Chapter 27. Motorcyclists
David J. Houston
University of Tennessee, Knoxville, TN, USA
Motorcyclists are the most vulnerable motorized vehicle users on the road. At a time when safety is improving for other road users, motorcyclists are not enjoying the same gains in safety. Substantial gains in global traffic safety will require reducing the risk that motorcyclists face. Much has been learned about factors that contribute to crashes and the characteristics of the individuals and behaviors involved. In recent years, transport psychology has focused on understanding why motorcyclists engage in risky behaviors. Toward this end, the field of transport psychology has begun to identify the psychosocial determinants of unsafe riding behavior. The theory of planned behavior has served as a useful theoretical framework to guide this research. Findings suggest that attitudes toward unsafe riding influence an individual's intention to perform risky behaviors that lead to an increased risk of a crash. Also, relevant attitudes are in part determined by personality traits. However, little attention has focused on rider/driver performance as related to hazard perception and “looked but failed to see” errors. Given the role that behavioral factors play in motorcyclist crashes, transport psychology has much to contribute for improving rider safety.

1. Introduction

The heightened vulnerability of motorcyclists is due to characteristics of the machine, the environment, and human behavior. In the event of a crash, motorcycles afford riders little physical protection, and their “single track” design and higher power-to-weight ratio make stability and handling more challenging. They can easily become unstable, making it difficult to keep the vehicle upright and under control. This is especially a concern during braking because motorcycles are susceptible to skidding and a loss of control if brakes are not properly applied (e.g., locking up the brakes). The higher power-to-weight ratio means that motorcycles can accelerate quickly and reach high speeds (Elliott, Armitage, & Baughan, 2003). Thus, operating a motorcycle is a more complex task than driving a car because it requires superb coordination, balance, and motor skills (Mannering & Grodsky, 1995). Furthermore, the small size of a motorcycle reduces its conspicuity (i.e., it is difficult for motorists to see), leading to “looked but failed to see” errors in which another motorist violates the rider’s right of way. Even when it registers to another motorist that a motorcycle is approaching, a motorcycle’s small mass makes it difficult to accurately judge its speed and time of arrival (Caird & Hancock, 1994).
The environment (e.g., weather, road condition, and road design) poses greater challenges for riders than for other vehicle users. Weather conditions are likely to make motorcycle travel more hazardous. The stability or handling of a motorcycle can be affected by poor and uneven road conditions or slick lane and road markings. Typically, roads and highways are designed to promote efficient, fast travel for larger vehicles, and they can place motorcyclists at greater risk. Multilane roads, the presence of an uncontrolled left turning lane, and wide medians have been found to increase the risk of a motorcycle crash (Haque, Chin, & Huang, 2010). In addition, crash barriers and guardrails designed to keep heavier vehicles from crossing into oncoming traffic or from leaving the roadway have been found to increase the severity of injury suffered when collided with by a motorcyclist (Brailly, 1998; as cited in Elliott et al., 2003).
Frequently, an engineering perspective is adopted to promote safer vehicle and road designs. Although technological safety developments for motorcycles have lagged behind developments for cars and trucks, several can be identified. Gains in safety have come in the form of enhanced stability offered by antilock braking systems (Watson, Tunnicliff, White, Schonfeld, & Wishart, 2007). Compensation for the lack of physical protection has led to the design of leg fairings that reduce the risk of a lower limb injury, increased motorcyclist conspicuity through the use of daytime running lights, and the development of helmets that substantially reduce the risk of a head injury or fatality (Bayly et al., 2006, Elliott et al., 2003, Watson et al., 2007 and Wells et al., 2004). Efforts to redesign the roadway environment entail altering road design and materials, designing safer crash barriers, and the placement of road signage that reduces obstructions.
Although an engineering perspective is very important for enhancing safety, behavioral factors play a major role in most traffic crashes (Evans, 2004 and Peden et al., 2004). In 2004, the Association des Contructeurs Eruopéens de Motocycles (ACEM) conducted the Motorcycle Accident In-Depth Study (MAIDS), which was an investigation of a sample of 921 motorcycle crashes that occurred in France, Germany, The Netherlands, Spain, and Italy. Behavioral factors were determined to be the primary cause in 87% of these crashes, with 37% being attributed to the motorcycle rider and 50% to the driver of another vehicle. In contrast, environmental and vehicle factors were considered the primary cause in only 8 and 4% of these crashes, respectively.
However, relatively little is known about the role of human behavior in motorcyclist safety. Early periods of research focused on crash analysis (1970s) and the process of riding (1980s) (Chesham, Rutter, & Quine, 1993a). Although correlates of crashes and the types of risky behavior that increase the likelihood of a crash have been identified, only recently have scholars turned attention to understanding the psychology of risky or unsafe motorcycle riding. Toward this end, in the early 1990s a conception of the rider as an “active agent” began to emerge, highlighting the importance of psychosocial influences on riding behavior (Chesham et al., 1993a).
Improving safety requires understanding the importance of the three components in motorcycle transport: the machine, the environment, and human behavior. Whereas the first two of these components are largely the realm of engineering, arguably it is human behavior that is the most important factor in motorcyclist crashes, and this is the realm of transport psychology.
The purpose of this chapter is to identify key issues relevant to motorcyclist safety with special attention to theoretical developments in transport psychology. The following section discusses key trends in motorcyclist safety. To better understand why crashes occur, the most frequent types of crashes are presented along with descriptive correlates of these crashes. Research on the psychosocial influences on riding behavior is then summarized. The final section discusses the implications of research in transport psychology on motorcyclist safety and suggests future topics to explore.

