Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
4 Speed Enforcement and Adjudication Thirty days Hath September April June and the Speed offender Burma Shave (Rowsome 1965) Compliance with any regulation such as a speed limit requires that it represent a reasonable constraint on behavior. In the preceding chap- ter, a range of methods for setting reasonable speed limits was dis- cussed, reflecting trade-offs among safety, travel time, and enforceability on different types of roads and roadway environments. In this chapter, two additional requirements for driver compliance are considered: public support and consistent enforcement. 139
MANAGING SPEED 140 Public support (i.e., willingness to obey) is closely linked with and should follow from the first requirement, that is, reasonableness of speed limits. For compliance with speed limits to be at a high level, a majority of the driving public must perceive them to be legitimate and comply with them voluntarily. Otherwise, large numbers of motorists will disregard the limits. In this situation, if attempts are made to enforce them, large numbers of violations will overwhelm the law enforcement community--the police, the prosecutors, and the courts. Moreover, without strong public support, law enforce- ment agents themselves may be reluctant to enforce the speed lim- its. This response is typical of laws that are viewed as nonlegitimate in the eyes of the public (Wilson 1983; Ross 1973; Ross 1982). Of course, a combination of reasonable speed limits, sustained enforce- ment, and education may change driver behavior toward speed com- pliance, but this requires the perception that speeding is a safety problem and a long-term effort to change driver attitudes and behavior. Even if speed limit regulations are generally viewed as legitimate by most motorists, enforcement is essential to ensure conformity of the remaining drivers who will obey the laws only if they believe they are likely to be apprehended and prosecuted for noncompliance. In this chapter, appropriate methods of speed enforcement and adjudi- cation are considered. The chapter begins with a discussion of the role of deterrence, surveillance, and sanctions in ensuring compliance with speed limits. Then, evidence of the effectiveness and the limita- tions of traditional enforcement methods on speed choice and safety outcomes is considered. Next, the potential of automated technolo- gies and public information programs to supplement traditional enforcement methods is reviewed. The role of the courts in adjudica- tion of speeding violations is then considered. Finally, key conclu- sions concerning enforcement of speed limits and related judicial procedures are summarized. LESSONS FROM DETERRENCE THEORY Enforcement works primarily through the principle of deterrence. The fundamental idea is that credible threats of punishment deter
141 Speed Enforcement and Adjudication unwanted behavior.1 More specifically, the proscribed behavior is dis- couraged by the perception that legal punishment is "swift, sure, and severe" (Ross and LaFree 1986, 132). Elements of the Deterrence Process The effectiveness of deterrence depends on several factors. First, the proscribed behavior must be definable, understandable, and detectable, not only by the individuals to be deterred but also by those who are expected to enforce compliance and penalize those who do not comply (TRB 1987, 91). In the case of speeding, police officers must be able to reliably verify vehicle speeds and provide evi- dence that will hold up in court. Second, the effectiveness of deterrence depends on the perceived risk of apprehension (TRB 1987, 91; Shinar and McKnight 1985, 387). For the risk to be credible, drivers must believe that they have a nontrivial chance of being apprehended if they engage in the pro- scribed behavior. Thus, some minimum level of enforcement leading to actual apprehension is necessary (Shinar and McKnight 1985, 407). A well-designed publicity campaign coupled with visible enforcement will expand the perception of risk to a large segment of the target population (Shinar and McKnight 1985, 409). Third, the effectiveness of deterrence depends on the swiftness, certainty, and severity of the punishment (TRB 1987, 91). Empirical research suggests that the perceived certainty of punishment is a more powerful deterrent than the severity of the penalty (Shinar and McKnight 1985, 387).2 One explanation is that, if the risk of pun- 1 The approach is based on the philosophic view that human behavior is rational and that human beings behave to maximize personal pleasure and minimize pain. Thus, unwanted behavior can be deterred by increasing the costs of an undesired behavior so that it outweighs the benefits (Ross and LaFree 1986, 130). 2 In principle, the behavior of potential scofflaws should be affected by both the cer- tainty and severity of punishment, and illegal behavior should decline when either the certainty or the severity of the penalty is increased. However, empirical research has found that, for a series of crimes, the response depends much more on the certainty than on the severity of the penalty (Waldo and Chiricos 1972; Sellin 1967; Tittle 1969;
MANAGING SPEED 142 ishment is so low that the violator regards the threat as negligible, then the severity of the punishment is irrelevant (Ross and LaFree 1986, 144).3 In addition, more severe penalties may actually reduce the deterrence effect by requiring more legal representation, which precludes swift punishment and may lessen the likelihood of receiving any penalty at all (Ross 1990 in Zaal 1994, 11). However, debate con- tinues about just how immediate the penalty must be to provide an effective deterrent (Harper 1991 in Zaal 1994, 12). Role of Police and the Courts The effectiveness of deterrence is also influenced by actions of the police and the courts who carry out the apprehension and punish- ment of lawbreakers. The traditional police role in controlling vehi- cle speeds is to detect, apprehend, and punish the speeding driver. Like other traffic violations, penalties for speeding include fines; some states assess penalty points that can lead to license revocation. If a court hearing is involved, judges have discretion to vary the penalties. Traditional police enforcement works in two ways: through detec- tion and punishment of specific drivers who exceed the speed limit and through deterrence of speeding behavior in general. The first method, often referred to as specific deterrence, is based on the idea that individual drivers who are caught and punished for speeding will be deterred from committing further speeding violations. The second method, known as general deterrence, is based on the assumption that the process of apprehending individual violators can influence the behavior of a larger segment of the driving population. More specifi- cally, police presence alerts drivers that traffic violations, including those related to speeding, are being enforced. In turn, increased Teeven 1972; Chauncey 1975; Silberman 1976; Piliavin et al. 1986; Blumstein et al. 1978; Wilson 1983; Brier and Fienberg 1980; Cameron 1988). 3 Increased severity of penalties can also produce undesired and unanticipated side effects through the discretion of legal agents who are perceived as unfairly apprehend- ing a random sample of a much larger population who were simply lucky enough not to get caught (Ross and LaFree 1986, 144).
