Effect of Distraction and Driving Behaviour to Traffic Accidents in Jakarta Using Partial Least Squares Structural Equation Modeling (PLS-SEM)

Traffic accidents are the eighth leading cause of death worldwide, and each year, Indonesia reports an increasing number of such incidents. Human error, specifically risky driving behaviour such as distraction, is the primary contributors to the accidents. A thorough understanding of the contributing factors to traffic accidents is crucial to enhancing road safety initiatives. Therefore, this study aimed to design a model to assess the effect of road distraction, driving behaviour, and perception of risk on self-reported crashes by private car drivers in Jakarta, Indonesia, as well as formulate strategies to improve safety. This study used a diverse group of 142 drivers from Jakarta as respondents, utilizing a combination of quantitative methods, such as Partial Least Squares Structural Equation Modelling (PLS-SEM) and Pearson's Chi-square tests, complemented by questionnaire instruments such as the Driving behaviour Questionnaire (DBQ), Road Distractions Scale (RDS), and Risk Perception and Regulation Scale (RPRS). The results showed that driver distractions significantly increase the possibility of lapses, while errors, violations, and risk perception significantly affect the incident of traffic incidents. Furthermore, chi-square analysis showed that men are more likely to commit violations and are more distracted by attractive roadside objects compared to women, who reported a higher incidence of lapses and greater disturbance from weather conditions. This study offered strategic recommendations with the potential to lower accident rates and improve driving safety overall.

Driving behaviour; Distracted driving; Driving Behaviour Questionnaire (DBQ); Road safety; Structural Equation Modelling (SEM)

    Traffic accidents are the eighth leading causes of death in the world. According to the Global Status Report on Road Safety published by the World Health Organization (WHO), approximately 1.35 million people die in road traffic accidents yearly (WHO, 2018). As a result, most countries suffer significant economic losses, amounting to about 3% of their gross domestic product (WHO, 2022). In Indonesia, the number of traffic accidents reached 103,645 in 2021, an increase of 3.62% compared to 100,028 cases in 2020. There were 25,266 fatalities and material losses of up to Rp246 billion in 2021 (Kementerian Perhubungan RI, 2022). Most accidents occurred during rush hour when commuters were heading home using highways and main roadways. When the number of vehicles on the road increases, the probability of accidents also rises (Zainy et al., 2023; Puspasari et al., 2015).
    The high number of traffic accidents can be attributed to the physical environment of the road and vehicle and human factors (Shope, 2006; Ulleberg and Rundmo, 2003). The term "physical environment factor" refers to unsupportive road conditions, including damaged or uneven roads, sharp curves, inadequate traffic signs, and faded road markings. Vehicle factors include issues such as malfunctioning brakes, tire blowouts, and defective lights. Human factors pertain to the abilities and characteristics of the drivers.

Data from the Indonesian National Police indicate that the predominant contributing factors to traffic accidents are human at 61%, followed by infrastructure and environment at 30%, and vehicle at 9%. Human factors contribute to the highest percentage of traffic accidents (Zuraida, Wijayanto, and Iridiastadi, 2022; Kementerian Komunikasi dan Informatika RI, 2017).

Manchester Driving Behaviour Questionnaire (DBQ) is popularly used to analyse driving behaviour (Hussain et al., 2023; Jomnonkwao et al., 2021; Shen et al., 2018). Driving behaviour assessed using DBQ can be divided into 3 subscales, including error, lapse, and violation (Parker et al., 1995; Bakhshi et al., 2022; Wang and Xu, 2019). The term 'error' denotes instances of judgment mistakes or failures in observation that may endanger others. 'Lapses' are defined as unintentional deviations or behaviours caused by inattention or similar shortcomings. 'Violations' refer to deliberate departures from legally mandated or socially expected safe vehicle operation norms (Wang and Xu, 2019; Zhao et al., 2012).

