Evaluating the Effectiveness of the United Kingdom's National Driver Improvement Scheme

Cris Burgess and Paul Webley

School of Psychology, University of Exeter.


Abstract - The National Driver Improvement Scheme (NDIS) is a driver rectification scheme currently employed by an increasing number of police districts in the UK as a direct alternative to a court prosecution for minor traffic offences. The scheme is described and a study that evaluated the effectiveness of NDIS in terms of changes in the self-reported attitudes and behaviour of clients is reported. Significant changes in self-reported driving behaviour and attitudes towards traffic violations in the appropriate direction are reported immediately after completion of the course and after three months. The nature of these changes and the mechanism that underlies them is discussed. Possible changes to the course content and structure are suggested.


This project was funded by the Department of the Environment, Transport and the Regions under contract DPU 9/31/35. The authors would like to thank Matthew Joint and Dianne Parker for their helpful advice on the research, Geoff Haddock for his comments on the draft manuscript, and the participating service providers for their co-operation and enthusiasm.



One of the recommendations made in the UK's Road Traffic Law Review (1988) was that:

"A pilot study of one day retraining in basic driving skills as a disposal should be undertaken to determine whether such retraining produces a lasting improvement in the driving skills of the offender undertaking it."


The recommendation was prompted by the supposition that a large number of motor vehicle collisions are caused by a skills deficit on behalf of those drivers involved. Furthermore, it was felt that this approach could also benefit drivers who persistently commit traffic violations. It was felt that these drivers' attitudes towards the commission of traffic violations could be modified, with a consequent reduction in their violating behaviour.

The National Driver Improvement Scheme (NDIS)

Although the driver improvement experiment was not incorporated into national legislation, Devon and Cornwall Constabulary, Devon County Council's Road Safety Unit and the Crown Prosecution Service met and agreed that such an experiment was worthy of further consideration. The police initially agreed to consider drivers for inclusion in the scheme who had been involved in a blameworthy collision within a clearly defined divisional area, and who would traditionally be facing the relatively minor charge of "driving without due care and attention" (Section Three; Road Traffic Act, 1988). After an initial operational evaluation in 1992, both Police and County Council agreed that the scheme should be made available throughout Devon for Section Three offenders. Those drivers accepting the offer and completing the course would not face a court prosecution.

The Devon initiative is receiving steadily increasing support nationally, and has become known as the National Driver Improvement Scheme (NDIS), incorporating the course structure and content that is now the National Model. Predominantly run by local government, thirty service providers currently receive client referrals from twenty-two police forces around the UK. Police sources estimate that approximately 85% of clients offered NDIS accept the offer of a course, in preference to a court prosecution (source: Devon & Cornwall Constabulary).

In 1996, the UK's Department of the Environment, Transport and the Regions commissioned a two-year study to evaluate the effectiveness of NDIS in the UK. The Evaluation Study aims to assess the effectiveness of the National Driver Improvement Scheme (NDIS) as an alternative to court prosecution for minor traffic offences.

Effectiveness of Driver Rectification Schemes

Driver rectification schemes of various types have been employed by a number of countries for some time, most notably in Germany and the US. However, evaluation studies to test the effectiveness of these schemes have revealed mixed results. Some studies concluded that such measures are ineffective in reducing subsequent violation frequency and crash involvement (e.g.: Koppa and Banning, 1981; Lynn, 1982), whilst others found that such schemes were effective in their aims (e.g.: Finigan, 1995; Wark, Raub and Reischl, 1998). Still other studies revealed improvements in the criterion measures, but only amongst particular age and gender groups (e.g.: Payne, Brownlea and Hall, 1984). McKnight and Tippets (1997) differentiate between 'accident-prevention' programs, which primarily seek to foster safe driving practices, and 'recidivism-prevention' programs, which attempt to encourage lawful driving behaviour. Results revealed that offenders participating in the recidivism prevention program were involved in significantly fewer collisions and committed fewer violations in the following year than those whose intervention focussed upon accident prevention. In a review of some sixty-five 'driver rectification' courses, Struckman-Johnson et al (1989) found wide variation in course content and delivery, which included warning letters and interview procedures as well as traditional classroom and practical driving sessions. Based on this evidence, it is clear that driver rectification courses must be taken on their own merits and that broad generalisations regarding the effects of this type of intervention may not be valid.

NDIS course structure - the National Model

A fundamental difference exists between NDIS and other similar schemes found elsewhere, in that NDIS is a direct alternative to court prosecution and not a sentencing option. NDIS courses run over one and a half days. The first morning is spent in a lecture room and is divided into a number of sections, each of which tackles specific driving issues.

The contractual relationship between service provider, police and client is explained, emphasising the independent status of the service provider from police and court authorities. Clients are invited to consider their motives for choosing the course in preference to a court appearance, and what they wish to gain from course attendance.

