Brake’s response to the Driving Standards Agency’s consultation
on learning to drive
October 2008
For all queries, please contact: Cathy Keeler, Deputy Chief Executive
tel: 01484 559909, email: ckeeler@brake.org.uk
About Brake
Brake is a national road safety charity, dedicated to stopping deaths and injuries on roads and caring for
people bereaved and affected by road crashes.
Brake carries out research into road users’ attitudes on a range of road safety issues, including aspects of
learning to drive; training and testing; driver attitudes and behaviour; traffic law and its enforcement; and
charges and penalties for traffic offences.
Through the ‘graduates’ of its Road Safety Academy, Brake delivers road safety education to thousands of
young people each year. This allows the organisation to collate information on the attitudes and self-reported
behaviour of young passengers and drivers.
Background to Brake’s consultation response
Brake’s policy recommendations on the issue of learning to drive have been developed in the light of
information on best practice, gathered through a number of forums.
1. Brake played an active part in a discussion group convened by the Association of British Insurers
(ABI) in 2006, which led to the publication of a report 1 endorsed by the ABI, Brake, the Make Roads
Safe Campaign, the Parliamentary Advisory Council for Transport Safety, the RAC Foundation and
RoadSafe. As a minimum, the report called for Government action to:
 set a 12-month minimum learning period before the driving test;
 implement a structured learning programme for all new drivers, covering the different road
and traffic conditions learner drivers should experience before taking the test;
 encourage young drivers to carry fewer passengers;
 encourage young drivers to avoid driving at night.
2. Brake encouraged public debate on the topic of young drivers and learning to drive through Road
Safety Week 2006, which took young driver safety as the national theme for the Week.
3. Brake has been gathering evidence on young people’s attitudes and behaviour as drivers and
passengers with other young drivers since April 2004, when it established a young driver education
scheme in London, the pre-cursor to the current Young Driver module of its Road Safety Academy,
sponsored by FedEx Express. Evidence about attitudes and behaviour is gathered prior to the
education session being delivered by a Brake volunteer and analyses to date have revealed
alarming levels of risk-taking among young drivers.2 3
4. Brake organised two international road safety conferences to disseminate information on best
practice policies and education of young drivers in May 2007 and January 2008, involving academic
and government speakers from a range of countries which have successfully managed to reduce
numbers of young driver and passenger casualties. 4
5. Brake provided written and oral evidence to Parliament’s Transport Select Committee inquiry on
novice drivers in 2006.
Summary of Brake recommendations for Government action
To tackle the high levels of young driver and passenger casualties and the numbers of fatal and serious
injury crashes involving younger, more inexperienced drivers, Brake recommends the following urgent action
by Government to improve young, learner and newly-qualified driver safety:
1
Young drivers: reducing deaths on the roads. Four actions to save lives, Association of British Insurers (2006)
Stradling S, Kinnear N, Mann H, Evaluation of Brake Young Driver Education Scheme: A Pilot Project funded by the Bridge House
Trust, Transport Research Institute, Napier University (2005)
3
An up-to-date analysis of this data can be found at Annex 2
4
A summary of some of the information presented at these conferences can be found at Annex 3
2
1
1. setting challenging targets to cut the numbers of young drivers and passengers being killed and
seriously injured on our roads and the number of fatal and serious injury crashes involving young
drivers
2. introducing a Graduated Driver Licensing system, involving:
a. a minimum learner driver period of at least 12 months, encouraging learner drivers to
gain supervised experience in all seasons and types of weather (this would also effectively
raise the age for driving solo to 18)
b. a minimum number of hours professional tuition that must be undertaken before
someone can sit their practical driving test
c. restrictions on newly-qualified drivers for a post-test period of at least 24 months,
during which time they should carry compulsory probationary ‘P’ plates. Restrictions
should include:
i. driving at night
ii. driving with young passengers (e.g. under-21s, or between the ages of 14-25)
iii. driving high-powered vehicles (e.g. over 1.4 litre engine)
3. introducing regular re-tests for all drivers (e.g. every five or ten years)
4. making road safety education a compulsory part of the National Curriculum, for all ages, to
foster safe behaviour and ingrain responsible attitudes towards risk-taking on the road before young
people get to the age where they start learning to drive
Brake also urges the Government to take the following actions, which would improve road safety generally,