2. Trends in Motorcycle Use and Safety

At a time when traffic safety is improving for most motorists, motorcyclists generally have experienced increased risk. For instance, a study of trends in 30 nations revealed that from 2000 to 2009, the overall number of annual road fatalities decreased for all but 1 of these nations. However, in 13 of these nations, the annual number of motorcyclist fatalities increased. Even in those nations that experienced a reduction in annual motorcyclists fatalities, these reductions lagged behind those of other motorists.
The United States provides an example of the trend of increased motorcyclist fatalities. From 1980 to 1997, the number of rider deaths declined from 5144 to 2116. However, motorcyclist fatalities have increased each year since and reached a recorded high of 5312 in 2008. During this same 30-year period (1980–2010), total traffic fatalities decreased from 51,091 to 37,423 (National Highway Traffic Safety Administration (NHTSA), 2010).
Several factors underlie these trends. First, the popularity of motorcycles began to increase in the mid-1990s and these vehicles now comprise a larger, albeit still small, proportion of the motorized vehicles in many developed nations. Second, motorcycles have gotten more powerful. Engine displacement (in cubic centimeters (cc)) is a common measure of the power of a motorcycle, with larger engines generally being able to produce more power. In the United States, Shankar and Verghese (2006) reported that in 1990, approximately 21% of on-highway motorcycles had an engine displacement less than 350cc and 41% had more than 750cc. By 2003, the smaller motorcycles accounted for only 7% of on-highway motorcycles, whereas 76% of motorcycles had an engine displacement more than 750cc. Other nations have experienced a similar shift in the motorcycle fleet.
Third, rider demographics have changed. Although motorcyclists are still predominantly male, they are becoming an older age group as more individuals in their 40s are picking up motorcycling as a new hobby or are returning to it after some time away (Elliott et al., 2003 and Jamson and Chorlton, 2009). As a consequence, motorcycle crashes involve proportionally more older riders than in past years. In Australia, for instance, in 1985, riders younger than age 26 years comprised 71% of motorcyclist fatalities, a figure that declined to 25% in 2008. Conversely, whereas riders 40+ years old accounted for only 5% of fatalities in 1985, they accounted for 39% in 2008 (Department of Infrastructure, Transport, Regional Development, and Local Government (DITRDLG), 2009). A similar trend is apparent in the United States: From 1998 to 2008, the percentage of fatally injured riders younger than age 30 years decreased from 40 to 33%, whereas fatalities for riders 40+ years old increased from 33 to 51% during this period (NHTSA, 2010). Motorcycling safety is no longer just a young person’s issue.
Fourth, the issue of motorcyclist safety is important because riders are overrepresented in crash statistics. Data from three nations in which motorcyclist crashes are frequently studied illustrate this point. In the United States, motorcycles comprise only 3% of registered motor vehicles and account for 0.5% of vehicle miles traveled, but they account for 14.2% of all traffic fatalities (NHTSA, 2010). In the United Kingdom in 2008, 3.8% of licensed vehicles were motorcycles, which accounted for 1% of vehicle miles traveled and 19.4% of all road fatalities (Department for Transport, 2009). Data from 2007 for Australia indicate that 3.4% of registered vehicles were motorcycles. It is estimated that motorcyclists accounted for 0.9% of kilometers traveled but comprised 14.8% of road fatalities (Australian Bureau of Statistics, 2008 and Department of Infrastructure, 2009).
A better picture of rider vulnerability is gleaned from comparing motorcyclist fatality rates to those experienced by other road users. For the years just referenced, based on the amount of travel, the motorcyclist fatality rate is 39.4 times higher than the car fatality rate in the United States and 39.4 and 17.5 times higher in the United Kingdom and Australia, respectively. Furthermore, in the European Union as a whole, the risk of a motorcyclist fatality is 20 times that of a car passenger (Organization for Economic Co-operation and Development, 2010).
Similar trends are occurring in developing nations as well. Due to their low cost, convenience, and ability to maneuver on congested roads, motorcycles are an especially important mode of transportation in middle- and low-income nations. In China, motorcycles accounted for 23% of total motorized vehicles in 1987, a figure that rose to 54% by 2005 (Xuequn, Ke, Ivers, Du, & Senserrick, 2011). Motorcycles account for 67% of vehicles in Taiwan (Chang & Yeh, 2007), 52% in Nigeria (Oluwadiya et al., 2009), 60% in Malaysia (Radin-Umar, Mackay, & Hills, 1996), and 95% in Vietnam (Hung, Stevenson, & Ivers, 2006).
Consequently, related injuries and fatalities have increased in these medium- and low-income nations (Ameratunga et al., 2006 and Lin and Kraus, 2009). In Singapore, motorcyclists account for 54% of traffic fatalities and 51% of injuries (Haque, Chin, & Huan, 2010). They account for more than 50% of fatalities in Malaysia and Taiwan (Radin-Umar et al., 1996), 80% of traffic injuries in Thailand (Ichikawa, Chadbunchachai, & Marui, 2003). In China, 22.2% of all 2007 road fatalities were motorcyclists, a three-fold increase over the 7.5% recorded in 1987 (Xuequn et al., 2011).

