Personality, risky driving and accident involvement among Norwegian drivers

https://doi.org/10.1016/S0191-8869(02)00010-7Get rights and content

Abstract

Despite the large body of studies, the role of personality in risk research still remains debatable and unclear. The objective of this study was to identify determinants of road user behaviour and accident involvement with the aim of developing effective accident countermeasures. Examining relationships between personality, risky driving and involvement in accidents can open up the possibility of early identification of those more likely to be involved in accidents. The aim is not to influence personality as such, but to develop measures constructed for specific groups. The results are based on a self-completion questionnaire survey carried out among a sample of Norwegian drivers in year 2000 and 2001 (n=2605). The Norwegian Directorate of Public Roads financed the study. The questionnaire included measures of risky driving, accident involvement, normlessness, sensation-seeking, locus of control and driver anger. Results showed that those who scored high on sensation seeking, normlessness and driver anger reported more frequent risky driving compared to those who scored low on these variables. They were more often involved in both speeding and ignorance of traffic rules. Respondents involved in risk taking-behaviour experienced near-accidents and crashes leading to both injuries and material damage more often than other drivers.

Introduction

Traditional considerations of traffic safety focus on the physical environment, the vehicle and the road user. Improvements in road environment and vehicles have achieved major safety gains. However, less progress has been made in understanding the behaviour of the road user (Rothengatter, 1997). In recent years psychology has become more involved in the study of risk behaviour and traffic safety, focusing on how emotional and personality factors influence driving behaviour and accident involvement. At present there is a renewed interest in analysing the role of personality in accident causation research. Elements of personality can not only compel individuals to commit particular acts but can also mediate effects of social influences designed to constrain these behaviours (Burgess, 2000). A major challenge is to find measures that influence groups of high-risk recipients more efficiently. Amongst the personality factors which have been found to be significantly associated with risky driving are locus of control, driver anger, sensation seeking and normlessness (Burns and Wilde, 1995, Deffenbacher et al., 1994, Montag and Comrey, 1987).

The first aim of the present paper is to examine the relationship between locus of control, driver anger, sensation seeking, and normlessness on the one hand and risky driving on the other. The second aim will be to analyse the relationship between these four factors, risky driving and accident involvement. Due to the fact that driver anger may be an important predictor of risky driving it is also of interest to consider determinants of driver anger. Therefore, the study aims to investigate the relationship between locus of control and driver anger.

The importance of behaviour in the prevention of vehicular accidents has been documented extensively (e.g. Elander et al., 1993, Parker et al., 1995a) and has led to many attempts to encourage a variety of safe driving behaviours (e.g. Juhnke et al., 1995, Martinez et al., 1996). The emphasis has shifted from performance-related capabilities to willingness to take risk.

Evidence indicates risk-taking as a major factor underlying high collision risk (Jonah, 1986). Several self-report measures of driver behaviour have been developed. One of the most prominent is the Driver Behaviour Questionnaire (Parker et al., 1995b, Reason et al., 1990) which is concerned with assessing the relative frequencies with which drivers engage in different types of aberrant driving behaviour. Self-reported violations, defined as the deliberate infringement of some regulated or socially accepted code of behaviour, have been shown to predict accident rates (Parker et al., 1995b, West et al., 1993). Manstead, Parker, Stradling, Reason, & Baxter (1992) found that speeding was perceived as the most prevalent of eight driving violations. Research reports associations between major deviations (both slower and faster) from the average traffic speed and an increase in crash risk (Parker, Manstead, Stradling, Reason, & Baxter, 1992). Speeding is not only a common violation; it is also regarded with a degree of tolerance by many. It is the propensity to violate (deliberate infringements), rather than the tendency to make errors of intention or action while driving, which is associated with involvement (Parker et al., 1995b, Parker et al., 1996). However, high scores on the error and lapse factor were found to be predictive of involvement in active accidents among elderly people, while passive accidents was associated with high scores on the lapse factor (Parker, McDonald, Rabbitt, & Sutcliffe, 2000). Considering these results, it is expected that drivers who have a high score on self-reported risky driving will be more frequently involved in traffic accidents than other respondents.

Research has emphasised personality factors as contributors to risky driving and accident causation. Despite the large body of studies, findings have been either conflicting or of little importance (Ranney, 1994). However, this does not imply a lack of capabilities of these measures to predict accidents. Conflicting results can arise from theoretical or methodological shortcomings (Lajunen & Summala, 1997). Mechanisms behind different types of accidents can differ, and these should be treated separately in analysis (Summala, 1996). Personality characteristics are often measured with general inventories, which do not include measures related to the specific targeted behaviour. Attempts to relate general constructs of personality to outside criteria have been more successful when the measures of these constructs were tailored more specifically to the targeted behaviour (Montag & Comrey, 1987). However, it is important to have in mind the definition of personality as a general, broad set of behaviours.

