The Driver Behavior Questionnaire (DBQ) is one of the most used tools for measuring self-reported aberrant driving behavior. Several versions of the DBQ exist varying in terms of the item number and factor structure. This study aimed to investigate the psychometric properties and the factorial structure of the DBQ for assessing aberrant behavior among Lebanese drivers and to examine the relationship between DBQ subscales and socio-demographic characteristics and traffic-related variables. A cross-sectional study was conducted among Lebanese drivers aged more than 18 years old from all Lebanese governorates. Using a convenience sampling technique, data were collected using an anonymous Arabic, self-administered questionnaire that includes socio-demographic characteristics, and the Arabic Lebanese version of the DBQ (DBQ-L). Statistical analyses were performed using SPSS version 24 and Amos software. Exploratory factor analysis (N = 568) and confirmatory factor analysis (N = 534) were performed to test the factorial structure of the DBQ-L. Linear regression analyses were performed to examine the association between DBQ-L, sociodemographic and relevant traffic variables. A total of 1102 drivers aged between 18 and 82 years participated in this study with an average age of 34.64 (SD = 12.33). Of the total, 68.4% of participants were male having an average driving experience of 3.46 years (SD = 12.33). The DBQ-L total scale revealed high internal consistency (Cronbach’s α = 0.892). Exploratory and confirmatory factor analysis supported the original four-factor structure reflecting aggressive violations, ordinary violations, errors, and lapses dimensions. Aggressive violations had the highest mean among all the DBQ subscales. Men were found to have a higher tendency to commit aggressive and ordinary violations than females. Besides, the ordinary violation score was positively associated with younger age and increased mileage. All DBQ subscales decreased with the highest educational level. Being a professional driver increased the likelihood of drivers reporting violations, errors, and lapses, and increased mileage amplified the tendency of displaying aggressive violations and lapses when driving. However, errors decreased for young drivers and increased with age and larger driver experience. The cross-cultural adapted version of the DBQ was found to be a reliable and valid tool for assessing aberrant behaviors among Lebanese drivers. Further studies are needed to examine the ability of DBQ to predict road accidents.
Article
Examining self-reported aberrant behavior among Lebanese drivers using the Driver Behavior Questionnaire (DBQ)
https://doi.org/10.21203/rs.3.rs-1888094/v1
This work is licensed under a CC BY 4.0 License
Version 1
posted
You are reading this latest preprint version
Driver Behavior Questionnaire
Lebanese drivers
psychometric properties
Road traffic crashes (RTCs) are recognized as a major cause of death and disability worldwide. The worldwide annual death toll due to RTCs has reached 1.35 million with millions of life-altering injuries and long-lasting effects [1]. Several studies have shown that human factors play a critical role in road traffic accidents [2, 3], in particular the driver’s behavior [4]. Therefore, a growing interest has emerged within the past years in exploring the basic human factors affecting driving behavior in road transport studies [5-7].
One of the most widely used instruments for measuring aberrant driving behaviors is the Driver Behavior Questionnaire (DBQ). It was tested in different cultural contexts and nationalities, such as among samples of British, Dutch [8], Irish and Finnish [9], Italian [10, 11], Australian [12], Turkish [13, 14], French [15], and Qatari and United Arab Emirates drivers [16]. The original version, developed by Reason et al. (1990), evaluated the driver behavior under a theoretical framework and reflected the metacognitive skills of the drivers [17]. Hence, DBQ could embody a potential milestone in building a model for understanding driver's behavior [18]. Of note, DBQ is based on the main distinction between errors and violations, which were presumed to originate from different psychological roots and appeal to different modes of remediation [17]. Errors are defined as a failure of a planned sequence of mental or physical activities in the selection of an action to achieve the intended goal [17]. However, their consequences could be potentially dangerous. Violations are defined as deliberate risky behavior, not essentially reprehensible but may have detrimental consequences. According to Reason et al., errors reveal a cognitive dysfunction in information processing whereas violations are derived from psychosocial and motivational origins [17]. Therefore, errors refer to driving skills, while violations confer to driving style. Of note, it is possible that a driver may commit violations without errors or may commit errors without violations.
Based on the original version of DBQ, a wide-ranging of DBQ versions were developed such as the 104-item version developed by Aberg and Rimmö in 1998 [19], the 28-item version [20], and the 9-item version, edited by Martinussen et al. [21]. In addition to the difference in the number of items, the factorial structures of the DBQ varied between different driving cultures and nations as well. However, this variety of DBQ versions, used in the literature to assess risky driving behavior, complicated the ability to compare and compare the findings of different studies. Although the most cited factorial structure of DBQ was the one displaying three factors, which is endorsing the original formulation of Reason et al. [17], while other complex factorial explanations have also been suggested, e.g., Rowe et al. [22].
A substantial regional difference between countries was observed in RTCs. Similarly, driver behavior and driving style might vary between countries due to a large number of potential interpersonal conflict areas specific to each country’s traffic environment [23]. Lebanon is a small country in the Eastern Mediterranean Region (EMR) where injury prevention has not been yet a high priority [24]. Many countries in the EMR lack simple measures to reduce road traffic injuries. According to the Lebanese Ministry of the interior and municipalities, 487 fatalities and 6101 injuries resulting from RTCs were recorded in 2019, mainly due to human factors [25]. However, a road safety tipping point cannot be reached if the human-made epidemic of road traffic injuries (RTI) continues to be neglected, underexplored, and marginalized. Since information related to driver behavior in Lebanon is scarce, exploring the driver behavior that could trigger RTCs could be interesting in pointing out gaps that should be addressed in future interventions. It might be a gateway to new studies to assess other human factors that may be associated with crash involvement as well. To the best of our knowledge, driver behavior has not been previously explored among Lebanese drivers using a psychometrically reliable and valid tool. Certainly, a self-reported evaluation of driving behavior cannot be addressed without the DBQ since it remains the most popular and used tool in traffic psychology that gained wide acceptance in various countries worldwide [26].Given the absence of previous publication on psychometric evaluation of this instrument in Lebanon, it is therefore important to adapt the DBQ scale robustly for use among Lebanese drivers before embarking to explore their behavior. The investigation of driver behavior using a validated scale could assist stakeholders in target interventions to improve road safety.
Aims of the present research
The present research has two objectives. The first one was to adapt and validate the DBQ to the socio-cultural characteristics of the Lebanese drivers and to evaluate its psychometric properties in assessing these drivers' aberrant behavior. The second one was to examine the relation between the Lebanese version of the DBQ (DBQ-L), sociodemographic characteristics, and relevant traffic variables.
The validation of The Arabic version of DBQ comprised two stages. The first stage included the translation and cross-cultural adaptation of the DBQ into Arabic and the second stage focused on testing the psychometric properties of the Arabic version of the DBQ among Lebanese drivers.
Adaptation of DBQ
Translation and back-translation:
The original 50-items version of DBQ was meticulously translated through a process of forward-backward translation including translation synthesis [27] (Figure 1). First, two independent bilingual translators whose mother tongue is Arabic and who were also proficient in English translated the English version of DBQ to Arabic. One of the translators was a road safety specialist who was aware of the DBQ concept. The second translator was selected independently from the langue department at the Lebanese university and was unaware of the concept. Any discrepancy found between the two translators was considered and discussed, then the two translated versions were synthesized into one. At that point, the initial |Arabic translated version was back-translated into English by two independent translators whose native language was English and who were also fluent in Arabic. Of note, these translators didn’t have a behavioral science or road safety background and they don’t have access to the original version of the DBQ as well.
A committee of experts which included a road safety specialist, a linguistic professional, a psychologist, and the principal investigator was composed to verify the clarity and the suitability of the items in terms of wording, and specific features of the Lebanese context. Inquiries raised in committee meetings were communicated to the forward and back translators. The two versions were reviewed and compared by the experts at the end of this procedure, , to discuss any discrepancy or inconsistency in the previous stage of translation. Any change made to the translated version was resolved by consensus. The latter was reached by removing five items and incorporating its contents in other items to avoid redundancy. The removed items were as follows: The following items were removed: (1) become impatient with a slow driver in the outer lane and overtake on the inside,(2) On turning left, nearly hit a cyclist who has come up on your inside, (3) Fail to notice someone stepping out from behind a bus or parked vehicle until it is nearly too late, (4) overtake a slow-moving vehicle on the inside lane or hard shoulder of a motorwayand (5) fail to notice pedestrians crossing when turning into a side street from the main road. The rest of the modifications suggested by the review committee are presented in Annex A1.
