Motorcycle Type Matters: Use of Helmet, Speeding, and Drinking in Motorcycle Crashes

Transcription

1 Motorcycle Type Matters: Use of Helmet, Speeding, and Drinking in Motorcycle Crashes *Chanyoung Lee, Ph.D., AICP, PTP Center for Urban Transportation Research University of South Florida 4202 East Fowler Avenue, CUT100 Tampa, FL Phone: (813) Fax: (813) Joan Pino, M.A. Center for Urban Transportation Research University of South Florida 4202 East Fowler Avenue, CUT100 Tampa, FL Phone: (813) Fax: (813) Pei-Sung Lin, Ph.D., PE Center for Urban Transportation Research Phone: (813) Fax: (813) Edie Peters FDOT Traffic Safety Specialist 605 Suwannee Street, MS 17 Tallahassee, Florida PH (850) Fax (850) TRB Annual Meeting Submitted November 15, 2012 Word Count: 4,108 words + 12 figures and tables (250*12) = 7,108 words * Corresponding author

2 ABSTRACT Motorcyclist traffic crashes and fatalities continue to remain a primary concern for traffic safety in the United States. In 2010, Florida had 7.3 percent of the registered motorcycles in the U.S., yet 8.7 percent of motorcycle related fatalities occurred in Florida. Many studies have been conducted to examine motorcycle crashes, causes, and outcomes to shed light on potential countermeasures to efficiently and effectively reduce motorcycle related injuries and fatalities. While previous research has controlled for several important factors in analyses, many studies have often examined data without recognizing motorcycle types. Different motorcycle types can represent certain characteristics unique to the people that ride them. This study analyzed Florida crash data and annual observational surveys of motorcyclists in Florida to understand the unique characteristics and crash outcomes according to motorcycle type. Statistical analysis and regression models are used to examine the factors that affect the level of injury severity by motorcycle type. The main findings of the study include the following: 1. Helmet use is highly associated with motorcycle type; 2. The composition of motorcycle type in the motorcycle crash data is different by age; 3. Young motorcyclists, as well as sport bikes, are overrepresented in motorcycle crashes; 4. Fatally injured sport bike riders were more likely to speed whereas cruiser and touring riders were more likely to be under the influence of alcohol; and 5. The same contributing factors increase the level of injury severity in motorcycle crashes regardless of motorcycle type.

3 Lee at al INTRODUCTION According to a recent report by the National Highway Traffic Safety Administration (NHTSA), traffic fatalities in the U.S. have been significantly reduced in past ten years (1). Not only has the absolute number of annual traffic fatalities been reduced, but the fatality rate per 100,000 registered vehicles and the fatality rate per 100 million vehicle miles traveled (VMT) has also been reduced during the same period. In the meantime, motorcycle fatalities still represent a much higher proportion of fatal traffic crashes compared to other vehicle types. The Centers for Disease Control and Prevention (CDC) reported that motorcycle crash-related injuries and deaths cost the U.S. as much as $12 billion in a one year period in medical care costs and productivity losses (1). For the year 2010, NHTSA reported a total of 4,502 motorcycle fatalities, which is a slight increase from the previous year (2). In 2010, Florida had 7.3 percent of registered motorcycles in the U.S., yet 8.7 percent of motorcycle related fatalities occurred in Florida (3). In 2010, motorcycle fatalities represented 15.7 percent of all vehicle crash fatalities in Florida while motorcycles accounted for less than 3 percent of Florida s motoring public (4). Figure 1 shows the trend of motorcycle crashes and fatalities in Florida over the past ten years. Motorcycle crashes and fatalities increased along with motorcycle registrations until 2008 and since then they have been reduced. Indexed (2001 = 1) 2.50 Indexed Motorcycle Registration in Florida Indexed Motorcycle Crashes 2.00 Indexed Motorcycle Fatalities (Source: Florida Department of Highway Safety and Motor Vehicles, (5)) FIGURE 1 Indexed Motorcycle Registration, Crashes, and Fatalities in Florida ( ) In recent years, state and federal highway safety offices have made ample efforts on motorcycle safety as crash data has revealed motorcyclists to be vulnerable road users. Numerous studies (6,7,8,9,10) have been conducted to examine motorcycle crashes, causes, and outcomes to shed light on potential countermeasures to efficiently and effectively reduce motorcycle related injuries and fatalities. NHTSA publishes studies regularly that examine trends in fatal motorcycle crashes by a variety of factors, including helmet use, alcohol involvement, age, speed, and engine size. In addition to federal efforts, many academic scholars and researchers have studied motorcycle crashes extensively using both national and statewide datasets and employing many different statistical evaluation techniques.