Clients' beliefs concerning the causes of traffic collisions are explored, and commonly held beliefs are challenged. Amongst these are the belief that the majority of collisions are chance events that occur at random, are determined by environmental factors and that the driver has minimal control over or responsibility for such events. Clients are encouraged to discuss, dispute or debate any beliefs that they hold regarding traffic collision causation throughout the course with the group, course leader and instructors.

The course leader reports the evidence of empirical studies that have suggested that the majority of 'accidents' are the result of deliberate acts on behalf of drivers that violate road traffic law (eg: Manstead, Stradling and Parker, 1991), human error and lapses in concentration (eg: Sabey and Taylor, 1980). The nature of these acts of violation is explored and the potentially lethal consequences are discussed.

The influences of peers are explored as causes of unsafe driving practices, ranging from peer pressures within groups of young male friends to the influence of corporate culture on professional drivers. Research findings in these areas are discussed. Social group influences are emphasised as a primary motivation in determining safe or unsafe driving practices. Group work, with the course leader as facilitator, encourages unsafe drivers to explore these issues and to learn techniques for coping with social pressure.

Clients are divided into small groups (3-5 individuals) and each group is given a different example of a commonly observed traffic collision to discuss. For example; a vehicle turning right at a junction from a minor road onto a major road (in the UK this involves crossing over one line of traffic on the major road to reach the opposite carriageway), obscured by parked vehicles and is subsequently struck by a vehicle travelling at 10mph above the legal maximum speed limit. Clients investigate the rational decisions and behaviour that led to the collision and are directed to discuss 'blame' for the incident in terms of the distinction between responsibility (active commission of inappropriate behaviour) and avoidability (passive ommission of appropriate behaviour). Alternative courses of action for each driver involved in the incident are put forward and discussed. Each small group is then required to present their example and relevant conclusions to the other groups, who are then asked to comment.

Theoretical viewpoints (eg: Wilde, 1982) regarding voluntary and calculated risk-taking are explored as a part of everyday driving in the modern traffic environment.

Hazard awareness and perceptions of risk are described in terms of an interactive cognitive activity. The high incidence of 'near misses' is emphasised as evidence for the failure of such processes.

The causes and effects of stress are discussed and coaching in developing coping strategies to reduce and manage stress is given. Time management is discussed; emphasis is placed on the minimal benefits, in terms of time saved, of speeding, close following, risky overtaking and other relatively high-risk manoeuvres compared with the high cost of crashes and collisions, as personally experienced by clients.

The effects of fatigue on collision likelihood (eg: Horne, 1991) are discussed and strategies to reduce the risk are developed by each client for their own particular lifestyle (eg: realistic time management).

During the afternoon and following morning a qualified approved driving instructor takes a group of three clients on a practical driving exercise.

The emphasis in both sections is on an interactive approach, with a condition of course completion resting on the clients' willingness to participate fully in classroom discussions. The classroom session itself more closely resembles McKnight and Tippetts' (1997) description of their recidivism prevention program, in that the nature of deliberate violations and unintentional errors is established, and the potential danger to the driver and other road users is assessed. In this way the antisocial and deviant nature of these behaviours is reinforced. In addition, conditions that lead to violations and errors are explored, methods of coping with these conditions discussed and personal strategies for implementing these coping strategies are agreed upon. The remainder of the classroom session is largely informational, exploring various topics including driver fatigue, hazard perception and differentiation between responsibility for vehicle collisions and avoidability.

The aim in the practical driving sessions is to highlight and eliminate any dangerous or inappropriate elements of the clients' driving style. Groups of three clients are taken out on public roads by a specially qualified driving instructor. Each client drives for a number of short periods, while the instructor directs them. After each turn at the wheel, with the vehicle stationary, the instructor will initiate a discussion between the clients, using examples of driving from the previous period to illustrate particular points. In addition, feedback from the two non-driving clients is elicited in order that the driving client may see the effects of particular behaviours on his or her passengers.

The NDIS Evaluation Study

Clients referred to NDIS are deemed by the police to have been driving in an inappropriate manner. It may be assumed that these individuals maintain inappropriate attitudes towards traffic law, or may be deficient in skills or ability. Therefore, the current evaluation study will compare clients' attitudes towards traffic violations and self-reported driving behaviour before and after course attendance. In this way, the effects of the intervention may be assessed.



1,821 clients contributed to the study, attending NDIS courses at eight Service Provider sites over a twelve-month period from October 1997 to September 1998. The typical NDIS client is young (over 20% fall into the 17-21 year-old category), male (70% of sample), inexperienced (almost 25% passed their driving test within the last two years) and covers an average of 10,000 miles per year. Individuals who drive for a living are over-represented (47% of sample), but they also record a significantly higher annual mileage (27,000 miles, on average, compared with 9,000 for ‘domestic and pleasure use' drivers). A small number of clients (approximately 4% of the sample) had not been involved in a blameworthy collision, but were referred to NDIS after being considered by the police to have demonstrated a standard of driving below the acceptable minimum.