143 Speed Enforcement and Adjudication enforcement raises the perceived probability of being caught for speeding and thus helps deter unwanted speeding behavior. Successful deterrence, however, may have the undesired effect of reducing the level of police surveillance and enforcement. Just as drivers increase their compliance as the perceived likelihood of apprehension rises, police may reduce their level of enforcement activity when the unwanted behavior diminishes (Tsebelis 1993, 366367).4 This close coupling of behavioral adaptation by the police and the driving public provides one explanation for the difficulty of sustaining the deterrence effects of traditional enforcement methods (Tsebelis 1993, 366; Bjørnskau and Elvik 1990, 139). It also suggests that random or automated surveillance methods that break this behavioral link may offer more effective ways of maintaining high levels of enforcement (Bjørnskau and Elvik 1990, 140141); this topic is discussed in a subsequent section. The courts can also deter speeding, but the effect on driver behav- ior is less direct. The lag between apprehension and punishment is likely to reduce the deterrence effect of sanctions on speeding drivers. Moreover, court discretion in assessing penalties for speeding, if viewed as arbitrary and unfair, can have the undesired effect of turn- ing public opinion against enforcement and adjudication methods. If judges perceive speed limits to be unreasonable and routinely dismiss speeding citations, the incentive of the police to enforce the limits may be reduced. Lessons for Enforcement To optimize use of their limited resources, the police and the courts attempt to achieve the widest possible deterrence through enforce- ment and sanctions. The effectiveness of deterrence depends on cre- ation of a widely perceived impression that noncompliers will be detected and apprehended if they engage in the proscribed behavior. 4 This interpretation derives from a game theoretic approach to enforcement. The approach suggests that the public and the police are engaged in a game in which the prob- ability of surveillance is not independent of the level of the crime (Tsebelis 1990; Tsebelis 1993; Cox 1994; Hirshleifer and Rasmusen 1992; Weissing and Ostrom 1991).
MANAGING SPEED 144 This impression, in turn, must be accompanied by swift, sure, and publicized punishment to establish a credible threat. The application of these principles to speed enforcement is discussed in the following section. APPLICATION OF DETERRENCE THEORY TO SPEED ENFORCEMENT Speed Enforcement Strategies Speed enforcement using mobile patrol vehicles measuring driving speeds with radar is the most popular means of conducting speed enforcement in the United States, according to a special survey con- ducted for this study (Figure 4-1).5 State police use aircraft as well as laser and VASCAR for speed detection.6 The mobile patrol method involves a police vehicle circulating through traffic and citing speeding drivers. Stationary patrol enforce- ment, where a marked or unmarked police car parked along the side of the roadway uses radar or LIDAR to measure speeds, is another common technique (Stuster 1995, A4A8). Apprehension of speed- ing drivers occurs downstream of the monitoring vehicle, sometimes with another patrol officer. The merits of mobile and stationary patrols have been a topic of study. The former is effective in detecting specific violators and slow- ing traffic in the immediate vicinity of the patrol car (Stuster 1995, A6). The latter is effective in deterring speeding at a particular loca- tion (Stuster 1995, A7). The advantages and disadvantages of visible and concealed enforcement have also been studied. One purpose of concealed enforcement is to increase the uncertainty of where and 5 The survey of state law enforcement agencies was conducted by the Illinois State Police. Responses were received from 34 (68 percent) of the 50 state agencies respon- sible for traffic law enforcement. 6 Laser speed guns use LIDAR technology, which stands for Light Distance and Ranging. VASCAR stands for Visual Average Speed Computer and Recorder (Coleman et al. 1996, xixii). VASCAR is a time-distance speed-measuring device that does not transmit a signal but computes vehicle travel time between two points. It can operate in a stationary or mobile mode in a patrol car, motorcycle, or airplane.
145 Speed Enforcement and Adjudication Figure 4-1 Percentage of state law enforcement agencies using the top 10 enforcement strategies, based on replies to a survey of 50 state agen- cies responsible for traffic law enforcement conducted for this study. Responses were received from 34 of the 50 agencies. when enforcement will occur. As a result of its limited visibility, how- ever, its general deterrence effect appears to be limited (Shinar and McKnight 1985, 393394). Moreover, in jurisdictions where radar detectors are permitted, concealed police vehicles may be "seen" and their location communicated to others by CB radios, thereby com- promising their concealment. Resources for Speed Enforcement When the 55-mph (89-km/h) National Maximum Speed Limit (NMSL) was in effect and states were required to document speed compliance levels or face penalties, only one on-duty state highway patrol officer, on the average, was available to patrol every 190 mi (306 km) of highways posted at the 55-mph limit (TRB 1984, 162). Comparable figures are not available today, but a special survey of
MANAGING SPEED 146 state agencies responsible for traffic law enforcement conducted for this study provides some data on current levels of effort on speed enforcement. Data on costs of traffic enforcement, if properly collected, could provide valuable additional information. However, they are not routinely collected because of the difficulty and cost of doing so. The percentage of law enforcement resources devoted to speed enforcement varies widely by state (Figure 4-2). More than half of the state law enforcement agencies contacted for this study reported that they spend less than 50 percent of their time in speed enforce- ment.7 One-quarter of the respondents spend 50 percent or more. In addition, since repeal of the NMSL in 1995, the majority of state law enforcement agencies reported "no change" (38 percent) or reduc- tions (24 percent) in the time expended on speed enforcement. Those who reduced effort cited budget cuts or substitution of other activi- ties, such as community policing and investigations. Nearly one-third of the agencies (32 percent), however, reported increased time devoted to speed enforcement.8 Given these data, if limited resources are not to be squandered, whatever speed enforcement is undertaken must be effective. Effectiveness of Traditional Speed Enforcement A critical difficulty in deterring speeding using traditional methods is maintaining the effect over time and space. The longevity of effects can be expressed in terms of a "halo," that is, the spatial or temporal extent of the deterrence effect from the enforcement officer. Hauer and Ahlin (1982) investigated both effects by measuring vehicle speeds before, during, and after enforcement using stationary, marked police vehicles on semirural, two-lane roads. Similar to earlier stud- ies, the researchers found a marked reduction in average traffic speeds in the vicinity of the enforcement site to speeds close to posted speed limits (Hauer and Ahlin 1982, 277). Reduction in speed dispersion, 7 Comparable figures could not be found for county or municipal law enforcement agencies. 8 The remaining 6 percent were unable to offer any comparative data.
Figure 4-2 Time expended on speed enforcement as a percentage of total enforcement efforts, based on replies to a survey of 50 state agencies responsible for traffic law enforcement con- ducted for this study. Responses were received from 34 of the 50 agencies.