Many studies relate personality factors to the number of accidents. Accidents are rare events, and it is difficult to obtain valid information about occurrence and preceding behaviour. The shortcomings of accident data are well known, and an alternative criterion measure could be to include the behavioural level, not only outcomes. The next section investigates personality characteristics specially targeted to risky behaviours and accident involvement.

Rotter's (1966) Internality–Externality scale (I–E) has been related to risky driving. This scale measures the degree an individual feels in control of behavioural outcomes. The rationale for the relationship between locus of control and safe driving is based on the reasoning that externality is related to lack of caution and a failure to take precautionary steps to avoid the occurrence of unfavourable outcomes (Hoyt, 1973, Phares, 1978, Williams, 1972). Internals attribute more responsibility for road accidents to internal, controllable causes, and report less anxiety while travelling via automobile. They report wearing seat-belts more often (Hoyt, 1973, MacDonald, 1970), and experience car travel as more interesting and involving (Hoyt, 1973). Empirical findings have been mixed (Arthur and Doverspike, 1992, Knapper and Cropley, 1981). Guastello and Guastello (1986) found no direct relation between the Rotter (1966) locus of control scale and accidents. However, this lack of relation appeared to be rooted in the transformation of the Rotter concept into specific beliefs about accidents and driving behaviours. Montag Driving Internality and Driving Externality (MDIE) scales are an attempt to relate internality and externality to locus of control measures specifically tailored to the target behaviour of driving (Montag & Comrey, 1987), rather than using the more general I–E scale itself. Results of their study showed Driving Internality (DI) to be negatively related and Driving Externality (DE) to be positively related to involvement in fatal accidents (Montag & Comrey, 1987). Results indicate that high DE respondents not only tend to believe in external causation, but also exhibit low conformity, low emotional stability, lower energy level, lack of compulsion, and egocentrism (Montag, 1992). Respondents scoring high on DI tend to be emotionally stable, conforming, compulsive, active, and empathic. This indicates that generalised internality rather than externality is related to cautious behaviour. However, half of the sample in the study of Montag and Comrey (1987) was drivers who had been involved in a fatal road accident as a driver. It can be argued that these respondents would alter responses on the DI and DE scales to make themselves appear less responsible for the accident. Arthur and Doverspike (1992) concluded that MDIE was generally not associated with driving accident involvement in both a predictive and postdictive design.

Zuckerman's (1979) Sensation Seeking Scale (SSS) is a personality trait of thrill seeking behaviour. It reflects a need for varied, complex and novel sensations and experiences and a willingness to take risks for these sensations and arousal. Gender differences have been found in the trait; males recording higher scores than females (Zuckerman, 1984). Drunk driving behaviour among adolescents has been related to sensation seeking as well as egocentrism (Arnett, 1989). This measure has been found to correlate with many types of risk taking behaviour, like driving speed and self-reported traffic violations (Jonah & Clement, 1984). Burns and Wilde (1995) suggest that sensation seeking personality can be used to predict an individual's tendency to be a repeat offender, because of the association with observed fast and careless driving and convictions for both speeding and traffic violations. However, none of these studies found a positive correlation between sensation seeking and records of previous collision involvement. This implies that sensation seekers may drive fast but they do it safely (Burns & Wilde, 1995). Arnett (1994) claims that certain limitations of concept and form are associated with the use of SSS. He states that the “forced choice” format may be perplexing and frustrating, some activities are age-related, items reflect the 1960s and 1970s period, and finally, the items are often precisely the types of behaviour studied. Jonah (1997) reviewed 38 studies, which examined the relationship between sensation seeking and risky driving. Reported correlations ranged from the 0.30 to 0.40 range, and only four studies failed to find a positive relationship. He concluded that it is clear that SS is related to risky driving, but the strength and biological basis of this link require further research (Jonah, 1997).

Another personality variable examined in this study is normlessness. Human behaviour is largely constrained by rules that govern particular situations and social interaction. Constantly we are obliged to behave in certain ways to avoid particular reactions. Some of these rules are laws developed by authorities; others are social, informal rules implicit within the specific situation. However, people don't always adhere to rules, and this reduces safety margins that rules provide and often increases the likelihood of accidents. In psychological research the concept of normlessness is often a measure of subjective variety, i.e. perceived normlessness in an individual-centred viewpoint. The idea is that certain people at certain times may not respect presumed norms, trust others to respect them, perceive consensus with respect to appropriate behaviour or are prepared to act in deviant ways to achieve given goals. Kohn and Schooler (1983) operationalized normlessness on a continuum ranging from the individual's belief that it is “acceptable to do whatever (one) can get away with” to “holding responsible moral standards”.