3. Characteristics of Crashes

One emphasis in research on motorcycle safety has been to identify situations in which crashes tend to occur. One notable typology was offered by Preusser, Williams, and Ulmer (1995) based on an examination of fatal crashes in the United States during 1992. Through their analysis, Preusser et al. identified five common crash types. The most common involved a motorcyclist running off the road and overturning or striking an off-road object. These crashes comprised 41% of the sample, often occurred on a curve, and tended to be due to excessive alcohol consumption or riding too fast for the conditions. The rider was at fault in nearly all these crashes.
The second most common type involved another vehicle that failed to stop or yield the right-of-way to a motorcyclist (Preusser et al., 1995). These accounted for 18% of the sample of crashes and often occurred at controlled road intersections. The primary fault for these crashes tended to be other motorists, who were responsible for 66% of these crashes. The third type of crash involved a head-on collision, usually a rider who crossed over into oncoming traffic. These were more likely to occur in rural areas, on higher speed roads, and on curves. In approximately 73% of these cases, it was the motorcycle rider who was primarily at-fault.
A fourth type of crash involves one vehicle turning across the path of another, which occurred in 9% of the cases (Preusser et al., 1995). In nearly all these crashes (approximately 99%), the other motorist was primarily at fault. These crashes are an example of “looked but failed to see” errors, in which the driver of the other motor vehicle claims to have looked but did not see the motorcycle approaching. The last type of crash that Preusser et al. identified involved a rider who lost control and came off the motorcycle. This last crash may involve a rider falling off a motorcycle when engaged in a risky maneuver or when trying to avoid a collision. The primary liability for these crashes could not be determined based on the available data.
An important distinction among these crashes is whether other vehicles are involved. In general, motorcycle-only crashes are less frequent than those that involve a motorcycle and another motor vehicle. For instance, an in-depth study of more than 900 crashes that occurred in Los Angeles during 1976 and 1977 revealed that 26% were single-vehicle crashes (Hurt et al., 1981). Similarly, Tunnicliff’s (2005) study of fatal motorcycle crashes in Australia reported that 41% did not involve another vehicle. Other studies have indicated that only slightly more than one-third of fatal crashes are in the single-vehicle category (Christie & Harrison, 2002). Riders largely are at fault in single-vehicle crashes, which are frequently due to rider error and often involve a motorcycle leaving the road. Shankar (2001) indicated that 80% of fatal single-vehicle crashes occurred on the shoulder, median, or off the side of the road. Riding off the road on a curve is common, and it has been found to be the cause in approximately one-third of single-vehicle crashes (Christie & Harrison, 2002). Excessive alcohol is more commonly found in these crashes than in multivehicle ones (ACEM, 2004).
In contrast, multivehicle crashes are more common, and most of the time the driver of the other motor vehicle is at fault. For instance, the MAIDS study concluded that the motorcycle rider was primarily at fault in only 32% of the European crashes that were examined (ACEM, 2004). Most typically, these crashes involve a nonrider failing to yield the right-of-way to the motorcyclist. Hurt et al. (1981) indicated that in two-thirds of multivehicle crashes, the other driver violated the motorcyclist’s right-of-way. Clarke, Ward, Bartle, and Truman (2007) reported that crashes involving a right-of-way violation comprised 38% of UK crashes during 1997–2002, and in less than 20% was a motorcyclist either partly or fully at fault. The majority of these crashes occur at a T-junction or a four-legged intersection (Clarke et al., 2007 and Haque et al., 2010). These crashes also occur when a motorist turns in front of a motorcycle approaching from the opposite direction. The cause of most of these crashes is a “looked but failed to see” error in which a driver states that he or she looked but failed to see the approaching motorcycle. Clarke et al. estimated that 65% of crashes involving a right-of-way violation were caused by a driver failing to see an approaching motorcycle that was clearly visible to others at the scene, and they seemed to occur with greater frequency among drivers 65+ years of age.
In terms of roadway type, motorcyclists are especially vulnerable to other motorists in complex traffic situations such as intersections (e.g., four-legged intersection and T-junctions) and uncontrolled left-turning lanes (or right-turn lanes in the United Kingdom) (Sexton, Baughan, Elliott, & Maycock, 2004). Curves or bends in roads also present a risk to motorcyclists because crashes in these road sections increase the likelihood of a motorcyclist death or serious injury by 2 and 1.5 times, respectively, compared to other crashes (Clarke et al., 2007). On these curved road sections, typically it is the leisure-oriented rider who places him- or herself at risk due to speeding and inexperience.
Regarding population density, accidents that occur in rural areas are more likely than not to be the result of poor handling skills and risk taking (Lin et al., 2003 and Lin et al., 2003), whereas accidents that occur in urban areas are more likely to be the fault of the other motor vehicle driver and typically involve a right-of-way violation (Clarke et al., 2007 and Sexton et al., 2004). For instance, in Singapore, 58% of motorcyclists involved in crashes that occurred at intersections were the victims in the crashes, whereas only 33% of riders in crashes on expressways were the victims (Haque, Chin, & Huang, 2009). Lastly, motorcyclist crashes are more common on rural roads with high posted speed limits. For instance, Savolainen and Mannering (2007) report that crashes on U.S. roads with speed limits exceeding 50 mph have a 132% higher likelihood of a fatal injury than crashes that occur on roads with lower posted speed limits.