but would be likely to have a particular impact on young, learner and newly-qualified driver safety:
5. increasing spending on road safety publicity, to enable year-round high-profile campaigns which
specifically target young drivers, including prime-time TV and cinema slots; and developing social
marketing techniques, particularly those that use new media such as viral emails, YouTube videos
and social networking sites to disseminate information
6. making roads policing a national policing priority and giving traffic police the powers and
resources they need to tackle risk-taking by young drivers (and others) on the roads, including
a. rolling out ANPR to help catch more unlicensed and uninsured drivers
b. enabling hit and run killer drivers to be prosecuted with tougher charges and receive
tough penalties
c. increasing the numbers of traffic police on our roads to police all types of dangerous driving
d. giving the police powers to stop drivers in random and targeted enforcement
campaigns to test for alcohol and drugs
e. introducing a charge of driving on illegal drugs which does not need police to prove
that driving is impaired
f. reducing the drink-drive limit for all drivers, including young drivers, from 80mg to 20mg
alcohol per 100ml blood
Brake’s response to specific questions from the consultation paper
Q1. What views do you have about our explanation of the high accident rate among newly-qualified drivers?
Brake agrees with the broad explanation given of the high crash and casualty rate among newly-qualified
drivers outlined in the consultation document: research from the UK and abroad indicates that age,
inexperience and attitudes to risk-taking are all important factors. It is clear from the high crash and casualty
rates of young drivers that the current system of learning to drive is not addressing these risk factors
adequately.
Age
While the consultation paper points out that newly-qualified drivers may be from any age group, it is
important to remember that the vast majority are young. TRL’s first cohort study clearly shows that young
novice drivers are more likely to have a crash in their first year of driving after passing the test than older
novice drivers: 18% of drivers who were aged between 17 and 19 years when they passed their test were
involved in a road crash in their first year of driving, whereas only 12% of drivers who were aged over 25
2
years when they passed their test had a crash in their first year of driving.5 The consultation paper notes that
‘drivers starting aged 27 years are about 30% safer than drivers starting aged 17 years’.
It is difficult to fully separate the effects of age and inexperience on the high crash rate of newly-qualified
drivers. (It is also possible that more of the people who delay learning to drive may have safer attitudes
towards driving.) However, research into the development of the brain, and in particular, the development of
the parts of the brain associated with judgement and decision-making in complex situations, confirm that age
does play a significant factor in the safety of drivers.
Crashes caused by young drivers often involve a lack of higher ‘executive’ cognitive functions associated
with the frontal lobe of the brain – such as emotion regulation, impulse control and decision processes, but
also functions that are considered as higher-level driving skills such as eye scanning, hazard anticipation
and risk management. Immature executive functioning may lie behind the poor hazard anticipation and
detection, decision-making and risk management skills that characterise many teenage drivers. Brain
imaging studies have shown that the frontal lobes are not fully developed until young people reach the age of
25 years, the same time when age stops being a significant risk factor for crashes.6
Analysis of international experience of changing the age at which people can learn to drive shows that crash
risk can be significantly reduced by delaying the age at which drivers have access to a full licence, even if
only by a matter of months.7
Inexperience
TRL studies and other research have clearly demonstrated that inexperience is a major risk factor for drivers
and gaining experience of driving helps to reduce crash risk. TRL’s first cohort study showed that just one
year’s driving experience reduced the risk of crashing by 38% for 17-year-old drivers, by 35% for 18-year-old
drivers and by 32% for 19-year-old drivers. Even for older novice drivers, inexperience is clearly still a major
risk factor: a 25-year-old’s crash risk reduces by 20% after one year’s driving experience, while a 30 year-old
novice driver’s crash risk decreases by 12%.8
The younger and more inexperienced a driver is, the greater their crash risk. Brake urges the Government to
implement changes that will ensure that drivers gain more experience before they are able to gain a full
driving licence. Introducing a GDL scheme, as outlined in our summary of recommendations for
Government, would bring the twin benefits of encouraging drivers to gain more experience, over time, while
automatically increasing the age at which they could gain a full licence.