4 Lee at al Motorcycles are a means of transportation and motorcyclists have the same rights and responsibilities as other motorists. However, it is well-known that many people also enjoy riding motorcycles as a hobby in the U.S. According to a recent survey of 2,773 Florida motorcyclists, 66.3 percent of respondents indicated that recreation is the primary purpose of riding their motorcycle while 18.4 percent of respondents stated their primary riding purpose was commuting (11). It seems that different types of motorcycles reflect different cultural and behavioral characteristics. However, few studies have been conducted to understand motorcycle crashes and injuries in the context of motorcycle type. This study analyzed Florida police crash reports and annual observational surveys of motorcyclists in Florida to understand the unique characteristics and crash outcomes according to motorcycle type that have been over looked in previous motorcycle related studies. MOTORCYCLE TYPE For legal purposes, motorcycles have been defined as motor vehicles with motive power having a seat or saddle for the use of the rider and designed to travel on not more than three wheels in contact with the ground (12). In the U.S., there are several classifications of motorcycles. Each class varies dramatically in design and has unique characteristics which appeal to the different interests and needs of motorcyclists. While there are a few ways to define the different types of motorcycles, Table 1 provides a summary of the major types in the U.S. and their corresponding characteristics. The authors decided on and defined the listed motorcycle types based on a combination of two sources, the U.S. Department of Transportation s motorcycle classifications and a paper by Teoh and Campbell which explored the role of motorcycle type in fatal crashes (13, 14). It is noted that while these are the classifications used in this paper, other motorcycle classifications exist and in some cases there is considerable overlap between motorcycle types in terms of their design characteristics. TABLE 1 Motorcycle Classifications (13, 14) Motorcycle Type Definition / Characteristics Currently the most popular motorcycle category in the U.S. They are heavy motorcycles with more emphasis on style and less on performance. The Cruiser profile is long with a low saddle height which places the motorcyclists feet forward of the seat and hands near shoulder height. Small, lightweight, and low power vehicles. They are designed for short Moped/Scooter distance travel in urban environments and typically not legal for use on high-speed roadways. Tall and lightweight vehicles with a high center of gravity which allows for On/Off Road good suspension on off-road conditions. They are suitable for both on- and off- road use. These motorcycles have an emphasis on performance (handling, Sport Bike acceleration, top speed, cornering, and braking). They tend to be lighter weight and more powerful than other motorcycle types. These motorcycles are designed as a means of practical transportation and Standard their popularity has declined over time. They have a basic design with an upright riding position.

5 Lee at al Touring Trike These are large motorcycles with big engines and room to haul luggage that are intended for long distance travel. They often have many features such as wind protection, antilock brakes, audio systems, and cruise control for added comfort. Three wheeled vehicles that do not handle or steer like a motorcycle. Some states do not apply motorcycle laws, such as helmet use and licensing requirements, to trikes. Some research has been conducted to investigate the differences among different motorcycle types. An analysis conducted by the Insurance Institute for Highway Safety (IIHS) uncovered that 71 percent of sport bike riders who died in 2005 wore helmets, compared with only 52 percent of touring riders (15). In addition, the performance capability of sport bikes may influence riders risk taking behaviors, such as speeding (14). In 2007, the IIHS reported that speed was a factor in 57 percent of sport bike riders fatal crashes, while it was only cited in 27 percent of fatal crashes of cruiser and 22 percent of touring fatal crashes (15). OBSERVED MOTORCYCLE TYPES IN FLORIDA This study analyzed the results of Florida s annual Observational Survey of Motorcyclists over the past three years ( ). The observational surveys aim to observe the use of personal protective equipment (PPE), including helmets, and the types of motorcycles ridden by motorcyclists in Florida. The surveys were conducted based on the NHTSA guidelines for seat belt use observational survey. They were conducted in Florida counties with the highest number of motorcycle fatalities and one hour field observations were done at intersections for each of the selected counties. Table 2 summarizes the number of selected counties for each year and the total number of observed motorcycles. TABLE 2 Observational Survey of Motorcyclists, Florida ( ) Average Selected number of counties Selected number of intersections Total number of observed motorcycles Average number of observed motorcycles per hour per site Percentage of helmeted motorcyclists (operator/ passenger) 5,959 5,967 9,613 7, (55% / 47%) (49% / 52%) (48% / 46%) (51% / 48%) As it can be seen in Table 3, on average, cruisers (43.4%) were the most frequently observed motorcycle type in Florida, followed by tourings (22.1%) and sport bikes (20.0%). For the analysis, mopeds/scooters were excluded as they are reported as a separate vehicle category than motorcycle in Florida police crash reports. Trikes and on/off road motorcycles were included in the other category as they represent a small percentage of motorcycles on Florida roadways.

