Running Head: Factors Affecting Highway Safety

Silent Killers: Non-Conventional Factors That Affect Highway Safety

Khris Downey

Harding University

Abstract At some point traffic safety will affect everyone. During his or her lifetime each individual will be a driver, passenger, or pedestrian. Modern day society is one dominated by the automobile. Motor vehicle accidents are the number one cause of death for people ages 1-34 (U.S. Department, 2002). In the year 2000 there were 41,611 traffic fatalities in the United States (U.S. Department, 2000) or approximately 2 fatalities every 20 minutes. During the time necessary to read this chapter two or three individuals will die in traffic related accidents.

Research Design

Over the past several decades, motor vehicle transportation has become the most popular mode of transportation in the United States. For the estimated 282,124,631 (U.S. Census Bureau) people living in the United States there are 221, 475, 173 registered vehicles (National Highway), that is 78.5 registered vehicles per 100 people. It is no wonder traffic safety is a growing concern throughout the United States. Americans are becoming increasingly sensitive to the dangers associated with motor vehicles. Automobile safety features are top considerations for many American automobile buyers. In a recent national survey (Home and Away) 83% of automobile buyers listed safety features as one of their top three considerations when purchasing an automobile. Traffic safety related articles frequently appear in news, journals and periodicals on both the local and national levels. For example, in a West Virginia newspaper the public was reassured as to legislative efforts addressing growing traffic safety problems. In the article, the author outlined a new statewide plan to reduce roadway deaths (Robertson). Another article explored the growing impact young drivers have on traffic safety (Pfieffer). The author noted that young drivers are involved in a greater number of traffic fatalities than all other drivers. Across the United States citizens and government officials alike recognize that traffic safety is a serious issue that merits attention, action, and funding. The licensed driving population is increasing annually. The number of age groups represented is also increasing. Licensed teenage drivers are far more likely to own an automobile today than they were thirty years ago. Life expectancy is on the rise and so is the number of elderly drivers. Drivers are starting at young ages and continuing to drive into their seventies, eighties and even nineties (Marshall). In the United States traffic fatalities have increased in some states and decreased in others (Nelson). Here, the phenomena will be referred to as the traffic safety differential. In this study the conventional and non-conventional factors associated with traffic safety will be analyzed. Conventional factors are those assumed by the majority of the public to be responsible for the traffic safety differential. Alcohol, speed and the number of licensed drivers per 100,000 persons are examples of conventional factors. Non-conventional factors are more subtle and receive less attention than conventional factors, or perhaps no attention at all. However, they could potentially have a significant impact on highway safety. Gas tax, funds disbursed for highway maintenance and miles of rural roads are examples of non-conventional factors. The research objective is to determine whether or not there are non-conventional factors that have an influence on traffic safety that merits study. Much time and money are spent by government agencies and private organizations on efforts to improve traffic safety. Research conclusions from this study may be used to affirm the current focus on conventional factors, or possibly direct attention to non-conventional factors that have a significant impact on traffic safety. Numerous studies have been conducted on conventional factors such as alcohol and speed (Fell, Burling, Jorgenson). Researchers have found that conventional factors, such as alcohol and speed, influence the traffic safety differential between states, but are not totally responsible for this difference. This gap implies that other factors may influence the traffic safety differential as much or more than those conventional factors. However, very little research has been done to identify these factors. In this research that gap will be addressed to provide a more complete understanding of the factors affecting traffic safety. The purpose of this research is to determine whether or not efforts to improve traffic safety are focused on factors that have the greatest impact on traffic safety. In this chapter the research design will be explained. An introduction to the overall research question and thesis will be followed by a description as to how the project will be conducted. The processes used for population identification, sample selection, data collection and statistical analyses will be explained, then secondary research questions and accompanying sub-hypotheses will be introduced. In the section which follows these explanations, terms important to this research will be defined. Since the purpose of this research is not to address every issue that relates to traffic safety, delimitations related to this research will be included in this chapter as well. Finally, this will be followed by a description of the research report organization. The traffic safety differential among states indicates traffic safety improvements in some states and declines in other states. The overall research question is this: what explains highway safety differential? The answer to this question will help identify what factors could in fact improve highway safety. Once such factors are identified, they