When a crash does occur, the most common injuries that motorcyclists sustain in a nonfatal crash are to the lower limbs. Watson et al. (2007) report 38% of injuries to be to the legs, 30% to the arms, 18% to the torso, and 12% involve the head and neck. Similarly, Elliott et al. (2003) state that 40–60% of all nonfatal injuries are to legs, approximately 25% to the head, and 20% to the arms. Although head injuries are less frequent in nonfatal motorcycle crashes, typically these are more serious, require greater time for recover, are more likely to be incapacitating, and require a greater financial expenditure (Eastridge et al., 2006). Furthermore, head injuries are the major cause of death following a motorcycle crash (Talving et al., 2010).
The most important behavior that riders can perform to reduce the risk of serious injury and death due to a head injury is to wear helmets, which have been found to reduce the risk of sustaining a head injury by 69% and decrease the risk of death by 42% (Liu et al., 2008). For this reason, mandatory helmet use laws have been adopted that require helmets to be used by all motorcyclists. In states with these compulsory laws, helmet use rates are higher and rider fatality rates are lower (Houston & Richardson, 2008). However, in the absence of more fundamental behavioral change, the effectiveness of a helmet law is in part a function of perceived enforcement effort. For this reason, helmet use often is lower in rural areas where enforcement is lax (Xuequn et al., 2011).

4. Correlates of Crashes

Beyond developing profiles of common crash circumstances and sustained injuries, investigations of motorcycle crashes have identified demographic and other attributes that correlate with an increased likelihood of a crash. The vast majority of motorcyclists are male (approximately 90% in many nations), who are also overrepresented in crash statistics (Christie and Harrison, 2002, Haworth et al., 1997, Mannering and Grodsky, 1995 and Watson et al., 2007). The overrepresentation is a function of the amount of travel that males account for and a stronger propensity for risky behaviors associated with increased crash risk (Fergusson et al., 2003, Lin et al., 2003, Rutter and Quine, 1996 and Savolainen and Mannering, 2007).
In addition, the risk of a crash has consistently been found to decline with rider age (Harrison and Christie, 2005 and Rutter and Quine, 1996), a relationship that is especially pronounced among male riders (Maycock, 2002). For instance, riders younger than 25 years of age are more likely than older riders to be involved in a motorcycle crash and to experience a moderate or fatal injury (Association des Contructeurs Eruopéens de Motocycles, 2004, Haworth et al., 1997, Sexton et al., 2004 and Watson et al., 2007). Similarly, a New Zealand study found that riders younger than 25 years of age had more than double the risk of a fatal or moderately serious injury when involved in a crash (Mullin, Jackson, Langley, & Norton, 2000). Zambon and Hasselberg (2006) conclude that the majority of riders involved in crashes are young men who typically adopt unsafe attitudes and behaviors that increase the risk of a crash and injury.
It is not just the risk of a crash that is associated with age: The likelihood of being at fault also declines with rider age (Clarke et al., 2007 and Elliott et al., 2007). This can in part be explained by a heightened propensity for risky behavior (Lin, Chang, Pai, et al., 2003; Rutter & Quine, 1996). Younger riders possess an increased willingness to break the law (Chang and Yeh, 2007 and Rutter and Quine, 1996), such as riding faster than the posted speed limit (Fergusson et al., 2003 and Teoh and Campbell, 2010) and riding without a valid operator’s license (Watson et al., 2007).
However, in recent years, the number of fatalities suffered by riders older than 40 years of age has increased substantially in nations such as the United States (Shankar, 2001 and Stutts et al., 2004), the United Kingdom (Sexton et al., 2004), and Australia (Australian Transport Safety Bureau, 2007), raising concerns that the nature of the “motorcycle safety problem” has evolved from one that primarily is concerned with young riders to one that is about older riders. The concern is with the middle-aged riders who are returning to motorcycling after years away (“born-again riders”) or who are picking it up for the first time. This age group possesses the economic means to afford larger, more powerful motorcycles but may have diminished physical skills that make them ill-suited for the larger sized machines. A few studies have found that older riders do have an increased risk of severe or fatal injury from a crash (Quddus et al., 2002 and Savolainen and Mannering, 2007; Shankar & Mannering, 1996). However, the risk of a crash is still higher for younger riders (Mullin et al., 2000 and Sexton et al., 2004), and Sexton et al. found that returning riders are not at greater risk than others. Although new riders have a higher crash risk regardless of age, the fatality trends regarding older riders appear to more appropriately reflect the growing number in this age group who now enjoy motorcycling (Haworth et al., 2002 and Watson et al., 2007).
In addition to rider demographics, risk exposure and riding experience are also correlated with the likelihood of a crash. The risk of a crash increases with the number of miles ridden (Lin et al., 2003, Lin et al., 2003, Mannering and Grodsky, 1995 and Sexton et al., 2004). Conversely, the more years that an individual has been riding, the lower his or her likelihood of being in a crash (Chesham et al., 1993b, Haworth et al., 1997, Lin et al., 2003, Lin et al., 2003, Savolainen and Mannering, 2007 and Sexton et al., 2004) and the less likely the individual is to be at fault for a crash in which he or she is involved (Sexton et al., 2004). In fact, a significant number of motorcycle accidents occur within the first 6 months of riding (Elliott et al., 2007 and Sexton et al., 2004). Although age and riding experience are correlated, studies that have sought to disentangle these effects have concluded that experience and age have independent influences on motorcyclist safety (Elliott et al., 2007, Maycock, 2002 and Sexton et al., 2004).