6 Lee at al TABLE 3 Observed Motorcycle Types (Florida, ) Motorcycle Type Average Percentage Cruiser 52.5% (3,127) 38.9% (2,319) 38.9% (3,736) 43.4% Sport Bike 20.8% (1,238) 19.3% (1,153) 19.9% (1,909) 20.0% Standard 7.5% (445) 14.8% (884) 7.8% (747) 10.0% Touring 16.1% (958) 22.1% (1,321) 28.2% (2,708) 22.1% Other 3.2% (191) 4.9% (290) 5.3% (513) 4.5% Total Number of Motorcycles 5,959 5,967 9, ( ): Number of Motorcycles The effectiveness of DOT (Department of Transportation)-compliant helmets is a commonly discussed subject of motorcycle safety. As of July 2012, 31 states in the U.S. do not have a universal helmet law requiring helmets for all motorcycle riders (16). The state of Florida repealed its universal helmet law in 2000 and a person can operate a motorcycle without a helmet if the individual is over 21 and carries a minimum of $10,000 in health insurance. Table 4 shows the observed use of DOT-compliant helmets by motorcycle type. It appears that the use of DOTcompliant helmets varies significantly by motorcycle type. The three-year average shows that sport bike riders have the highest DOT-compliant helmet use (76.4%) while cruiser riders have the lowest (41.5%). TABLE 4 Observed Use of DOT-compliant Helmets by Motorcycle Type (Florida, ) Motorcycle Type Average Cruiser 45.1% 41.0% 38.5% 41.5% Sport Bike 76.1% 75.0% 78.0% 76.4% Standard 68.8% 52.3% 62.1% 61.1% Touring 57.6% 49.1% 44.5% 50.4% Other 51.2% 54.4% 41.3% 49.0% Note: Includes motorcycle operators and passengers The relationship between motorcycle type and helmet use was explored by conducting the Pearson chi-square test (χ 2 ) for independence to test the hypothesis that they are independent of each other. The results confirm that helmet use has a statistically significant association with motorcycle type (p=0.000, χ 2 = , df = 4). In other words, the percentage of overall helmet use can vary depending upon the proportional distribution of motorcycle types. For example, a higher proportion of sport bikes in the motorcycle population may result in a higher percentage of overall helmet use. MOTORCYCLE TYPES IN FLORIDA CRASH DATA Police crash reports are a primary source of collecting motorcycle crash information for motorcycle safety studies in the U.S. In many states, police crash reports are coded and stored in a crash database. In general, motorcycle type information is not collected on the crash reports. However, the information can be obtained by performing a post processing using vehicle identification numbers (VIN) which are included as required information in most crash reports.