may be used by states which have lower traffic safety levels to improve traffic safety and ultimately lower the number of traffic fatalities. Given the existing traffic safety differential, current efforts to improve overall traffic safety are not operating at their full potentials. The conventional factors addressed in current studies do not represent all factors which have a potentially significant impact on traffic safety. The overall thesis of this research is, that the non-conventional factors have the same impact on the highway safety differential and the number of traffic fatalities as the conventional factors. There are factors that have not been the focus of current studies, but they may have a significant impact on traffic safety. The overall dependent variable for this study is traffic fatalities by state. In order to manipulate this dependent variable, states will be divided into two groups. The states where traffic fatalities have increased over time is one population. The states where traffic fatalities have decreased over time is the second population. Data from these two groups will be analyzed to identify factors affecting traffic fatalities. As the thesis contends, if non-conventional factors do effect traffic safety, researchers could use them to improve traffic safety. As mentioned above, the first step in conducting this research is to identify appropriate populations based on the dependent variable, and the dependent variable for this research is traffic fatalities per state. This variable was selected to represent comparative traffic safety among states. The following assumptions are critical to this study. If traffic fatalities in a state increase, traffic safety is presumed to decrease. Conversely, if traffic fatalities in a state decrease, traffic safety is presumed to increase. The populations for this research then are states where traffic fatalities increased and states where traffic fatalities decreased. To identify these populations, data on percent change in traffic fatalities were collected for each state from 1995 to 2000. These data from the six-year period were averaged. The calculated data yielded either a percent increase or a percent decrease over time. A six year period was used instead of a one year period to avoid the impact of a deviant year which could yield unusual data if it had been affected by a specific event such as a major bus crash where many lives were lost in one accident. States were then divided into two populations. Population one was states where traffic fatalities increased and population two was states where traffic fatalities decreased. The entire populations will not be used to test hypotheses related to this research project. A computerized randomization program was used to select a random sample from each population. Each random sample included ten states. The actual samples will be mentioned in greater detail in chapter 3. Data collected for this research project will be used to test sub-hypotheses directly related to the thesis. The data needs to be relevant. Therefore a clear connection to the main research objective to determine factors that affect the traffic safety differential will be established. Research data were collected from the National Highway Traffic Safety Administration, U.S. Department of Transportation and Federal Highway Administration. These sources were used because they yielded data on the non-conventional factors related to traffic safety. The information provided by these sources was available by state, year, traffic safety topic or factor. These sources provided accessible reliable data in these categories directly related to this research. A majority of the data was collected for the six-year period 1995-2000. The data was averaged to insure that it represents the same period covered by the populations and samples. Several statistical analysis methods will be used to analyze the data. The research data are either nominal or interval. For nominal data the Chi Square statistical analysis will be used, and for interval data, t-test statistical analyses will be used. The independent variables in this research were selected based on the criteria that each had a direct connection to traffic safety or a potential effect on the traffic safety differential. A separate secondary research question and sub-hypothesis are presented for each independent variable. Some independent variables are associated with conventional factors that could explain the traffic safety differential, while others are associated with non-conventional factors potentially connected to the traffic safety differential. Both categories are represented to provide a comparative analyses of factors which could effect the traffic safety differential and determine if efforts to improve traffic safety are focused on the factors which may have a significant impact on traffic safety. An analysis of conventional variables is necessary to fully understand the relative significance of non-conventional variables. Variables associated with conventional factors that could explain the traffic safety differential will now be introduced beginning with alcohol, the most well known conventional variable. Alcohol is the most popular conventional factor used to explain the traffic safety differential. The independent variable in this research associated with alcohol is percent alcohol related traffic fatalities. The secondary research question is this: is the percent alcohol related traffic fatalities greater in states where traffic fatalities increased than in states where traffic fatalities decreased? The assumption is that alcohol related fatalities be higher in states where traffic fatalities increased. It is documented that alcohol slows reaction time and decreases awareness. When driving, either