Regarding behavior, much of the literature has found that engaging in risky behavior (e.g., speeding, overtaking, and performing stunts) increases the likelihood of a motorcyclist crash and death. For instance, speed is the most frequently cited contributing factor to motorcycle crashes. Studies of Australian accidents have found that speed is a factor in more than half of multivehicle crashes (Federal Office of Road Safety, 1999). Similarly, the difference in the traveling speed of motorcycles compared to surrounding traffic was found to be a direct contributory factor in 66% of crashes involving a motorcycle in Europe (ACEM, 2004). Fast speeds also increase the likelihood that the rider is at fault (Elliott, Armitage, et al., 2007), a conclusion that is consistent with the finding that riders are more likely to be at fault in crashes that occur on roads with higher speed limits (Clarke et al., 2007).
It is the very nature of the thrill that can come from speed that attracts some motorcyclists. For instance, a survey of U.S. motorcyclists found that 70% of those sampled had ridden at more than 100 mph on a public road, and approximately 40% of these anticipated doing so again (Mannering & Grodsky, 1995). In a qualitative study of motorcyclist motivations and attitudes, Watson et al. (2007) found that some ride a motorcycle because of the thrill that it can evoke, including “testing one’s limits,” which frequently involves traveling at fast speeds. Consequently, motorcyclists do indeed travel at faster speeds than car drivers (Horswill & Helman, 2003), and crashes involving motorcycles tend to occur at higher speeds than those involving only cars (Carroll & Waller, 1980). Predictably, the safety effect of higher speeds in motorcycle crashes is more severe injuries and a greater likelihood of a fatality (Lin et al., 2003, Lin et al., 2003, Quddus et al., 2002 and Savolainen and Mannering, 2007; Shankar & Mannering, 1996).
The consumption of alcohol before riding is another risky behavior that has been identified. Studies in the United States and the United Kingdom show that alcohol-related crashes more commonly involve motorcyclists than drivers of other vehicles (Bednar et al., 2000, Fell and Nash, 1989, National Highway Traffic Safety Administration, 2010, Soderstrom et al., 1993 and Subramanian, 2005). Whereas in Australia, several studies did not find an increased likelihood of impairment among motorcycle crashes (Diamantopoulou et al., 1995 and Queensland Department of Transport, 2003), Haworth et al. (1997) did find that alcohol was involved in 26% of single-vehicle crashes.
The effects alcohol has on riders is that it increases risk-taking behavior (e.g., traveling at higher speeds) (Haworth et al., 1997 and Soderstrom et al., 1993) and impairs basic handling skills (Colburn et al., 1993 and Creaser et al., 2009). Consequently, drinking riders are more likely to be in crashes that involve running off the road, especially on curves, and a loss of control (Kasantikul and Ouellet, 2005 and Ouellet and Kasantikul, 2006). Inattentiveness and slower response times are other effects alcohol has on rider performance (Creaser et al., 2009).
The size of a motorcycle, in terms of engine capacity, is positively correlated with the likelihood of a crash and injury severity (Broughton, 1988, Lin et al., 2003, Lin et al., 2003, Quddus et al., 2002, Shankar and Verghese, 2006 and Teoh and Campbell, 2010). Also, riders of motorcycles with large engine sizes are more likely to be at fault in an accident. Lynam, Broughton, Minton, and Tunbridge (2001) conclude that although riders of large motorcycles have more experience than others, they are more likely to travel at faster speeds and be at risk for loss of control. Crashes involving large motorcycles are most likely to occur on rural expressways and result from the capability to travel at faster speeds (Clarke et al., 2007).
However, two literature reviews conclude that engine size/power does not increase risk of an accident (Mayhew and Simpson, 1989 and TNO, 1997). Instead, it is the amount and the type of travel that increase crash risk. In comparison to smaller bikes, larger ones are more likely to be used by the recreational rider, who is likely to travel greater distances during a trip, and are more likely to travel on rural expressways that accommodate faster travel (Elliott et al., 2003).
Instead of being an accident risk, it may be that the type of motorcycle reflects the riding style of the rider. Thus, riding style may be the true causal factor that explains the higher crash and casualty statistics associated with more powerful engines. As Teoh and Campbell (2010) stated, “Riders prone to higher risk driving behavior may choose more powerful and performance-oriented motorcycles” (p. 507). Previous research has found that supersport or racing design motorcycles have crash rates that are four times higher than that for touring motorcycles (Kraus, Arzemanian, Anderson, Harrington, & Zador, 1988). A similar finding was reported by Teoh and Campbell, who also reported that the supersport motorcycles are more likely to be ridden by younger riders more prone to risky behavior.

5. Understanding Riding Behavior

Research on motorcyclist safety has largely focused on understanding the factors that contribute to crashes and the profile of individuals who are likely to be involved. Although this knowledge is useful for identifying behaviors that must be modified to enhance safety and the groups that are the most likely targets for these efforts, it does not provide insight to modify the unsafe, risky behavior. This point was made in reference to speeding, frequently identified as a contributing factor in crashes, but it applies to unsafe riding generally: “There is a notable dearth of studies that have sought to identify variables that both underpin motorcyclists’ speeding behavior and are potentially amenable to change via safety interventions” (Elliott, 2010, p. 718). Toward this end, recent research is devoted to identifying the psychosocial determinants of unsafe riding behaviors that increase crash risk.