7 Lee at al This study obtained three years ( ) of motorcycle crash data using the Florida Department of Transportation (FDOT) Crash Analysis Reporting (CAR) System. The VIN numbers were decoded to identify motorcycle make and model by a vehicle valuation company. Then, motorcycle makes and models were matched to a motorcycle type. The post processing conducted on the data identified motorcycle types for 22,751 out of 33,237 crash records. Due to incomplete crash reporting, some motorcycle VINs were unavailable and thus no motorcycle type could be matched. Table 5 displays the motorcycle types involved in crashes in Florida from 2007 through On average, sport bikes appeared most frequently in the crash data, followed by cruisers and tourings. It is noted that sport bikes represent nearly 40 percent of crashed motorcycles, while they only represent about 20 percent of motorcycles in Florida based on the observational surveys (see Table 3). TABLE 5 The Percentage of Motorcycle Crashes by Motorcycle Type (Florida, ) Motorcycle Type Average Percentage Cruiser 34.8% (2,618) 37.4% (3,026) 37.0% (2,640) 36.5% Sport Bike 43.0% (3,234) 39.1% (3,164) 37.4% (2,666) 39.8% Standard 4.3% (322) 4.5% (361) 3.9% (278) 4.2% Touring 15.3% (1,149) 16.7% (1,352) 19.2% (1,367) 17.0% Other 2.6% (198) 2.3% (190) 2.6% (186) 2.5% Total 7,521 8,093 7, ( ): Number of motorcycles Figure 2 shows the age of motorcyclists and the percentage of each motorcycle type involved in crashes, as well as the percent of endorsed motorcyclists by age in Florida. As it can be seen, crashed motorcyclists under the age of 45 are disproportionately overrepresented compared to the endorsed motorcyclist population. Sport bikes represent 71 percent of motorcycle crashes involving motorcyclists age 20 and younger. Furthermore, the proportion of sport bikes decreases while the proportion cruiser and touring motorcycles increases as the age of motorcyclists increase.

8 Lee at al FIGURE 2 Crashed Motorcycle Types by Agee and the Proportion of Endorsed Motorcyclists by Age (Florida Crash Dataa , Endorsement Data July 2009) Table 6 shows the percent of motorcyclists without the required endorsement based on crash data. It appears that sport bikes have a relatively higher unendorsed rider population compared to cruiser and touring motorcycles s. TABLE 6 The Percentage of Motorcyclists s without required endorsement in Crash Data Motorcycle Type Age Cruiser Sport Bike Standardd Touring 18.9% 22.6% 21.1% 20.3% 21.8% 25.2% 14.3% 22.2% 21.3% 23.1% 22.9% 10.8% 15.5% 19.9% 18.9% 11.3% 16.5% 18.1% 17.9% 11.2% 12.2% 11.1% 18.8% 9.0% 8.0% 17.6% 8.3% 7.5% 5.1% 8.8% Table 7 shows the observed proportion of motorcycle type on Florida roadways and in the crash data. As can be seen, sport bikes represent 20 percent of observed motorcycles in Florida while they represent 40.1 percent of all motorcycle crashes and 47.6 percent of fatal motorcycle crashes. Therefore, the probability of fatal motorcycle crashes among different motorcycle types can be obtained as follows:

9 Lee at al / /.32 / / / / /.045 is unknown. However, an expected relative frequency by motorcycle type can be calculated as follows:.32 : : : : The calculation shows that sport bikes are over three times more likely to be involved in fatal motorcycle crashes than cruisers in Florida. TABLE 7 Expected Relative Frequency by Motorcycle Type in Fatal Crashes Motorcycle Type Observed All Crashes Fatal Expected Relative Frequency Crashes (Reference to Cruiser) Cruiser 43.4% 36.5% 32.0% 1.00 Sport Bike 20.0% 40.1% 47.6% 3.23 Standard 10.0% 4.2% 3.6% 0.49 Touring 22.1% 17.1% 14.7% 0.90 Other 4.5% 2.1% 2.1% 0.66 Florida police crash reports collect the estimated speed of the motorcycle at the time of the crash and the post speed limit on the roadway where the crash occurred. Table 8 shows the average of the speed difference (miles per hour) by motorcycle type and by age in fatal motorcycle crashes. The speed difference was calculated by subtracting estimated speed from post speed limit. The average speed of sport bikes involved in fatal crashes was over 20 mph above the posted speed limit for all age groups under 45. TABLE 8 Average Speed Difference (mph) by Motorcycle Type and Age in Fatal Motorcycle Crashes Motorcycle Type Age Cruiser (241) Sport Bike (390) N/A 10.0 N/A Standard (36) N/A N/A N/A Touring (127) N/A N/A: The age cells in which there were less than 5 samples were excluded from calculation.

