NATIONAL SURVEY OF SPEEDING
AND OTHER UNSAFE DRIVING ACTIONS

VOLUME II: DRIVER ATTITUDES & BEHAVIOR

Chapters 1 & 2

EXECUTIVE SUMMARY

BACKGROUND

Speeding has been cited as a contributing factor in nearly one-third of all fatal motor vehicle crashes. In 1996, the cost of crashes involving speeding was estimated to be $28.8 billion. However, only limited information is available on driver attitudes and behavior regarding speeding and other forms of unsafe driving behavior, including those typically identified as aggressive driving, e.g., tailgating, weaving, running red lights, and making angry, insulting, or obscene gestures to other drivers. To help provide information in this important area, the National Highway Traffic Safety Administration (NHTSA) commissioned a national survey of the driving public to determine:

• the wide range of driver attitudes about speeding and other forms of aggressive/unsafe driving behavior;
• commonly occurring situations in which unsafe driving occurs;
• driver characteristics associated with those who commit these types of infractions; and
• the types of countermeasures the public believes are acceptable and effective for countering such behaviors.

Research of this nature supports NHTSA-sponsored efforts to more precisely specify targets (e.g., drivers, situations), and develop new or refine existing countermeasures that, ultimately, may reduce the occurrence of fatalities and injuries resulting from unsafe driving practices. (See Volume III: Countermeasures, for more detailed information about possible solutions.)

The survey was conducted by telephone by the national survey research organization, Schulman, Ronca and Bucuvalas, Inc. (SRBI). A national household sample was constructed using random digit dialing. Each household was screened to determine the number of adult (16 years of age or older) drivers in the household and one eligible driver was selected in each household to be interviewed for the survey. The interviews were conducted by professional interviewers, using computer-assisted telephone interviewing (CATI) to reduce interview length and minimize recording errors. A Spanish-language translation and bilingual interviewers were used to minimize language barriers to participation. The interviews, conducted between February 20 and April 11, 1997, averaged 30 minutes in length. A total of 6,000 interviews were completed with a participation rate of 73.5%. (For a detailed discussion of the methodology employed in this study, refer to Volume I: Methodology Report.)

Since this was the first national survey of speeding and unsafe driving practices, the number of issues to be covered was extensive. In order to accommodate the number of questions required without unduly burdening the public, two versions of the questionnaire were developed. One questionnaire focused primarily on speeding issues and the other focused primarily on other forms of unsafe driving. Each version is an independent national sample, constructed in an identical fashion. In addition, each version of the questionnaire used half-samples for some questions to extend the number of questions that could be covered in a 30 minute interview. This random assignment of questions to half of the sample within the two national cross-sectional samples effectively created four national samples. Hence, for some questions, we have national estimates based on sample sizes of about 1,500 or 3,000, while estimates for core questions about speeding and unsafe driving, as well as driver and driving characteristics shared by both versions are based on sample sizes of 6,000.

FINDINGS

The majority of drivers in the United States consider speeding and other forms of unsafe driving to be a major threat to the personal safety of themselves and their families. More than six out of 10 drivers (61%) say that speeding by other people is a major threat to personal safety of themselves and their families. Two-thirds (66%) of drivers say that other drivers' unsafe driving actions (other than speeding) on the roads they drive is a major threat to themselves and their families.

The threat of unsafe driving is real, rather than hypothetical for many drivers. More than six out of 10 drivers (62%) report that the behavior of another driver has been a threat to them or their passengers within the past year. Those who have felt threatened by the behavior of other drivers in the past year were asked the nature of the action they found threatening. Most commonly, these drivers reported that, "another driver had cut very closely in front of me" (36%), "drove very closely behind me" (19%), "passed me in a dangerous manner" (15%), "cut me off at an intersection or exit" (13%), "made an obscene or threatening gesture" (5%), "wove in and out of traffic" (4%), "ran a red light" (3%) or "ignored a stop sign" (1%).

Consequently, the majority of drivers (52%) believe that it is very important that something be done to reduce speeding, while another 41% feel that it is somewhat important that something be done. Even more dramatically, 75% of drivers feel that it is important to do something about unsafe driving, while another 23% feel that it is somewhat important to do something.

