Elsevier

Accident Analysis & Prevention

Volume 40, Issue 6, November 2008, Pages 2058-2064
Accident Analysis & Prevention

An application of the theory of planned behaviour to truck driving behaviour and compliance with regulations

https://doi.org/10.1016/j.aap.2008.09.002Get rights and content

Abstract

A questionnaire study was conducted with truck drivers to help understand driving and compliance behaviour using the theory of planned behaviour (TPB). Path analysis examined the ability of the TPB to explain the direct and indirect factors involved in self-reported driving behaviour and regulation compliance. Law abiding driving behaviour in trucks was related more to attitudes, subjective norms and intentions than perceived behavioural control. For compliance with UK truck regulations, perceived behavioural control had the largest direct effect. The differing results of the path analyses for driving behaviour and compliance behaviour suggest that any future interventions that may be targeted at improving either on-road behaviour or compliance with regulations would require different approaches.

Introduction

The importance of occupational health has increased dramatically in recent years, with growing public and political pressure for corporate and individual accountability (HSE, 2004). In countries such as the United States, Australia and the United Kingdom road traffic fatalities are one of the main contributors to work-related fatalities, leading to a significant cost at both an individual and societal level. For example, in the United States, fatal highway incidents have consistently been the leading cause of fatal work-related events over a 15-year period from 1992 to 2006, accounting for approximately one in four fatal work injuries (Bureau of Labor Statistics, 2007). In the UK alone, road traffic crashes during working hours account for the greatest number of work-related deaths per year (Clarke et al., 2005), with commercial vehicles involved in approximately 25% of all road traffic fatalities (WRSTG, 2001). Similarly, a quarter to a third of all road traffic incidents involve people at work at the time (HSC, 2001). In addition, road-traffic crashes during working hours affect not only company employees but the wider public (HSC, 2007), and the annual UK cost from fatalities and injuries on the road is estimated at £3.5 billion (TUC, 2004). As such, road transport is a major risk factor for organisations that requires effective assessment and management, and the challenge remains of how to reduce the number of road traffic crashes at work. To date there has been limited research exploring the human factors underlying risky driving behaviour in occupational settings.

The current study concentrates on factors relating to the crash liability of a sub-group of occupational drivers, namely drivers of large goods vehicles (LGV), defined as goods vehicles with a gross weight over 3500 kg. As a group truck drivers are distinguishable from other road users for reasons beyond vehicle type. For example, according to the latest UK statistics available, in 2005 the average annual distance travelled by vehicles over 3500 kg (53,000 km, DfT, 2006a) far exceeds the average annual distance travelled by cars (14,450 km, DfT, 2006b). Greater exposure might suggest that road traffic crash involvement would be greater for truck drivers than non-commercial drivers, but when mileage is taken into account truck drivers (42 crashes per 100 million kilometres travelled) are involved in fewer crashes than non-commercial drivers (71 crashes per 100 million kilometres) in the UK in 2005 (DfT, 2006c). However, despite lower crash involvement than other vehicle types, trucks are more likely to be involved in a crash that results in a fatality due to the weight and relative size of the vehicle compared to other road users, as well as increased length of stopping distances (Campbell, 1991, Chang and Mannering, 1999, Huang et al., 2005, Clarke et al., 2005, Björnstig et al., 2008). The fatal crash rate for LGV drivers is 1.8 per 100 million vehicle kilometres for LGVs, which is double the crash rate of cars (0.9 per 100 million vehicle kilometres) (DfT, 2006c). In addition, injury severity is worse if a crash involves a truck (Chang and Mannering, 1999). Finally, with regards to responsibility for crash-involvement Clarke et al. (2005) analysed over 2000 crash reports involving a range of work-related vehicles. LGV drivers scored highest among occupational drivers for ‘blameworthiness’ in their crash involvement (a ratio of ‘to blame’ and ‘partly to blame’ crashes compared to ‘not to blame’ crashes), with the casual factors of fatigue and vehicle defects most prevalent in truck crashes. Clarke et al. also found that trucks were the most likely work-related vehicle group to be involved in crashes where people were killed or seriously injured.