Driver distraction includes activities that divert attention away from the primary task of driving (Carney, Harland, and McGehee, 2018; Regan, Hallett, and Gordon, 2011). These distractions are a significant factor in the loss of concentration on the road. Distractions can generally be classified into four categories: visual, cognitive, auditory, and manual, which include interacting with the vehicle or environment (Regan, Lee, and Young, 2008).

Numerous studies examined driving behaviour (Hussain et al., 2023; Wang and Xu, 2019; Zhao et al., 2012) and distractions (Carney, Harland, and McGehee, 2018; Ortiz et al., 2018; Regan, Hallett, and Gordon, 2011; Kass, Cole, and Stanny, 2007) as separate topics. Some studies on driving behaviour focused on significant differences in DBQ variables across countries (Hussain et al., 2023). Meanwhile, studies on driving distractions often examined cell phone use (Ortiz et al., 2018; Kass, Cole, and Stanny, 2007). Carney, Harland, and McGehee (2018) examined the relationship between types of distractions and crashes. Additionally, other studies investigated the relationship between driving behaviour and self-reported crashes (Wang and Xu, 2019; Zhao et al., 2012). However, study that integrates driving behaviour with distraction variables to assess traffic crashes is limited. Arevalo-Tamara et al. (2022) investigated a model that included distractions about risky road behaviours and traffic crashes in Bogota. The model did not account for the 'lapses' variable in driving behaviour. Definitively, lapses are unintentional actions resulting from inattention or a deficit, such as taking the wrong exit (Wang and Xu, 2019). The original DBQ comprises three subscales: errors, lapses, and violations (Parker et al., 1995), which means including 'lapses' as a variable is crucial. Several studies showed traffic disruption was a critical factor that should be considered in road safety policymaking (Arevalo-Tamara et al., 2022; Stanojevic, Jovanovic, and Lajunen, 2013).

This study aims to develop a model for assessing the impact of road distractions, driving behaviour, and perceived risk on self-reported crashes among private car drivers in Jakarta. Specifically, violations, errors, and lapses as the driving behaviour variables were considered and expected to help suggest strategies for enhancing road safety, drawing on insights from the driving behaviour model. The objectives can be summarised into two, including (1) creating a model that connects distractions and driving behaviour with traffic accidents, and (2) recommending effective strategies to bolster road safety.

2.1. Study Object and Subject

This study focused on private car drivers in Jakarta who use the vehicles for daily activities. The participants included 142 respondents, aged between 26 and 59. All respondents were confirmed to hold valid driving licenses (SIM A) and were deemed suitable for this study. This sample size surpasses the minimum requirement of 112 respondents for PLS-SEM, as calculated by the inverse square root method (Kock and Hadaya, 2016). Convenience sampling, a non-probability sampling method where participants self-select in response to an open invitation, was used (Stratton, 2021).

`Six latent variables were used, including Distractions (DS), Risk Perception (RP), Errors (E), Lapses (L), Violations (V), and Traffic Incidents (TI). DS evaluates the extent to which various commonly observed road disturbances influence drivers. RP assesses driver awareness of the risks and the understanding of traffic regulations (Arevalo-Tamara et al., 2022; Useche et al., 2018). E, L, and V are the subscales of DBQ that quantify driving behaviours related to inattention and distraction. TI captures the history of the respondent traffic incidents, including accidents/collisions and incidents/near misses.

2.2. Model and Study Hypothesis

This study adopts the study model from Arevalo-Tamara et al. (2022) to evaluate the effects of road distractions, driving behaviour, and risk perception on self-reported crashes. It hypothesizes that road distractions influencing drivers may significantly predict the risky driving behaviour, potentially leading to traffic accidents. Figure 1 shows the conceptual model, which has been augmented by including the 'Lapses' variable from DBQ as developed by Wang and Xu (2019). Additionally, the model features a refined 'Traffic Incidents' (TI) component that includes two indicators, accidents and incidents/near misses. A total of 11 hypotheses were developed as shown in Table 1.