Driver anger has been related to traffic violations and accidents (Deffenbacher, Huff, Lynch, Oetting, & Salvatore, 2000). Trait anger reflects a broad predisposition to experience anger more frequently and intensely across situations (Spielberger, 1988). Deffenbacher et al. (1994) suggested a general dimension of driving anger as well as anger related to specific driving-related situations. Driver anger was defined as more frequent and intense anger while operating a motor vehicle. Anger while driving may interfere with attention, perception, information processing and motor performance and may increase the likelihood of an accident directly or indirectly through the increased probability of other risk behaviours. Deffenbacher et al. (1994) developed a Driver Anger Scale (DAS), which can help to explore anger as a person-related variable in accidents or driving risks and also assess how anger relates to different driving behaviours. Ellison-Potter, Bell, and Deffenbacher (2001) examined the effects of trait driving anger, aggressive stimuli and anonymity on aggressive driving behaviour. Participants drove more aggressively when they were anonymous and exposed to aggressive stimuli, and males drove more aggressively than females. Results suggested that situational factors affecting other forms of aggression are also important in aggressive driving (Ellison-Potter et al., 2001). It is important to note that this study focuses on driver anger (emotional state), not behaviour.

Section snippets

Sample

The present study was based on a self-completion questionnaire survey carried out in 2000–2001 among a sample of Norwegian drivers randomly selected from the driver's licence register. All the respondents received the questionnaire by mail, and replied anonymously. The response rate was 48%, and the final sample comprised a total of 2605 respondents, 48% men and 52% women. Mean age of the respondents was 45 years (S.D.=15.67). Forty-four per cent had a college/university education, 47% work

Dimensions of personality and risky driving

Five principal component analyses using maximum-likelihood method and varimax rotation were applied to determine the dimensions of each of the main elements (locus of control, driver anger, sensation seeking, normlessness and risky driving). The factor analysis of Montag Driving Internality and Driving Externality (MDIE) scales reproduced the same solution as originally (Montag & Comrey, 1987), and the scales were scored and analysed as two separate scales measuring DI (eigenvalue=3.71) and DE

Discussion

The results of the present study imply that some of the personality variables could be used to predict an individual's tendency to commit risky driving. The intention is not to suggest interventions aimed at changing personality. However, many interventions have failed to show any effects. One of the reasons can be that they try to influence too large and too heterogeneous groups of drivers. Identification of the groups representing specific personality traits associated with risky driving and

References (48)

  • Burgess, C. (2000). The Wilful Traffic Offender Profile and its implications for education and training. PhD Research...
  • C.P. Burns et al.

    Risk taking in male taxi drivers: relationships among personality, observational data and driver records

    Personality and Individual Differences

    (1995)
  • P. Chapman et al.

    Forgetting near-accidents: the role of severity, culpability and experience in the poor recall of dangerous driving situations

    Applied Cognitive Psychology

    (2000)
  • J.L. Deffenbacher et al.

    Characteristics and treatment of high-anger drivers

    Journal of Counseling Psychology

    (2000)
  • J.L. Deffenbacher et al.

    Development of a driving anger scale

    Psychological Reports

    (1994)
  • J. Elander et al.

    Behavioural correlates to individual differences in road-traffic crash risk: an examination of methods and findings

    Psychological Bulletin

    (1993)
  • P. Ellison-Potter et al.

    The effects of trait driving anger, anonymity and aggressive stimuli on aggressive driving behaviour

    Journal of Applied Social Psychology

    (2001)
  • R. Elvik et al.

    Trafikksikkerhetshandbok

    (1997)
  • S.J. Guastello et al.

    The relation between locus of control construct and involvement in traffic accidents

    The Journal of Psychology

    (1986)
  • J.E. Gustafsson et al.

    STREAMS user's guide. Version 2.5 for Windows

    (2000)
  • M. Hatakka et al.

    Self-reported driving habits are valid predictors of violations and accidents

  • Jonah, B. A. (1997). Sensation seeking and risky driving. In T. Rothengatter, & E. Carbonell Vaya (Eds.), Traffic and...
  • B.A. Jonah et al.

    Field dependence, sensation seeking and driving behaviour

    Personality and Individual differences

    (1984)
  • G.A. Juhnke et al.

    Attitude changes in DWI offenders: a study of short-term treatment program

    Journal of Addictions and Offender Counselling

    (1995)
  • Cited by (0)

    View full text