1. Introduction
Safe driving entails a complex combination of behaviors in which the driver is required to successfully perform a number of complex activities concurrently [
1]. Such activities include (but are not limited to) route detection and tracking, speed and lateral position controlling, collision avoiding, road laws obeying, and vehicle systems status monitoring [
2]. Drivers are often observed as engaging in activities that will likely divert their attention away from the main crucial activity required for safe driving [
1]. Pashler [
3] pointed out that humans normally have limited capability to carry out more than two tasks in parallel.
The definitions of distracted driving reported in the literature vary. Distracted driving occurs when the recognition of the information required for safe driving is delayed due to something such as an individual, an event, or an activity inside or outside the vehicle, forcing the driver to divert his attention away from the driving task [
4,
5]. Distraction can undermine driving performance and impact driver behavior in different ways and to various degrees. Dingus et al. [
6] maintained that driver distraction and slips of attention have a significant influence on road crashes and other related road safety issues. Distraction reduces the performance of the driver and is an important factor in the occurrence of traffic accidents [
7,
8]. Louw et al. [
9] indicated that over 90% of reported traffic accidents were the outcome of human error and caused by issues related to the acquisition of visual information. Moreover, drivers can be held responsible for causing 90% of critical traffic conditions [
6,
10]. Khattak et al. [
11] indicated that human factors still contribute to 93% of crashes, whereas recognition errors (including distraction) are found to be the driving error contributing to the highest percentage of crashes.
Regan et al. [
12] defined driver inattention as “…insufficient, or no attention, to activities critical for safe driving”. They also listed several forms of driver inattention such as biological factors that block the driver from noticing crucial information for safe driving; misprioritized attention of drivers (i.e., drivers’ attention focused on one side of driving while excluding another); and driver distraction, which is considered as one form of driver inattention where a competing or a secondary task (which can be driving-related or non-driving-related) diverts the attention of the drivers away from crucial safe driving tasks.
Goodsell et al. [
1] conducted a comprehensive review of studies on driver distraction and listed four major moderating factors that might influence the occurrence of driver distraction. The first moderating factor is driver characteristics (age and driving experience). The second moderating factor is the level of demand of the driving task. Weather, traffic, and road conditions were found to influence the demand of driving tasks [
13]. The third moderating factor is the level of demand for attention of the competing activity (i.e., visual–manual vs. auditory–vocal). The fourth moderating factor is the drivers’ ability to change their behavior to keep satisfactory driving performance in the face of competing tasks (also known as self-regulation) [
14].
Shortfalls of driving performance due to engagement in competing tasks have been found to include the following: poor lane keeping and speed control, increase in reaction time, skipping traffic signals, inadequate headway distances, risky gap acceptances, reduction in situational awareness, and visual scanning of the roadway [
15,
16,
17,
18]. Many of these shortfalls in driving performance are correlated with a substantial risk of crashes [
6]. For example, poor speed control, when examined separately from other causal factors such as driver distraction, might increase crash risk by 12.8 times, and short headway distances can increase crash risk by 13.5 times [
6].
Young et al. [
13] indicated that driver distraction might be affected by the demand of the driving task, which in turn may be influenced by factors such as weather, traffic, and road conditions. Generally, more demanding tasks impair driving performance more. Accordingly, it is expected that distractions in different areas and road types may vary.
Driving tasks might be affected by road aggressiveness [
19]. Spanish drivers’ perceptions of aggressive behavior and the most aggressive acts performed when driving were studied by Alonso et al. [
19]. Their results show that both infractions and dangerous behaviors correspond to aggressive driving, which definitely becomes a source of distraction. It has also been found that drivers who tend to have several accidents give less importance to elements that can affect the driving task such as distraction.
Xuedong et al. [
20] indicated that crash rates in rural sections are regularly lower than those in urban ones. They also indicated that rural areas have lower crash risk when compared to urban places. Additionally, it was found that an increase in traffic volume and section length would increase the frequency of crashes and lower the chances of the occurrence of severe crashes. Research indicates that fatality rates in rural areas are higher than in urban locations [
21,
22,
23]. Similarly, it is expected that suburban roadways will have similar characteristics of crash potential. Suburban roadways are transitional high-speed roadways between low-speed urban streets and high-speed rural highways. They have the characteristics of both rural (i.e., high-speed) and urban (i.e., with curb and gutter used for drainage) roadways [
24]. Territorial variables (built environment and physical severance) can affect the occurrence of accidents, even due to distraction. In their study on traffic crashes on rural roads with older pedestrians, Casado-Sanz et al. [
25] indicated that a small number of studies in the literature have discussed territorial variables (built environment and physical severance). Casado-Sanz et al. [
25] aimed at determining the significant factors that increase the probability of a fatal outcome in the case of a crash involving at least one seriously injured individual at rural crosstown roads. The results revealed that the physical severance index (territorial variable) was very significant and suggested that the probability of a fatal outcome on a lateral rural crosstown road is almost two times higher than that at the baseline condition of a central rural crosstown road.