Figure 1 Conceptual model for predicting traffic incidents

Table 1 Study Hypotheses

Data collected from surveys, including DBQ, Road Distraction Scale (RDS), and Risk Perception Rating Scale (RPRS), along with traffic incident histories were analysed using Partial Least Squares Structural Equation Modelling (PLS-SEM) and the Chi-square test. The results helped design strategies, which would be developed through a literature review and validated by experts and stakeholders.

2.3. Data Collection

The data collection phase includes discussing the types and methods of data collection, designing conceptual models and hypotheses based on a literature review, creating questionnaires, and analysing the results obtained from these questionnaires. The data processing phase includes testing the validity and reliability of the questionnaires, specifying the model, analysing SEM model, which includes testing both the measurement model (outer model) and the structural model (inner model) and conducting the Chi-square test.

Data were collected through questionnaires comprising five parts, including driver demographics and characteristics, DBQ, RDS, RPRS, and information on driving experience and accident/incident history. According to Arevalo-Tamara et al. (2022), RDS includes eight types of distractions, specifically text messaging/chatting (DS1), phone calls (DS2), billboards (DS3), attractive roadside objects (DS4), personal thoughts/concerns (DS5), weather conditions (DS6), behaviour of other road users (DS7), and road obstacles (DS8). In addition to questionnaire data, further information was collected to inform the formulation of strategic recommendations at the conclusion. This additional data collection involved validating the prepared strategies based on the results of hypothesis testing, Chi-square analysis, and literature review with relevant experts.

Validation with these experts aimed to ensure the content of the questionnaire was accurate and relevant before distribution. This validation process included assessing the accuracy and relevance of the content and incorporating any significant aspects that may have been initially overlooked. The experts who validated this study included an Associate Expert Researcher from the BRIN Transportation Safety Research Group and the Head of the Work System Engineering & Ergonomics Laboratory at the Bandung Institute of Technology.

2.4. Validity and Reliability Test

After collection, the data is tested for validity and reliability using IBM SPSS Statistics 29 to ensure it is consistent and accurately reflects the conditions being measured. An item is considered valid in case it has a positive correlation value (r-value) and the calculated r (r count) is greater than the critical r (r table). Meanwhile, an item is invalid suppose the r count is less than or equal to the r table (Silvia and Irwansyah, 2023). The Pearson Correlation test results for 41 items all exceed 0.1743, indicating that the questionnaire is valid. Following the validity test and confirming that all indicators are valid, a reliability test is performed. The reliability test results for the 41 indicators yield a Cronbach’s alpha value of 0.875, which is above the acceptable threshold, confirming the reliability of the questionnaire as a measurement instrument.

2.5. PLS-SEM Processing

PLS-SEM processing is carried out using SmartPLS 4 application. PLS-SEM processing consists of three stages, including model specification, evaluation of the measurement model (outer model), and evaluation of the structural model (inner model). The circle symbols on the model represent latent variables, while the rectangular symbols represent indicators. In the model, arrows represent the relationships between indicators and latent variables, as well as the among the latent variables themselves. The type of model used in this study is a reflective measurement model. Figure 2 shows the model while Table 2 explains the codes used for each latent variable.

Figure 2 PLS-SEM model for predicting traffic incidents

Table 2 Codes used for each latent variable

SEM processing was carried out on 142 respondent data, a number that already exceeds PLS-SEM minimum sample requirements calculated using the inverse square root method. The formulation of SEM processing stages is as follows:

2.5.1. Measurement Model (Outer Model)

This test is conducted to determine the validity and reliability of the constructs used. Reflective indicator testing includes indicator reliability, internal consistency reliability, convergent validity tests, and discriminant validity tests with each measurement having different approaches and requirements. The steps consist of indicator and internal consistency reliability, as well as convergent and discriminant validity (Ahdika, 2017; Hair et al., 2017).