There also was some concern that the dangers of driving, including aggressive driving, were increasing. Although most drivers (54%) felt that driving was neither more dangerous nor safer than a year ago, one-third of drivers (33%) reported that they feel driving is more dangerous now than it was a year ago. Some of the factors cited by those who feel driving is more dangerous now are: heavier traffic and more cars (33%), careless and inattentive drivers (20%), faster drivers (18%), increased speed limits (16%), aggressive driving (14%), young drivers (10%) and drinking drivers (10%). Regarding aggressive driving, the majority of drivers (65%) reported that other motorists in their area drove no more or less aggressively than they did a year ago. Nonetheless, 30% of drivers felt that other drivers in their area drive more aggressively now, and, of these, 13% believed that they drive a lot more aggressively now as compared to a year ago.

Unsafe speeds on our nation's roadways were commonly reported. Over half of drivers reported that they see vehicles traveling at unsafe speeds all or most of the time when they drive on residential streets (53% urban, 53% rural); three-fifths on non-interstate highways (59% urban, 55% rural); and two-thirds on interstate highways (70% urban, 67% rural). When asked what other (than speeding) types of unsafe driving they usually encounter on the roads they regularly drive, drivers most commonly reported: cars weaving in and out of traffic (24%), tailgating (17%), driver inattention (15%), unsafe lane changes (10%), unsafe passing (9%), ignoring stop signs (8%), failing to yield (6%), drinking and driving (5%), and running red lights (5%). Only 16% of all drivers reported that they usually did not see unsafe driving on the roads they drove. A majority of those who reported seeing unsafe driving on the roads they regularly travel also reported that all or most of those who were doing these unsafe actions were also speeding.

This national sample of drivers was also asked about their personal driving behaviors over the past year. Drivers were asked when was the most recent time in the past year they had personally committed certain types of driving actions, which would be classified as unsafe driving by many traffic safety experts. Three out of 10 drivers (30%) reported entering an intersection just as the light was turning red within the past week. A quarter (26%) reported slowing but not completely stopping at a stop sign in the past week. More than one in five reported that they have driven 10 miles per hour over the speed limit on an interstate in the past week (23%) or have driven 10 miles per hour than most other vehicles were going (22%) in the past week. How recently 17 other unsafe driving behaviors was also reported by the survey. The least commonly reported types of unsafe driving in the past year were driving when affected by alcohol (8%) and racing another driver (6%).

The survey confirms that age and gender are two important factors associated with of unsafe driving. Men are more likely than women to report committing all 21 types of unsafe behaviors examined in the survey. There was a 2-to-1 difference between the genders in past year performance of some unsafe driving behaviors (e.g., driving when affected by alcohol), while for others (e.g., entering an intersection on a red light) the difference was marginal. Age is an even more striking factor for unsafe driving. The proportion of drivers who engage in virtually all of the unsafe driving actions examined in the survey declines as age increases. The survey findings suggest that unsafe driving declines on a continuous basis as drivers age, rather than after a specific age.

The decline in unsafe driving as age increases is mirrored in the perceived dangerousness of these driving actions. In one striking example, when rating the safety of driving 10 miles an hour over the speed limit on two-lane rural roads, the average rating of the dangerousness of that act increases from 3.4⁽¹⁾ for drivers aged 16-20, to 3.5 for 21-24 and 25-34 year-olds, to 3.7 for 35-44 year-olds, to 3.8 for 45-54 year-olds, to 4.0 for 55-64 year-olds, to 4.2 for drivers aged 65 and older. Most strikingly, is that 16 to 30 year olds rated nearly every unsafe behavior as less dangerous than older drivers: e.g., of 22 behaviors examined the 16 to 20 year olds rated 17 of these as less dangerous than any other age group. Hence, it appears that as drivers age, there is an ongoing reevaluation of the dangerousness of these driving actions.