Two established instruments were employed in the study; the Driving Behaviour Questionnaire, (DBQ; Parker, Reason, Manstead and Stradling, 1995) and the Driver Attitudes Questionnaire (DAQ; Parker, Stradling & Manstead, 1996).

The DBQ (Parker et al, 1995) records the self-reported frequencies of three types of inappropriate driving behaviour; lapses, errors and violations (from Reason, Manstead, Stradling, Baxter and Campbell, 1990). Lapses are those behaviours that do not present any immediate danger to other road-users, but do not typify efficient or effective driving practice; for example, attempting to drive away with the handbrake still engaged, or forgetting where one parked in a car-park. Reason et al (1990) defined errors as "slips and mistakes in the highest risk category", and imply a lack of effort or attention rather than a positive act on behalf of the driver; for example, misjudging an overtaking gap or failing to see another vehicle while pulling out of a junction. Finally, violations are considered to be "deliberate... ...[and]...involving a definite risk to others" (Reason et al, 1990). Respondents indicate on a six-point scale how frequently they engage in the behaviours ('0' indicates 'never', while '5' indicates 'all the time').

The DAQ (Parker et al, 1996) assesses attitudes towards four common traffic violations; drinking and driving, close-following (tailgating), dangerous overtaking and speeding. The questionnaire consists of two counterbalanced scales of twenty statements, each requiring the respondent to indicate their degree of agreement on a five-point Likert-type scale ('1' indicates strongly disagree', while '5' indicates ' strongly agree').


Three measurement points were used; four weeks before attending the course (after involvement in a blameworthy accident or incident, and having formally accepted the NDIS alternative to court prosecution), immediately after successful course completion, and a three month follow-up.


Clients' perception of NDIS course

Most clients (over 70%) perceived NDIS as a way of avoiding court prosecution. Less than one third of the whole sample (29%) felt that they had been treated unfairly by the authorities, and the majority (almost 90%) felt that the course offered them an opportunity to look at areas of their driving that may need attention.

Although over 70% of the sample considered NDIS to be a way of avoiding a court appearance, and the accompanying likelihood of penalty points and a fine, most drivers (87%) were optimistic about the opportunity to improve particular aspects of their driving. It seems apparent that a large proportion of the drivers themselves recognise a need for some form of retraining, particularly those in the older age categories.

Self-reported behaviour

The pre-course measure relates to acts of inappropriate driving behaviour committed in the previous three months, while the follow-up measure relates to the three months since course completion. Significant reductions in the self-reported frequency of commission of lapses (t=2.062, df=608, p=0.04), errors (t=4.097, df=608, p<0.001) and violations (t=1.911, df=608, p=0.05) are evident in the three month period following the course, compared with the three months prior to completing the first measure (Figure One).

Figure One: Self-reported behaviour (DBQ subscales; pre-course vs follow-up)


The frequencies of commission show relatively strong correlations between the measurement points (Table One), indicating that it is largely the same clients committing the acts prior to and following the course. The significance of the decrease in all three behaviours indicates that although it is largely the same drivers committing the acts, they are less frequent in the three months following the course than in the same period before it.














Table One: Inappropriate driving behaviour; correlation between pre-course and follow-up measures (with significance level)


Attitude modification

The immediate effect of the course on attitudes towards traffic violations is an overall significant improvement in the clients’ general safety-orientation, with the overall DAQ score increasing from 72.97 to 73.76 (t=3.586, df=1438, p<0.001). Three months after completing the course, there is still a significant improvement in general attitude score with DAQ scores, for those remaining in the sample, from 74.07 to 74.75 (t=2.121, df=608, p=0.03).

Attitudes assessed by the first measure reveal significant differences between clients' attitudes towards the four violations (F3,4599=505.296; p<0.001) prior to attendance on the course (Figure Two). Post-hoc contrast analyses reveal that significant differences exist between clients' attitudes towards dangerous overtaking and speeding. The largest difference is between attitudes toward speeding offences, and the rest combined (F1,1533=1574.95, p<0.001). Differences between the violation types are still significant immediately following the course (Figure Three), but these differences are not as marked (F1,1688=106.750, p<0.001). Attitudes relating to speeding offences show a large and significant improvement maintained three months after the course (t=-3.797, df=608, p<0.001). Attitudes towards close-following are also significantly more appropriate three months after the course (t=-1.984, df=608, p=0.048). Attitudes towards drink-drive and dangerous overtaking do not change significantly.

Figure Two: Attitude scores (DAQ subscales; pre-course vs post-course)


Figure Three: Attitude scores (DAQ subscales; pre-course vs follow-up)


Non-improving clients

Those clients who did not demonstrate a change in attitude scores, or whose scores became less appropriate after attending the NDIS course, reported significantly more appropriate scores prior to the course than those who showed an improvement (Table Two).