2.5.2. Structural Model (Inner Model)

a.   Multiple collinearity test. All indicators show a VIF value of <3 which means there is no collinearity problem in the study model. All VIF tests on the hypothesis show good and acceptable results.

b.   Coefficient of determination. According to Cohen (1988), the value of R² can be categorized into: > 0.26 (Strong), 0.13 - 0.26 (Moderate), and < 0.13 (Weak). R² values in social and behavioural study tend to have low values (Hair et al., 2017). Traffic Incidents have moderate predictive power (R2 = 0.228), while Errors, Lapses, and Violations have weak predictive power (R² below 0.13).

c.   Predictive relevance. Hair et al. (2017) recommends cross-validated redundancy as chosen as the best approach. According to Hair et al. (2017), a good Q² value is > 0 and that can be said to have good predictive ability. The Q² value on the latent variable listed already shows a value > 0. Therefore, these endogenous latent variables have good predictive relevance and are acceptable.

d.    Path coefficient. The Rule of Thumb for path coefficient value is that the hypothesis will be accepted if the p-value < 0.05 and t-value > 1.96. 

3.1. Analysis of Significance Test

The analysis of the significance test is conducted to determine whether the relationship between latent variables has statistical significance (Sarstedt, Ringle, and Hair, 2021). Hypothesis testing was conducted using the bootstrapping method using a two-tailed test scheme with a significance level of 5% (= 0.05).

Table 3 shows all the hypotheses for this study. A total of 4 hypotheses out of 11 were accepted, specifically H2, H5, H9, and H11. The final PL et alS-SEM model is shown in Figure 3.

Table 3 Significance Test

                                                           ^**p < 0.010; ^*p < 0.050

Figure 3 Final PLS-SEM model

 The hypotheses confirmed a direct impact of distractions on lapses (H2). Additionally, cognitive limitations in middle-aged drivers may influence the response to distractions, manifesting as delayed reactions in intermediate or middle-aged drivers, which is evident through deviant road behaviour. The results were in line with Feng, Marulanda, and Donmez (2014), reporting that involuntary distractions significantly correlated with lapses. Similarly, Chen et al. (2016) found a positive correlation between self-reported distraction involvement and all four categories of unsafe driving behaviours identified in DBQ, including lapses. According to Zhao et al. (2012), there is a correlation between a high frequency of lapses increased steering reversal rates and inconsistent throttle control, both of which can compromise driving safety.

The results of this study show that driving errors have a significant effect on traffic incidents (H5). Such errors, including near-misses, misinterpreting traffic signs, or failing to use the rearview mirror, directly influence the frequency of traffic incidents. These mistakes heighten the risk of accidents or lead to other hazardous situations if drivers are unaware of the actions. However, the results were not in line with Arevalo-Tamara et al. (2022), which did not observe a significant relationship between driving errors and traffic accidents. This discrepancy may highlight the differences between the driving contexts in Bogota, Colombia (Arevalo-Tamara et al., 2022), and Jakarta (the current study), where drivers in Jakarta appear to commit more judgment-impairing driving errors, leading to traffic incidents. Additionally, this study supports the results of Wang and Xu (2019), establishing that high-risk drivers could commit errors due to inattention.

The results show that risk perception has a negative direct impact on traffic incidents (H9). Specifically, an improved awareness of risk correlates with a reduction in unwanted traffic events. Drivers with a keen sense of risk are usually more vigilant, can recognize potential hazards, assess the consequences of their actions, and act accordingly to mitigate risks. The results are in line with Arevalo-Tamara et al. (2022), which indicated a negative correlation between risk perception and the frequency of traffic accidents. Additionally, the study showed that violations had a direct positive impact on traffic incidents (H11). Traffic violations, which reflect non-adherence to road rules, can lead to an increase in dangerous situations. These behaviours were observed both in younger drivers who were associated with higher-risk driving as well as in more seasoned middle-aged drivers. This is in line with Zhao et al. (2012) and Arevalo-Tamara et al. (2022), which established that drivers who commit violations demonstrate poorer lateral control, more frequent sudden changes, and increased rates of sudden acceleration. Behaviours associated with traffic violations are significantly correlated with an increased rate of traffic accidents.