of these conditions could result in failure to safely operate a motor vehicle. This could easily lead to increased fatalities. Could enforcement or legislation, both non-conventional factors impact the number of alcohol related fatalities? There could be less obvious non-conventional factors that help explain the number of alcohol related traffic fatalities and also help explain the traffic safety differential. Speed is another conventional factor frequently connected to the traffic safety differential. The independent variable related to speed is percent speeding related traffic fatalities. The secondary research question is: Do speeding related traffic fatalities occur more often in states where traffic fatalities increased than in states where traffic fatalities decreased? The assumption is that speeding related traffic fatalities occur more often in states where traffic fatalities increased. Increased speeds make it much easier for a driver to lose control of the vehicle. Also, as speed increases so does injury severity in the event of a crash (White). However, there are other factors which explain speeding related traffic fatalities. Enforcement, funding and legislation are all non-conventional somewhat hidden factors that could have a significant affect on the number of speeding related traffic fatalities and subsequently the traffic safety differential. Many think the traffic safety differential can be explained by numbers alone. When there is a greater population there is a greater chance for traffic fatalities. Population has become a conventional factor. The relevant independent variable is millions population. The secondary research question is: “Are populations greater in states where traffic fatalities increased than in states where traffic fatalities decreased? The assumption here is that populations are greater in states where traffic fatalities increased. It is logical to assume that if there are several million more people in an area, then there will be more drivers, more cars and therefore a greater risk of traffic fatalities. Prevalent thought is that the more drivers there are in an area, the more traffic fatalities there will be. The independent variable is number of licensed drivers per 100,000 persons. The secondary research question is: Are licensed drivers per 100,000 persons greater in states where traffic fatalities increased than in states where traffic fatalities decreased? As with population, it is naturally assumed that if there are more licensed drivers then there is a greater chance for a traffic fatality. If more people are driving there are a greater number of vehicles with which to collide. Since no traffic fatality can occur without motor vehicle involvement, a common factor that could be associated with the traffic safety differential is the number of registered vehicles. The independent variable is registered vehicles per 100,000 population. The secondary research question is: Do states where traffic fatalities increased have more registered vehicles per 100,000 population than states where traffic fatalities decreased? Logically, there should be more registered vehicles in states where traffic fatalities increased. If there are more vehicles on the road then there is a greater chance of one being involved in a traffic fatality. As a vehicle increases its miles traveled, it also increases the risk of becoming involved in a traffic fatality. The independent variable is vehicle miles traveled per 100,000 population. The research question is: Are the vehicle miles traveled per 100,000 greater in states where traffic fatalities increased than in states where traffic fatalities decreased? Greater travel increases the risk of traffic fatalities. Vehicle miles traveled should be higher in states where traffic fatalities increased. Safety belts are designed to save lives when a motor vehicle crash occurs. Safety belt usage is a conventional factor associated with the traffic safety differential. The independent variable is percent safety belt usage. The secondary research question is: Is the percent safety belt usage higher in states where traffic fatalities decreased than in states where traffic fatalities increased? Safety belt usage should be higher in states where traffic fatalites decreased. However, safety belt usage does not necessarily explain the traffic safety differential. There are many other factors that help explain the differential. Safety belt usage might not be as significant as it is perceived to be. In many recent studies, the focus has been on driver age. Special attention has been given to young drivers and elderly drivers. In the past thirty years the impact of these two groups on traffic safety has increased dramatically. The independent variable is percent drivers 19 and younger. The secondary research question is: Is the percent drivers 19 and younger greater in states where traffic fatalities increased than in states where traffic fatalities decreased? Younger drivers are presumed to be inexperienced and reckeless. These characteristics put them at greater risk for traffic fatalities. The percent drivers 19 and younger should be greater in states where traffic fatalities increased. However, it is important to note that there are non-conventional factors that could influence the number of young drivers involved in fatalities. For example, driver education funding, curriculum and instruction. The conventional factor associated with traffic safety is not always the only explanatory factor. Another independent variable associated with driver age is percent drivers 75 and older. The secondary research question is: Is the percent drivers 75 and older greater in states where traffic fatalities increased than in states where traffic fatalities decreased? Older drivers are more