Translators were responsible to address these changes and updating the translated version once those variations were communicated to them.
The pre-final version of the translated DBQ-L was piloted on a small sample of 35 drivers to evaluate the questionnaire’s comprehensibility and provide final input on its language. After completing the questionnaire, each participant was asked to elaborate on what they thought about each questionnaire item and what their corresponding response meant. Based on their feedback, minor revisions were made. This included changing the confusing words and potentially misleading items in particular those related to technical terms not commonly used by the Lebanese drivers. Hence, they were changed to the slang language while conserving their meanings and ensuring that they can still feed into the relevant scale. Finally, the Arabic version of DBQ was produced and ready for psychometric testing.
Psychometric testing
Participants and sampling
As part of a large project exploring driver behavior in Lebanon, a cross-sectional study was carried out using a convenience sampling technique among Lebanese drivers aged 18 years old and above from all Lebanese provinces over the period extending from October to December 2019. To ensure the representativeness of the sample in terms of age, gender, and location and to align the sample distribution with the population for those variables, a weighting procedure was used following an iterative proportional fitting. This procedure involved setting predetermined target figures across gender and age for each Lebanese governorate (Bekaa, Baalbeck-Hermel, Mount-Lebanon, Beirut, North, Akkar, South, and Nabatyeh).
All Lebanese active drivers having 18 years or over, having a driver's license, driving regularly in Lebanon, Arabic literate, and agreeing to participate were eligible to be part of this study.
This study excluded drivers who are not practicing driving activities currently, illiterate drivers who could not understand the questions, non-Lebanese drivers, and drivers who refused to participate in the study. Rarely driving referred to drivers who drive less than once per month were also excluded. The research protocol was properly evaluated and approved by IPNET. All methods were performed following the relevant guidelines and regulations. The study design assured adequate protection of study participants, and neither included clinical data about patients nor configured itself as a clinical trial. None of the survey’s queries questioned for information that could harm the respondent in any way.
Minimal sample size calculation
According to Comrey and Lee, sample size guidance indicated that five to ten observations for each scale item were found necessary for establishing sufficient evidence of scale validity and reliability [28]. The original version of the DBQ scale consists of 50 items therefore, 500 patients were needed, for the run of the exploratory factor analysis. To increase the validity of the study, we used two different samples: one for the exploratory factor analysis (EFA) and one for the confirmatory factor analysis (CFA). Therefore, considering losses to withdrawal, follow-up, or protocol violation, we set to recruit ≥more than 500 participants for each analysis (exploratory and confirmatory).
Reliability
Internal consistency reliability
The reliability of the DBQ was evaluated using internal consistency which looks to the degree to which every test item measures the same construct. The internal consistency reliability was estimated using Cronbach’s alpha where its value α ≥ 0.70 was considered satisfactory based on the Rule of Thumb.
Test-retest reliability
The–retest reliability measures the degree to which the measurement's results in one sitting are consistent and stable over time (usually over two-time points (T1, T2)) through the correlation between scores from one administration of the DBQ to another. In terms of the sample size required for test-retest reliability, a small sample size is required and a minimal sample size of 22 drivers is considered sufficient given that such a test is commonly conducted during the initial pilot study. Then, 40 drivers were asked to fill out the questionnaire for the second time after almost 3 weeks, which is a relatively short period, to mitigate against conclusions being due to memory bias. Test-retest reliability was evaluated using Pearson correlation ((Pearson’s r) where its value ≥ 0.70 was considered satisfactory for ruling on the correlation between the retest and the initial study. However, to ascertain the magnitude of agreement between the time points rather than the relationship, we calculate the difference between each data point and the mean (mean difference) alongside the standard deviation. The latter allowed us to determine how agreeable the measures are based on how close the data points deviate from the line of equality, 95% of differences were expected to be less than two standard deviations away from the mean.
Validity
Content validity:
Content validity is defined as “the extent to which elements of an assessment tool are relevant and fairly representative to the entire domain the tool seeks to measure [29]. The content validity of the DBQ was determined using the viewpoints of the panel of experts including content experts as well as lay experts. Content experts are professionals who have research experience or work in the field and lay experts who are the potential research subjects (drivers). Of note, using subjects of the target group as expert ensures that the population for whom the instrument is being developed is represented
Therefore, this panel was composed of eight members and included a road safety specialist (two), an epidemiologist (two), drivers (two), a psychologist/behavior specialist (one), and the principal investigator. In this study, content validity was tested using qualitative and quantitative approaches to confirm that each item in the construct was necessary. In the qualitative content validity method, content experts and target groups reviewed the items and suggested recommendations which were adopted on observing grammar, using appropriate and correct words, applying correct and proper order of words in items. In the quantitative content validity method, the experts are requested to specify whether an item is necessary for operating a construct in a set of items or notusing the Lawshe method [30].
. To this end, they are requested to score each item from 1 to 3 with a three-degree range of “not necessary, useful but not essential, essential” respectively. The confidence is maintained in selecting the most important and correct content in an instrument, which is quantified by the content validity ratio (CVR). Then, the content validity ratio (CVR) for each item was calculated using the following formula: 
.“Ne” referred to the number of experts who rated an item as “relevant.” And N referred to the total number of experts [31]. The content validity ratio varies between 1 and -1. The higher score indicates further agreement of members of the panel on the necessity of an item in an instrument. The acceptable range in the CVR depends on the number of experts on the panel which, in the present study, was based on the judgments of the 8 panelists. According to criterion values provided by Lawshe, CVR is equal to or larger than 0.49 for each item indicating an acceptable level of significance, therefore item was retained The Content Validity Index (CVI) is the mean score of those retained items having CVR≥0.49.
Floor and ceiling effects:
To determine the sensitivity of the DBQ, the bottom (Floor) effects and roof (Ceiling) effects were calculated. Terwee and colleagues considered that 50 participants could be used to adequately assess the floor and ceiling effects [32]. The floor and ceiling effects of more than 15% were considered to be significant [33]. Floor and ceiling effects were defined by calculating the number of respondents who scored the lowest status or the highest status, respectively, on the Arabic-DBQ [34].
Construct and factorial validity:
The two tests of Kaiser-Meyer-Olkin (KMO) and Bartlett were performed to evaluate the adequacy of data for factor analysis. To examine the appropriateness of the factor structure and to increase the validity, we split our data into 2 random samples, one for the exploratory factor analysis (EFA) and one for the confirmatory factor analysis (CFA). A sample of 580 drivers was used for conducting a factor analysis using the principal axis factoring (PAF) method with varimax rotation on retained items from the DBQ through SPSS to identify DBQ-L dimensions. Based on Eigenvalues >1 and scree plot, it was decided on the number of factors. A minimum factor loading of 0.40 per item is required to be included in the factor. A parallel analysis (PA) was performed to determine the number of components or factors to retain from factor analysis. To evaluate the internal consistency of the DBQ, Cronbach’s alpha reliability coefficients were calculated.
Confirmatory Factor Analyses (CFA) and psychometric analyses
The 45-item version of DBQ-L was subjected to confirmatory analysis using SPSS Amos to examine the good fit of the model to the Lebanese drivers and to confirm its factorial structure. Confirmatory factor analyses (CFA) were used to compare items organization of 2 models: (a) exploratory factor model, and (b) Lebanese adapted model. To identify the best fitting model, the following goodness-of-fit statistics [35] were calculated using IBM AMOS 24.0. The structural models were considered as a good fit to the data when Chi-squared value (χ 2 )/ degree of freedom(df) χ2/df < 5[36], normative fit index (NFI)>0.9, goodness of fit (GFI)>0.9, the comparative fit index (CFI)>0.9 and the root mean square residual (RMSEA)<0.08 [37]. In case of a poor fit, modification indexes were observed to identify additional parameters that improved the goodness of fit of the models. Covariances were permitted to be freely estimated and items that loaded 0.40 on two or more factors (cross-loading items) were eliminated in the modified models [38].