The importance of examining the determinants of riding behaviors is illustrated by research that has identified the various motivations individuals have for riding. Motorcycling is not just an instrumental activity. It also has a strong affective, expressive component that has implications for risky behavior. In reference to the relationship between an affinity for speed and committing behavioral errors, Sexton et al. (2004) write, “Such relationships lend support to the view that an important part of the motorcycle safety problem stems directly from the motivations that lead people to ride motorcycles in the first place” (p. 31).
Early efforts to understand motorcyclist behavior sought to identify the motives that characterize riders. Based on 100 in-depth interviews with riders in Wales, Walters (1982; as cited in Sexton et al., 2004) classified riders into three categories: those riding for practical reasons (cost and convenience), riding enthusiasts (pleasure), and those who enjoy the excitement and freedom of motorcycling. Those who were motivated by practical reasons comprised 35% of the sample and tended to ride small bikes for short distances, often for commuting. Riding enthusiasts comprised 48% of the interviewees, tended to be younger, rode for work and pleasure, and were confident in their riding skills. The smallest group (10%)—those motivated by excitement or to gain attention—tended to be young, were overconfident in their skills, perceived themselves as “invincible,” and engaged in riding behavior that most others would regard as irresponsible. The remaining 7% could not be easily classified.
A study of 376 German motorcyclists identified three general motivational categories: biking for pleasure, biking as a fast competitive sport, and control over the motorbike (Schulz, Gresch, & Kerwien, 1991). These categories differed in terms of rider age and bike type. Young riders were found to be more influenced by pleasure and thrill seeking. Riders of sports bikes were more influenced by motives relating to competition and exhibition. In addition, riding for pleasure was more likely to characterize riders of sports bikes and other specialized bikes (e.g., choppers).
Some of the first research on this topic was published by Rutter and Quine (1996) and Rutter, Quine, and Chesham (1995), who conducted two surveys of riders in the United Kingdom. The first survey asked riders about their attitudes and behaviors, whereas the second survey queried the same respondents 1 year later about riding crashes and mishaps they experienced during the intervening period. Analysis of survey responses revealed four categories of riding behavior: breaking laws and rules, taking care, carelessness, and safety equipment and training. Of these behaviors, breaking the laws and rules of safe riding was found to be positively related to accident involvement.
Elliott, Baughan, and Sexton (2007) created the 43-item Motorcycle Rider Behaviour Questionnaire (MRBQ) to measure the self-reported frequency of riding behaviors. The MRBQ was based on the Driver Behaviour Questionnaire developed by Reason, Manstead, Stradling, Baxter, and Campbell (1990), which identified types of driving behavior. Five categories of riding behavior were identified by Elliott, Baughan, et al.: traffic errors, control errors, speed violations, stunts, and safety equipment. Traffic errors were the most consistent predictor of crash involvement, likely reflecting the physical challenge of motorcycle riding, whereas speed violations were associated with involvement in a crash that a rider admitted to at least be partly at fault.
Among the psychosocial determinants of risky riding behavior that have been considered in research are the attitudes, beliefs, intentions, and personality traits of riders. The most pronounced theoretical influence on this research comes from the theory of planned behavior developed by Ajzen (1991), which has been demonstrated to be useful for explaining risky driving behavior (e.g., drinking and driving and also speeding) (Armitage and Conner, 2001, Elliott and Armitage, 2009, Elliott et al., 2007, Elliott and Thomson, 2010, Lawton et al., 1997 and Manstead and Parker, 1995).
The premise of the theory of planned behavior is that intentions are the best predictors of individual behavior. Intentions are formulated by a reasoned process whereby, either implicitly or explicitly, an individual considers the consequences of his or her actions and chooses the action that is most likely to generate a desired outcome. These intensions are themselves determined by an individual’s attitudes and subjective norms regarding the behavior. The attitudes represent the positive and negative assessments of a behavior that an individual possesses. Subjective norms represent the social pressure an individual feels and are determined by whether “important others” would approve or disapprove of the behavior.
However, in some instances an individual may feel as though they have little volitional control over their actions. For example, a rider who wants to perform a “wheelie” may not possess the skill or may be limited by the bike’s mechanical condition (Watson et al. 2007). Thus, perceived behavioral control is a third determinant of intentions. In addition to influencing intentions, perceived behavioral control may also directly influence behavior. When an individual does not feel in complete volitional control, attitudes and subjective norms determine intentions along with perceived behavioral control. Conversely, in situations where an individual feels that they have control over what they do, attitudes and subjective norms alone will determine intentions.
One of the earliest efforts to apply the theory of planned behavior to motorcyclists was by Rutter and Quine (1996) and Rutter et al. (1995). In addition to identifying that the involvement in a crash is related to traffic errors, they also examined the structural relationships among concepts at the base of an early version of the theory of planned behavior that Fishbein and Ajzen (1975) labeled the theory of reasoned action. Rutter et al. and Rutter and Quine found that attitudes toward obeying laws and rules, and attitudes toward taking care while riding, predicted the self-reported breaking of laws and rules of safe riding. Consistent with the theory of reasoned action, they concluded, “Beliefs about safe riding do predict riding behavior, which in turn predicts accident involvement, and that beliefs are best seen as mediators between demographic inputs, such as age and experience, and behavioral outcomes” (Rutter et al., 1995, p. 369).