In Jordan, the rapid growth in population was accompanied by an increase in vehicle ownership. According to Jordan’s Department of Statistics [
26], there were about 10.5 million inhabitants in 2019. Additionally, Driver and Vehicle License Department (DVDL) [
27] records show that the number of licensed drivers in 2019 exceeded 2.5 million, and that there were around 1.6 million registered vehicles. The majority of Jordan’s population (90.33%) live in urban or suburban areas (areas are classified as follows: urban when the population > 10,000 inhabitants, and suburban when the population falls between 5000 and 10,000 inhabitants [
28]). Consequently, a rapid growth in population and vehicle ownership has contributed, among other factors, to the increasing number of crashes.
Accident statistics clearly show that when compared to developed countries, Jordan suffers from a relatively low level of traffic safety, which results in great socio-economic losses. The Annual Report of Traffic Accidents in Jordan [
29] showed that 161,511 traffic accidents occurred in Jordan in 2019, of which 10,857 were human injury accidents that resulted in 643 fatalities at an estimated financial cost of around USD 454 million. When compared with accidents worldwide, Jordan ranked in the top four countries in the year 2017 in relation to deaths per 10,000 vehicles. In the year 2019, human-related factors were responsible for 98.2% of accidents in Jordan that resulted in human injuries. Driver fault (not taking the necessary precautions) accounted for a high percentage (around 40%) of injuries and deaths in 2019 [
26].
Due to the significant impact of distraction on driver performance and the great losses resulting from it, this study aimed to identify the most salient driver faults that cause crashes on nine segments of Jordan’s rural and suburban roadways using the available crash data. The study also sought to examine crashes with distracted driving as the driver’s fault and to investigate the differences between crashes caused by distracted driving on rural and suburban roadways. Knowing which road factors influence distracted driving on rural and suburban roadways can help researchers, traffic experts, and authorities to identify the most effective measures that may reduce driver distraction and target different driver groups with the best practices that fit their driving behaviors and keep them attentive. Likewise, the best practices can be implemented on different roadway types, under varying weather conditions, and at different times of the year.
4. Discussion
The driver fault analysis conducted in this study proves the significant influence of distraction on driver performance and the great losses resulting from it. The analysis revealed that distracted driving was responsible for causing nearly one-fourth of the reported crashes included in this study. Moreover, the analysis of the severity of crashes showed that distracted driving was responsible for 22.7% of all reported fatalities, 22.7% of severe injuries, and 22.4% of PDO crashes. These results are in line with previous findings [
6,
7,
8,
9,
10,
11] indicating that human error was responsible for a very high percentage of crashes, and distraction had a significant influence on road crashes. These results also prove that humans continue to have limited capability to perform more than two tasks in parallel, as reported by Pashler [
3].
The analysis of distracted driving on rural and suburban roads was performed based on three different groups of factors: human- and time-related factors, road- and environment-related factors, and crash-type- and vehicle-related factors. Distracted driving in rural areas seemed to be more fatal, whereas suburban roads witnessed more severe injuries. The higher frequency of fatalities on rural roads could be attributed to the fact that traffic on rural roads runs at higher speeds, resulting in more severe and fatal crashes. This finding is in line with those of Eberhardt et al. [
21], Kmet et al. [
22], and Gonzalez et al. [
23], who indicated that fatality rates in rural areas are higher than those in urban locations.
The analysis of distracted driving based on human factors revealed that males were responsible for almost two-thirds of the crashes on both rural and suburban roadways. This percentage is reasonable considering that the number of licensed male drivers in Jordan is almost twice that of licensed female drivers [
23]. The differences between genders in terms of distracted driving were investigated using an independent samples
t-test. Statistically significant differences with a small effect size were found between male and female drivers under three conditions, namely road surface conditions, weather conditions, and lighting conditions. As reported by Young et al. [
13], distractions and road and environmental factors are expected to make the driving task more demanding, thus leading to impaired driving performance.
Differences were noted between drivers’ age groups in that younger age groups were causing crashes due to distracted driving almost three times more often than older groups. This finding is in line with that of Regan et al. [
12], who suggested that older and/or more experienced drivers are more skillful at conducting multiple activities while driving, unlike younger and/or less experienced drivers. The slight difference in percentages of crashes between age groups for rural and suburban areas can be attributed to the more demanding driving tasks on suburban roads due to higher traffic volumes and extra accessing maneuvers.