Questionnaire development
An anonymous Arabic, self-administrated questionnaire was developed in the Arabic language. It included a brief introduction of the study objectives and instructions on how to fill out the questionnaire as well.
It comprised mainly closed-ended questions and consisted of three main sections: (a) socio-demographic characteristics; (b) traffic-related variables; and (c) the DBQ scale.
The first section collected socio-demographic data of the participants, including gender, age, marital status, profile, education level, and residency.
The second section covered the topic of traffic variables where drivers were asked about their driving experience and their annual mileage. The third section of the questionnaire included the over mentioned adapted Arabic version of DBQ to objectively assess Lebanese aberrant drivers' behaviors. It consisted of 45 items categorized in four domains intended to measure the main forms of self-reported aberrant behaviors. Each item describes a particular aberrant driving behavior. Participants were requested to consider their driving behavior and to report the frequency of engagement in these behaviors on a six-point Likert scale (0= never; 1=hardly ever; 2=occasionally; 3=quite often; 4= frequently; 5=nearly all the time). Items scores were summed, with higher scores indicating more frequent aberrant behavior.
Data collection procedure:
Since no updated official list of drivers is available, potential respondents were recruited from public places such as shopping malls and parking stations via a face-to-face approach. Four well-trained data collectors were responsible for the dissemination of questionnaires. These investigators were students in the traffic major at the Lebanese Higher Technical School collectors. The entire procedure included the administration of the preliminary questions to assess first the eligibility of the participant based on the preset inclusion criteria. Before participation, study objectives and general instructions were delivered orally by data collectors to the eligible participants. Written informed consent was obtained from eligible respondents who were willing to complete a questionnaire. No reward was given to the drivers for their involvement in the study which was entirely voluntary. Drivers were also free to withdraw from the study at any time. Since the study was observational, participants' anonymity and confidentiality were respected. The completion of the questionnaire took around 10–15 minutes.
Statistical analysis
The collected data was entered and analyzed using the statistical software SPSS (Statistical Package for Social Sciences), version 24.0. Since missing data constituted < 10% of the total database, then it was not substituted. Before analysis, the distribution of each DBQ item was checked for normality. Descriptive analyses were done using counts and percentages for categorical variables and mean and standard deviation for continuous measures. A bivariate analysis was conducted using the ANOVA test to compare the means of DBQ subscales and the categorical variables. Pearson correlation was used for linear correlation between continuous variables. All variables that showed a p < 0.2 in the bivariate analysis were included in the model as independent variables. Four stepwise linear regressions were performed taking the DBQ subscales as the dependent variable and the sociodemographic and traffic variables as dependent measures. To perform linear regression, the aforementioned variables had to be modified and the number of categories was reduced (categories were merged in cases where there was no significant difference). Significance was set at p < 0.05.
Baseline information
A total of 1102 questionnaires were collected. The response rate was 80.71%. The age ranged from 18 to 82, with a mean age of 34.64 ± 12.33 years. Of the total, 68.4% of participants were male. Around 57.6% of them had an education degree higher than secondary level and were practicing a work other than “professional driver”. The mean driving experience was 13.46 ± 10.76. Moreover, more than 30% of respondents traveled a distance higher than 10000 kilometers annually (Table 1). Of note, 43.2% of surveyed drivers were involved in RTCs and 46.8% of them got penalized for traffic offenses in the previous three years (Table 1).
Table 1: Socio-demographics characteristics of surveyed Lebanese drivers (N=1102)
|
|
N |
% |
|
Gender |
|
|
|
Male |
754 |
68.40% |
|
Female |
348 |
31.60% |
|
Age (mean ± S.D) |
34.64± 12.33 |
|
|
Marital status |
|
|
|
Single |
466 |
42.30% |
|
Married |
576 |
52.30% |
|
Other (Widowed, Separated..) |
60 |
5.60% |
|
Education level |
|
|
|
Secondary and below |
468 |
42.40% |
|
University or above |
634 |
57.60% |
|
Occupation |
|
|
|
Student |
272 |
24.70% |
|
Professional driver (taxi) |
116 |
10.50% |
|
Working other than a professional driver |
630 |
57.20% |
|
Other (Not working, Housewife) |
84 |
7.60% |
|
Years of driving experience (mean± S.D) |
13.46±10.76 |
|
|
Annual mileage (Km) |
|
|
|
<6000 km |
524 |
47.5% |
|
6000-10000 km |
242 |
22.00% |
|
> 10000 km |
336 |
30.50% |
|
Road traffic crashes in the previous 3 years |
|
|
|
No |
626 |
56.80% |
|
Yes |
476 |
43.20% |
|
Number of RTCs in the previous 3 years |
0.76±1.488 |
|
|
Fines in the last 3 years |
|
|
|
No |
586 |
53.20% |
|
Yes |
516 |
46.80% |
|
Number of fines in the previous 3 years |
0.7±1.078 |
|
|
*S.D: Standard deviation, n frequency, % percentage, RTCs: road traffic crashes, km: Kilometers |
||
DBQ-L items:
All items had an appropriate intra-class correlation (ICC) with other items, and none of the 45 items was deleted. The high level of internal consistency for the 45 items, suggested that they were homogenous. The DBQ scale has good reliability (Cronbach’s α =0.892). Table 2 showed the mean and the standard deviation of all 45 items of the DBQ scale.
Table 2: Mean and standard deviation of the Lebanese version of the Driver behavior questionnaire items
|
All drivers (n=1102) |
|||
|
# |
Mean |
S. D |
|
|
DBQ1 |
Attempt to drive away from traffic lights |
2.52 |
0.708 |
|
DBQ2 |
Check your speedometer and discover that you are unknowingly traveling faster than the legal limit |
2.55 |
0.711 |
|
DBQ3 |
Lock yourself out of your car with the keys still inside |
2.52 |
0.687 |
|
DBQ4 |
Drive as fast along country roads at night on dipped lights as on a full beam |
2.55 |
0.731 |
|
DBQ5 |
Attempt to drive away without first having switched on the ignition |
2.54 |
0.584 |
|
DBQ6 |
Drive especially close or “flash” the car in front as a signal for that driver to go faster or get out of your way |
2.67 |
0.677 |
|
DBQ7 |
Forget where you left your car in a car parking |
2.53 |
0.714 |
|
DBQ8 |
Distracted or preoccupied, realize belatedly that the vehicle ahead has slowed, and have to slam on the brakes to avoid a collision |
2.55 |
0.637 |
|
DBQ9 |
Intend to switch on the windscreen wipers, but switch on the lights instead, or vice versa |
2.64 |
0.723 |
|
DBQ10 |
Turn left onto the main road into the path of an oncoming vehicle that you hadn’t seen, or whose speed you had misjudged |
2.58 |
0.577 |
|
DBQ11 |
Misjudge your gap in a car park and nearly (or actually) hit an adjoining vehicle |
2.57 |
0.524 |
|
DBQ12 |
‟Wake up‟ to realize that you have no clear recollection of the road along which you have just traveled |
2.50 |
0.680 |
|
DBQ13 |
Miss your exit on a motorway and have to make a lengthy detour |
2.42 |
0.624 |
|
DBQ14 |
Forget which gear you are currently in and have to check with your hand |
2.49 |
0.739 |
|
DBQ15 |