Watson et al. (2007) and Elliott (2010) provide two additional examples that invoke the theory of planned behavior to explain risky riding. In the study by Watson et al., surveys administered to 227 motorcycle riders in Australia were analyzed using hierarchical regression analysis. Key dependent variables were intentions to engage in three safe behaviors (handle the motorcycle skillfully, maintain 100% awareness, and avoid riding impaired) and three unsafe behaviors (bend road rules, push limits, and perform stunts and/or ride at extreme speeds), along with self-reported behaviors (handling errors, awareness errors, ride impaired, bend road rules, push limits, and stunts or speed). In general, it was found that attitudes had a significant influence on risky behavior intentions, whereas intentions to perform safe behaviors were more likely to be influenced by perceived behavioral control. In addition, self-reported behavior was consistent with intentions, confirming the thesis of the theory of planned behavior that other influences on behavior are mediated through intentions.
Elliott (2010) sampled 110 riders from motorcycle clubs in Scotland who were asked about their intent to speed on roads with low and high speed limits. In general, the findings are consistent with the theory of planned behavior in that attitudes and perceived behavioral control were found to be important predictors of speeding intentions. However, several modifications and extensions to the theory of planned behavior are suggested by these and other studies.
For instance, Elliott (2010) draws a distinction between instrumental and affective attitudes. Instrumental attitudes represent cognitive evaluations about the benefits of performing a behavior, whereas affective attitudes are an emotional evaluation. Affective attitudes are strong predictors of intention across most social behaviors (Trafimow et al., 2004), and motorcycle riding is recognized as having a strong affective motivational component (Christmas et al., 2009). Previous research on risky motorcycle riding has not drawn this distinction. For instance, Watson et al. (2007) measured only instrumental attitudes, whereas Jamson et al. (2005) used a mixed measure that combined instrumental and affective attitudes into one scale that they found to be positively correlated with risky riding behavior. In contrast, Elliott (2007) used a specific measure of affective attitudes and reported it to be significantly correlated with motorcyclist intentions to speed on roads with either low or high speed limits.
In terms of its application to explain safety behavior, at times the concept subjective norm has not had the predictive power that is expected according to the theory of planned behavior. It is speculated that this is because the concept is understood as pertaining to “what important others would think” (Elliott, 2010 and Watson et al., 2007). Given the social nature of motorcycle riding, it may be that the norms that matter are those that are set by the group in which the riding takes place. The opinions of parents, spouses, and significant others in a rider’s life may matter less than what the other riders one rides with think about certain behaviors. When operationalized in this way, subjective norms have been more useful for predicting intentions (Watson et al., 2007).
In addition to the subjective norms hypothesized to affect intentions, Watson et al. (2007) included a measure of personal norms. Drawing upon the study of traffic violations by Parker, Manstead, and Stradling (1995), who contended that internalized moral norms are useful for explaining deviant behavior, Watson et al. hypothesized that personal norms affect risky behavior in addition to the effect of socially derived norms. However, the personal norms construct proved difficult to measure and consequently was not included in the final analysis (Watson et al., 2007).
A last suggested extension of the theory of planned behavior is the inclusion of additional concepts to represent the weight of social influence on intentions. Although subjective norms tap the influence of groups, Elliott (2010) contends that Ajzen’s theory of planned behavior does not go far enough to represent the importance of the social context of motorcycle riding. For this reason, Elliott includes self-identity and social identity constructs. Self-identity refers to an individual’s self-concept that is defined in terms of the societal roles with which an individual identifies and associated role-appropriate behavior. In contrast, social identity theory highlights the role of group memberships and explains that social identities stem from the social groups to which individuals belong or with which they identify. Because of this group identity, an individual is likely to align behavior so that it is consistent with the norms that characterize a salient group. Elliott reports that social identity has a significant influence on speeding intentions for individuals with strong group identifications. However, Watson et al (2007). did not find self-identification as a “risky rider” to be an important influence on intentions.
Beyond being the result of planned or intentional behavior, personality traits have long been thought to influence a rider’s tendency to engage in risky behavior that may lead to a crash. This approach builds off the work of Zuckerman (1994), who regarded sensation seeking as a biological trait that predisposes one to either underestimate risk or regard risk as merely the cost that must be paid to enjoy a thrilling experience. A correlation between a sensation-seeking trait and risky driving behavior is well-established (Jonah, 1997a and Jonah, 1997b). However, sensation seeking alone is unlikely to distinguish risky riders from safer ones because the population of motorcyclists is regarded as being higher in sensation seeking compared to the nonriding population. Thus, sensation seeking in combination with high aggression is likely to be more predictive of risky riding behavior (Watson et al., 2007 and Zuckerman, 1994). In addition, Ulleberg (2001) contends that high-risk populations are characterized by anger, normlessness, and sensation seeking—traits frequently found to be related to crash involvement (Ulleberg & Rundmo, 2003). The causal relationship is likely to be such that personality traits influence risky driving behavior through individual attitudes about unsafe behavior (Ulleberg & Rundmo, 2003).
As hypothesized, Watson et al. (2007) found both sensation seeking and a propensity for aggressive behavior to be significant predictors of self-reported risky riding behaviors. A measure of risk taking was also included in a study by Haque, Chin, and Lim (2010), who found both impulsive sensation seeking and aggression to be correlated with having been involved in a motorcycle crash. Defining personality types based on impulsive sensation seeking and aggression, Haque, Chin, and Lim concluded that “extrovert” and “follower” personality types of motorcyclists are more prone to crashes.