Measurement point



mean score




























Table Two: Differences in attitude measures (DAQ total score) between clients recording a positive improvement in general attitudes (positive change) compared with a negative inappropriate change (negative change)


Small but significant reductions in the reported frequencies of commission of lapses, errors and violations are all evident in the three months following the course. These frequencies are quite highly correlated before and after the course, implying that it is largely the same individuals committing the acts, although they are committing them less frequently than they were prior to course attendance.

The data reveal a significant overall improvement in attitudes towards traffic violations (DAQ total). Again, the changes in overall attitude scores are small, but if the changes in attitudes towards particular violations are considered, some interesting patterns emerge. Prior to the course, attitude measures reflect that speeding offences are not perceived as serious transgressions when compared with other offences. Attitude scores related to drink-drive offences, dangerous overtaking and close-following are all near the upper (safety-orientated) end of the scale. While changes in attitudes towards these three offences are small (possibly as a result of a ceiling effect), attitudes towards speeding become significantly more appropriate immediately after the course; the perceived seriousness of speeding offences is now more in line with the remaining three violations. This significant improvement in attitudes towards speeding offences is retained after three months.

The results of this study indicate that attending a NDIS course has a significant effect on self-reported behaviour and attitudes towards traffic offences. However, the exact nature of this effect is not particularly clear. What mechanism causes these changes?

Decisional control

It is important to remember that clients choose to attend NDIS courses as an alternative to court prosecution. Clients therefore have decisional control over their fate. There is considerable evidence that suggests that when people have committed themselves to a course of action that subsequently turns out to be relatively laborious, they reduce post-decisional conflict by re-appraising their decision in a more positive light (Festinger and Carlsmith, 1959; Kiesler, Collins and Miller, 1969; Zimbardo, 1969). In the present case, it is arguable that a forced compliance factor and a contradictory element of decisional control are present. Some clients may feel that they had no choice but to accept the offer of NDIS in preference to a court appearance, penalty points and a fine. However, less than one third of clients felt unfairly treated by the authorities prior to attendance on the course, and almost ninety percent felt that the course offered them the opportunity to look at areas of their driving that may need attention. Therefore, we may assume that the majority of these individuals consider that they made an active, unforced and positive choice in attending the course. Dissonance theory (Festinger, 1957) suggests that NDIS clients will justify their physical, financial and psychological investment by changing their attitudes in such a way that they become consistent with those endorsed during the course.

Changes in driving-related attitudes in the period following the course may well be partly due to dissonance effects, but another mechanism may affect these attitudes. The follow-up questionnaire, arriving three months after attending NDIS, will have acted as a 'cue'; a reminder of the personal investment that the clients made in attending the course, the instruction that they received and the 'good resolutions' made by the clients to their instructors or to themselves. This is likely to reinforce their appropriate attitudes, and it may be that the arrival in the post of the final questionnaire is partially responsible for the improvement in attitudes at the follow-up measurement point. This cueing effect could be enhanced to increase the effect of the course on clients' attitudes. However, there is a problem with using measures of driving-related attitudes to assess course effectiveness.

Habitual driving style

In measuring attitudes, as with this study, we are measuring an individual's 'behavioural intention'; how that person would like to behave in a particular set of circumstances. But how many times have drivers travelled from home to their place of work, only to realise that they remember nothing of the journey? What part have their behavioural intentions played in guiding their behaviour during this journey? Current research strongly suggests that a large proportion of human performance is guided by routinised behaviours. These routinised behaviours are largely functional, in that they allow the achievement of valued goals or end states. Habits are established when a particular element of behaviour is repeated sufficiently and satisfactorily, in response to an environmental cue (Ronis et al, 1989; Gollwitzer & Schaal, 1998). The habit becomes a behaviour that is regularly carried out in response to particular environmental cues, which is efficient and requires little in the way of conscious attention or awareness. Verplanken and Aarts (in press) describe a habit as "goal-directed automaticity", and suggest that "we may well carry out the most complicated behaviours (eg: driving a car) by force of habit under conditions of heavy mental load, such as time pressure, distraction, or information overload". They argue that when an individual develops a strong behavioural habit, the relationship between attitudes and behaviour is weak, whereas individuals with weak habits demonstrate a strong attitude-behaviour link (Verplanken, Aarts, van Knippenberg and van Knippenberg, 1994). Using Ajzen's (1991) Theory of Planned Behaviour to test their model they found that, compared to weak habit individuals, strong habit individuals acquired less information and gave evidence of less elaborate choice strategies (Verplanken, Aarts, van Knippenberg and Moonen 1998; Aarts, Verplanken and van Knippenberg, 1997). In other words, individuals with strong habits tend not to look around them to gain information with which to develop a behavioural strategy; they just behave the way that they always behave.