3.2. Analysis of Chi-Square Test

Pearson's Chi-square test is used to examine the relationship between two or more categorical variables or nominal data in the form of contingency tables. This test is meant to determine whether there is a significant relationship or association between the variables tested. Chi-square test processing was conducted based on gender groups, specifically male and female, as shown in Table 4.

Table 4 Chi-Square Test Results

                                                  ^**p < 0.010; ^*p < 0.050

The analysis showed gender-based correlations with latent variables. Men tend to commit more traffic violations and have a higher incidence of traffic accidents. Additionally, men were more susceptible to distraction by DS4 (Attractive objects). This in line with Arevalo-Tamara et al. (2022), which established a significant distraction in men when encountering visually appealing objects while driving. Meanwhile, women were more prone to lapses and reported being more affected by DS6 (Weather Conditions), contrasting with Arevalo-Tamara et al. (2022). This suggests that in Jakarta, female respondents can be distracted by weather conditions and are more susceptible to lapses. The strategy recommendations to minimize traffic accidents from driving distractions are built based on the significant relationships between each latent variable, a study from the literature review, and expert validations, as shown in Table 5.

Table 5 Proposed Strategies

    The results of this study offer valuable insights for the development of new traffic policies for policymakers. These policies aim to substantially reduce traffic accidents in Jakarta while carefully considering critical factors such as driver distraction, age group, and driving behaviour.

    In conclusion, this study aimed to design a model to assess the effect of road distraction, driving behaviour, and risk perception on traffic accidents using PLS-SEM and Chi-square analysis as well as to develop strategies for improving road safety for private car drivers in Jakarta, Indonesia. The study novelty lay in the inclusion of the 'lapses' variable within the model of distraction and driving behaviour, a distinction that differentiated it apart from previous studies. The results show that both errors and violations contributed to traffic incidents, while a heightened risk perception negatively correlated with such incidents. Additionally, the role of distractions in causing lapses was emphasized. Chi-square analysis showed that violations and susceptibility to distractions from attractive roadside objects were higher in men than women. Meanwhile, women were more prone to lapses and more affected by weather conditions. The theoretical implications of this study included providing new insights into the relationship between distraction and driving behaviour on the road, focusing on distractions that interfere with driver abilities. Several strategies for improving road safety were proposed in this study. The practical implications related to policy measures that stakeholders could adopt include law enforcement, system development, interventions to enhance road safety skills, and the use of sensor-based applications. These recommendations presented viable options to reduce accident rates, improve driving safety, as well as contributing to the evolution of previous studies and providing a reference for future ones. There were certain limitations in this study, such as focusing only on private car drivers in DKI Jakarta, not considering factors such as fatigue and exhaustion, and was conducted over a brief period from April to June 2023. Future studies should consider including other types of road users and different regions. It was also essential to include a larger number of experts from various fields to obtain more representative data and broader insights.

    This study is supported by the PUTI Q1 Grant in 2023, funded by the Directorate of Research and Community Service (DRPM) Universitas Indonesia, Number NKB-524/UN2.RST/HKP.05.00/2023.

Ahdika, A., 2017. Improvement Of Quality, Interest, Critical, and Analytical Thinking Ability of Students Through The Application of Research-Based Learning (RBL) In Introduction to Stochastic Processes Subject. International Electronic Journal of Mathematics Education, Volume 12(2), pp. 167–191

Albert, G., Musicant, O., Oppenheim, I., Lotan, T., 2016. Which Smartphone's Apps May Contribute to Road Safety? An AHP Model to Evaluate Experts' Opinions. Transport Policy, Volume 50, pp. 54–62

Arevalo-Tamara, A., Caicedo, A., Orozco-Fontalvo, M., Useche, S.A., 2022. Distracted Driving in Relation to Risky Road Behaviors and Traffic Crashes in Bogota, Colombia. Safety Science, Volume 153, p. 105803

Bakhshi, V., Aghabayk, K., Parishad, N., Shiwakoti, N., 2022. Evaluating Rainy Weather Effects on Driving Behaviour Dimensions of Driving Behaviour Questionnaire. Journal of Advanced Transportation, Volume 2022, 6000715