The association of personality traits with both attitudes and behavior was examined in a study of 257 students at a Taiwanese university (Chen, 2009). Attitudes toward unsafe riding were found to directly influence self-reported risky riding behavior (speeding, rule violations, and self-assertiveness). It was reported that the role of personality traits on behavior is indirect because they are mediated by attitudes. Specifically, the traits of anger, sensation seeking, and normlessness are positively associated with risk-taking riding attitudes, whereas anxiety is negatively associated. Of these traits, normlessness had the largest influence on attitudes. Although not addressed by the author, the influence of normlessness on attitudes about risky riding supports the need to consider personal norms for understanding unsafe behavior, as suggested by Watson et al. (2007). A last finding from the study by Chen is that altruism (i.e., a concern for others) was the only personality trait that had a direct influence on behavior, indicating that a being predisposed to think of others diminishes the likelihood of risky riding.
Similarly, a study of young Taiwanese motorcyclists by Wong, Chung, and Huang (2010) concluded that personality traits (sensation seeking, amiability, and impatience) only indirectly influence self-reported risky riding behavior (fast riding and riding violation) because personality traits are mediated through affective risk perception and utility perception. In addition, personality traits have direct effects on riding confidence, which indirectly affects behavior through attitudes toward unsafe riding and being aware of traffic conditions. This latter finding suggests that confident riders are more likely to perform risky riding behavior but are attentive to the hazards involved. These results are consistent with the theory of planned behavior in that attitudes toward unsafe riding are related to self-reported risky behavior. The constructs of utility perception and affective risk perception suggest that a distinction between instrumental and affective attitudes is useful, and planned control behavior (similar to Wong et al.’s riding confidence construct) is also relevant for risky riding behavior.
In summary, transport psychology research devoted to understanding risky motorcycle riding is in a nascent stage. Although considerable attention has been devoted to risky driving behavior, relatively little research has considered explaining the risky behavior of motorcyclists. Based on the research that has been conducted, it is clear that a variety of motives attract motorcyclists, including affective ones. The theory of planned behavior provides a promising theoretical framework to direct future research. Coupled with an understanding of personality traits, the theory of planned behavior may provide a sound understanding of unsafe motorcycle riding and inform the development of interventions that will create safer behavior.

6. Conclusion

Motorcyclists are the most vulnerable motorized vehicle users on the road. At a time when safety is improving for other road users, motorcyclists are not enjoying the same gains in safety. The motorcyclist safety problem is even more serious in developing nations, where these vehicles are more important for commuting and for commerce. Substantial gains in global traffic safety will require reducing the risk that motorcyclists face.
Early periods of motorcycle safety research focused on crash analysis and the technical facets of motorcycle travel. As a result, much has been learned about factors that contribute to crashes and the characteristics of the individuals and behaviors involved. This knowledge has suggested the need to engineer safer motorcycles, to redesign roads that more safely accommodate riders alongside other vehicles, and to design safety equipment (e.g., helmets and protective clothing) to protect riders involved in a crash.
Several implications for safety interventions can be drawn from the extant transport psychology literature. First, attitudes toward risky riding are likely the key to modifying behavior, meaning that interventions should focus on altering personal and social norms or encouraging the development of alternative group identities that value less aggressive riding styles (Chen, 2009 and Elliott et al., 2007). Second, rider training programs not only need to focus on rider skill and road rule knowledge but also should address the attitudinal and motivational influences on rider behavior to encourage greater personal control (Watson et al., 2007). Third, motorcyclists are a heterogeneous population. Different riders pose different hazards, requiring interventions to be tailored to the target group and behavior (Wong et al., 2010).
In addition to the study of unsafe or risky riding behavior, rider/driver performance is in need of attention by transport psychology. Only recently has research begun to shed light on the way in which riders process information when faced with potential hazards (Hosking et al., 2010 and Liu et al., 2009). Given that motorcycle riding is such a challenging activity that exposes the rider to a wide variety of dangers, a greater understanding of hazard detection and response is needed to assist motorcyclists in developing these skills.
Less obvious is the importance of research to study driver performance. Motorcyclists are more often the victim in a crash, and “looked but failed to see” errors committed by the driver of a motor vehicle are a common cause. An engineering approach has dominated efforts to reduce these errors by attempting to increase the visibility of motorcycles by making bikes more readily stand out from the background. This approach treats conspicuity as a trait inherent in the motorcycle and the rider. Although some gains in rider conspicuity have been achieved through motorcycle design (e.g., daytime running lights and vehicle and rider apparel color and reflectivity), physical conspicuity aids have not solved the problem posed by “looked but failed to see” errors. Wulf, Hancock, and Rahimi (1989) distinguished between “sensory conspicuity” and “cognitive conspicuity,” but it is the former that has characterized most research on this issue and encouraged an engineering perspective. Studying cognitive conspicuity entails thinking of the automobile driver as an information processor that is central to these errors and necessitates a perspective rooted in transport psychology. Langham (1999) provides one example of the type of research needed to address “looked but failed to see” errors.
Given the role that behavioral factors play in motorcyclist crashes, transport psychology has much to contribute for improving rider safety. The increased attention that motorcyclist safety has received in recent years is timely given trends in both developed and developing nations. However, much more needs to be done to understand rider behavior and rider/driver performance to sufficiently inform interventions and strategies that will make riders on the road safer.