10 Lee at al Table 9 shows the percent of motorcyclists judged under the influence of alcohol by motorcycle type and by age in fatal motorcycle crashes. It shows a high proportion of fatally injured riders on cruiser and touring motorcycles were under the influence of alcohol, especially for riders in the age group of 35 through 54. TABLE 9 Percentage of Riders Under the Influence of Alcohol in Fatal Motorcycle Crashes Motorcycle Type Age Cruiser (241) 33.3% 27.3% 19.0% 33.3% 39.5% 25.0% 8.3% 25.0% Sport Bike (390) 16.0% 22.6% 29.0% 19.8% 21.1% N/A N/A N/A Standard (36) N/A 12.5% 25.0% 16.7% N/A 33.3% 25.0% N/A Touring (127) N/A N/A N/A 35.7% 33.3% 24.4% 7.7% N/A N/A: The age cells in which there were less than 5 samples were excluded from calculation. The above descriptive statistical analysis of motorcycle crash data with motorcycle types showed that different characteristics exist among motorcycle types. The proportion of crashed motorcycle types is different by age group. In addition, contributing factors to fatal crashes seem to be different as the observed frequency of speeding and drinking is quite different by motorcycle type and the age of the motorcyclists. REGRESSION ANALYSIS BY MOTORCYCLE TYPES To examine the relationship between injury severity and motorcycle type, four different regression models were developed. In one model, cruiser, sport bike, and touring were all included as independent variables and cruiser was used as the reference category. In the other three trials, regression models were developed specific to each motorcycle type. The same independent variables were included in each model to allow comparisons to be made between the models regarding the capacity of each independent variable in explaining injury severity in a motorcycle crash. A linear multiple regression model adopted injury severity measured on an ordinal scale from 1 to 5 (See the bottom of Table 1 for the scale description) as the dependent variable. Independent variables included in each model are: motorcyclist s age; estimated speed (miles per hour); speed difference (estimated speed posted speed); helmet use (1 = Yes; 0 = No); total number of vehicles involved in the crash; total number of harmful events in the crash (e.g. collision with another motor vehicle in transport, overturned, explosion, etc.); alcohol and drug use; being ejected in a crash (1 = Ejected; 0 = Not ejected); total number of traffic controls on a crash scene (e.g. use of stop sign, yield sign, speed control sign, etc.). Results from each regression model are summarized in Table 10.

11 Lee at al TABLE 10 Regression of Injury Severity by Motorcycle Type Independent variable All Motorcycles Cruiser Sport Bike Touring Constant (p<0.01) (p<0.01) (p<0.01) (p<0.01) Motorcyclist's age (p<0.01) (p<0.01) (p<0.01) (p<0.01) Sport bike (p = 0.887) Touring (p<0.01) Estimated speed (mph) (p<0. 01) (p<0.01) (p<0.01) (p<0.01) Speed difference (p<0.01) (p = 0.645) (p<0.05) (p<0.05) Helmet use (p<0.01) (p = 0.24) (p<0.01) (p<0.01) Total number of vehicles in crash (p<0.01) (p = 0.409) (p<0.05) (p=0.112) Total harmful events (p<0.05) (p = 0.079) (p = 0.484) (p=0.12) Alcohol and drug use (p<0.01) (p<0.01) (p<0.01) (p<0.01) Being ejected or not (p<0.01) (p<0.01) (p<0.01) (p<0.01) Total traffic controls (p<0.01) (p<0.01) (p<0.01) (p<0.01) R-squared Dependent variable: Injury severity from 1 to 5. (1: None; 2: Possible; 3: Non-incapacitating; 4: Incapacitating; 5: Fatal From all regression models, the following relationships between injury severity and the independent variables were found. Motorcyclist s age, estimated speed (mph), speed difference, total harmful events, alcohol and drug use, being ejected in crash, and total traffic controls are positively related to injury severity as the dependent variable. Only helmet use and the total number of vehicles involved in the crash are negatively related to injury severity. The following independent variables are consistently significant across the four models, all with p-values smaller than These variables are: motorcyclist s age, estimated speed, alcohol and drug use, being ejected or not, and total number of traffic controls. These findings confirmed the results of previous studies (17, 18). It is noted that in the inclusive motorcycle type model, sport bike as an independent variable does not achieve statistical significance in explaining injury severity. Based on the results in this model, sport bikes should be ignored in the analysis of injury severity. However, in the model specific to sport bikes, independent variables achieving statistical significance outnumber those in the cruiser model. Also, the parameter estimates with statistical significance from the sport bike model are very close in value to those from the inclusive model.