Stuck behind a slow-moving vehicle on a two-lane highway, you are driven by frustration to try to overtake in risky circumstances |
2.50 |
0.573 |
|
DBQ16 |
Intending to drive to destination A, you „wake up‟ to find yourself on route to B, where the latter is the more usual journey |
2.50 |
0.720 |
|
DBQ17 |
Take a chance and cross on lights that have turned red |
2.56 |
0.797 |
|
DBQ18 |
Angered by another driver's behavior, you give chase to give him/her a piece of your mind |
2.61 |
0.607 |
|
DBQ19 |
Have an aversion to a particular class of road user and indicate your hostility by whatever means you can |
2.61 |
0.547 |
|
DBQ20 |
Try to overtake without first checking your mirror, and then get hooted at by the car behind which has already begun its overtaking maneuver |
2.63 |
0.543 |
|
DBQ21 |
Deliberately disregard the speed limits late at night or very early in the morning |
2.52 |
0.715 |
|
DBQ22 |
Forget when your road tax/insurance expires and discover that you are driving illegally |
2.57 |
0.783 |
|
DBQ23 |
Lost in thought, you forget that your lights are on the full-beam until „flashed‟ by other motorists |
2.52 |
0.786 |
|
DBQ24 |
In a queue of vehicles turning left onto the main road, pay such close attention to the traffic approaching from the right that you nearly hit the car in front |
2.48 |
0.758 |
|
DBQ25 |
Drive back from a party, restaurant, or pub, even though you realize that you may be over the legal blood-alcohol limit |
2.52 |
0.814 |
|
DBQ26 |
Lost in thought or distracted, you fail to notice someone waiting and crossing a pelican crossing light that has just turned red |
2.59 |
0.634 |
|
DBQ27 |
Park in a no parking zone and risk a fine |
2.58 |
0.706 |
|
DBQ28 |
Misjudge the speed of the oncoming vehicle when overtaking |
2.56 |
0.685 |
|
DBQ29 |
Hit something when reversing that you had not previously seen |
2.50 |
0.718 |
|
DBQ30 |
Plan your route badly, so that you meet traffic congestion you could have avoided |
2.60 |
0.632 |
|
DBQ31 |
Overtake a single line of stationary or slow-moving vehicles, only to discover that they were queuing to get through a one-lane gap or roadwork lights |
2.57 |
0.549 |
|
DBQ32 |
Cut the corner on a right-hand turn and have to swerve violently to avoid an oncoming vehicle |
2.54 |
0.571 |
|
DBQ33 |
Get into the wrong lane at a roundabout or approaching a road junction |
2.53 |
0.671 |
|
DBQ34 |
Fail to read the signs correctly, and exit from a roundabout on the wrong road |
2.54 |
0.704 |
|
DBQ35 |
Fail to give way when a vehicle is signaling its intention to pull out |
2.53 |
0.687 |
|
DBQ36 |
Ignore „give way‟ signs, and narrowly avoid colliding with traffic having right of way |
2.53 |
0.614 |
|
DBQ37 |
Fail to check your mirror before pulling out, changing lanes, turning, etc. |
2.57 |
0.587 |
|
DBQ38 |
Attempt to overtake a vehicle that you hadn't noticed was signaling its intention to turn right |
2.55 |
0.574 |
|
DBQ39 |
Deliberately drive the wrong way down a deserted one-way street |
2.62 |
0.801 |
|
DBQ40 |
Disregard red lights when driving late at night along empty roads |
2.60 |
0.760 |
|
DBQ41 |
Drive with only „half-an-eye‟ on the road while using the mobile, changing radio channels, turning on the air conditioner… |
2.58 |
0.761 |
|
DBQ42 |
Get involved in unofficial „races‟ with other drivers |
2.56 |
0.760 |
|
DBQ43 |
Race‟ oncoming vehicles for a one-car gap on a narrow or obstructed road |
2.57 |
0.666 |
|
DBQ44 |
Brake too quickly on a slippery road and/or steer the wrong way in a skid |
2.55 |
0.606 |
|
DBQ45 |
Misjudge your crossing interval when turning right and narrowly miss a collision |
2.55 |
0.586 |
Flooring and Ceiling results
No floor or ceiling effects were detected since all the dimensions had low values of bottom and ceiling effects (< 15%). Ten respondents obtained the highest status scores of 10%.
Test-retest reliability:
The value of the Pearson correlation ((Pearson’s r) was satisfactory (0.82 ≥ 0.70). Furthermore, the mean difference and 95% LOA were calculated as 3.18 (-5.18 to 12.14) after 21 days.
Exploratory factor analysis
The original sample was split into 2 samples using random sampling. A sample of 580 drivers was used for conducting exploratory analysis. Initially, using varimax rotation, items converged over a solution of four factors that had eigenvalues over 1.0 and the examination of the scree plot suggested that the four-factor solution was the most interpretable one. (Factor 1 = Aggressive violations, Factor 2 = Simple/ordinary violation, Factor 3 = Errors and Factor 4 = Lapses). The result of the KMO test (KMO = 0.935) was satisfactory and that of Bartlett’s test was highly significant( p<0.001). The total variance explained was 75.36%, and the Cronbach’s alpha = 0.892. Parallel analysis (PA) revealed also that the four factors explained 75% of the total variance would have been selected. Both analyses yielded the same results in terms of the higher factor coefficient for each of the items selected. Table 3 presented the loadings of the 45 items on each of these four factors. The first factor explains 28.45% of the total variance and consists of 15 items measuring lapses. The second factor includes 15 items that tapped drivers’ errors and accounted for 21.94% of the variance. We labeled this factor as “errors”. The third factor includes 10 items about simple traffic violations and accounted for 17.41% of the variance and was named ‘‘Ordinary violations’’. The fourth factor included 5 items that tapped aggressive reactions to driving situations. This factor accounted for 7.55% of the variance and was named ‘‘Aggressive violations’’. Of note, one item related to ordinary violation (DBQ33: Cut the corner on a right-hand turn and have to swerve violently to avoid an oncoming vehicle) loaded in the “aggressive violation component.
|
Table 3: Four-factor solution of DBQ-L items, eigenvalues, and Cronbach’s alpha coefficients, and variance of the DBQ subscales both among Lebanese drivers |
|||||
|
# |
|
DBQ Components |
|||
|
Lapses |
Errors |
Ordinary violations |
Aggressive violations |
||
|
DBQ34 |
Fail to read the signs correctly, and exit from a roundabout on the wrong road |
0.923 |
|
|
|
|
DBQ12 |
‟Wake up‟ to realize that you have no clear recollection of the road along which you have just traveled |
0.899 |
|
|
|
|
DBQ33 |
Get into the wrong lane at a roundabout or approaching a road junction |
0.896 |
|
|
|
|
DBQ3 |
Lock yourself out of your car with the keys still inside |
0.886 |
|
|
|
|
DBQ23 |
Lost in thought, you forget that your lights are on the full-beam until „flashed‟ by other motorists |
0.884 |
|
|
|
|
DBQ23 |
Check your speedometer and discover that you are unknowingly traveling faster than the legal limit |
0.884 |
|
|
|
|
DBQ13 |
Miss your exit on a motorway and have to make a lengthy detour |
0.866 |
|
|
|
|
DBQ1 |
Attempt to drive away from traffic lights |
0.862 |
|
|
|
|
DBQ24 |
In a queue of vehicles turning left onto the main road, pay such close attention to the traffic approaching from the right that you nearly hit the car in front |
0.851 |
|
|
|
|
DBQ22 |
Forget when your road tax/insurance expires and discover that you are driving illegally |
0.847 |
|
|
|
|
DBQ16 |
Intending to drive to destination A, you „wake up‟ to find yourself on route to B, where the latter is the more usual journey |
0.842 |
|
|
|
|
DBQ29 |
Hit something when reversing that you had not previously seen |
0.840 |
|
|
|
|
DBQ14 |
Forget which gear you are currently in and have to check with your hand |
0.826 |
|
|
|
|
DBQ35 |
Fail to give way when a vehicle is signaling its intention to pull out |
