.
The availability of actionable analysis products provides the operational executive or supervisor the ability to pinpoint the use of available resources to the locations with the greatest need for intervention. More specifically, it allows for the day, the time, the cause, and a variety of other factors to be targeted for countermeasures.
High-visibility enforcement. One of the primary countermeasures used to target traffic crash and violation problems is highvisibility enforcement (HVE). HVE has been shown to change driving behavior and deter violations that contribute to crash activity.10 Given that enforcement resources (that is, personnel) are shrinking, it is critical that HVE be deployed as effectively as possible. A good analysis, as described above, will provide the information necessary for this purpose. All available information should be used to identify not only the best places for HVE, but also the specific times and days that it should target.
While the use of HVE to deter traffic violations has the potential to reduce crashes in a general sense, the type of enforcement used is important. A substantial amount of traffic enforcement is focused on speeding. This has the important and positive effect of reducing vehicle speeds, which contributes to fewer crashes and reduced severity of crashes. Speeding, however, is not the only violation that contributes to crash problems. Impaired driving remains an issue nationwide. Other issues often identified in crash analysis include aggressive driving, right-of-way violations, red light violations, and distracted driving. Of course, the proper use of occupant protection—seat belts, safety seats, booster seats, and so on—has tremendous potential to reduce death and injury. These issues point to the need for enforcement that targets the specific violations that contribute to traffic safety problems.
Enforcement tolerance. Another important area of consideration for operational supervisors is the issue of tolerance. This includes two parts. First, at what point are officers making a stop? This includes the number or type of violations against which an officer will take action or the amount of miles per hour over the speed limit that are allowed before a car is pulled over. It may also relate to the type of violator. Commercial vehicles present unique hazards to traffic safety, but some officers are reluctant to stop a large vehicle or may be intimidated by the complex regulations that apply. Similarly, bicyclists and pedestrians may also contribute to safety problems but are not often stopped by officers. The importance of this issue cannot be understated. Law enforcement knows that making the stop is critical to making a difference. Without the stop, officers do not gain the advantages of visibility, deterrence, enforcement action against the driver, and the intelligence information or even evidence of criminal behavior that may result. By lowering this initial tolerance, total stops can be increased and their positive effects can be maximized.
The second issue related to tolerance is the point at which a citation is issued. Traffic citations and the penalties that result do affect driving behavior. Drivers whose behavior jeopardizes the safety of others need to be cited. Over time, it seems that police tolerance for violations has increased. A review of Maryland’s citation data from 2006 found that, in some jurisdictions, more than half of speeding citations were issued to drivers exceeding the limit by 20 or more miles per hour.11 This is well above the speed that presents a safety hazard. On the other hand, the use of a zero tolerance approach, which would entail issuing citations for virtually all violations, is not likely to be acceptable to either officers or communities. The use of strict enforcement should be considered when violations are found to be contributing to traffic crashes. This does not imply that officer discretion should be eliminated. Supervisors and commanders, however, have the obligation to limit discretion in order to further the achievement of operational objectives. For instance, supervisors should consider setting guidelines for the issuance of citations versus warnings for hazardous traffic violations that are contributing to traffic crashes.
Another example relates to the enforcement of seat belt laws. Seat belts are universally required and have been for many years. It is not likely that the behavior of a violator will be changed by the educational effects of a warning. Again, the effects of ensuring an appropriate tolerance in this regard have the potential to provide positive results in the reductions of violations. In addition, ensuring appropriate tolerance and reaping the potential rewards does not require additional funds or resources—just good communication and a focus on documented problem areas.
Automated enforcement. The use of automated enforcement is expanding constantly and has the potential to act as a force multiplier. The most common types of automated enforcement include red lights and speeding, although some jurisdictions are also enforcing seat belt laws and registration and insurance violations through this method. While there are a number of legal requirements to consider, and legislative action is normally required to authorize this enforcement, the use of traffic safety data and thorough analysis is important to ensure adequate justification so that this method can receive public support.