It appears that LGV driving shares some of the risks faced by other road users, but has its own characteristics and risks that require specific attention. It is clear that the consequences of a road traffic crash are more serious when it involves a LGV, but there has been relatively little examination of the human factors in truck driving compared to the substantial literature on the general driving population. While there have been several studies examining risk factors in truck driving, the majority have focused on the relationship between fatigue and crash involvement (e.g., Summala and Mikkola, 1994, Adams-Guppy and Guppy, 2003, Morrow and Crum, 2004, McCartt et al., 2000) with little or no focus on psychological and behavioural processes involved in truck driving. As such, there is a need to identify psychological precursors to behaviour in order to help inform future interventions with LGV drivers that are aimed at reducing risk and crash involvement.

Based on a review of the literature and detailed pilot work with the Vehicle and Operator Services Agency (VOSA), the regulatory body for the UK truck industry, and with UK truck operators and truck drivers, two primary factors were identified as connected to crash involvement within truck driving, specifically inappropriate driver behaviour and non-compliance with vehicle and driver safety protocols. For example, with regard to driver behaviour, previous investigations of aberrant driving behaviour in truck drivers have demonstrated that driving violations are significantly correlated with crash involvement in truck drivers (e.g., Sullman et al., 2002). Furthermore, in line with previous research on driving behaviour within car drivers (e.g., Parker et al., 1995a, Parker et al., 1995b), driving violations by truck drivers were a significant predictor of crash involvement once annual mileage and other demographic variables (e.g., age) were accounted for (Sullman et al., 2002). Finally, research on US commercial drivers found that driving violations and prior convictions were significant predictors of crash involvement (Murray et al., 2006). Reckless driving and improper turns were the violations associated with the highest increase in crash likelihood (325% and 105%, respectively), and improper/erratic lane changes and failures to yield right of way were the convictions associated with the greatest increase in crash likelihood (100% and 97%, respectively).

In relation to the second factor, non-compliance with statutory regulations is an ongoing issue with LGVs. Over 465,000 roadworthiness tests were conducted on LGV vehicles in the UK in 2006–2007, with a failure rate of 22.1% (VOSA, 2007). The leading offences detected were driving hours and tachograph related offences, followed by overloading, driver and operator licence offences among others (VOSA, 2007).

Research identifying underlying psychological factors involved in LGV driving behaviour has not yet been conducted. One social psychology model proposed to understand volitional and non-volitional human behaviour is the theory of planned behaviour (TPB, Ajzen, 1985, Ajzen, 1988). In short, according to the theory the best predictor of a person’s behaviour is their intention to perform the behaviour. This includes their intentions to commit violations, and their intention to perform safe behaviours that would avoid making errors. These behavioural intentions are determined by three preceding factors: the person’s attitude towards the behaviour (e.g., whether the driver believes the behaviour to lead to good outcomes); their subjective norms (their beliefs about the attitudes and behaviours of socially relevant others); and their perceived behavioural control (the degree to which they feel they can personally influence the behaviour in question). Perceived behavioural control can also directly influence the behaviour. If no opportunity is available to perform the behaviour, then a person’s attitude, subjective norm, and intention is rendered irrelevant. Therefore, if one is interested in understanding why drivers do or do not engage in risky behaviour, previous research shows that psychological antecedents of behaviour are good predictors of actual behaviour. As a consequence one can develop a greater understanding of behaviour by measuring drivers’ attitudes towards a behaviour, their perception of the social pressure associated with the behaviour, and the level of confidence they have in controlling that behaviour.

The conceptual framework of TPB, along with its predecessor the theory of reasoned action (Fishbein and Ajzen, 1975), has been applied to a wide variety of settings in order to account for and understand people’s behaviour, including healthy eating (Armitage and Conner, 1999), physical activity (French et al., 2005, Armitage, 2005), and pedestrian road crossing behaviour (Evans and Norman, 1998). In the application of the TRA/TPB model specifically to driving, there has been a range of behaviours examined including committing driving violations (e.g., Parker et al., 1992a, Parker et al., 1992b), speeding behaviour (De Pelsmacker and Janssens, 2007, Warner and Åberg, 2006, Letirand and Delhomme, 2005), seatbelt use (Thuen and Rise, 1994, Şimşekoğlu and Lajunen, 2008), and drink driving (Armitage et al., 2002, Sheehan et al., 1996).