prone to infirmities which could impair driving. The percent drivers 75 and older should be greater in states where traffic fatalities increased. Non-conventional factors are often found beneath the surface in such areas as legislation and appropriations. For example, the motor vehicle gas taxes. Revenues from gas taxes go toward highway maintenance and improvements. The independent variable is cents per gallon gas tax. The secondary research question is: Is the cents per gallon gas tax greater in states where traffic fatalities decreased than in states where traffic fatalities increased? Since the gas taxes go towards maintenance and improvements, states with a greater gas tax should have safer, better maintained roads. Therefore, the cents per gallon gas tax should be greater in states where traffic fatalities decreased. Highway maintenance is an important traffic safety consideration. The assumption could be made that more funds disbursed for maintenance would result in safer roads. Therefore, states where traffic fatalities decreased should have more funds disbursed for highway maintenance. The independent variable is percent highway funds disbursed for maintenance of roads. The secondary research question is: Is the percent highway funds disbursed for maintenance of roads greater in states where traffic fatalities decreased than in states where traffic fatalities increased? It takes funds to police the roadways. More funds would allow for more officers to monitor traffic safety and therefore could reduce drunk driving, speeding, reckless driving and overall traffic fatalities. This could potentially be more important than factors such as speeding and alcohol. More funds should be disbursed for enforcement in states where traffic fatalities decreased. The independent variable is percent highway funds disbursed for highway police and safety. The secondary research question is: Is the percent highway funds disbursed for highway police and safety greater in states where traffic fatalities decreased than in states where traffic fatalities increased? It is assumed that the funds disbursed are greater in states where traffic fatalities decreased. There are two types of safety belt laws, primary and secondary. A primary seatbelt law authorizes an officer to issue a fine or enforce a penalty on the sole basis of failure to wear a safety belt. A secondary seatbelt law does not allow an officer to issue a fine or enforce a penalty on the sole basis of failure to wear a seatbelt, the driver must have been cited primarily for another violation. A primary seatbelt law is intended to promote greater safety belt compliance. The independent variable is the type of safety belt law. The secondary research question is: Are primary safety belt laws more common in states where traffic fatalities decreased than in states where traffic fatalities increased? It is assumed that states with primary safety belt laws will have higher safety belt usage and therefore fewer traffic fatalities. Primary laws could be the factor that explains seat belt usage and is therefore more important than the variable percent safety belt usage. Primary laws should be more common in states where traffic fatalities decreased. A lower illegal blood alcohol concentration level is thought by many to have a positive effect on traffic safety. Some states have recently lowered their illegal blood alcohol concentration level from .10 to .08 in efforts to decrease alcohol related traffic accidents. The independent variable is illegal blood alcohol concentration level. The secondary research question is: Is an illegal blood alcohol concentration level of .08 more common in states where traffic fatalities decreased than in states where traffic fatalities increased? Here it will be argued that lowering the illegal blood alcohol concentration level has a significant impact on highway safety. 0.08% Illegal blood alcohol levels should be more common in states where traffic fatalities decreased. Urbanization leads to an increase in miles of urban roads. Urban roads have increased traffic volume and congestion. They also have better lighting, more lanes, wider lanes and are generally better maintained. The miles of urban roadway may have an effect on traffic safety. The independent variable is percent urban public roads. The secondary research question is: Is the percent urban public road greater in states where traffic fatalities increased than in states where traffic fatalities decreased? Overall Urban roads are safer than rural roads, so states with more miles urban roads should have decreased traffic fatalties. Rural roadways have many potential traffic safety hazards. Rural roadways are often two lane divided highways instead of four lane, emergency vehicles take longer to respond to calls in rural areas and rural roadways present more challenging driving conditions such as mountains, curves and wildlife. The independent variable is percent rural lane miles. The secondary research question is: Do states where traffic fatalities increased have more miles rural roads than states where traffic fatalities decreased? Because of the driving hazards associated with rural roads, traffic fatalities probably increased in states with more miles rural roads. Each independent variable and research question have corresponding sub-hypotheses. These sub-hypotheses will be tested by statistical methods to determine the significance of the associated factor. The null sub-hypotheses for the \ independent variables in this research are as follows:
H1 There is no statistically significant difference between the mean percent alcohol related traffic fatalities in states where traffic fatalities increased and the mean percent alcohol related traffic fatalities in states where traffic fatalities decreased.