Training. Another way to maximize the use of limited resources toward traffic safety is training. An evaluation of the training levels of patrol and traffic enforcement personnel may reveal needs that, if met, could increase the level of traffic enforcement in a community. Specific areas to examine would be the number of officers certified to use radar and other speed-measuring devices and to administer the standardized field sobriety test batteries. Additionally, advanced training as drug recognition experts or in problem solving could be beneficial. In many areas, programs such as these are available for nothing more than the personnel costs associated with attendance.
Educational programs. Police officers have the absolute responsibility for conducting traffic enforcement. In fact, policing is the only profession that has the authority to do so. Enforcement, while tremendously important, is not the only effective traffic safety countermeasure. There are many excellent educational and informational programs that have been used effectively to supplement enforcement efforts and to encourage voluntary compliance with the law. Resources for these are limited, so it is important that they, too, are directed by data and analysis to the areas and populations where they can have the most effect. State highway safety offices can help in these efforts, as can any number of potential partners in the community. A good source of reference is the Highway Safety Desk Book, published by the IACP’s Highway Safety Committee.12
Engineering. Another important countermeasure is traffic engineering. In some cases, physical modifications to roadways can be made to discourage traffic violations and reduce the potential for future crashes. In difficult budget situations, these modifications can be hard to fund. The use of comprehensive data and analysis will help to identify priority areas for changes and may provide justification for the expense. It is also important to note that, in the long term, physical changes are generally less expensive than the cost of response to traffic crashes or the use of human resources for enforcement.
Resources. When personnel are scarce, it can seem difficult, if not impossible, to engage enforcement resources in problem areas. Several agencies that have implemented the DDACTS model have found it possible to make efficient use of existing patrol resources for this purpose. Traffic enforcement should be an expectation of all uniformed personnel, particularly those in patrol functions. The question is how much time, on average, during a patrol shift is it reasonable to expect an officer to spend on HVE or other self-initiated activities? The answer obviously depends on the agency and community; however, even modest expectations can result in substantial investment over a period of time.
Traffic enforcement efforts may be supplemented through the use of grant-funded overtime. Each state receives federal funds through NHTSA for traffic safety purposes that are administered by state highway safety offices. Executives and supervisors should make it a point to be involved with representatives of both of these agencies in order to learn about the availability of these funds and their allowable uses. In many cases, they are targeted toward specific issues, such as impaired driving, but the use of analysis to identify traffic safety problems will help to determine if this type of funding will be available and will help to support any proposal that might be submitted.
Operational planning. Once a data-driven approach has been developed, the use of a detailed operational plan will help to ensure not only that planned activities are carried out but that better results are achieved. One method of doing this is through the development of objectives and action plans. Objectives should be specific, measurable, action-oriented, and timely and can be used to target crash reduction, occupant protection usage, speed levels, or other problems. A specific action plan should be developed to document all the steps that will be taken to achieve the objectives and to assign responsibility to those who will be accountable for their completion. This provides a mechanism to monitor progress and make any necessary adjustments.
Once an action plan is implemented, first-line supervision is critical. With support from executives, the first-line supervisors’ commitment, presence, and leadership will improve the response of line officers. In most cases, supervisors who lead by example by being active in target areas and holding subordinates responsible for doing likewise will produce better results. For this reason, their involvement in the planning process and commitment to the final plan is an important consideration.
Sound data collection, a detailed analysis, and specific plans for traffic safety are important components to a successful approach. This is not, however, the end of the process. Law enforcement must examine the outcomes of its efforts to determine if it is being effective. Evaluation can be completed at many different levels of complexity, depending on the resources available. A basic evaluation does not need to be complicated. It should focus on asking several basic questions.
- Were the objectives met?
- Were the steps in the action plans completed?
- Were there any side effects of the plan?
- Are any adjustments necessary for future plans?
These will help law enforcement to identify strategies and tactics that are effective and should be used for the future. They will also help to eliminate unproductive efforts and ensure accountability for the completion of required tasks. A written evaluation report also serves as a reference as future activities are considered.
The evaluation process should be part of the operational plan. This will help to ensure that targeted incidents are tracked and that necessary activity data are available. It will also help to ease the anxiety that is sometimes associated with evaluations since all of its elements and requirements will be known to all participants.