MANAGING SPEED 148 however, was less pronounced. Moreover, the distance-halo effect decayed quickly downstream of the enforcement site, following the gen- eral decay pattern of earlier studies (Hauer and Ahlin 1982, 277).9 With regard to the time-halo effect, average traffic speeds remained depressed for 3 d following a single episode of enforcement activity and for considerably longer (up to 6 d) with repeated enforcement (Hauer and Ahlin 1982, 275). Similar results were found for the effect of police presence on urban driving speeds (Armour 1986, 45). Increasing enforcement intensity should, according to deterrence theory, boost the deterrence effect by increasing the perceived risk of apprehension. In a review of European studies, Østvik and Elvik con- cluded that enforcement intensity must be increased significantly--to more than three times the initial level--before there is an appreciable effect on the perceived risk of detection or reduction in the number of speeding offenses (1991, 57).10 Studies of experience from other enforcement campaigns that have significantly increased the certainty of apprehension, such as anti-drunk-driving programs, have found the same effect. Positive behavioral and safety effects are evident immedi- ately after the adoption of the program. With a major initiative, the effect can last from a few months to a few years (Ross 1984, 29). However, deterrence effects of even a major program can diminish with time (Ross 1984, 29). One explanation for this effect is the pre- viously discussed behavioral adaptation of the police to the success of the program (Tsebelis 1993, 366). Another reason is that enforcement may deter the unwanted behavior but does not necessarily change the underlying attitudes that ultimately determine the behavior (De Waard and Rooijers 1994, 764; Rothengatter 1988, 599). There is some evidence that those who drive well in excess of the speed limit are the most impervious to the deterrence effects of tra- ditional enforcement methods. A recent study of an intensive police 9 Hauer and Ahlin found that the effect of enforcement was reduced by half for approximately every 2,953 ft (900 m) downstream from the enforcement site (Hauer and Ahlin 1982, 277). 10 According to the reviewers, increasing the level of enforcement by more then five times the initial level increases the perceived risk of detection, reduces the percentage of offenders, and may reduce the number of crashes by up to 20 to 30 percent (Østvik and Elvik 1991, 57).
149 Speed Enforcement and Adjudication intervention to reduce speeding on a 40-mph (64-km/h) urban road in northern England found a greater reduction in the number of driv- ers breaking the speed limit by a small amount than in the number exceeding the limit by 20 mph (32 km/h) or more (Holland and Conner 1996, 595). This finding corroborates a related drivers' sur- vey (part of the same study), which indicated that those who admit- ted to breaking the speed limit by a large amount in the past showed more intention to speed in the future than did those who admitted to speeding by smaller amounts (Holland and Conner 1996, 595). As discussed in Chapter 2, driving well in excess of average speeds is associated with both higher crash probability and greater crash sever- ity. This finding is confirmed by recent evidence from British Columbia that drivers with four or more excessive speed convictions had almost twice the overall crash rate of drivers whose most serious multiple offenses were simply exceeding the posted speed limit (Cooper 1997, 94). Roadway environment and traffic density also affect the success of traditional enforcement methods. Poor roadway geometry makes it difficult to station a patrol vehicle to take accurate speed measure- ments. In urban areas, high traffic volumes can hamper efforts to monitor speeding and apprehend violators (Shinar and McKnight 1985, 398). In very congested traffic, speed enforcement becomes irrelevant. Implications for Enforcement Strategies The foregoing evidence on the deterrence effect of traditional enforcement methods has several implications for both the uses and limitations of these methods. First, traditional enforcement "works" when the level of enforcement is sufficient to convince most drivers of the strong likelihood of detection and sanctions if they exceed the speed limit. Moreover, the level of effort must be maintained, placing a strain on most enforcement agency budgets, if the deterrence effect is to be sustained. Drivers generally revert to standard behavior once enforcement is reduced. Thus, to ensure a high level of compliance, speed limits have to be set at levels that are largely self-enforcing, or at the lowest speed the police are able to enforce.
MANAGING SPEED 150 Second, there is some evidence that enforcement does not deter those high-speed drivers who travel well in excess of the speed limit and pose a hazard to both themselves and other road users. These drivers obviously pose a special challenge for law enforcement. Finally, traditional enforcement methods are limited in certain contexts, particularly where road geometry is poor or when traffic is congested. In the following section, alternative ways of addressing these enforcement challenges are discussed. ALTERNATIVE METHODS FOR INCREASING EFFECTIVENESS OF SPEED ENFORCEMENT Optimizing Traditional Enforcement Methods The deterrence effect of enforcement clearly depends on creating the impression that road users who violate the law have a high probabil- ity of being apprehended. One way to achieve a credible level of enforcement without overstraining enforcement resources is to enforce speed regulations where and when risk-taking behaviors are most evident and traffic volumes are sufficient to justify the effort (Zaal 1994, 1617). Planned patrols on commuter routes at varying time intervals and locations, for example, were effective in extending the time- and, to a lesser extent, the distance-halo effects of enforce- ment (Brackett 1977). Patrol vehicle presence was reduced without disturbing the speed suppression effect, but only after an initial min- imum 6 weeks of continuous speed control activity (Brackett 1977, 48, 71). Varying the location of police patrols on commuter routes appeared to extend the distance-halo effect, but the evidence was inconclusive (Brackett 1977, 67, 72).11 Selective deployment strategies can also target particular types of unwanted risk-taking behavior, such as driving well in excess of the speed limit. Care must be exercised, however, that drivers do not receive the wrong message (i.e., that the de facto speed limit is well 11 The effect could have been created by CB radio reporting or the warnings of the presence of patrol vehicles by motorists flashing the headlights of their automobiles (Brackett 1977, 67).