0.807 |
|
|
|
|
DBQ9 |
Intend to switch on the windscreen wipers, but switch on the lights instead, or vice versa |
0.805 |
|
|
|
|
DBQ11 |
Misjudge your gap in a car park and nearly (or actually) hit an adjoining vehicle |
|
0.944 |
|
|
|
DBQ10 |
Turn left onto the main road into the path of an oncoming vehicle that you hadn’t seen, or whose speed you had misjudged |
|
0.927 |
|
|
|
DBQ26 |
Lost in thought or distracted, you fail to notice someone waiting and crossing a pelican crossing light that has just turned red |
|
0.917 |
|
|
|
DBQ45 |
Misjudge your crossing interval when turning right and narrowly miss a collision |
|
0.902 |
|
|
|
DBQ5 |
Attempt to drive away without first having switched on the ignition |
|
0.842 |
|
|
|
DBQ38 |
Attempt to overtake a vehicle that you hadn't noticed was signaling its intention to turn right |
|
0.839 |
|
|
|
DBQ28 |
Misjudge the speed of the oncoming vehicle when overtaking |
|
0.825 |
|
|
|
DBQ7 |
Forget where you left your car in a car parking |
|
0.812 |
|
|
|
DBQ20 |
Try to overtake without first checking your mirror, and then get hooted at by the car behind which has already begun its overtaking maneuver |
|
0.807 |
|
|
|
DBQ36 |
Ignore „give way‟ signs, and narrowly avoid colliding with traffic having right of way |
|
0.802 |
|
|
|
DBQ44 |
Brake too quickly on a slippery road and/or steer the wrong way in a skid |
|
0.802 |
|
|
|
DBQ8 |
Distracted or preoccupied, realize belatedly that the vehicle ahead has slowed, and have to slam on the brakes to avoid a collision |
|
0.789 |
|
|
|
DBQ37 |
Fail to check your mirror before pulling out, changing lanes, turning, etc. |
|
0.787 |
|
|
|
DBQ31 |
Overtake a single line of stationary or slow-moving vehicles, only to discover that they were queuing to get through a one-lane gap or roadwork lights |
|
0.786 |
|
|
|
DBQ30 |
Plan your route badly, so that you meet traffic congestion you could have avoided |
|
0.775 |
|
|
|
DBQ40 |
Disregard red lights when driving late at night along empty roads |
|
|
0.953 |
|
|
DBQ17 |
Take a chance and cross on lights that have turned red |
|
|
0.941 |
|
|
DBQ4 |
Drive as fast along country roads at night on dipped lights as on full beam |
|
|
0.931 |
|
|
DBQ21 |
Deliberately disregard the speed limits late at night or very early in the morning |
|
|
0.929 |
|
|
DBQ41 |
Drive with only „half-an-eye‟ on the road while using the mobile, changing radio channel, turning on A.C etc. |
|
|
0.925 |
|
|
DBQ25 |
Drive back from a party, restaurant, or pub, even though you realize that you may be over the legal blood-alcohol limit |
|
|
0.923 |
|
|
DBQ39 |
Deliberately drive the wrong way down a deserted one-way street |
|
|
0.921 |
|
|
DBQ42 |
Get involved in unofficial „races‟ with other drivers |
|
|
0.886 |
|
|
DBQ27 |
Park in a no parking zone and risk a fine |
|
|
0.867 |
|
|
DBQ43 |
Race‟ oncoming vehicles for a one-car gap on a narrow or obstructed road |
|
|
0.744 |
|
|
DBQ33 |
Cut the corner on a right-hand turn and have to swerve violently to avoid an oncoming vehicle |
|
|
|
0.908 |
|
DBQ18 |
Angered by another driver's behavior, you give chase to give him/her a piece of your mind |
|
|
|
0.864 |
|
DBQ15 |
Stuck behind a slow-moving vehicle on a two-lane highway, you are driven by frustration to try to overtake in risky circumstances |
|
|
|
0.858 |
|
DBQ19 |
Have an aversion to a particular class of road user and indicate your hostility by whatever means you can |
|
|
|
0.854 |
|
DBQ6 |
Drive especially close or „flash‟ the car in front as a signal for that driver to go faster or get out of your way |
|
|
|
0.682 |
|
|
Eigenvalue |
12.80 |
9.87 |
7.84 |
3.40 |
|
|
Cronbach alpha |
0.73 |
0.81 |
0.79 |
0.680 |
|
|
Variance (%) |
28.445 |
21.942 |
17.416 |
7.553 |
|
|
Note: Factor loadings below .30 were omitted for the sake of clarity, 75.36% of the variance was explained, Extraction Method: Principal Component Analysis. Rotation Method: Varimax with Kaiser Normalization. |
||||
Reliability analyses and inter-correlations between factors
Table 4 showed the reliability estimates. The α-values obtained from subscales ranged between 0.68 and 0.814, indicating good internal consistency. The overall reliability of the DBQ-L scale was good (α=0.85). Similarly, all the subscales were reliable. Alpha reliabilities for aggressive violations, ordinary violations, errors, and lapses subscales were 0.68, 0.79, 0.81, and 0.73 respectively. Skewness [.075, 0.967] and kurtosis [0.521, 0.714] estimates for the four factors allowed the use of parametrical correlational analyses. Males ranked higher than females in the subscales of DBQ-L related to violations in particular the one related to aggressive violations. Of the four subscales, the highest item mean was for aggressive violations followed by ordinary violations, lapses, and then errors. The correlation between DBQ-L factors ranged from 0.435 to 0.572, most of them being highly statistically significant (p <0 .001) with a moderate correlation (Annex, Table A2).
Table 4: Summary statistics for the DBQ-L subscales
|
|
|
All drivers (n=1102) |
Male (n=754) |
Female (n= 348) |
|||||||||||
|
# |
Number of items |
Item mean |
Scale Mean |
S.D |
Min |
Max |
Alpha α |
Skewness |
Kurtosis |
Mean |
S.D |
Mean |
S.D |
||
|
Aggressive violations |
5 |
2.99 |
14.98 |
2.04 |
11 |
24 |
0.68 |
0.075 |
0.654 |
15.03 |
2.11 |
14.8 |
1.9 |
||
|
Ordinary violations |
10 |
2.65 |
26.56 |
4.72 |
19 |
49 |
0.79 |
0.213 |
0.634 |
26.46 |
4.69 |
25.8 |
4.79 |
||
|
Errors |
15 |
2.21 |
33.27 |
5.9 |
23 |
57 |
0.81 |
0.967 |
0.589 |
33.23 |
5.84 |
33.34 |
6.08 |
||
|
Lapses |
15 |
2.52 |
37.8 |
4.8 |
28 |
65 |
0.73 |
0.579 |
0.521 |
37.84 |
4.97 |
37.73 |
4.43 |
||
|
DBQ-L scale |
45 |
2.51 |
113.14 |
14.5 |
91 |
188 |
0.85 |
0.336 |
0.714 |
113.0 |
14.8 |
113.3 |
14.89 |
||
|
N.B: S.D refers to standard deviation, alpha referred to Cronbach alpha, min for minimum and max for maximum, Vehicle/Env refers to the vehicle and environment dimension |
|||||||||||||||
Confirmatory factor analysis
The default model corresponding to the result of the exploratory analysis was not acceptable based on the resulted fit indicators (Table 5). Based on modification indices, a high correlation between errors (e2 and e1, e3 and e4, e1 and e5) was found. Hence, several model specifications were performed based to adjust the model to achieve a better fit. The latter included drawing eight covariates between the following errors e8-e15, e1-e9, e16-e18, e18-e30, e27-e29, e21-e22, e16-e17 and e17-e18 respectively. The model used in confirmatory factor analyses is presented schematically in the annexed figure FA1. This resulted in a better fitting than the default model. The model fit measures of the data analysis were as follow χ2=3929.5, χ2/df = 4.221 <5; NFI = 0.912>0.9, CFI = 0.907>0.9, GFI = 0.906>0.9, AGFI = 0.902>0.9, RMSEA = 0.071 <0.08 therefore suggesting a reasonable model fit.
Table 5: Model fit indicators for DBQ-L
|
χ2 |
χ2/df |
NFI |
CFI |
GFI |
AGFI |
SRMR |
RMSEA |
|
|
Model based on EFA result |
5939.232 |
6.325 |
0.822 |
0.845 |
0.703 |
0.736 |
0.0456 |
0.097 |
|
Adapted model |
3929.514 |
4.221 |
0.912 |
0.907 |
0.906 |
0.902 |
0.048 |
0.071 |
DBQ-L and socio-demographic variables
As result of the bivariate analysis, gender differences revealed that men reported a higher tendency to commit aggressive violations (p=0.048 <0 .05), and ordinary violation (p=0.049 < 0.05), than female. Cohen’s d was very small indicating a limited effect size. However, errors and lapses did not reveal significant differences between men and women.