The utility of the TPB in accounting for a significant proportion of variance in driving behaviour has previously been demonstrated in the scientific literature. Parker et al. (1992b) first tested the applicability of TPB in explaining driving behaviour, finding that the model accounted for large and significant amounts of variance in intentions to drink and drive (42.3%), speed (47.2%), tailgate (23.4%), and overtake dangerously (31.7%). Attitudes and subjective norms accounted for one fifth to one third of the variance across intentions to commit violations, and the addition of perceived behavioural control significantly improved prediction of behavioural intention. Using the same sample of drivers, Parker et al. (1992a) found that drivers who had been involved in a crash in the previous 3 years were distinguishable from non-crash-involved drivers by measures of subjective norms only, with drivers involved in a crash generally perceiving significant others as more likely to expect them to commit driving violations. This suggests that crash-involved drivers might perceive less social pressure to avoid committing driving violations, or perhaps even a personality type that leads drivers to define themselves among their peer groups as ‘risky drivers’ in return for perceived social status.

Other research has supported the finding that attitudes, subjective norms and perceived behavioural control can independently account for variance in driving behaviour. These three factors have explained a significant proportion of variance in intention to comply with speed limits (Elliott et al., 2003) and intention to exceed the speed limit (Letirand and Delhomme, 2005), as well as variance in observed speed choice in a driving simulator (Warner and Åberg, 2006, Elliott et al., 2007). Strong correlations between self-reported and observed behaviour have also been demonstrated which suggests that self-report measures are a reasonably good surrogate for observed behaviour (Elliott et al., 2007), and support previous evidence of a strong relationship between externally observed speed and self-reported speed (Haglund and Åberg, 2000).

The aforementioned studies have demonstrated the utility of the TPB factors in accounting for variance in general driving behaviour, but not specifically for occupational driving behaviour. There has only been one previous study assessing the TPB model in relation to commercial drivers. Newnam et al. (2004) investigated intentions to speed in a company owned versus personally owned vehicle. Contrary to expectation they found drivers had a lower intention to speed in work vehicles than in their own personal vehicles. Hierarchical regression analysis of TPB factors and anticipated regret (the contemplation of having possibly made the wrong choice) were more likely to predict intentions to speed in a personal vehicle (27%) than intentions to speed in a work vehicle (16%) after drivers’ age, gender and annual kilometres travelled had been accounted for. TPB factors alone accounted for a significant, but small amount of variance in intentions to speed in a personal vehicle (15%) and work vehicle (8%). It was also noted that safety policies and practices within organizations affected employee driving intentions, with drivers from the companies with more extensive safety policies and practices reporting greater perceived control over speeding in a work vehicle, and more feelings of anticipated regret after speeding in a work vehicle, than other companies with less strong safety cultures. This is supported by other research demonstrating the influence of a strong safety climate on driver performance. Specifically, Morrow and Crum (2004) found that perceptions of a strong company safety management practice among commercial motor-vehicle drivers accounted for significant variance in fatigue and near-crashes due to fatigue, although it did not explain variance in crash involvement.

Given the demonstrated success of the TPB model in accounting for variance in driving behaviour within the general population, the current study aimed to test the success of the TPB model in accounting for behavioural intention and self-reported behaviour of professional LGV drivers. To date there has been no investigation of LGV drivers using the model of the TPB to explain variance in driving behaviour. The model was applied concurrently to observation of road traffic laws (driving behaviour) and compliance with road traffic regulations (compliance behaviour). The application of a model that explained a significant proportion of variance in intentions and behaviour would assist in helping develop interventions to reduce risky behaviour, and ultimately crash involvement.