Botzer, A., Musicant, O., Perry, A., 2017. Driver Behavior with a Smartphone Collision Warning Application–A Field Study. Safety Science, Volume 91, pp. 361–372

Carney, C., Harland, K.K., McGehee, D.V., 2018. Examining Teen Driver Crashes and The Prevalence of Distraction: Recent Trends, 2007–2015. Journal of Safety Research, Volume 64, pp. 21–27

Chen, H.Y.W., Donmez, B., Hoekstra-Atwood, L., Marulanda, S., 2016. Self-Reported Engagement in Driver Distraction: An Application of The Theory of Planned Behaviour. Transportation Research Part F: Traffic Psychology and Behaviour, Volume 38, pp. 151–163

Cohen, J., 1988. Statistical Power Analysis for The Behavioral Sciences. New York: Academic Press

Feng, J., Marulanda, S., Donmez, B., 2014. Susceptibility to Driver Distraction Questionnaire: Development and Relation to Relevant Self-Reported Measures. Transportation Research Record, Volume 2434(1), pp. 26–34

Hair, J.F., Hult, G.T.M., Ringle, C.M., Sarstedt, M., 2017. A Primer on Partial Least Squares Structural Equation Modeling (PLS-SEM). 2^nd Edition. SAGE Publications

Hussain, B., Miwa, T., Sato, H., Morikawa, T., 2023. Subjective Evaluations of Self and Others’ Driving Behaviors: A Comparative Study Involving Data from Drivers in Japan, China, and Vietnam. Journal of Safety Research, Volume 84, pp. 316–329

Jomnonkwao, S., Uttra, S., Ratanavaraha, V., 2021. Analysis of a driving behavior measurement model using a modified driver behavior questionnaire encompassing texting, social media use, and drug and alcohol consumption. Transportation Research Interdisciplinary Perspectives, Volume 9, 100302

Kass, S.J., Cole, K.S., Stanny, C.J., 2007. Effects of Distraction and Experience on Situation Awareness and Simulated Driving. Transportation Research Part F: Traffic Psychology and Behaviour, Volume 10(4), pp. 321–329

Kementerian Komunikasi dan Informatika RI, 2017. Rata-rata Tiga Orang Meninggal Setiap Jam Akibat Kecelakaan Jalan (An Average of Three People Die Every Hour Due to Road Accidents). Kementerian Komunikasi dan Informatika Republik Indonesia

Kementerian Perhubungan RI, 2022. Focus Group Discussion: Sidang Para Pakar Keselamatan Transportasi Jalan (Focus Group Discussion: Meeting of Road Transportation Safety Experts). Kementerian Perhubungan Republik Indonesia

Kimura, T., Imai, Y., Moriizumi, S., Yumoto, A., Taishi, N., Nakai, H., Renge, K., 2022. An Experimental Study on Errors Regarding the Driving Behavior of Young Males Caused By Temporal Urgency On Open Roads: A Bayesian Estimation. IATSS Research, Volume 46(1), pp. 147–153

Kock, N., Hadaya, P., 2018. Minimum Sample Size Estimation In PLS-SEM: The Inverse Square Root and Gamma-Exponential Methods. Information Systems Journal, Volume 28(1), pp. 227–261

Liu, J., Wang, C., Liu, Z., Feng, Z., Sze, N.N., 2021. Drivers’ Risk Perception and Risky Driving Behavior Under Low Illumination Conditions: Modified Driver Behavior Questionnaire (DBQ) and Driver Skill Inventory (DSI). Journal of Advanced Transportation, Volume 2021, pp. 1–13

Ortiz, C., Ortiz-Peregrina, S., Castro, J.J., Casares-Lopez, M., Salas, C., 2018. Driver Distraction By Smartphone Use (Whatsapp) in Different Age Groups. Accident Analysis & Prevention, Volume 117, pp. 239–249