Moreover, the four dimensions of aberrant behaviors (aggressive violations, ordinary violations, errors, and lapses) were significantly affected by the occupation and the educational level of the surveyed drivers. Drivers with a secondary educational level or less were more prone to involve in aggressive and ordinary violations, errors, and lapses behavior than their counterparts with the highest educational level. In regards to the relations between DBQ-L, age and driving experience, Pearson correlations revealed that ordinary violation (r = -.559, p < .001) and lapses (r = -.544, p < .001), were negatively correlated with age. Younger age is related to an increase in ordinary violations and lapse subscales. However, errors (r =.442, p < .001) were positively associated with age and had a different pattern where errors decreased for young drivers. In addition, aggressive violations (p<0.001), ordinary violations (p<0.001), errors (p<0.001) and lapses (p=0.016<0.001) increased with annual mileage. Despite that violations(r=-0.142, p<0.001) and lapses (r=-0.135, p<0.001) decreased with driving experience, errors were found increasing with driving experience (r=0.161, p<0.001) (Annex Table A3).
Multivariable analysis:
In terms of aggressive violations, older age, female gender, large driving experience and high education level were found negatively associated with the Lebanese drivers involvement in aggressive violations. However, increased annual mileage and being professional driver were both positively associated with aggressive violations among these drivers. Similar associations were found between the overmentioned sociodemographic characteristics and ordinary violations. As for errors , professional and older drivers and those who had an increased mileage were more prone to be involved in erroneous behavior when driving. However, larger driving experience and higher education level were found negatively associated with such an erroneous behavior. Lapses were also found positively associated with increased annual mileage and being professional drivers and negatively associated with larger driving experience and higher education level.
Table 6: Linear regression for each of the DBQ-L subscale
|
|
|
Standardized Coefficients Beta |
t |
Sig. |
F |
sig |
|
|
Aggressive Violations |
Age |
-0.081 |
-1.499 |
0.034 |
4.891 |
<0.001 |
|
|
|
Gender (female vs male) |
-0.003 |
-0.109 |
0.024 |
|||
|
|
Driving experience (large vs small) |
-0.010 |
-0.184 |
0.044 |
|||
|
|
Education level (high vs low) |
-0.029 |
-0.909 |
0.363 |
|||
|
|
Annual mileage |
0.058 |
1.785 |
0.045 |
|||
|
|
Occupation (Professional driver vs nonprofessional driver) |
0.126 |
3.902 |
<0.001 |
|||
|
Ordinary violations |
Age |
-0.052 |
-1.003 |
0.016 |
20.856 |
<0.001 |
|
|
|
Gender (female vs male) |
-0.069 |
-2.237 |
0.025 |
|||
|
|
Driving experience (large vs small) |
-0.163 |
-3.026 |
0.003 |
|||
|
|
Education level (high vs low) |
-0.071 |
-2.309 |
0.021 |
|||
|
|
Annual mileage |
0.128 |
4.061 |
<0.001 |
|||
|
|
Occupation (Professional driver vs nonprofessional driver) |
0.227 |
7.304 |
<0.001 |
|||
|
Errors |
Age |
0.019 |
0.357 |
0.021 |
23.247 |
<0.001 |
|
|
|
Gender (female vs male) |
0.031 |
1.004 |
0.315 |
|||
|
|
Driving experience (large vs small) |
-0.236 |
-4.410 |
<0.001 |
|||
|
|
Education level (high vs low) |
-0.087 |
-2.849 |
0.004 |
|||
|
|
Annual mileage |
0.075 |
2.396 |
0.017 |
|||
|
|
Occupation (Professional driver vs nonprofessional driver) |
0.244 |
7.874 |
0.000 |
|||
|
Lapses |
Age |
0.054 |
1.005 |
0.315 |
10.855 |
<0.001 |
|
|
|
Gender (female vs male) |
-0.012 |
-0.375 |
0.708 |
|||
|
|
Driving experience (large vs small) |
-0.248 |
-4.502 |
<0.001 |
|||
|
|
Education level (high vs low) |
-0.089 |
-2.836 |
0.005 |
|||
|
|
Annual mileage |
0.104 |
3.215 |
0.001 |
|||
|
|
Occupation (Professional driver vs nonprofessional driver) |
0.098 |
3.075 |
0.002 |
|||
Given that human factors including driver behavior are one of the major contributors to RTCs[5-7], improving driver behavior is one of the overriding goals of road safety. DBQ is a prominent measurement scale based on self-reported behavior which has been used in many countries to usefully and efficiently examine drivers’ aberrant behaviors despite minor cultural nuances. Thus, having an accurately translated and rigorously validated tool to assess driver aberrant behavior and its associated risk in Lebanon and Arabic-speaking communities is extremely needed.
Our findings indicate that the DBQ-L has adequate psychometric properties, revealing good reliabilities of all subscales. The DBQ-L factorial structure was comparable to the structure proposed by Harrison et al. [39], revealing four dimensions namely aggressive violations, ordinary violations, errors, and lapses. Despite small differences in the theoretical four-factor structure of the DBQ, the overall factor structure has been confirmed in different samples [8, 16, 40, 41]. The four factors had also acceptable reliabilities. The lowest one was shown in the subscale related to “aggressive violation”. The DBQ-L multidimensional structure is statistically supported by satisfactory fit indices when running a confirmatory analysis. The factorial solution of the DBQ-L also matches the meaning of the four broad facets of the 28-items version among Australian drivers [42]. However, other studies have established diverse item and factor structures that fit better their data using other forms of the DBQ such as the 32-item four-factor Swedish DBQ [43]; the 22-item four-factor Persian DBQ [43]; and the 50-item four-factor solution based on data from drivers in Australia[12]. Such discrepancies in factor configuration could be attributed to the misunderstanding and different interpretations of certain items by participants.
Of the four subscales, the highest item mean score was for aggressive violations followed by ordinary violations, lapses, and then errors. Our results were inconsistent with the findings of a study conducted among Australian drivers that showed that the highest mean score was for lapses followed by aggressive violations and violations. The scores for the current sample were higher than those found by Harrison et al. [39]. In addition, the frequencies of the DBQ responses among Lebanese drivers ranged from “never” to “hardly ever” and rarely “occasionally” which was in line with the findings of previous studies [12, 19, 40], However, the Lebanese drivers, in general, scored higher in all DBQ items than the Arab Gulf drivers (Qatari and United Arab Emirati) [16], much higher than British [17], Australian [12], Finnish and Dutch [8], and even higher than Greek and Turkish (e.g., Ozkan et al., 2006b ¨ ) drivers. This is coherent with such a tendency found in Ozkan et al. (2006b) ¨ study [23]. The DBQ factor structure might get more indistinct while the occurrence of the behaviors gets higher when going far from West/North to East/South. The interpretation of errors and aggressive violations factors differs from country to country. Shinar et al. suggested that aggressive driving would be highly influenced by cultural context [44], violations are affected by contextual factors [17] and errors were mainly related to cognitive processes of the individual or the lack of supportive social and physical context [23].
No floor or ceiling effects were detected for the DBQ-L. This indicates a good distribution of scores for the -DBQ-L, good content validity, and additional support of adequate reliability In the present study, the obtained internal consistency indicates good homogeneity of items of the adapted questionnaire which is a significant feature of a questionnaire, especially if all items are assessing the same concept. The test-retest reliability of the DBQ-L was found good as well.
Gender differences revealed that men reported a higher tendency to commit aggressive violations and ordinary violations. Similar results were found in previous studies that also reported that men exhibited violation behaviors more frequently than women [12, 17, 19, 39, 45] and when considered separately, aggressive violations [39]. In our study, no significant difference across gender was found concerning self-reported errors. This was consistent with the findings of the study conducted by Reason et al [17]. Of note, some studies have reported a higher tendency in females to report lapses [17, 24, 45, 46] which were not consistent with our study.