151 Speed Enforcement and Adjudication above the posted limit plus a small tolerance). The targeted enforce- ment approach should have two effects. First, police presence should reduce unwanted behavior at high-risk locations and times. Second, planned but varying deployment schedules should extend the deter- rence effect to a broader segment of the driving population by increasing the expectation that enforcement may be present but leav- ing drivers uncertain exactly when the enforcement will occur (Zaal 1994, 17). A review of selective enforcement programs in the United States ( Jernigan 1986) found that the most successful programs are (a) deployed at specific localities and times when unwanted behavior is most likely to occur, (b) made highly visible to the public, and (c) maintained for more than a single year ( Jernigan 1986, 26). Making systematic safety gains from targeted enforcement strategies is diffi- cult, however, because of the relative infrequency of crashes. New Speed Enforcement Technologies Laser Speed Measurement Several new technologies have recently been introduced that could enhance the enforcement capabilities of police officers. Radar mea- surements of vehicle speeds have been the primary technology used by the police to detect speeders, but their effectiveness is compro- mised by radar detectors. Laser speed measurement presents an attractive alternative to law enforcement agencies because it offers the ability to target individual vehicles more accurately on multilane roads and is more difficult to detect than conventional radar (Figure 4-3).12 A recent study of the comparative effectiveness of radar and laser speed-measuring devices ( Jones and Lacey 1997) suggested that laser devices should complement rather than replace conventional radar. With its widespread and easily detected signal, radar may have a bet- 12 The narrow laser beam width (less than 0.5 degree) provides a high level of accuracy and thus is particularly effective for use in situations in which vehicle targeting is crit- ical, for example, on multilane roads ( Jones and Lacey 1997, xi). Moreover, the nar- rowness of the beam reduces the probability that a radar detector can identify the beam in time for the driver to slow down and avoid apprehension ( Jones and Lacey 1997, 1).
MANAGING SPEED 152 Figure 4-3 Comparison of conventional "down-the-road" radar tech- nology with "cross-the-road" laser speed measurement (Fitzpatrick 1991, 7). 1 ft = 0.305 m. ter general deterrence effect and thus is superior for general-purpose enforcement. Laser devices should be considered for situations where vehicle targeting is critical, such as monitoring speeders in high- traffic-volume conditions or in the left lane of a multilane facility ( Jones and Lacey 1997, xi). Automated Speed Enforcement Automated speed enforcement (ASE) equipment has been in use for over 30 years. Recent technology improvements have enhanced its effectiveness.13 With computer technology advances, sophisticated photographic and video equipment is now available for speed detec- tion purposes. Most ASE systems incorporate some form of radar, which can determine vehicle speeds under most conditions; supple- mentary photographic equipment is used to record the speed and doc- ument information on the violator (Blackburn and Gilbert 1995, 4). The system can be mounted in a patrol car for conventional mobile 13This section draws heavily on a review of automated technologies for speed man- agement and enforcement, which was commissioned for this study and appears in its entirety as Appendix D.
153 Speed Enforcement and Adjudication or stationary operation, or it can be attached to structures, such as poles or overpasses, for unattended operation (Blackburn and Gilbert 1995, 4344). ASE serves as a useful complement to traditional enforcement methods. It helps maintain an enforcement level that provides a meaningful deterrent to drivers by increasing the probability of detection for speeding violations. Because it can be deployed without police presence, ASE can increase the perceived level of risk to driv- ers and hence compliance levels without producing a reduction in police surveillance levels. It can be used in locations where patrol vehicles cannot be safely and effectively deployed. Moreover, when ASE is operated without police presence, it frees police for other traf- fic and law enforcement activities. The primary ASE technology is photo radar, which combines a radar unit14 and a computer that triggers a camera (or a video) to photograph a vehicle and its license plate if its speed exceeds some preset limit (Blackburn and Gilbert 1995, 4). The photograph records the time, date, location, and speed; the license plate is used to identify the vehicle owner. A citation is then mailed to the owner, who, depending on local laws, may be required to pay the fine or identify the offending driver. Owners can review the film at a police station and appeal the fine, but experience suggests that there are few challenges (Blackburn and Gilbert 1995, 22). Photo radar has been used extensively abroad, and there has been some use in U.S. cities; this experience is reviewed in Appendix D. Some comprehensive efforts will be discussed here to illustrate the effect of photo radar on driving speeds and safety. One of the earli- est evaluations of the effects of photo radar on speeds and crashes involved a high-crash location on a long downgrade section of the German autobahn between Cologne and Frankfurt (Lamm and Kloeckner 1984). The introduction of a 62-mph (100-km/h) speed limit, approximating the design speed for this highway segment, along with photo radar resulted in the desired reduction in driving 14 Despite the greater precision of laser speed-measuring devices, the review of ASE technologies conducted for this study (Appendix D) did not find any examples of ASE systems that use laser technology.
MANAGING SPEED 154 speeds and a dramatic improvement in safety. Injury crash rates15 dropped by a factor of 20 between 1971, before photo radar was introduced, and 1982, when an evaluation of the photo radar pro- gram was conducted (Lamm and Kloeckner 1984, 14). Victoria, Australia, has perhaps the most extensive photo radar enforcement program in the world. The program was launched with a massive publicity campaign in 1989, primarily on arterial roads with 37-mph (60-km/h) speed limits in metropolitan Melbourne and rural Victoria, where there had been serious injury collisions and val- idated complaints of excessive speeding (Coleman et al. 1996, 43). Evaluation of the program's first 2 years found a statistically signifi- cant decline in casualty crash frequency16 of 30 percent on arterial roads in Melbourne and 20 percent on rural roads, which corre- sponded to a greater level of speed camera enforcement on metro- politan Melbourne roads (Cameron et al. 1992, vi). The researchers attributed the decline to the deterrence effect of a dramatic increase in speeding offender detection and a related publicity effort (Cameron et al. 1992, ii). Speeding tickets in Victoria increased from about 20,000 per month before the program was launched to between 40,000 and 80,000 per month when photo radar was in use. Over the 2-year period, more than 20 percent of all drivers received at least one speeding ticket (Coleman et al. 1996, 42). Speed data suggest that the program was also successful in virtually eliminating very high-speed driving in the vicinity of photo radar deployment. In 5 years of operation, the percentage of vehicles exceeding the speed limit tolerance of nearly 10 mph (16 km/h) was reduced from 23 per- cent to 3 percent (Coleman et al. 1996, 44). A study of speed data taken near the permanent speed-monitoring sites confirmed these results for excessive speeding but found little change in average speeds and 85th percentile speeds (Rogerson et al. 1994, 33). Norway introduced photo radar in 1988. A carefully designed before-and-after study found a statistically significant 20 percent 15 Rates are measured as injury crashes per 106 vehicle-km. 16 This crash reduction occurred during "low-alcohol" hours. Crash times were sepa- rated into low- and high-alcohol hours to help distinguish the effects of the speed pro- gram from a concurrent drinking/driving campaign.