12 Lee at al All four models show that alcohol/drug use and being ejected in the crash have the greatest impact on injury severity. Not only do they achieve statistical significance across all four models (p-value < 0.01), but their estimated regression parameters are also much larger in magnitude than the other independent variables achieving statistical significance. In fact, the relatively small parameter values of these latter independent variables should be borne in mind in the interpretation of their impact on injury severity, in spite of their statistical significance in the regression models. SUMMARY OF FINDINGS 1. Helmet use is highly associated with motorcycle type. More than 75 percent of sport bike riders wear a helmet, while around 50 percent or less of cruiser and touring riders wear a helmet. This trend has been observed in all three years of the observational survey. In Florida, motorcyclists under the age of 21 are required to wear a helmet and this age group has a strong preference to ride sport bikes over other motorcycle types. This is evidenced by the fact that 71 percent riders age 16 to 20 who were involved in crashes were riding sport bikes. Although this partially explains why helmet use is high among sport bike riders under 21, they only account for less 3 percent of the endorsed rider population in Florida. It seems that the varied usage of helmets by different motorcycle types stems from a difference in culture between riders of different motorcycle types. 2. The composition of motorcycle type in the motorcycle crash data is different by age. The majority of riders age 34 and younger who were involved in crashes rode sport bikes. Cruiser and touring rider involvement in crashes increased to the majority as the age of crashed riders increased. The underlying cause for this finding is most likely the differences in the types of motorcycles that are popular among different age groups. In a 2011 survey of Florida motorcyclists, the majority of respondents age 34 and younger (57%) reported the motorcycle type they ride most often is a sport bike. In contrast, 38 percent of respondents 35 and older indicated they ride cruisers and 40 percent indicated they ride touring motorcycles most often (11). 3. Young motorcyclists as well as sport bikes are overrepresented in motorcycle crashes. Young motorcyclists over representation in crash data has been a major concern of federal and state highway safety offices for several years. Young riders prefer sport bikes over other motorcycle types, whereas older riders prefer cruiser and touring motorcycles. This fact, combined with the finding from the simple probability estimation that sport bikes are three times more likely to be involved in a fatal crash compared to cruiser motorcycles, provides some clarity as to why young motorcyclists are overrepresented. In addition, sport bikes may be overrepresented as risk seekers may choose to ride more powerful and performance-oriented motorcycles, such as sport bikes (11). 4. Fatally injured sport bike riders were more likely to have been speeding whereas cruiser and touring riders were more likely to be under the influence of alcohol. The study s descriptive statistical analysis uncovered differences in the frequency of contributing factors in fatal crashes by motorcycle type. The majority of fatally injured sport bike riders had estimated crash speeds 20 mph above the posted speed limit. Their speeds were significantly higher than both cruiser and touring riders, who had estimated crash speeds below, at, or slightly

13 Lee at al above the posted speed limit. While cruiser and touring riders generally crashed at the posted speed limit, more than one third of fatally injured cruiser and touring riders between ages of were reported under the influence of alcohol. In general, the prevalence of alcohol use was much lower among fatally injured sport bike riders. 5. Regression analysis found the same contributing factors increase the level of injury severity in motorcycle crashes regardless of motorcycle type. Motorcyclist s age, estimated speed (mph), speed difference, total harmful events, alcohol and drug use, being ejected in crash, and total traffic controls are positively related to injury severity as the dependent variable. These variables were also statistically significant when separate regression models were developed for each motorcycle type. This implies that the same behaviors, such as speeding and drinking, contribute to increase the level of injury severity regardless of motorcycle type. However, it is interesting to note that the frequency of those risky behaviors is different by age and motorcycle type. CONCLUSION The study recognized that motorcycles are utilized as a means of transportation as well as for recreation in the U.S. As this is the case, different motorcycle types can represent certain characteristics unique to the people that ride them. This study attempted to analyze the use of helmets and motorcycle crashes in the context of motorcycle type. Helmet use by riders of different motorcycle types was investigated by analyzing the result of Florida s annual observational survey of motorcyclists. The use of helmet has a statistically significant association with motorcycle type. According to the crash data analysis, it seems that motorcyclists who are under age 45 are disproportionally overrepresented in motorcycle crashes compared to the endorsed rider population. The majority of people in this age group ride sport bikes. Regression analysis showed that the same contributing factors are related to the level of injury severity regardless of motorcycle types. However, it is also noted that the observed frequencies in motorcycle crashes of those contributing factors such as speeding and drinking are quite different by motorcycle types. Unfortunately, motorcycle type information is not readily available from police crash reports which are a primary data source for many motorcycle safety studies. This study confirms that motorcycle type information can uncover many hidden layers of information for motorcycle safety research and can help to develop more adequate countermeasures to reduce motorcycle crashes and fatalities.