Habits thus arise from repeated execution of a particular behavioural act in response to environmental cues, becoming automated and thus requiring minimal conscious effort. However, situations are frequently encountered in everyday life that require conscious decision-making, but under considerable pressure in terms of time and mental load. In these circumstances, are individuals likely to display a similar deficit in information gathering and planning, even though the process is under conscious control? Gollwitzer (1993) defines 'implementation intentions' as plans of action that link particular responses to specific cues. Gollwitzer suggests that these take the form "I intend to 'X' when situation 'Y' is encountered", and thus install contingencies between situational cues and goal-directed responses. Actions that lead to goal fulfilment thus gain a degree of automaticity by being under the control of situational or environmental cues (Gollwitzer, 1993; Gollwitzer & Moscowitz, 1996). Verplanken & Aarts (in press) propose that the difference between habit and implementation intentions is that contingencies in habits originate from experience and systematic reinforcement, whereas those contingencies related to implementation intentions arise as a result of deliberate analysis and planning. However, they contend that the effects on overt behaviour of these contingencies are similar, regardless of their origin, leading individuals with strong habits or implementation intentions to limit the amount of situational information that is sought out before a behavioural act takes place. Indeed, these authors go on to suggest that the formation of implementation intentions may be the first step in establishing a habitual behavioural response.

Although suggesting that further research must be done in order to clarify the similarities between these concepts, Verplanken & Faes (1999) have provided some evidence that supports the view that both processes initiate a similar situation-specific cognitive orientation. This mind-set (Gollwitzer, 1990) directs conscious attention only towards particular types of relevant information in the environment. Gollwitzer (1990) distinguishes several types of mind-set, which vary according to the phase of the goal-setting or goal-completion track described in the model of action phases (Gollwitzer, 1990, 1996; Verplanken & Aarts, in press). For example, when an individual is considering a choice of goals, a deliberative mind-set may be activated that comprises an open mind for new information and promotes relatively objective information processing. However, an implemental mind-set will focus attention on particular information regarding where, when and how to act, and is characterised by closed-mindedness. Once a behavioural act is initiated, an actional mind-set will focus the individual's attention exclusively on aspects of the self and environment that sustain that behaviour. Drawing upon Gollwitzer's mind-set concept, Verplanken & Aarts (in press) provide evidence for their suggestion that a habitual mind-set may enhance the perceptual readiness for habit-related cues, and prevent the individual from being distracted and adopting other, less efficient courses of action. This mind-set acts as a kind of enduring 'default' cognitive orientation that is inherently associated with the habitual behaviour, although it may be present on a relatively chronic basis, and not exclusively during the time that a habitual action is actually executed.

The preceding discussion suggests mechanisms whereby routinised behaviours arise in response to environmental and situational factors, both through direct and repeated experience of such factors, and through deliberate processing of relevant information. The evidence suggests that these routinised behaviours will become particularly apparent in situations of high mental load. Driving a vehicle on public roads creates a considerable mental load, with constant distractions and demands on the driver's attention. It is likely that novice drivers quickly learn to routinise their behaviour, these automated procedures extending beyond mechanical operations such as changing gear and road position, to habitual responses to a variety of environmental cues. The manner in which implementation intentions give rise to routinised behaviours suggests that each individual will develop an individual style of driving, with consistent biases across categories of situation, their habitual mind-set dictating how much and what information in the driving environment is attended to. Breaking habits will often require time and repetition of interventions, while the new, more acceptable behaviour is likely to relapse (e.g.: Marlatt, 1982). The likelihood is that drivers will revert to their inappropriate habitual style after any kind of intervention. Indeed, Wark et al (1998) identified a ninety-day 'window' following course attendance, during which offenders attending a rectification course were distinguished from offenders attending court by a significantly lower re-conviction rate. However, beyond three months there was no difference between the groups. In order to maintain clients' behavioural intentions, their habitual behaviours must be altered and their new, more appropriate attitudes maintained.

Habitual driving style is a constant influence on any driver; it is psychologically 'ever-present' and persistently influences how the individual behaves on the road. In order to change that driving style, any corrective influence must also be continually present or the driver will revert back to their previous inappropriate style. In order to maximise the effects of interventions of this type, methods must be determined that ensure that the clients receive constant reminders of the instruction given on the course, emphasising the 'probationary' nature of the period following the course. Indeed, knowing that a further questionnaire would be sent out and that their progress checked up on, clients may have felt that they were 'on probation' in the three months following the course. In addition, it is made clear to the clients at the start of the course that if they are apprehended committing another similar offence in the three years following course completion, they will automatically go to court. The period immediately following the course, while these considerations are fresh in the minds of the clients, has a probationary quality that may well constrain offending behaviour throughout this period. If this is the case, it supports research from the US which shows that longer terms of licence suspension lead to improved attitudes and behaviour in banned drivers once they regain their licence (Hurst, 1980). It is estimated that the great majority of suspended drivers in the US do not stop driving during their period of suspension (Peck, 1991; Ross, 1991), but they are considerably more law-abiding throughout this period, as the penalties for driving while disqualified are severe. This more sedate and safety-orientated behaviour becomes accommodated in the habitual driving style of the individual; indeed it redefines the driving style. The longer the licence suspension, the more likely it is that this new style will be retained, even after the licence is regained (Hurst, 1980). If drivers with a strong habitual driving style do not seek information from their environment, mechanisms must be developed that ensure that elements of the driving environment impinge on the drivers' awareness and remind them of elements of their training and the negative consequences of allowing themselves to revert to inappropriate habits. If drivers' habitual mind-sets prevent external information from being processed, then one possible solution might be to establish internal behavioural cues signalled by an element of the environment.