H2 There is no statistically significant difference between the mean percent speeding related traffic fatalities in states where traffic fatalities increased and the mean speeding related traffic fatalities in states where traffic fatalities decreased.

H3 There is no statistically significant difference between the mean population in millions of states where traffic fatalities increased and the mean population in millions of states where traffic fatalities decreased.

H4 There is no statistically significant difference between the mean licensed drivers per 100,000 population in states where traffic fatalities increased and the mean licensed drivers per 100,000 population in states where traffic fatalities decreased.

H5 There is no statistically significant difference between the mean registered vehicles per 100,000 population in states where traffic fatalities increased and the mean registered vehicles per 100,000 population in states where traffic fatalities decreased.

H6 There is no statistically significant difference between the mean vehicle miles traveled per 100,000 population in states where traffic fatalities increased and the mean vehicle miles traveled per 100,000 population in states where traffic fatalities decreased.

H7 There is no statistically significant difference between the mean percent safety belt usage in states where traffic fatalities increased and the mean percent safety belt usage in states where traffic fatalities decreased.

H8 There is no statistically significant difference between the mean percent drivers 19 years and younger in states where traffic fatalities increased and the mean percent drivers 19 years and younger in states where traffic fatalities decreased.

H9 There is no statistically significant difference between the mean percent drivers 75 years and older in states where traffic fatalities increased and the mean percent drivers 75 years and older in states where traffic fatalities decreased.

H10 There is no statistically significant difference between the mean cents per gallon state motor vehicle gas tax rate in states where traffic fatalities increased and the mean cents per gallon motor vehicle gas tax rate in states where traffic fatalities decreased.

H11 There is no statistically significant difference between the mean percent highway funds disbursed for maintenance of roads in states where traffic fatalities increased and the mean percent highway funds disbursed for maintenance of roads in states where traffic fatalities decreased.

H12 There is no statistically significant difference between the mean percent highway funds disbursed for highway police and safety in states where traffic fatalities increased and the mean percent highway funds disbursed for highway police and safety in states where traffic fatalities decreased.

H13 There is no statistically significant difference between the frequency of primary seatbelt laws in states where traffic fatalities increased and the frequency of primary seatbelt laws in states where traffic fatalities decreased.

H14 There is no statistically significant difference between the frequency of .08 illegal per se blood alcohol concentration limits in states where traffic fatalities increased and the frequency of .08 illegal per se blood alcohol concentration limits in states where traffic fatalities decreased.

H15 There is no statistically significant difference between the mean percent urban public road in states where traffic fatalities increased and the mean percent urban public road in states where traffic fatalities decreased.

H16 There is no statistically significant difference between the mean percent rural lane miles in states where traffic fatalities increased and the mean percent rural lane miles in states where traffic fatalities decreased.

Employed throughout this research is various terminology unique to traffic safety. This terminology is introduced and defined below. 1. BAC is an acronym for Blood Alcohol Concentration.