Road aggressiveness is one important human factor that was not included in the crash records. Alonso et al. [
19] indicated that drivers who tend to have several accidents give less importance to elements that can affect the driving task. The number and types of traffic violations (accident history) of drivers can be used to indicate the road aggressiveness factor. Therefore, the availability of the driver’s accident record in the crash report can be very helpful in studying driving behaviors and can contribute to a better understanding of driver distraction for reducing road crashes.
The study of the effect of time-related factors on distracted driving showed that the majority of crashes occur on “no holiday” days and during weekdays, which can be attributed to the higher exposure to risky situations resulting from higher traffic volumes that occur on roads on regular working days. Rural roadways showed a slightly higher occurrence of distracted driving crashes during weekdays, although suburban roads showed more crashes during weekends. Morning and night-time crashes were slightly more frequent on suburban roadways; this could be due to the higher traffic volumes on suburban roadways during these times. Meanwhile, rural roadways showed a higher percentage of crashes due to distracted driving at midday; this could also be attributed to the active movement of traffic on rural roads during this time.
Young et al. [
13] indicated that distraction on different areas and road types may vary. Weather, traffic, and road conditions influence the demandingness of the driving task. Therefore, the characteristics of crashes caused by distracted driving according to road- and environment-related factors were investigated. The result of the independent samples
t-test revealed that four variables were statistically significant with a small effect size: at junction, road grade, number of lanes, and weather condition. The percentage of at-junction crashes was slightly higher on suburban roads, which could be attributed to the higher demandingness level of the driving task due to higher traffic volumes. The majority of distracted driving crashes were seen on level grades, under dry surface conditions, in clear weather, and on two-lane roads for both rural and suburban areas. Suburban roads witnessed more crashes with distracted driving as the drivers’ fault during clear weather and daylight times, whereas more crashes were recorded on rural roadways when the lighting conditions were dark. This is consistent with Stutts et al. [
32], who found that those who were distracted at the time of their crashes were less likely to be traveling on multi-lane roadways and were less likely to have crashed at an intersection or other road junction.
Casado-Sanz et al. [
25] suggested using the physical severance index in future research, which considers the distribution of urban land as a relationship between the difference in the built areas on both sides of the road and the total built area. Territorial variables (built environment and physical severance) can affect the occurrence of accidents. Since the available data provide the coordinates of crashes, it might be possible to consider such territorial variables in future studies.
The analysis of the crashes caused by distracted driving as the drivers’ fault according to crash-type- and vehicle-related factors showed that collision was the most common crash type in both rural and suburban areas. The crash types and the number of vehicles involved were statistically significantly different between rural and suburban roads. Collision and run-off-road (ROR) crashes occurred in higher percentages on rural roads, whereas more pedestrian crashes were seen on suburban road sections. The high percentage of pedestrian crashes in suburban areas makes sense, as pedestrians are seen in urban and suburban areas more often and are only occasionally seen on rural roadways. Crashes that involved more than one vehicle were seen in higher percentages in suburban road sections, which could be explained by the existence of traffic-congested sections in suburban areas. The number of distracted driving crashes involving one vehicle was higher in rural areas due to less traffic and higher speeds.
As presented in the previous section, the highest percentages of crashes on suburban roads occurred at speeds of 40–80 km/h, while for rural roads, the highest percentages of crashes occurred at speeds of 60–100 km/h. Crashes at speeds of higher than 100 km/h seemed to occur on rural roads only. This is consistent with earlier findings by Thomason [
24], who reported the same result about suburban roadways and rural highways. As for vehicle type, we noticed that both automobiles and light vehicles comprised more than 80% of vehicles involved in the reported crashes. Rural roads witnessed a higher percentage of crashes caused by distracted driving involving automobiles, medium vehicles, and tractor vehicles. A similar finding was reported by Stutts et al. [
32], who indicated that distracted drivers were more likely to be driving during non-daylight hours and were more likely to be driving a pickup truck, van, or sport utility vehicles.
One of the major limitations of the present study is that the obtained data lack, in some respects, important details that, if they existed, would deliver more thorough and useful results about distraction. Therefore, it is recommended that accident reports be carefully designed to include all specific details, such as the type of distraction factor (visual, manual, cognitive, and auditory) that caused the driver’s distraction. Providing such details in the crash report would help eliminate confusion caused by interrelated information and would deliver complete information that will be significantly helpful to this field of research. It is also suggested that crash reports should include information on the number of accidents that drivers were previously involved in. Having this factor available can help researchers examine the effect of aggressive driving, as recommended by Alonso et al. [
19], on all kinds of crashes, including distraction.