The expanding capabilities of crime analysis have provided an excellent opportunity for the law enforcement profession toincrease its effectiveness in addressing the important problem of traffic safety. The use of analytical techniques can help focus resources on the most important areas and issues. For that reason, even with budgetary challenges, law enforcement can continue to improve effectiveness.
The availability of data and analysis alone does not solve problems, but it is a starting point. The many years of experience of police officials involved in traffic safety activities must be used to translate these into action and results. With the application of proven countermeasures and sound planning and evaluation, this is now more possible than ever. Many resources and partners, some of which have been mentioned in this article, are available to assist. It is up to the executive to take the step forward, set the expectation, and address traffic safety issues in our communities. Citizen safety depends on it. ■
1National Highway Traffic Safety Administration (NHTSA), “Traffic Fatalities in 2010 Drop to Lowest Level in Recorded History: DOT Estimates Three Percent Drop beneath 2009 Record Low,” press release, April 1, 2011, http://www.nhtsa.gov/PR/NHTSA-05-11 (accessed May 6, 2011).
2Jeff Godown, “The CompStat Process: Four Principles for Managing Crime Reduction,” The Police Chief 76 (August 2009): 36–42, http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=1859&issue_id=82009 (accessed May 6, 2011).
3David Weisburd, “Place-Based Policing,” Ideas in American Policing, no. 9 (January 2008): 4, http://www.policefoundation.org/pdf/placebasedpolicing.pdf (accessed May 6, 2011).
5Howard B. Hall, “Targeting Crash and Crime Hot Spots in Baltimore County,” The Police Chief 76 (July 2009): 24–28, http://www.policechiefmagazine.org/magazine/index.cfm?fuseaction=display_arch&article_id=1840&issue_id=72009 (accessed May 6, 2011).
6NHTSA, Data-Driven Approaches to Crime and Traffic Safety (DDACTS) Operational Guidelines, August 2009, DOT HS 811 185, version 1.1, http://www.nhtsa.gov/DOT/NHTSA/Traffic%20Injury%20Control/Articles/Associated%20Files/811185.pdf (accessed May 6, 2011).
7Additional information may be found at “Louisiana Crash Data Reports,” http://www.lacrashdata.lsu.edu (accessed May 6, 2011).
8NHTSA, executive summary to Data-Driven Approaches to Crime and Traffic Safety (DDACTS) Data Suggestions for Implementation, http://www.stko.maryland.gov/LinkClick.aspx?fileticket=BiinpkkKdnk%3d&tabid=131&.pdf (accessed May 6, 2011).
9“CrimeStat III,” National institute of Justice, http://www.nij.gov/nij/maps/crimestat.htm (accessed May 6, 2011).
10NHTSA, Guidelines for Developing a High-Visibility Enforcement Campaign to Reduce Unsafe Driving Behaviors among Drivers of Passenger and Commercial Motor Vehicles: A Selective Traffic Enforcement Program (STEP) Based on the Ticketing Aggressive Cars and Trucks (TACT) Pilot Project, October 2007, DOT HS 810 851, 25, http://www.nhtsa.gov/DOT/NHTSA/Traffic%20Injury%20Control/Articles/Associated%20Files/HS810851.pdf (accessed May 6, 2011).
11Michael Dresser, “Up to Speed: Penalties for Md. Motorists Vary by County,” Baltimore Sun, February 3, 2008, http://articles.baltimoresun.com/2008-02-03/news/0802020179_1_montgomery-county-speeding-tickets-baltimore-county (accessed May 6, 2011).
12IACP Highway Safety Committee, Highway Safety Desk Book (September 1, 2004), http://www.theiacp.org/LinkClick.aspx?fileticket=qJ0Qs%2f6MtRY%3d&tabid=87 (accessed May 17, 2011).
Please cite as:
Howard B. Hall and Emily N. Varga, "Data-Driven Practices Maximize Resources amid Shrinking Budgets," The Police Chief 78 (July 2011): 18–27.