155 Speed Enforcement and Adjudication reduction in injury crashes for 64 road sections with ASE (Elvik 1997, 17).17 The largest reductions were found on sections that met two criteria that are currently used for ASE road selection: (a) a crash rate criterion, that is, crash rates higher than for similar road classes, and (b) a crash density criterion, that is, injury crash levels greater than or equal to 0.5 injury crashes per kilometer per year (Elvik 1997, 14). Nonrandom selection of study sites, however, limits generaliza- tion of the results. Data from Norway were combined with results from 15 other data sets of reported effectiveness of ASE, primarily in other European countries.18 The combined data indicated that a sta- tistically significant 17 percent average injury crash reduction accom- panied the introduction of photo radar (Elvik 1997, 18). Photo radar has been used sparingly in the United States. Programs are currently operating in four states--Arizona, Colorado, Oregon, and California (see Appendix D for details). Two of the best-known programs--in Paradise Valley, a small community in the Phoenix metropolitan area, and in Pasadena, California--operated for several years. Photo radar was deployed in staffed police vehicles on residential and arterial streets in both locations (Blackburn and Gilbert 1995, 34). Police data suggest that crashes were reduced, but scientifically designed studies of program effectiveness have not been conducted (Blackburn and Gilbert 1995, 37). Public opinion surveys conducted in both communities found high awareness of and support for the programs; nearly 60 percent of the respondents from both communities approved of the use of photo radar (Freedman et al. 1990, 62). However, a sizable minority (37 percent) disapproved. Since the survey was conducted, the Pasadena program has been ter- minated because of erosion of public support, loss of the equipment 17 The analysis controlled for potential bias from regression to the mean. This adjust- ment was necessary because abnormally high crash rates are one of the criteria for using ASE; thus selection bias is potentially a serious source of error (Elvik 1997, 15). 18 The countries are Germany, Sweden, England, the Netherlands, and Australia (Elvik 1997, 18). A statistical technique known as the logodds method of meta-analy- sis was used to combine the results. The method was essentially the same as combin- ing the data from the 64 individual road sections in Norway. The results were then weighted by the crash sample size from each country to derive a weighted mean change in injury crashes (Elvik 1997, 18).
MANAGING SPEED 156 vendor, and a reduction in police personnel (for more details see Appendix D). The primary issues regarding photo radar use in the United States are not technical but rather legal and political. Constitutional issues, such as the right to privacy and protection against illegal search and seizure, are frequently raised by opponents of photo radar use. However, state and Supreme Court decisions have ruled that driving on a public highway does not afford protections cited in the Fourth Amendment to the U.S. Constitution. It is difficult to prove that photo radar is an unreasonable exercise of police power because of its minimal intrusiveness and its legitimate public safety purpose (Gilbert 1996, 9). Owner versus driver issues are another legal concern. Deployment of photo radar systems is simplest when the law holds vehicle own- ers responsible for speeding violations whether or not they were driv- ing at the time of the infraction.19 This reduces the burden of providing a clear frontal photograph of the driver.20 Owner liability typically requires enabling legislation such as has been passed in Australia, the Netherlands, and some U.S. jurisdictions (Gilbert 1996, 7).21 This legislation is easier to enact if the sanctions are civil rather than criminal. This approach treats the violation much the same as a parking ticket. In many European countries and in Australia, however, fines are graduated according to the amount of the speed violation; points can be assessed against the driver's record, and licenses can be suspended for excessive speeding--presumably with a somewhat greater deterrence effect. Public opinion is another important issue. Surveys conducted in Paradise Valley and Pasadena found that the primary objections to photo radar were the possibility of error (i.e., the wrong person gets 19 Recent research on red light running in the United States found that most drivers who were ticketed were driving their own cars (Retting et al. 1998). 20 Of course, the owner has a clear defense if the vehicle was reported stolen prior to the time of the violation (Gilbert 1996, 7). 21 The Paradise Valley Ordinance, for example, provides that the owner or person in whose name the vehicle is registered pursuant to Arizona state law shall be held prima facie responsible for any speeding violation. The District of Columbia, Maryland, Michigan, Oregon, California, Virginia, and Utah have similar statutes (Gilbert 1996, 7).
157 Speed Enforcement and Adjudication the ticket) and perceived fairness (e.g., entrapment, lack of personal contact to explain mitigating circumstances) (Freedman et al. 1990, 65). The efficiency of photo radar in detecting speed violations can easily turn public opinion against use of the system and has been responsible for terminating some programs (Blackburn and Gilbert 1995, 3336). One method of addressing the fairness and "big brother" concerns is to publicize any program using photo radar at its inception, perhaps use warnings rather than fines at first, and to pro- vide signs informing the public that ASE is in force along highways where the devices may be in use (Gilbert 1996, 9). There are costs associated with the acquisition, deployment, oper- ation, and maintenance of ASE equipment. However, photo radar can generate revenue, which can partially or completely offset such costs.22 From the public perspective, one issue is who receives the revenues and for what purpose. In many U.S. communities, fines go to the jurisdiction rather than the local police department, with the result that the police are unable to cover program costs and thus drop the program. Another issue is use of the revenues; the public will not be supportive if they believe the devices are being deployed as fund- raising "speed traps." Current experience suggests that photo radar's success depends heavily on how it is introduced. First, it is important to deploy photo radar at sites where the safety record indicates and the public per- ceives there is a problem, perhaps including school and construction zones, locations with high crash rates, and sites where traditional police enforcement can be dangerous (e.g., urban Interstates) (Streff and Schultz 1992). In some locations, communities have been involved in selection of these locations. To avoid any perception of entrapment, it would be wise to set high enforcement thresholds at first, targeting excessive speed limit violators [e.g., 20 mph (32 km/h) 22 Most U.S. cities with ASE programs contract with a vendor to provide and support the equipment, train police officers, handle and process the film, research department of motor vehicle records, prepare citations, provide special photographic evidence for trials, and even provide expert testimony during trials (Blackburn and Gilbert 1995, 24). Depending on the penalty structure for speeding offenses, the program can be self- supporting.