A total score was calculated for each driver based on the reported frequency of the 8 to 9 unsafe driving acts about which that driver had been asked. The average unsafe driving score was nearly 40 percent higher for male drivers (92) than female drivers (66). The mean unsafe driving score falls nearly fourfold between drivers aged 16-20 (150) to those aged 65 and over (37). One of the most striking differences reported in unsafe driving appears to be related to geography. The average unsafe driving score is nearly twice as high for drivers from New England, compared to drivers from the Pacific Northwest and the Mountain states.

Enforcement of traffic safety laws is the primary countermeasure for speeding and other forms of unsafe driving. About one in seven drivers (14%) report that they have been stopped by police for traffic-related reasons in the past year. Most commonly, these drivers report that they have been stopped for speeding (64%), followed by stop signal violations (8%), no lights (8%) and stop sign violations (7%). Over half of those stopped by police in the past year (57%) report that they received a ticket, while most of the rest (34%) received a warning on the most recent occasion.

A relationship was found between reported unsafe driving behavior and being stopped by the police. The likelihood of being stopped by police in the past year increased from 5% of those with no reported unsafe driving, to 38% of those with the highest unsafe driving scores. The good news from the survey is that the likelihood of being stopped by the police for safety violations increased with the frequency of those violations. The bad news is that a majority of those who drive in an unsafe and illegal manner -- 62% of those with the highest unsafe driving scores -- were never stopped by the police.

Nonetheless, the survey finds that a majority of drivers are satisfied with the current amount of police enforcement of speeding laws on the types of roads they normally drive -- from 52% who feel it is about right on residential streets, to 67%, who feel it is about right on non-interstate highways in predominantly urban areas. About four in 10 drivers (38%) say that they see police every day or nearly every day on the roads they drive most frequently. Three-quarters see police once a week or more often when driving.

Less than 1% of drivers felt that the police issued tickets whenever they saw a driver just exceed the speed limit. A small number, 4%, felt drivers could go no more than 4 miles above the limit before a ticket would be issued. About two-thirds of all drivers felt the police would issue a ticket if they saw someone going 5 miles per hour above the limit. This drops to about one in five who feel tickets are not issued until police see someone going more than 10 miles per hour above the limit.

A majority of drivers are satisfied with the current amount of police enforcement of traffic laws related to running red lights (52%), failure to stop at signs (53%), failure to yield (52%) and speeding (50%). However, a majority of drivers feel that there is too little police enforcement of traffic laws in the areas of tailgating (61%) and weaving (58%). Moreover, a very substantial minority of drivers, ranging from 40% to 44% believe that there is too little police enforcement of traffic laws in the other areas.

CHAPTER I.
BACKGROUND AND OBJECTIVES

BACKGROUND

Speeding has been implicated as a contributing factor in about one-third of all fatal motor-vehicle crashes. In addition, increased attention has been given to other unsafe driving behavior -- running red lights, tailgating, cutting other drivers off, etc. -- which may lead to crashes. However, very little information is available on when, where, and under what conditions drivers engage in speeding and other unsafe driving actions or behaviors; nor is there adequate information on the types of drivers who engage in these behaviors.

To help fill this information gap, the National Highway Traffic Safety Administration (NHTSA) of the U.S. Department of Transportation (DOT) contracted with Schulman, Ronca, & Bucuvalas, Inc., a national survey research firm, to conduct a survey of the driving public's attitudes and experience related to speeding and other unsafe driving actions. Research of this nature supports NHTSA-sponsored efforts to more precisely specify targets (e.g., drivers, situations), and develop new or refine existing countermeasures that, ultimately, may reduce the occurrence of fatalities and injuries resulting from unsafe driving practices.

The unsafe driving behaviors examined in the survey, including tailgating, weaving, and making obscene gestures to other drivers, are sometimes used as examples of "aggressive driving." There is increased public concern about the role of aggressive driving and "road rage" in crashes and traffic fatalities. Unfortunately, there is no general agreement among traffic safety experts as to what constitutes aggressive driving. Consequently, the survey focuses more on specific unsafe driving acts rather than on aggressive driving.