Parker, D., Reason, J.T., Manstead, A.S., Stradling, S.G., 1995. Driving Errors, Driving Violations and Accident Involvement. Ergonomics, Volume 38(5), pp. 1036–1048

Puspasari, M.A., Muslim, E., Moch, B.N., Aristides, A., 2015. Fatigue Measurement in Car Driving Activity using Physiological, Cognitive, and Subjective Approaches. International Journal of Technology, Volume 6(6), pp. 971–975

Regan, M.A., Hallett, C., Gordon, C.P., 2011. Driver Distraction and Driver Inattention: Definition, Relationship and Taxonomy. Accident Analysis & Prevention, Volume 43(5), pp. 1771–1781

Regan, M.A., Lee, J.D., Young, K., 2008. Driver Distraction: Theory, Effects, and Mitigation. CRC Press

Sarstedt, M., Ringle, C.M., Hair, J.F., 2021. Partial Least Squares Structural Equation Modeling. In: Handbook of Market Research. Springer International Publishing, pp. 587–632

Shen, B., Ge, Y., Qu, W., Sun, X., Zhang, K., 2018. The different effects of personality on prosocial and aggressive driving behaviour in a Chinese sample. Transportation Research Part F: Traffic Psychology and Behaviour, Volume 56, pp. 268-279

Shope, J.T., 2006. Influences On Youthful Driving Behavior and Their Potential for Guiding Interventions To Reduce Crashes. Injury Prevention, Volume 12(1), p. 011874

Silvia, M., Irwansyah, I., 2023. Validity and Reliability Test of Content Creator Strategy Management. Jurnal Kajian Jurnalisme, Volume 6(2), pp. 158–170

Stanojevic, P., Jovanovic, D., Lajunen, T., 2013. Influence of Traffic Enforcement on The Attitudes and Behavior of Drivers. Accident Analysis and Prevention, Volume 52, pp. 29–38

Stratton, S.J., 2021. Population Research: Convenience Sampling Strategies. Prehospital and Disaster Medicine, Volume 36(4), pp. 373–374

Sullman, M.J., Stephens, A.N., Taylor, J.E., 2019. Dimensions of Aberrant Driving Behaviour and Their Relation to Crash Involvement for Drivers in New Zealand. Transportation Research Part F: Traffic Psychology and Behaviour, Voume 66, pp. 111–121

Ulleberg, P., Rundmo, T., 2003. Personality, Attitudes and Risk Perception as Predictors of Risky Driving Behaviour Among Young Drivers. Safety Science, Volume 41(5), pp. 427–443

Useche, S.A., Alonso, F., Montoro, L., Esteban, C., 2018. Distraction of Cyclists: How Does it Influence Their Risky Behaviors and Traffic Crashes? PeerJ, Volume 6, p. e5616

Wang, X., Xu, X., 2019. Assessing the Relationship Between Self-Reported Driving Behaviors and Driver Risk Using a Naturalistic Driving Study. Accident Analysis & Prevention, Volume 128, pp. 8–16

World Health Organization (WHO), 2018. Global Status Report on Road Safety. World Health Organization

World Health Organization (WHO), 2022. Road Traffic Injuries. World Health Organization

Zainy, M.L.S., Pratama, G.B., Kurnianto, R.R., Iridiastadi, H., 2021. Fatigue Among Indonesian Commercial Vehicle Drivers: A Study Examining Changes in Subjective Responses and Ocular Indicators. International Journal of Technology, Volume 14(5), pp. 1039–1048

Zhao, N., Mehler, B., Reimer, B., D’Ambrosio, L.A., Mehler, A., Coughlin, J.F., 2012. An Investigation of The Relationship Between The Driving Behavior Questionnaire and Objective Measures of Highway Driving Behavior. Transportation Research Part F: Traffic Psychology and Behaviour, Volume 15(6), pp. 676–685

Zuraida, R., Wijayanto, T., Iridiastadi, H., 2022. Fatigue during Prolonged Simulated Driving: an Electroencephalogram Study. International Journal of Technology, Volume 13(2), pp. 286–296