MANAGING SPEED 158 or more over the speed limit].23 Coupling the introduction of photo radar programs with high-visibility publicity campaigns can increase public awareness and inform motorists of how, where, and why the system will operate. Although these measures cannot guarantee suc- cess, they can ameliorate key obstacles to successful photo radar use. Publicity Whether new or conventional enforcement technologies are used, the effectiveness of speed enforcement can be enhanced by well-designed public information programs. Publicizing an enforcement program should increase driver awareness and expectation of intensified enforcement levels (Zaal 1994, 23). Of course, the publicity must be followed up with an active enforcement effort if the general deter- rence effect is to be sustained. A further benefit of publicity programs is to increase public awareness of the reasons for the enforcement effort, which may help change underlying attitudes about the pro- scribed behavior or, at a minimum, create a more supportive climate for the enforcement program (Zaal 1994, 23). Empirical evidence corroborates these effects. For example, expe- rience with well-publicized anti-drunk-driving programs both abroad and in the United States suggests that sustained programs can deter drinking and driving (TRB 1987, 149150). At least part of the decline in alcohol-related fatalities over the past decade can be attrib- uted to a change in attitudes from greater publicity about the hazards of driving drunk and intensified enforcement.24 Publicity campaigns in combination with primary belt use laws25 and stepped-up enforce- ment have also resulted in sustained increases in safety belt use in several communities (Williams et al. 1996; Jonah et al. 1982). 23 Again, this can be a two-edged sword if drivers perceive that the de facto speed limit is 20 mph (32 km/h) or more over the posted speed limit. 24 The 17,126 alcohol-related fatalities in 1996 (41 percent of total traffic fatalities for the year) represents a 29 percent reduction from the 24,045 alcohol-related fatalities reported in 1986 (52 percent of the total) (NHTSA 1997, 1). 25 Primary belt use laws enable law enforcement officers to stop and fine drivers who are not wearing safety belts.
159 Speed Enforcement and Adjudication A study of special speed enforcement programs in two California cities with extensive publicity programs to increase public awareness found statistically significant reductions in speed-related crashes (Stuster 1995, iii). Moreover, the general deterrence effect appeared to spill over and contribute to reductions in certain types of crimes (Stuster 1995, iv). However, permanent behavior changes are not easy to achieve, as the experience with drunk driving and other public health initiatives such as antismoking campaigns has shown. A large-scale review of 87 road safety mass media campaigns for which some scientific evalua- tion had been performed found that, on the average, publicity cam- paigns can be expected to achieve about a 30 percent increase in awareness, but only a 5 percent change in attitudes and a 1 percent change in driver intentions (Elliott 1993, iv). Publicity is most effec- tive when it is combined with enforcement in a long-term effort. SANCTIONS AND ADJUDICATION The courts play a key role in speed enforcement. Judges can undermine police enforcement by throwing out speeding violations or reducing fines when they believe that the limits are arbitrary or unreasonable or that the fines are too harsh. Thus the police and the traffic judges must agree that the speed limits are sensible if they are to be enforced. Consistent treatment of speeding violations by the courts is also important to defuse any public perception that traffic regulations for speeding are arbitrary or capricious. Development of sentencing guidelines and training for traffic court judges who handle speeding violations can help ensure consistent treatment of violators. Licensing point demerit systems, which impose a system of gradu- ated penalties for speeding and other traffic violations, have already assisted in reducing inconsistencies in penalty assessments. The deterrence effect of sanctions for speeding violations is limited if lengthy court backlogs create a substantial lag between detection and punishment. New speed enforcement technologies, particularly photo radar, can simplify and streamline the adjudication process. For example, if speeding violations are treated as civil infractions, like parking tickets, with comparably priced fees, adjudication can be han-
MANAGING SPEED 160 dled administratively for the most part, with a high potential for reducing court hearings and costs. Further, if legislation holds the owner of the vehicle responsible for the speeding violation as opposed to the driver, greater efficiencies can result. For example, when such vicarious liability legislation was passed in New South Wales, Australia, the cost of processing speeding offenses was reduced by approximately 50 percent (South et al. 1988 in Zaal 1994, 21). A system that involves only civil sanctions and administrative adju- dication procedures, however, is likely to have a limited deterrence effect. A fine for a speeding violation would not be reported on a driv- er's record; hence, no penalty would be imposed for repeated offenses. In addition, if fines are kept low to discourage costly court hearings and appeals, drivers may receive the message that the infraction is minor. Thus, efficiency may be gained at the price of effectiveness. Some successful photo radar programs use a combined approach. For example, the Victoria, Australia, photo radar program introduced an automated traffic infringement notice (TIN) penalty system to allow efficient processing of offenses. The TIN informs the vehicle owner of the details of the offense as well as the penalties. The latter increase with the level of speeding over the posted speed limit; a combination of fines, license demerit points, or immediate license suspension may be imposed depending on the severity of the offense (Cameron et al. 1992, i). The vehicle's registered owner is liable for the penalties unless the owner identifies the driver at the time of the offense (Cameron et al. 1992, i). The U.S. Paradise Valley photo radar program offers a similar com- bined approach. Drivers caught speeding 20 mph (32 km/h) or less over the posted speed limit are charged with a civil infraction. Those caught speeding more than 20 mph over the posted speed limit are charged with a misdemeanor (i.e., a criminal traffic offense) (Blackburn and Gilbert 1995, 22). If the civil infraction is ignored, a second notice is sent and the owner's driver's license is suspended until the fine is paid. If the criminal offense is ignored, the owner's driver's license is sus- pended and an arrest warrant is issued (Blackburn and Gilbert 1995, 22). An analysis of citations issued under the Paradise Valley ASE pro- gram in 1988 and 1989 found that most of those cited opted to pay the fine (31 percent) or attend a defensive driving class (37 percent). Fewer