Moral vs social-conventional transgressions - a distinction

In trying to prevent the commission of traffic offences, law-enforcement agencies are sustaining a system of rules, the transgression of which may be perceived in different ways. Moral transgressions result in direct infringements of people's rights and welfare. For example, we have formal laws forbidding assault and theft, and informal rules about not cheating on one's partner or shouting unwarranted verbal insults. Social-conventional transgressions are considered to be less serious (Nucci, 1981). They violate the arbitrary and agreed-upon conventions that co-ordinate the behaviour of individuals within social systems; for example, a bar owner continuing to serve alcohol beyond his licensed hours, failing to make a tax return, or talking to yourself in public places. There is ample evidence that children are able to distinguish between these types of transgression from an early age (Smetana, 1981; 1985) and throughout early adulthood (Nucci, 1981; Turiel, 1983), and that moral transgressions are considered far more serious than infringements of social-conventional rules.

If we consider traffic offences in the light of this distinction, there are clear differences between the types of traffic offences. Close-following and dangerous overtaking have obvious and immediate implications for other road-users. Driving through red traffic lights, and not stopping at pedestrian crossings are equally clear in their effects on other people. Extended police and media campaigns in the US have ensured that drinking and driving is seen as immoral (Piquero and Paternoster, 1998) and the same has been found in Japan (Deshapriya and Iwase, 1996). The consequences of such behaviour are seen to have direct implications for the lives of other road-users. Police and the Government's efforts to reduce the levels of drinking and driving are considered by the majority as commendable. However, speeding is not seen in the same way. Although it has been the subject of some media campaigning, the maximum speed limit is still likely to be considered merely the convention that drivers maintain on each category of road, and therefore the methods employed by the authorities to enforce such behaviour are seen as revenue generators, not as attempts to make the roads safer.

Moral transgressions have negative consequences for other people. Breaches of social-conventional regulations, like the rules in a computer game, do not. What is needed, therefore, is to establish the moral justification for such laws. The general population is likely to consider the probability of their being involved in a vehicle collision as quite low. The great majority of clients who are referred to NDIS have been involved in a blameworthy collision and are considered by the police to have been partly or wholly responsible. The moral justification of traffic law, based on the effects of transgressions on other people, is thus particularly salient to NDIS clients.

Behaviour vs attitudes

It must be remembered that changing attitudes does not necessarily change behaviour, but changes in behaviour have been shown to reliably change attitudes. Howarth (1988) suggests that:

"It has been frequently demonstrated that attitudes are easier to change than behaviour and that a verbally expressed belief... ...may not be reflected in any increase in the related and easily observed behaviour. In contrast it has frequently been demonstrated that changes in behaviour, induced by environmental pressure, can lead to a change in verbally expressed attitudes, usually in the direction which justifies the new form of behaviour ".


In other words, changing attitudes does not necessarily affect behaviour, but if we are able to impose changes in behaviour, breaking habitual responses to elements of the driving environment, attitudes are likely to follow if the individual is able to justify those changes. It has long been established that humans 'search for meaning' in their own actions, as well as the actions of others (Heider, 1958). In retrospect, we try to understand why we behaved in a particular way. If our behaviour changes, due to factors beyond our control, we will look for reasons to justify those changes (Bem, 1972; Ross, McFarland, Conway and Zanna, 1983). If we cannot justify the new modes of behaviour, it is likely that we will revert to our old ways.

The NDIS course already tackles the issues of legitimacy of Road Traffic Law, by establishing the causes of inappropriate or sub-standard driving, and their effects on other road-users. The intention is to modify clients' attitudes towards these offences and to induce a subsequent change in behaviour. In establishing the justification for traffic laws, the clients are given a justifiable reason to maintain any behavioural changes based on those. However, it is unlikely that simply modifying attitudes will be enough to generate permanent changes in driving behaviour. In addition to giving the reasons behind the regulations, it is necessary to induce those behavioural changes directly, if for only a relatively short period of time.

Increased enforcement is an obvious method of imposing appropriate changes in driving behaviour. However, unless increased enforcement levels are undertaken over a prolonged period of time, appropriate behavioural changes will not have sufficient time to become incorporated into the individual's driving style, and their behaviour will revert to the established, inappropriate style. Prolonged increased enforcement is expensive and unpopular, and so alternative methods must be found.