2. Fa-D is the symbol used in tables to represent states where traffic fatalities decreased.

3. Fa-I is the symbol used in tables to represent states where traffic fatalities increased.

4. FHWA is an acronym for the Federal Highway Administration of the United States.

5. MADD is an acronym for Mothers Against Drunk Driving, a citizens action group founded by mothers who have lost children in alcohol related traffic fatalities.

6. NHTSA Is an acronym for the National Highway Traffic Safety Administration.

7. SADD is an acronym for Students Against Drunk Driving, a student organized citizen action group.

8. Traffic safety differential - The variance in traffic safety among states with regard to traffic fatalities. Between the year 1995 and the year 2000 traffic fatalities increased in some states and decreased in others.

9. USDOT is an acronym for the United States Department of Transportation. This research project is intended to serve a specific purpose by assessing the effects of some popular conventional factors on traffic safety as contrasted with specific non-conventional factors which affect traffic safety. It is not intended to address every issue relating to traffic safety. Included in the research are enough independent variables to sufficiently answer the research question. The independent variables do not represent every factor which could potentially affect traffic safety. The object of the research is to raise awareness as to the existence and powerful potential effects of non-conventional factors related to traffic safety. All factors affecting traffic safety are not necessary to identify the existence of a few. The most current publications of NHTSA and FHWA are statistics from the year 2000. Data from 2001 has yet to be compiled by the agencies and is not available for use in this report. Therefore data are limited to the year 2000. General non-conventional factors were selected for this research project. There are many specific programs left out of this study, such as driver education courses, which could be used in later comparative studies to learn more about specific non-conventional factors. Subscriptions to several scholarly sources related to highway safety were not available for this study. These sources are filled with studies on highway safety and could have added even more to the review of the literature.

References

Burling, John, Robyn Hagler, Carolyn Miller and Cheri Smith. The Influence of Legislation on the Driving and Alcohol Consumption of College Students. The Chronicle of the American Driver and Traffic Safety Education Association, 44, 1-5.

Fell, James and Delmas Johnson. The Impact of Lowering The Illegal BAC Limit to .08 In 5 States In The U.S. Proceedings from the 13th International Conference on Alcohol, Drugs and Traffic Safety. Adelaide, Australia, 13, 1-6.

Home and Away (AAA). (2001). Car Corner: Auto Safety Sells. Retrieved August 14, 2001 from {http://www.aaa.com).

Jorgenson, Darren L., Karlaftis, Matthew G., Sinha, Kumares C. (2001) Vehicle Speed Considerations in Traffic Management: Development of a New Speed Monitoring Program. Journal of Transportation and Statistics, 42, 69-81.

Marshall, S., Srickland, R. (2001). Age Expectancy of Liscensed Drivers On The Rise. Journal of Transportation and Statistics. 42, 188-194

National Highway Traffic Safety Administration (2000), Traffic Safety Facts. Retrieved January 17, 2002 from {http://www.nhtsadmin.org}

Nelson, P. (2001). Incidence of Traffic Related Deaths in the United States. The Chronicle of the American Driver and Traffic Safety Education Association. 49, 17-21

Pfeiffer, M.B.(1999). The Deadliest Drivers of All. Retrieved September 5, 2001, from [www.USAWeekend.com]

Robertson, G. (2001, September 8). State Plans New Campaign To Reduce High Road Death Rate. The Dominion Post, pp. C1, C7

United States Department of Transportation Statistics. (2002). Retrieved May 12, 2002, from [http://safety.fhwa.dot.gov/facts_data/facts_data.htm].

U.S. Department of Transportation Highway Statistics (2000). Table IN-6.
U.S. Census Bureau, Table US-2001EST-01-Time Series of National Population Estimates: April 1, 2001 - July 1, 2001.

White, George L. and Martin J. Spellicy. (2000). The Physics of Motor Vehicle Injury The Chronicle of the American Driver and Traffic Safety Education Association. 48, 1-3.