161 Speed Enforcement and Adjudication than 1 percent of the cases emanating from the citations went to trial, and, of these, the city prosecutor's office had an 82 percent conviction rate (Blackburn and Gilbert 1995, 22). In contrast, the Pasadena photo radar program, which did not have owner liability legislation and involved only monetary fines with limited provision for nonpayment, had large increases in speeding violators who ignored citations and declines in conviction rates as the program matured, which helped con- tribute to its demise (Blackburn and Gilbert 1995, 23). SUMMARY Enforcement and sanctions are necessary complements to speed lim- its. Simply posting a speed limit sign will not achieve desired driving speeds. Even if most drivers believe the limits are reasonable and com- ply with them, enforcement is essential to ensure conformity of the remaining drivers. Because the police have other important enforce- ment priorities and limited resources, their preferred strategy is to cre- ate a widely perceived impression that those who exceed the speed limits beyond some small tolerance band will be detected and appre- hended. This approach is successful only if motorists perceive that they have a nontrivial chance of being apprehended if they speed. The problem with traditional enforcement methods is their short- lived effect in deterring noncompliers. Extending the effect typically requires a level of enforcement intensity that exceeds the resources provided to the police for speed enforcement and other priorities. Policy makers can increase the resources directed toward speed enforcement, but providing adequate enforcement levels is expensive. One approach is to stretch existing resources by increasing the effectiveness of traditional enforcement methods. Planned patrols at varying time intervals and locations on a highway section can extend the time- and, to a lesser extent, the distance-halo effects of enforce- ment. The effect can be sustained even with reduced patrols but only after a lengthy period of continuous initial enforcement. Some suc- cess has also been documented from selective targeting of enforce- ment at high hazard locations and on roads and at times when high-risk, speed-related behaviors are most common and traffic vol- umes are sufficient to justify the effort. Systematic safety gains from
MANAGING SPEED 162 this approach are difficult to sustain because of the relative infre- quency of crashes. ASE technologies, particularly photo radar, offer an effective means of substantially increasing the intensity of enforcement and thereby the deterrence effect on speeding. Photo radar is widely used for speed control in Europe and Australia, with dramatic reductions in excessive speeding [i.e., exceeding the speed limit with a 10-mph (16-km/h) tolerance] and related reductions in casualty crash fre- quency. Photo radar use in the United States is limited because of legal and political, rather than technical, issues. If introduced selec- tively at first--at especially hazardous roads and locations with high crash rates--photo radar is likely to gain essential public support. Whether new or traditional enforcement technologies are used, a well-designed public information program can help boost the deter- rence effect. Publicity increases awareness and expectations of inten- sified enforcement. It can educate the public about the reasons for enforcement. To be credible, publicity must be followed up with an active enforcement effort. Although not part of the enforcement process, the courts play a major role in ensuring motorist compliance with speed limits. Support of traffic court judges is essential to ensure that speeding violations are treated seriously and that offenders are handled consis- tently. Automated enforcement has the potential to reduce court backlogs and shorten the time between detection and punishment by substituting administrative adjudication procedures and civil sanc- tions (like parking fines) for some speeding infractions. The current system of speed limits, enforcement, and adjudication, however, is not effective for all situations. Other speed management strategies, some of which are reviewed in the next chapter, may be required. REFERENCES ABBREVIATIONS NHTSA National Highway Traffic Safety Administration TRB Transportation Research Board Armour, M. 1986. The Effect of Police Presence on Urban Driving Speeds. ITE Journal, Feb., pp. 4045.
163 Speed Enforcement and Adjudication Bjørnskau, T., and R. Elvik. 1990. Can Road Traffic Law Enforcement Permanently Reduce the Number of Accidents? VTI Rapport 365A. Institute of Transport Economics, Norway, pp. 122145. Blackburn, R.R., and D.T. Gilbert. 1995. NCHRP Synthesis of Highway Practice 219: Photographic Enforcement of Traffic Laws. Transportation Research Board, National Research Council, Washington, D.C., 67 pp. Blumstein, A., J. Cohen, and D. Nagin. 1978. Deterrence and Incapacitation: Estimating the Effects of Criminal Sanctions on Crime Rates. National Academy of Sciences, Washington, D.C. Brackett, R.Q., Jr. 1977. Comparative Evaluation of Speed Control Strategies. Dissertation. Texas A&M University, Dec., 99 pp. Brier, S.S., and S.E. Fienberg. 1980. Recent Econometric Modeling of Crime and Punishment: Support for the Deterrence Hypotheses? Evaluation Review, Vol. 4, pp. 147191. Cameron, M., A. Cavallo, and A. Gilbert. 1992. Crash-Based Evaluation of the Speed Camera Program in Victoria 1990-1991. Phase 1 and Phase 2. Report 42. Accident Research Centre, Monash University, Victoria, Australia, Dec. Cameron, S. 1988. The Economics of Crime and Deterrence: A Study of Theory and Evidence. Kyklos, Vol. 41, pp. 301323. Chauncey, R. 1975. Deterrence: Certainty, Severity, and Skyjacking. Criminology, Vol. 12, pp. 447473. Coleman, J.A., et al. 1996. FHWA Study Tour for Speed Management and Enforcement Technology. FHWA-PL-96-006. Federal Highway Administration, U.S. Department of Transportation, Feb., 83 pp. Cooper, P.J. 1997. The Relationship Between Speeding Behaviour (as Measured by Violation Convictions) and Crash Involvement. Accident Analysis and Prevention, Vol. 28, No. 2, pp. 8395. Cox, G. 1994. A Note on Crime and Punishment. Public Choice, Vol. 78, Jan., pp. 115124. De Waard, D., and T. Rooijers. 1994. An Experimental Study To Evaluate the Effectiveness of Different Methods and Intensities of Law Enforcement on Driving Speed on Motorways. Accident Analysis and Prevention, Vol. 26, No. 6, pp. 751765. Elliott, B. 1993. Road Safety Mass Media Campaigns: A Meta Analysis. CR 118. Elliott & Shanahan Research, Sydney, New South Wales, Australia, Aug., 143 pp. Elvik, R. 1997. Effects of Accidents on Automatic Speed Enforcement in Norway. In Transportation Research Record 1595, Transportation Research Board, National Research Council, Washington, D.C., pp. 1419. Fitzpatrick, K. 1991. A Review of Automated Enforcement. FHWA/TX-92/1232-5. Texas Transportation Institute, Texas A&M University System, College Station, Nov., 72 pp. Freedman, M., A.F. Williams, and A.K. Lund. 1990. Public Opinion Regarding Photo Radar. In Transportation Research Record 1270, Transportation Research Board, National Research Council, Washington, D.C., pp. 5965. Gilbert, D.T. 1996. Photographic Traffic Law Enforcement. Legal Research Digest No. 36, National Cooperative Highway Research Program, Transportation Research Board, National Research Council, Washington, D.C., Dec.