Psychological probation

If it is not practical to physically impose new forms of behaviour on individuals, methods must be found to achieve this indirectly. One possible way of doing this is to ensure that somewhere in the environment is a reminder of the elements of training that constituted the intervention. Reminding the individual of the justification and legitimacy of a particular rule will make compliance significantly more likely, as it acts as a reminder of the psychological investment they made in attending the course and their behavioural intentions upon course completion, and justifies those changes in their behaviour. This approach has been shown to be effective in seat-belt use in the US (Geller, Berry, Ludwig, Evans, Gilmore and Clarke, 1990), and it indicates a need to carefully consider the construction of the NDIS course model to maximise this ‘cueing’ effect.

The effect of probationary measures is to bring into line the individual's attitudes with those changes in behaviour which are imposed through external means. The changes in behaviour become routinised and, once this is achieved, are slow to revert. If it is possible to maintain the new, appropriate behaviours for a certain amount of time, a 'probationary period', the likelihood that those behaviours will become habitual are significantly increased, particularly if the individual is able to justify those changes to themselves on moral grounds. There are a number of measures that could be used in following up a client who has completed a NDIS course. They range from simply sending out further information about other training courses or services offered by a service provider, to a legal requirement to re-attend the training centre a set period after course completion and increased penalties for offences committed within that period. The stronger the incentive to maintain the new behaviours, the better, as this will increase the chances that they are routinised. However, any extended 'probationary' period will reduce the attractiveness of the scheme in comparison with the alternative option of a court appearance. This point, as well as the legal and political considerations, must obviously be taken into account when making any changes to the current system.


Attending the NDIS courses has a significant effect on changing clients' attitudes and self-reported behaviour in the desired direction. If we are to understand the nature of those changes, and to maintain them over any length of time, it is essential that we establish how habitual driving style mediates the attitude-behaviour relationship. If we are to ensure that schemes such as NDIS are successful in modifying inappropriate driving behaviour, internal cues must be established that will result in changes in inappropriate behaviour, and those changes in behaviour must be induced over as long a period of time as possible. Using a variety of techniques it should be possible to reproduce the effects of a probationary period without further legislation. Combined with an educational approach based on justifying those changes in behaviour on a moral basis, inappropriate habitual driving styles will be challenged and offenders will have the incentive and the justification to maintain a more appropriate driving style. Further research must now identify those elements of training that are most effective in the context of a 'remedial' driver intervention.



Ajzen, I. (1991) The theory of planned behaviour. Organisational Behaviour and Human Decision Processes. 50, 179-211.

Bem, D.J. (1972) Self-perception theory. In Advances in Experimental Social Psychology. Ed. L. Berkowitz, Vol. 6, 1-62. Academic Press: New York.

Deshapriya, E.B.R. and Iwase, N. (1996) Are lower blood alcohol limits and a combination of sanctions desirable in reducing drunken driver-involved traffic fatalities and traffic accidents? Accident Analysis & Prevention. 28, 6, 721-731.

Festinger, L. (1957) A theory of cognitive dissonance. Stanford University Press: Palo Alto, California, USA.

Festinger, L. and Carlsmith, J.M. (1959) Cognitive consequences of forced compliance. Journal of Abnormal & Social Psychology. 58, 203-210.

Finigan, M. (1995) Traffic violator school research report, California Department of Motor Vehicles/Santa Clara County Municipal Court, Northwest Professional Consortium, West Linn, Oregon 97068.

Geller, E.S.; Berry, T.D.; Ludwig, T.D.; Evans, R.E.; Gilmore, M.R. and Clarke, S.W. (1990) A conceptual framework for developing and evaluating behaviour change interventions for injury control. Health Education Research. 5, 2, 125-137

Gollwitzer, P.M. (1990) Action phases and mind-sets. In: E.T. Higgins & R.M. Sorrentino (Eds.) Handbook of motivation and cognitions: Foundations of social behaviour (vol. 2). Guilford: New York.

Gollwitzer, P.M. (1993) Goal achievement: The role of intentions. European Review of Social Psychology, 4, 141-185.

Gollwitzer, P.M. (1996) The volitional benefits of planning. In P.M. Gollwitzer & J.A. Bargh (Eds.) The psychology of action: Linking cognition and motivation to behaviour. Guilford Press: New York.

Gollwitzer, P.M., & Moscowitz, G.B. (1996) Goal effects on action and cognition. In E.T. Higgins & A.W. Kruglanski (Eds.) Social Psychology: Handbook of basic principles. Guilford Press: New York.

Gollwitzer, P.M., & Schaal, B. (1998) Metacognition in action: The importance of implementation intentions. Personality and Social Psychology Review, 2, 124-136.

Heider, F. (1958) The Psychology of Interpersonal Relationships Wiley: New York.

Horne, J. (1992) Motorway menace: asleep at the wheel. New Scientist, 1802.