Younger age increased annual mileage, and less driving experience were found to be related to an increased violation score. Hence, violations decreased with age and increased with larger driving experience. Similar results were reported by a study conducted among Australian drivers [42]. However, the increase of violations with experience could be anticipated since beginner drivers lack confidence in their driving performance and are still in the phase of building confidence, hence avoiding perpetrating traffic violations [47]. When they get more experience, they feel more familiar with the traffic environment and increased traffic violations.
In regards to aggressive violations and lapses, they increased with the distance traveled annually. Remarkably, lapses increased also with the younger age. However, drivers who have more driving exposure are more prone to aberrant behavior including all its aspects (violations, errors….) [17, 47]. Higher education was found to be related to a decreased aberrant behavior (violations, errors..) Drivers with a secondary educational level or less were more prone to involve in aggressive and ordinary violations, errors, and lapses behavior than their counterparts with the highest educational level. Being a professional driver increases the likelihood to report violations, errors, and lapses. This could be understood under the umbrella of high exposure to traffic and longer driving periods.
Errors showed a different pattern where it was found that errors decreased for young drivers and increased with age and with larger driver experience. Compared to a previous study, errors were found to decrease for young drivers but remained quite unchangeable with age within the older samples [42]. Concerning the confounding roles of age and experience, a more specific insight was provided by Wells and al., their results showed that compared to drivers with 6, 12, 24, and 36 months of experience, older drivers had consistently lower violation scores than the younger drivers. In this case, the effects of experience on errors could be explained by the fact that novice drivers are still in the learning phase, hence improving their skills and reducing their driving errors. A driving simulator study about errors and violations during training of pre-licensed drivers showed similar results [47]. At a later age, acquired driving skills can be weakened and a further increase in errors could be expected.
Strengths and limitations of the study:
This study is the first to examine the reliability and the factor structure of the DBQ in a large, diverse sample of Lebanese drivers and to examine its relationship with socio-demographic characteristics. The validated scale used had very good to excellent reliability, and the sample size had adequate power for conducting the factorial analysis. However, several limitations related to methodological and qualitative biasing sources should be acknowledged. First, the study's cross-sectional design in nature does not allow deriving causal relationships of the main associations. Despite weighing over gender, age, and geographical regions, selection bias might be present given the use of the convenience sampling technique as the sample included a high percentage of university-educated participants, therefore, there are concerns related the representativeness of the sample. Besides, the data were based merely on drivers' self-reported behavior without combination with any observation. Similar to research in this area, concerns remain regarding the reliability of the reported behaviour due to the self-report methodology, such as the propensity of participant to provide socially desirable responses. [48], particularly for answers related to positive and aggressive driving behaviors. Of note, Lajunen and Summala directly examined DBQ responses and found that the influence of social desirability on responses is not considerable and even less when completed in privacy [8]. Also, Sullman and Taylor (2010) have shown that social desirability is only a minor issue in DBQ responses and aberrant driving behaviors. Moreover, it has been found that the driving behaviors measured with DBQ are strongly linked with the observed driver behaviors on the road (Rolls et al., 1991). Given that participants in the current study completed the questionnaires anonymously and were assured of the confidentiality of their responses, thus, the impact of social desirability on responses is expected to be minimal. Future studies linking self-report data with participants’ driving records are recommended.Additionally, the target group of the present study was Lebanese drivers. Although Syrian and Palestinian refugees might represent a notable proportion of the drivers' population in our country, the present study was restricted to the Lebanese drivers because of difficulties in data collection and probable differences in the behavior of drivers originating from different countries though they are sharing the same traffic environment. Finally, residual confounding is still possible. In summary, further research is required to establish the concurrent and predictive validity of the DBQ scale for the Lebanese setting as well actorial invariance of the DBQ across gender, age groups and driver profile. Endeavours into identifying additional factors associated with traffic offences and crashes involvement among both private and professional drivers can only benefit current road safety initiatives.
Implications:
This study provides important information that could help in implementing effective road safety strategies to reduce violations and erroneous behaviors. Effective countermeasures targeting aberrant driver behavior such as strict law enforcement, driver education, booster driving sessions, counseling, and training programs in multifaceted driving situations are recommended. to avoiding driver’s violations and distractions, and reducing their involvement in RTCs. Further studies examining the extent to which each of the DBQ subscales can predict crash involvement are recommended. This will deliver key information for road safety strategies to reduce violations and aberrant behaviors.
The Lebanese version of DBQ, with its confirmed multifactorial structure, is a reliable and valid instrument appropriate for use in the Lebanese driving context to measure aberrant driving behaviors. Further studies are needed to examine the invariance of DBQ structure across age, gender, and occupation of Lebanese drivers and its ability to predict RTCs in each category.
Ethics approval and consent to participate
All methods were performed in accordance with the ethical standards as laid down in the Declaration of Helsinki and its later amendments. As research involving the use of survey procedures are exempted from ethical approval in Lebanon and the risks of this research are no more than minimal, this study was exempted from ethical approval by the Ministry of Public Health. The study design assured adequate protection of study participants, and neither included clinical data about patients nor configured itself as a clinical trial. Written informed consent was obtained for each participant. They were reassured that their participation is voluntary and that they were free to withdraw at any time. In addition, all information was gathered anonymously and handled confidentially.
Funding
No funding was received.
Availability of data and materials
Data are available from the corresponding authors upon reasonable request.
Consent for publication
Not applicable.
Competing interests
The authors declare that they have no competing interests.
Acknowledgments
The authors acknowledge all drivers who participated in this study as well as thestudent of traffic major in IPNET Lebanese Higher Institute of Technical & Professional, Bir Hassan who helped us in the data collection.
Availability of data and materials:
The datasets used and analyzed during the current study are available from the corresponding author on reasonable request.
Authors’ contributions
Conception and design: D.Y
Analysis and interpretation of the data: D.Y and L.A.A
Drafting of the article: D.Y,
Critical revision of the article for important intellectual content: D.Y, P.S and LR.S
Final approval of the article: D.Y, P.S, L.A.A and L.R.S
Authors information
Dalal Youssef, MSc, MPH, RSM, Ph.D., Bordeaux Research Center for Population Health, Institut de santé publique, d'épidémiologie et de développement (ISPED), Bordeaux University, France & Clinical trial Program, Ministry of Public Health, Beirut, Lebanon & Lebanese Higher Institute of Technical & Professional, Bir Hassan, Ministry of Education, Beirut, Lebanon ORCID IDs: https://orcid.org/0000-0003-3085-6849 Email: [email protected]; [email protected]; Phone number: +96171239633
Pascale Salameh, PharmD, PhD, Lebanese university
Linda Abou Abbas, MPH, PhD, Ministry of Public Health, Epidemiological surveillance unit
Louis-Rashid Salmi, MD, PhD, ISPED Bordeaux
DBQ
Driver Behavior Questionnaire
DBQ-L
Lebanese version of the Driver Behavior Questionnaire
EMR
Eastern Mediterranean Region
RTI
Road traffic injuries
C.I
Confidence interval
RTCs
Road traffic crashes
P
P-value
M
Mean
SD
Standard deviation
SPSS
Statistical Package for Social Sciences
CFA
Confirmatory factor analysis
EFA
Exploratory factor analysis
KMO
Kaiser–Meyer–Olkin
CVI
Content Validity Index
CVR
Content validity ratio
RMSEA
Root mean square error of approximation
SRMR
Standardized root mean square residual
GFI
Goodness-of-Fit Index
CFI
Comparative Fit Index
NFI
Normed Fit Index
α
Cronbach’s alpha
r
coefficient de correlation r
d
Cohen’d
IPNET
Lebanese Higher Institute of Technical & Professional
- Organization, W.H., Global status report on road safety 2015. 2015: World Health Organization.
- Rumar, K., The role of perceptual and cognitive filters in observed behavior, in Human behavior and traffic safety. 1985, Springer. p. 151–170.
- Petridou, E. and M. Moustaki, Human factors in the causation of road traffic crashes. European journal of epidemiology, 2000. 16(9): p. 819–826.