MANAGING SPEED 164 Harper, J.G. 1991. Traffic Violation Detections and Deterrence: Implications for Automatic Policing. Applied Ergonomics, Vol. 23, No. 3, pp. 189197. Hauer, E., and F.J. Ahlin. 1982. Speed Enforcement and Speed Choice. Accident Analysis and Prevention, Vol. 14, No. 4, pp. 267278. Hirshleifer, J., and E. Rasmusen. 1992. Are Equilibrium Strategies Unaffected by Incentives? Journal of Theoretical Politics, Vol. 4, pp. 343357. Holland, C.A., and M.T. Conner. 1996. Exceeding the Speed Limit: An Evaluation of the Effectiveness of a Police Intervention. Accident Analysis and Prevention, Vol. 28, No. 5, pp. 587597. Jernigan, J. 1986. How To Make Selective Enforcement Work: Lessons from Completed Evaluations. VHTRC 870R15. Virginia Highway and Transportation Research Council, Charlottesville, 17 pp. Jonah, B.A., N.E. Dawson, and G.A. Smith. 1982. Effects of a Selective Traffic Enforcement Program on Seat Belt Usage. Journal of Applied Psychology, Vol. 67, pp. 8996. Jones, R.K., and J.H. Lacey. 1997. The Effectiveness of Laser and Radar Based Enforcement Programs for Deterrence of Speeding. DOT-HS-808-530. Mid America Research Institute, Winchester, Mass., Feb., 70 pp. Lamm, R., and J.H. Kloeckner. 1984. Increase in Traffic Safety by Surveillance of Speed Limits with Automatic Radar Devices on a Dangerous Section of a German Autobahn: A Long-Term Investigation. In Transportation Research Record 974, Transportation Research Board, National Research Council, Washington, D.C., pp. 816. NHTSA. 1997. Alcohol Traffic Safety Facts 1996. U.S. Department of Transportation, 7 pp. Østvik, E., and R. Elvik. 1991. The Effects of Speed Enforcement on Individual Road User Behaviour and Accidents. In Enforcement and Rewarding: Strategies and Effects. Proc., International Road Safety Symposium (M.J. Koornstra and J. Christensen, eds.), Copenhagen, Denmark, Sept. 1921, 1990, pp. 5659. Piliavin, I., R. Gartner, C. Thornton, and R.L. Matsueda. 1986. Crime Deterrence and Choice. American Sociological Review, Vol. 51, pp. 128135. Retting, R.A., A.F. Williams, and M.A. Greene. 1998. Red Light Running and Sensible Countermeasures: Summary of Research Findings. Presented at 77th Annual Meeting of the Transportation Research Board, Washington, D.C. Rogerson, P., S. Newstead, and M. Cameron. 1994. Evaluation of the Speed Program in Victoria 19901991. Phase 3 and Phase 4. Report 54. Accident Research Centre, Monash University, Victoria, Australia, Feb. Ross, H.L. 1973. Law, Science and Accidents: The British Road Safety Act of 1967. Journal of Legal Studies, Vol. 2, pp. 178. Ross, H.L. 1982. Deterring the Drinking Driver. Lexington Books, Lexington, Mass. Ross, H.L. 1984. Social Control Through Deterrence: Drinking and Driving Laws. Annual Review of Sociology, Vol. 10, pp. 2135. Ross, H.L. 1990. Drunk Driving: Beyond the Criminal Approach: United States. In Effective Strategies To Combat Drinking and Driving, International Congress on Drinking and Driving, Edmonton, Alberta, Canada, March 2830, pp. 217220.
165 Speed Enforcement and Adjudication Ross, H.L., and G.D. LaFree. 1986. Deterrence in Criminology and Social Policy. In Behavioral and Social Science, Fifty Years of Discovery (N. J. Smelser and D.R. Gerstein, eds.), National Academy Press, Washington, D.C., pp. 129152. Rothengatter, J.A. 1988. Risk and the Absence of Pleasure: A Motivational Approach To Modelling Road User Behaviour. Ergonomics, Vol. 31, No. 4, April, pp. 599607. Rowsome, F., Jr. 1965. The Verse by the Side of the Road. The Penguin Group. Sellin, T. 1967. Homicides in Retentionist and Abolitionist States. In Capital Punishment (T. Sellin, ed.), Harper and Row, New York, pp. 135138. Shinar, D., and A.J. McKnight. 1985. The Effects of Enforcement and Public Information on Compliance. In Human Behavior and Traffic Safety (L. Evans and R.C. Schwing, eds.), General Motors Research Laboratories, pp. 385419. Silberman, M. 1976. Toward a Theory of Criminal Deterrence. American Sociological Review, Vol. 41, pp. 442461. South, D., W.A. Harrison, and I. Portans. 1988. Evaluation of the Red Light Camera Program and the Owner Onus Legislation. SR/88/1. Victoria Transport, Road Traffic Authority, Hawthorn, Victoria, Australia, 35 pp. Streff, F.M., and R.H. Schultz. 1992. Field Test of Automated Speed Enforcement in Michigan: Effects on Speed and Public Opinion. UMTRI-92-39. The University of Michigan Transportation Research Institute, Ann Arbor, Sept., 135 pp. Stuster, J.W. 1995. Experimental Evaluation of Municipal Speed Enforcement Programs. DOT-HS-808-325. Anacapa Sciences, Inc., Santa Barbara, Calif. , Aug. Teeven, J.J., Jr. 1972. Deterrent Effects of Punishment: The Canadian Case. Canadian Journal of Criminology and Corrections, Vol. 14, pp. 6882. Tittle, C.R. 1969. Crime Rates and Legal Sanctions. Social Problems, Vol. 16, pp. 409423. Tsebelis, G. 1990. Penalty Has No Impact on Crime: A Game Theoretic Analysis. Rationality and Society, Vol. 2, pp. 255286. Tsebelis, G. 1993. Penalty and Crime: Further Theoretical Considerations and Empirical Evidence. Journal of Theoretical Politics, Vol. 5, No. 3, pp. 349374. TRB. 1984. Special Report 204: 55: A Decade of Experience. National Research Council, Washington, D.C., 262 pp. TRB. 1987. Special Report 216: Zero Alcohol and Other Options: Limits for Truck and Bus Drivers. National Research Council, Washington, D.C., 196 pp. Waldo, G.P., and T.G. Chiricos. 1972. Perceived Penal Sanction and Self-Reported Criminality: A Neglected Approach to Deterrence Research. Social Problems, Vol. 19, pp. 522540. Weissing, F., and E. Ostrom. 1991. Crime and Punishment: Further Reflections on the Counterintuitive Results of Mixed Strategy Equilibria Games. Journal of Theoretical Politics, Vol. 3, pp. 343350. Williams, A.F., D. Reinfurt, and J.K. Wells. 1996. Increasing Seat Belt Use in North Carolina. Journal of Safety Research, Vol. 27, No. 1, pp. 3341. Wilson, J.Q. 1983. Thinking About Crime (revised edition). Vintage Books, New York. Zaal, D. 1994. Traffic Law Enforcement: A Review of the Literature. Report 53. Monash University Accident Research Centre, Victoria, Australia, April, 188 pp.