14 Lee at al REFERENCES 1. Traffic Safety Facts: 2010 Data. Publication DOT HS NHTSA, U.S. Department of Transportation, Press Release: CDC Study Finds Annual Cost of Motor Vehicle Crashes Exceeds $99 Billion. Centers for Disease Control and Prevention, Traffic Safety Facts: 2010 Data. Publication DOT HS NHTSA, U.S. Department of Transportation, U.S. Department of Transportation, Federal Highway Administration. Highway Statistics Accessed July 23, U.S. Department of Transportation, Federal Highway Administration. National Household Travel Survey, Accessed July 23, Florida Department of Highway Safety and Motor Vehicles. Recurring/Periodic Reports. Accessed July 27, Kim, K, S. Kim, and E. Yamashita. Alcohol-Impaired Motorcycle Crashes in Hawaii, 1986 to In Transportation Research Record: Journal of the Transportation Research Board, No. 1734, Transportation Research Board of the National Academies, Washington, D.C., 2000, pp Lin, M.R., and J.F. Kraus. A Review of Risk Factors and Patterns of Motorcycle Injuries. Journal of Accident Analysis and Prevention, Vol. 49, 2009, pp Jones, S.L, and S. Gurupackiam. Who What When Where of Motorcycle Crashes in Alabama. Presented at 91 st Annual Meeting of the Transportation Research Board, Washington, D.C., Turner, P.A., and N. Georggi. Analysis of Alcohol-Related Motorcycle Crashes in Florida and Recommended Countermeasures. In Transportation Research Record: Journal of the Transportation Research Board, No. 3588, Transportation Research Board of the National Academies, Washington, D.C., 2011, pp Geedipally, S., P.A. Turner, and S. Patil. An Analysis of Motorcycle Crashes in Texas Using a Multinomial Logit Model.. In Transportation Research Record: Journal of the Transportation Research Board, No. 1779, Transportation Research Board of the National Academies, Washington, D.C., 2001, pp Lee, C., and J. Pino. Florida Motorcyclist Survey, Project Number MC Florida Department of Transportation.

15 Lee at al Federal Register. Reclassification of Motorcycles (Two and Three Wheeled Vehicles) in the Guide to Reporting Highway Statistics. Accessed July 24, U.S. Department of Transportation, National Highway Traffic Safety Administration. Background: Motorcycle Types and Characteristics. motorcycle/motorcycle39-41.html. Accessed July 2, Teoh, E.R. and M. Campbell. Role of Motorcycle Type in Fatal Motorcycle Crashes. Journal of Safety Research, Vol. 41, 2010, pp Insurance Institute for Highway Safety. Special Issue: Motorcycle Status Report. Vol. 42, No. 9, Governors Highway Safety Association. Helmet Laws. Accessed July 2, Schneider, W.H., and P.T. Savolainen. Comparison of Severity of Motorcyclist Injury by Crash Types. In Transportation Research Record: Journal of the Transportation Research Board, No. 2019, Transportation Research Board of the National Academies, Washington, D.C., 2011, pp Shankar, V., and F. Mannering. An Exploratory Multinomial Logit Analysis of Single- Vehicle Motorcycle Accident Severity. Journal of Safety Research, Vol. 27, No. 3, 1996, pp