- Motevalian, S.A., et al. Validation of a persian version of motorcycle rider behavior questionnaire. in Annals of Advances in Automotive Medicine/Annual Scientific Conference. 2011. Association for the Advancement of Automotive Medicine.
- De Ona, J., et al., Analysis of traffic accidents on rural highways using Latent Class Clustering and Bayesian Networks. Accident Analysis & Prevention, 2013. 51: p. 1–10.
- Lambert-Bélanger, A., et al., Aggressive driving behaviour in young drivers (aged 16 through 25) involved in fatal crashes. Journal of safety research, 2012. 43(5–6): p. 333–338.
- Tractinsky, N., E.S. Ram, and D. Shinar, To call or not to call—That is the question (while driving). Accident Analysis & Prevention, 2013. 56: p. 59–70.
- Lajunen, T., D. Parker, and H. Summala, The Manchester driver behaviour questionnaire: a cross-cultural study. Accident Analysis & Prevention, 2004. 36(2): p. 231–238.
- Mattsson, M., et al., Measurement invariance of the Driver Behavior Questionnaire across samples of young drivers from Finland and Ireland. Accident Analysis & Prevention, 2015. 78: p. 185–200.
- Smorti, M. and S. Guarnieri, Do aggressive driving and negative emotional driving mediate the link between impulsiveness and risky driving among young Italian drivers? The Journal of social psychology, 2016. 156(6): p. 669–673.
- Lucidi, F., et al., Personality and attitudes as predictors of risky driving among older drivers. Accident Analysis & Prevention, 2014. 72: p. 318–324.
- Blockey, P. and L.R. Hartley, Aberrant driving behaviour: errors and violations. Ergonomics, 1995. 38(9): p. 1759–1771.
- Özkan, T. and T. Lajunen, A new addition to DBQ: Positive driver behaviours scale. Transportation Research Part F: Traffic Psychology and Behaviour, 2005. 8(4–5): p. 355–368.
- Sümer, N., Personality and behavioral predictors of traffic accidents: testing a contextual mediated model. Accident Analysis & Prevention, 2003. 35(6): p. 949–964.
- Obriot-Claudel, F. and C. Gabaude, The Driver Behaviour Questionnaire: A French study applied to elderly drivers. Proceedings of the ICTTP, 2004.
- Bener, A., T. Özkan, and T. Lajunen, The driver behaviour questionnaire in Arab gulf countries: Qatar and United Arab Emirates. Accident Analysis & Prevention, 2008. 40(4): p. 1411–1417.
- Reason, J., et al., Errors and violations on the roads: a real distinction? Ergonomics, 1990. 33(10–11): p. 1315–1332.
- Ranney, T.A., Models of driving behavior: a review of their evolution. Accident analysis & prevention, 1994. 26(6): p. 733–750.
- Aberg, L. and P.-A. Rimmo, Dimensions of aberrant driver behaviour. Ergonomics, 1998. 41(1): p. 39–56.
- Mattsson, M., Investigating the factorial invariance of the 28-item DBQ across genders and age groups: an exploratory structural equation modeling study. Accident Analysis & Prevention, 2012. 48: p. 379–396.
- Martinussen, L.M., et al., Age, gender, mileage and the DBQ: The validity of the Driver Behavior Questionnaire in different driver groups. Accident Analysis & Prevention, 2013. 52: p. 228–236.
- Rowe, R., et al., Measuring errors and violations on the road: A bifactor modeling approach to the Driver Behavior Questionnaire. Accident Analysis & Prevention, 2015. 74: p. 118–125.
- Özkan, T., et al., Cross-cultural differences in driving behaviours: A comparison of six countries. Transportation Research Part F: Traffic Psychology and Behaviour, 2006. 9(3): p. 227–242.
- Kobeissy, F. and L. Carnis, The excuse is worse than the fault: It's time now to improve road safety in Lebanon. Case Studies on Transport Policy, 2021. 9(2): p. 500–510.
- Ramadan, T., Beirut, Car Accidents Caused 487 Deaths In Lebanon During 2019, https://www.beirut.com/l/59196. 2019.
- Martinussen, L.M., Facilitating improved road safety based on increased knowledge about driving behaviour and profiling sub-groups of drivers. Kgs. Lyngby: Technical University of Denmark. 2013: p. 123.
- Beaton, D.E., et al., Guidelines for the process of cross-cultural adaptation of self-report measures. Spine (Phila Pa 1976), 2000. 25(24): p. 3186–91.
- Comrey, A.L. and H.B. Lee, A first course in factor analysis. 2013: Psychology press.
- Rossiter, J.R., Content Validity of Measures of Abstract Constructs in Management and Organizational Research. British Journal of Management, 2008. 19(4): p. 380–388.
- Ayre, C. and A.J. Scally, Critical Values for Lawshe’s Content Validity Ratio. Measurement and Evaluation in Counseling and Development, 2014. 47(1): p. 79–86.
- Lawshe, C.H., A quantitative approach to content validity. Personnel psychology, 1975. 28(4): p. 563–575.
- Terwee, C.B., et al., Quality criteria were proposed for measurement properties of health status questionnaires. J Clin Epidemiol, 2007. 60(1): p. 34–42.
- Flora, D.B., C. Labrish, and R.P. Chalmers, Old and new ideas for data screening and assumption testing for exploratory and confirmatory factor analysis. Front Psychol, 2012. 3: p. 55.
- Adamu, A.S., et al., Cross-cultural Adaptation and Validation of the Hausa Version of the Oswestry Disability Index 2.1a for Patients With Low Back Pain. Spine (Phila Pa 1976), 2019. 44(18): p. E1092-e1102.
- Hair, J., Anderson, RE/Tatham, RL/Black, WC (1998): Multivariate data analysis. PrenticeYHall, Upper saddle river, NJ, 2005.
- Schermelleh-Engel, K., H. Moosbrugger, and H. Müller, Evaluating the fit of structural equation models: Tests of significance and descriptive goodness-of-fit measures. Methods of psychological research online, 2003. 8(2): p. 23–74.
- MacCallum, R.C., M.W. Browne, and H.M. Sugawara, Power analysis and determination of sample size for covariance structure modeling. Psychological methods, 1996. 1(2): p. 130.
- Byrne, B.M. and S.M. Stewart, Teacher's corner: The MACS approach to testing for multigroup invariance of a second-order structure: A walk through the process. Structural equation modeling, 2006. 13(2): p. 287–321.
- Harrison, W.A., Reliability of the Driver Behaviour Questionnaire in a sample of novice drivers. 2009, Austroads.
- Mesken, J., T. Lajunen, and H. Summala, Interpersonal violations, speeding violations and their relation to accident involvement in Finland. Ergonomics, 2002. 45(7): p. 469–483.
- Gras, M.E., et al., Spanish drivers and their aberrant driving behaviours. Transportation research part F: traffic psychology and behaviour, 2006. 9(2): p. 129–137.
- Stephens, A. and M. Fitzharris, Validation of the driver behaviour questionnaire in a representative sample of drivers in Australia. Accident Analysis & Prevention, 2016. 86: p. 186–198.
- Rimmö, P.-A., Aberrant driving behaviour: homogeneity of a four-factor structure in samples differing in age and gender. Ergonomics, 2002. 45(8): p. 569–582.
- Shinar, D., Aggressive driving: the contribution of the drivers and the situation. Transportation Research Part F: traffic psychology and behaviour, 1998. 1(2): p. 137–160.
- Parker, D., et al., Elderly drivers and their accidents: the Aging Driver Questionnaire. Accident Analysis & Prevention, 2000. 32(6): p. 751–759.
- Xie, C.-q. and D. Parker, A social psychological approach to driving violations in two Chinese cities. Transportation Research Part F: Traffic Psychology and Behaviour, 2002. 5(4): p. 293–308.
- de Winter, J.C. and D. Dodou, The Driver Behaviour Questionnaire as a predictor of accidents: A meta-analysis. Journal of safety research, 2010. 41(6): p. 463–470.
- af Wåhlberg, A.E., Social desirability effects in driver behavior inventories. Journal of safety research, 2010. 41(2): p. 99–106.
No competing interests reported.