Howarth, C.I. (1988) The relationship between objective risk, subjective risk and behaviour. Ergonomics. 31, 4, 527-535

Hurst, P.M. (1980) Can anyone reward safe driving? Accident Analysis and Prevention. 12, 217-220

Kiesler, C.A., Collins, B.E. and Miller, N. (1969) Attitude change: A critical analysis of theoretical approaches. Wiley: New York.

Koppa, R. J. and Banning, K.R. (1981) Young problem driver improvement project; report no. DOT-HS-704-104. US Department of Transportation, NHTSA, Washington DC, US.

Lynn, C. (1982) An evaluation of the Virginia driver improvement program on negligent driving: twelve-month report, Virginia Department of Transportation Safety, US.

Marlatt, G.A. (1982) Relapse prevention: A self-control program for the treatment of addictive behaviours. In R.B. Stuart (Ed.) Adherence, compliance, and generalisation in behavioural medicine. Brunner/Mazel: New York.

McKnight, A.J. and Tippetts, A.S. (1997) Accident prevention versus recidivism prevention courses for repeat traffic offenders. Accident Analysis & Prevention. 29, 1, 25-31.

Nucci, L. (1981) The development of personal concepts: A domain distinct from moral or societal concepts. Child Development. 52, 114-121

Parker, D., Reason, J.T., Manstead, A.S.R. & Stradling, S. (1995) Driving errors, driving violations and accident involvement. Ergonomics. 38, 1036-1048.

Parker, D, Stradling, S.G,. and Manstead, A.S.R. (1996) Modifying beliefs and attitudes to exceeding the speed limit: An intervention study based on the theory of planned behaviour. Journal of Applied Social Psychology. 26, 1-19.

Piquero, A. and Paternoster, R. (1998) An application of Stafford and Warr's reconceptualisation of deterrence to drinking and driving. Journal of Research in Crime and Delinquency. 35, 1, 3-39.

Payne, S., Brownlea, A. and Hall, A. (1984) Evaluation of Queensland defensive driving course, Griffith University, Nathan, Australia.

Peck, R.C. (1991) The general and specific deterrent effects of DUI sanctions: A review of California's experience. Alcohol, Drugs and Driving. 7, 1, 13-42

Reason, J., Manstead, A., Stradling, S., Baxter, J. and Campbell, K. (1990) Errors and violations on the roads: a real distinction? Ergonomics. 33, 10/11, 1315-1332

Road Traffic Act (1988) HMSO: London

Road Traffic Law Review (1988) HMSO: London

Ronis, D.L., Yates, J.F. and Kirscht, J.P. (1989) Attitudes, decisions and habits as determinants of repeated behaviour. In Attitude Structure and Function eds. A.R. Pratkanis, S.J. Breckler, and A.G. Greenwald. Lawrence Erlbaum: Hillsdale, NJ, USA.

Ross, H.L. (1991) Licence deprivation as a drunk-driver sanction. Alcohol, Drugs and Driving. 7, 1, 63-68

Ross, M., McFarland, C., Conway, M., and Zanna, M.P. (1983) Reciprocal relationships between attitudes and behaviour recall: Committing people to newly-formed attitudes, Journal of Personality and Social Psychology. 45, 257-267

Smetana, J. (1981) Pre-school children's conceptions of moral and social rules. Child Development. 52, 1333-1336.

Smetana, J. (1985) Children's impressions of moral and conventional transgressors. Developmental Psychology. 21, 4, 715-724.

Struckman-Johnson, D.L., Lund, A.K., Williams, A.F. and Osborne, D.W. (1989) Comparative effects of driver improvement programs on crashes and violations, Accident Analysis and Prevention. 21, 3, 203-216

Turiel, E. (1983) Development of social knowledge: morality and convention. Cambridge University Press: Cambridge, England.

Verplanken, B., Aarts, H., van Knippenberg, A. and van Knippenberg, C. (1994) Attitude versus general habit: Antecedents of travel mode choice. Journal of Applied Social Psychology. 24, 4, 285-300.

Verplanken, B., Aarts, H. and van Knippenberg, A. (1997) Habit, information acquisition, and the process of making travel mode choices. European Journal of Social Psychology. 27, 539-560.

Verplanken, B., Aarts, H., van Knippenberg, A. and Moonen, A. (1998) Habit versus planned behaviour: A field experiment. British Journal of Social Psychology. 37, 111-128.

Verplanken, B. and Aarts, H. (in press) Habit, attitude and planned behaviour: Is habit an empty construct or an interesting case of goal-directed automaticity? In press: European Journal of Social Psychology. 10.

Verplanken, B.S., & Faes, S. (1999) Good intentions, bad habits and the effects of forming implementation intentions on healthy eating. European Journal of Social Psychology, in press.

Wark, R., Raub, R. A. and Reischl, B.E. (1998) Psychological and organisational variables relating to the effectiveness of a remedial driver training program, 24th International Congress of Applied Psychology, San Francisco, US.

Zimbardo, P.G. (1969) The cognitive control of motivation: The consequences of choice and dissonance. Scott, Foresman: Glenview, Ill., US.