That the American public is very concerned about the consequences of speeding and other unsafe driving actions, can be seen from the results of NHTSA's 1997 Customer Satisfaction Survey where 87% of the driving age public said it was important that something be done to reduce speeding on highways and fully 97% said it was important to do something about speeding on residential streets.⁽²⁾ In the earlier 1995 Customer Satisfaction Survey, 90% said it was important for the federal government to conduct public education campaigns to increase compliance with stop signs and signals.⁽³⁾ The 1997 Customer survey also showed that the public believes the problem of unsafe driving is becoming worse -- 60% of the driving-age public said they believe drivers were driving less safely now than 10 years ago, compared with only 8% who thought drivers are driving more safely now.

OBJECTIVES

The specific objectives of this survey were to determine:

The first three objectives are the focus of this report. The fourth objective is discussed in the Volume III: Countermeasures report. Overall, the survey provides a status report on public attitudes and behavior related to speeding, aggressive and unsafe driving behaviors as well as provides information that can aid in the development of appropriate countermeasure activity.

SAMPLE DESIGN

The survey was conducted by telephone by the national survey research organization of Schulman, Ronca & Bucuvalas, Inc. (SRBI). A national telephone household sample was constructed using random digit dialing. Each household was screened to determine the number of adult (age 16 or older) drivers in the household. One eligible driver was systematically selected in each eligible household by the interviewers. The survey was conducted using computer-assisted telephone interviewing (CATI) to reduce interview length and minimize recording errors. A Spanish-language translation and bilingual interviewers were used to minimize language barriers to participation.

Since this was the first national survey of speeding and unsafe driving practices the number of issues to be covered was extensive. In order to accommodate the number of questions required without unduly burdening the public, two versions of the questionnaire were initially developed. One questionnaire (Version 1) focused primarily on speeding issues. The other questionnaire (Version 2) focused primarily on other forms of unsafe driving. Each version was fielded as an independent national sample, constructed in an identical fashion. Hence, for some questions we have national estimates based on sample sizes of 3,000, while estimates for core questions about speeding and unsafe driving behavior, as well as driver and driving characteristics shared by both versions, are based on sample sizes of 6,000.

Each of the two questionnaire versions used split-half samples for some questions to extend the number of questions that could be covered in a 30 minute interview (see Table 1-1, Below). This random assignment of questions to half of the sample within the two national cross-sectional samples effectively created four national samples. Hence, the total sample size of 6,000 drivers in the survey is comprised of four independent samples of approximately 1,500 respondents, each. Individual questions may be asked of 1,500 drivers (one national sample), 3,000 drivers (two national samples) or all 6,000 drivers.

TABLE 1-1

The survey was conducted between February 20 and April 11, 1997. The telephone interviews averaged 30 minutes in length. A total of 6,000 interviews were completed with a participation rate of 73.5 percent.

The completed interviews were weighted to correct for selection bias as a result of the number of telephone lines and eligible respondents in the household. The complete weighting procedure and other aspects of the survey methodology are described in greater detail in Volume I: Methodology Report. The questionnaires used in this survey also appear in Volume I.

All sample surveys are subject to sampling variability or sampling error. The sampling error is the range within which sample estimates are expected to vary from true population values. At the 95 percent confidence level, the maximum expected sampling error for a simple random sample declines with size from + 2.5 percentage points for a sample of 1,500 (i.e., 47.5%-52.5% for a sample estimate of 50%), to + 1.8 percentage points for a sample of 3,000, to + 1.3 percentage points for a sample of 6,000. The formula for calculating sampling variances and a table of expected sampling errors by sample size is included in Volume I: Methodology Report.

Some percentages in the report are based on the total sample of survey participants (6,000), while others are based on one or two of the independent samples which comprise the total sample. Each table is labeled to show the appropriate, unweighted base. Due to rounding, the percentages in some tables may add to slightly more or less than 100%. We have labeled questions that permit multiple responses because they will add to more than 100%.

CHAPTER II.
DRIVERS, VEHICLES AND ROADS

The population for this survey was limited to drivers, i.e., persons aged 16 and older who drive a motor vehicle at least a few times a year. The basic contextual factors associated with the "driving experience" -- the characteristics of drivers, the vehicles they drive and the types of roads on which they travel -- are presented here. Other contextual factors -- the presence of other adults or children in the car, the need to be some place at a specific time, traffic control device (stop lights, stop signs, school bus stop arms, etc.), time of day, etc. -- likely to affect unsafe driving experiences are discussed in later chapters.