Section snippets

Participants

A total of 2943 questionnaires were distributed across a variety of outlets. Of the total figure, 2483 questionnaires were sent to drivers from three truck operators who agreed to participate in the study and distribute questionnaires to their fleet of drivers. The operators comprised of a waste management company (n = 1000), a dairy foods company (n = 1000), and a fleet management and logistics company (n = 483), and were primarily line-haul and long-haul operations. Furthermore, 460 additional

Demographic data

Mean age of the LGV drivers was 46.8 years (S.D. = 9.4 years), with annual truck mileage of 49,524 miles (S.D. = 39,092 miles). Drivers had an average of 19.95 years (S.D. = 11.58 years) experience driving trucks, and one crash every 4 years (M = 0.56, S.D. = 0.96 biennially). A total of 33.9% of truck drivers reported having a crash in the last 2 years, with 52.6% of crash-involved drivers partly or totally to blame for the crash.

Path analysis of TPB factors

Summary statistics and inter-item correlations for all driving behaviour

Discussion

This study set out to increase understanding of risky behaviour among truck drivers using path analysis to assess psychological precursors to driving and compliance behaviour. Interestingly, there appears to be different underlying psychological motivators for rule following behaviour on the road and rule compliance regarding LGV regulations. Intentions appear to be best direct predictor of self-reported driving behaviour. Subjective norms and perceived behavioural control, in that order, have

Acknowledgements

This work was funded in part by the Vehicle and Operator Services Agency (VOSA). We would like to thank VOSA’s Research & Development Team for their support and commitment to the project.

References (55)

  • A.T. McCartt et al.

    Factors associated with falling asleep at the wheel among long-distance truck drivers

    Accid. Anal. Prev.

    (2000)
  • P.C. Morrow et al.

    Antecedents of fatigue, close calls, and crashes among commercial motor-vehicle drivers

    J. Saf. Res.

    (2004)
  • S. Newnam et al.

    Factors predicting intentions to speed in a work and personal vehicle

    Transp. Res. Part F: Psychol.

    (2004)
  • D. Parker et al.

    Determinants of intention to commit driving violations

    Accid. Anal. Prev.

    (1992)
  • D. Parker et al.

    Behavioural characteristics and involvement in different types of traffic accident

    Accid. Anal. Prev.

    (1995)
  • O. Şimşekoğlu et al.

    Social psychology of seat belt use: A comparison of theory of planned behavior and health belief model

    Transp. Res. Part F: Psychol.

    (2008)
  • M.J.M. Sullman et al.

    Aberrant driving behaviour amongst New Zealand truck drivers

    Transp. Res. Part F: Psychol.

    (2002)
  • D. Walton

    Examining the self-enhancement bias: professional truck drivers’ perceptions of speed, safety, skill and consideration

    Transp. Res. Part F: Psychol.

    (1999)
  • H.W. Warner et al.

    Drivers’ decision to speed: a study inspired by the theory of planned behavior

    Transp. Res. Part F: Psychol.

    (2006)
  • J. Adams-Guppy et al.

    Truck driver fatigue risk assessment and management: a multinational survey

    Ergonomics

    (2003)
  • I. Ajzen

    From intentions to actions: a theory of planned behavior

  • I. Ajzen

    Attitudes, Personality, and Behavior

    (1988)
  • C.J. Armitage

    Can the theory of planned behavior predict the maintenance of physical activity?

    Health Psychol.

    (2005)
  • C.J. Armitage et al.

    Predictive validity of the theory of planned behaviour: the role of questionnaire format and social desirability

    J. Commun. Appl. Soc. Psychol.

    (1999)
  • C.J. Armitage et al.

    Can the theory of planned behaviour mediate the effects of age, gender and multidimensional health locus of control?

    Br. J. Health Psychol.

    (2002)
  • Bureau of Labor Statistics (BLS)

    National Census of Fatal Occupational Injuries in 2006

    (2007)
  • S. Clarke

    Perceptions of organizational safety: implications for the development of safety culture

    J. Organ. Behav.

    (1999)
  • Cited by (82)

    • Meta-analysis of driving behavior studies and assessment of factors using structural equation modeling

      2023, International Journal of Transportation Science and Technology
    View all citing articles on Scopus
    View full text