DRIVING CHARACTERISTICS

Almost nine out of 10 (88%) drivers drove almost every day and an additional 9% drove a few days a week (Table 2-1, next page). The remaining handful of drivers drove either a few days a month (2%) or a few days a year (1%).

Driving frequency varied by gender. Where 92% of male drivers reported driving almost every day, only 85% of female drivers reported the same frequency of driving. Conversely, the rates for driving a few days a week were twice as high for women as they were for men -- 14% vs. 6%. Taken together, the frequencies for both men and women who report driving at least a few times a week are almost identical -- 98% for men and 97% for women.

The frequency of driving almost every day also varied by age. Fully 85% of those in the 16 to 20 age group reported driving every day. This increased to 88% of those 21 to 24 years-old, to 93% for those 25 to 34, peaked at 94% for those 35 to 44 and then decreased to 91% for those 45 to 54, 85% for those 55 to 64, and finally 72% for those 65 and over.

Educational attainment also appeared to be related to frequency of driving. Almost four out of five (78%) of those with less than a high school education reported driving almost every day. This increased to 87% of those who graduated high school, to 90% of those with some college and to 92% of those with a college degree.

The frequency of driving almost every day did not vary greatly from the overall proportion of 88% with community type. Urban residents reported driving almost every day at very near the nationwide proportion (87%). Suburban residents reported a slightly higher proportion (90%) and rural residents reported the lowest frequency of driving almost every day (85%).

TABLE 2-1

* Detail excludes non-responses for specific demographic groups.

Two drivers in three (67%) drove with other adults almost every day (19%) or a few days a week (48%) (Figure 2-1). Additionally, one driver in four (23%) drove with other adults a few days a month and less that one in 10 (7%) drove with an adult only a few times a year. Only 3% of drivers reported that they never drive with other adults. The proportion of men and women who drove with other adults at least a few days a week is almost the same -- 68% and 66% respectively.

Slightly more than three drivers in five (63%) age 16 to 20 drove with other adults at least a few days a week. This increased to three in four (76%) of drivers 21 to 24, and slowly decreased for each successive age cohort until it reached 66% for the 55 to 64 age group; it then dropped to 60% for those 65 and older.

FIGURE 2-1

Qx: How often do you drive with children in your [vehicle]?

Base: Vehicle drive most often is not a motorcycle; adults=Group A questionnaire; children=Group B questionnaire.

Unweighted N(adults)=2,952; N(children)=3,039

A very different pattern appeared for driving with children. Overall, slightly less than two in five (38%) drivers drove with children almost every day (20%) or a few days a week (18%). Additionally, one driver in seven (14%) drove with a child only a few times a month and one in four (24%) drove with children only a few times a year. One driver in four (24%) reported never driving with a child. Unlike driving with adults, where there was no difference by gender, there was a large difference between men and women who reported driving with children. Women reported driving with children one-third more than men -- 44% and 33% respectively.

The pattern by age is even more dramatic. Only one in four (25%) of drivers 16 to 20 reported driving with children at least a few days a week. This proportion doubles to 29% for drivers 21 to 24 and almost doubles again to 53% for drivers 25 to 34. Driving with children at least a few days a week peeks at 64% for drivers 35 to 44 before it starts to drop precipitously -- 30% for drivers 45-54, 17% for drivers 55 to 64, and 10% for drivers 65 and over.

VEHICLE CHARACTERISTICS

Drivers were asked to identify their primary vehicle, that is the type of vehicle they drove most often. Fully two-thirds (68%) of drivers drove a car most often (Figure 2-2). The second most often driven vehicle was the pickup truck, which was mentioned by 16% of drivers. This is followed by van or minivan, mentioned by one in 10 (9%), sport utility vehicles, mentioned by one in twenty (5%) and other trucks, mentioned by 2%. Motorcycles and other vehicles were mentioned by a handful of drivers who accounted for less than 0.5%.

FIGURE 2-2

Base: Total population of drivers.

Unweighted N=6,000

This distribution hides major differences that exist between the genders for vehicle most often driven. While cars were mentioned overall by two-thirds of all drivers, they were mentioned by a little more than half (55%) of male drivers, but four out of five (79%) women drivers. Similarly, while pickup trucks were mentioned by 16% of all drivers, they were mentioned by almost twice as many men (28%) and one-third as many women (5%).

One area where a difference did not appear was in the selection of the van/minivan. The minivan has been characterized as the vehicle of choice of the "soccer mom," yet is the vehicle most often driven by 10% of women and 8% of men.

Fully 30% of motor vehicles (other than motorcycles) driven most often were from the last four model years (Figure 2-3). An additional 37% were five to nine years old. Together, two-thirds of the vehicles on the road at the time of the study were less than 10 years-old. Additionally, nine vehicles in 10 (90%) were less than 15 years old. The median model year of all vehicles was 1992.

FIGURE 2-3

Base: Vehicle driven most often is not a motorcycle.

Unweighted N=5,991

Approximately one driver in 14 (7%) had a radar detector in their vehicle (Figure 2-4). Radar detectors were used by almost twice as many males (8%) as females (5%). In addition, they were used by about one driver in 10 (9%) in the 21 to 24 age groups, followed by 8% of drivers in the 25 to 34 age group. Radar detectors were used least by drivers in the 55 to 64 age group (5%) and those drivers age 65 and over (4%).

FIGURE 2-4

Base: Total population of drivers.

Unweighted N=6,000

One driver in eight (12%) in the northeastern states of Connecticut, Maine, Massachusetts, New Hampshire, Rhode Island, and Vermont (NHTSA Region I) had a radar detector in his/her vehicle. This was followed by the south central states of Arkansas, Louisiana, New Mexico, Oklahoma, and Texas (Region VI) with 10% of drivers having a radar detector in their vehicle. States in the Gulf and south Atlantic areas -- Alabama, Florida, Georgia, Kentucky, Mississippi, North Carolina, South Carolina, and Tennessee -- (Region IV), with 8%, was the only other region above the national average.

PRIMARY ROAD TRAVELED CHARACTERISTICS

Slightly more than half (54%) of drivers reported that they most often drove on primarily urban rather than primarily rural roads (Figure 2-5). Conversely, two drivers in five (40%) reported that they most often drive on rural rather than urban roads. The remaining 6% reported that they drove equal time on both.

FIGURE 2-5

Base: Total population of drivers.

Unweighted N=6,000

Drivers who primarily used urban roads increased with their amount of education, going from 44% of those with less than a high school education to 63% of college graduates. The fact that the proportion of drivers who drive most often on urban roads increases as educational attainment increases, says more about the educational levels of the residence of urban and rural areas than it does about the driving habits of these drivers.

Slightly more than five drivers in six (86%) drive at least weekly on residential or neighborhood streets with posted speed limits of 35 miles per hour or less (Figure 2-6). Between 83% and 89% of all major sub-populations analyzed in this study drove on residential or neighborhood streets at least weekly.

FIGURE 2-6

Base: Total population of drivers.

Unweighted N=6,000

As was shown in Figure 2-6 (page 17), fully two-thirds of all drivers drove at least weekly on major roads with posted speed limits between 40 and 55 miles per hour. These roads were used most by residents of rural areas (71%) and least by residents of urban areas (62%).

FIGURE 2-7

Base: Total population of drivers.

Unweighted N=6,000

SUMMARY

Nine out of 10 drivers drove almost every day. Two-thirds of drivers rode with other adults at least a few days a week and nearly two-fifths drove with children at least a few days a week. Cars were the most frequently driven vehicles, outnumbering all other vehicles (trucks, vans or minivans and sport utility vehicles) combined by two to one. These cars were generally newer models, with two-thirds of all cars being no more than eight years old.

1. This series used a 5-point scale where 1 = Extremely Safe and 5 = Extremely Dangerous.

2. U.S. Department of Transportation, National Highway Traffic Safety Administration, 1997 Customer Satisfaction Survey, April 1998.

3. U.S. Department of Transportation, National Highway Traffic Safety Administration, 1995 Customer Satisfaction Survey, May 1996.