Young driver attitudes
Brake agrees with the Government’s assessment that the majority of young drivers are not intentionally
dangerous, although our own and other surveys of young drivers show that more of them report taking risks
and acting illegally while driving than older drivers.9
This research, together with research into the attitudes of young drivers (by Stradling and others) suggests
that it is not just the case that young drivers do not realise their limitations; their attitudes towards driving also
lead them to under-estimate the consequences of behaviour that they know to be risky.
 Young drivers differ from older drivers in their attitudes towards driving. Young drivers report a
greater enjoyment of driving and are more likely to drive for pleasure. They get a stronger sense of
personal identity or status from driving; this includes feelings of pride, power, and confidence.10
 Young drivers, especially young male drivers, are also more likely to seek thrills from driving, are
more fearless and more compliant with peer pressure. 11
 Young male drivers are also far more likely to indulge in competitive behaviour by driving
dangerously around other vehicles. Between 1999 and 2003, 17-18 year-old males had 70% more
5
E Forsyth et Al, Cohort Study of learner and Novice Drivers: Part 3, Accidents, Offences and Driving Experience in the First Three
Years of Driving, TRL Project Report 111, 1995
6
See Annex 3, “Assessing the road ahead – hazard perception and risk management among young and novice drivers”, Dr Robert
Isler, Director of Traffic And Road Safety research group (TARS), University of Waikato, Hamilton, New Zealand
7
See Annex 3, “The road to young driver safety – research and recommendations for developed countries”, Divera Twisk, Chair of the
Young Drivers Working Group, Organisation for Economic Co-operation and Development, the Netherlands; “The components of GDL –
analysing the US experience”, Susan Ferguson, Former Senior Vice President, Research, Insurance Institute for Highway Safety, USA
8
E Forsyth et Al, “Cohort Study of learner and Novice Drivers: Part 3, Accidents, Offences and Driving Experience in the First Three
Years of Driving”, TRL Project Report 111, 1995
9
See Annex 2
10
Young Driver Attitudes, S. Stradling, M. Meadows (DfT, 2001)
11
Young Driver Attitudes, S. Stradling, M. Meadows (DfT, 2001)
3

crashes involving “interaction or competition with other road users” than 30-59 year-old male
drivers.12
Young drivers have a high opinion of their own skills on the road and rate their own performance as
above average. They are also more likely to equate ‘good’ driving with the ability to master the
controls of the car at higher speeds.13
Stradling concludes that these attitudes make young drivers more inclined than older drivers to take risks on
the road, such as driving fast, accelerating hard and cornering at high speed. 14
To improve young driver safety, it is important to tackle the attitudes that make young drivers more inclined
to take risks on the road and to minimise their exposure to these risks.
Speed, drink and drugs
Young drivers are more likely to seek thrills from driving fast and cornering at high speed than older drivers. 15
Even sticking to the speed limit can be too fast in the wrong conditions – such as in ice or snow, or on bendy
country roads - but young drivers, particularly male drivers, may be reluctant to drive under the speed limit
for fear of ‘losing face’ in front of friends.
Compared to drivers of all ages, young drivers aged 17-19 are ten times as likely to have a drink-drive crash
and young drivers aged 20-24 are four and a half times as likely to have a drink-drive crash per mile driven
(17-19 year olds have 41 drink-drive crashes and 20-24 year olds have 18 drink-drive crashes per 100
million miles driven, compared to all drivers, who have 4 drink-drive crashes per 100 million miles driven). 16
Young drivers under 25 are more than twice as likely to fail a breath test as older drivers (6.3% failure rate,
compared to 3.6%). When they are involved in a crash where someone is killed or injured, young drivers
under 25 are twice as likely to fail a breath test as older drivers (4.0% failure rate, compared to 2.0)%. 17
Brake urges the Government to introduce powers for police to enable blanket and targeted roadside breath
testing and ensure that there are adequate resources to enable forces to carry out enforcement checks to a
level at which drivers would expect it to be more likely than not for them to be caught if they chose to drink
and drive. This would improve the safety of all drivers, but given the additional crash risk for young drivers
who choose to drink and drive, would be likely to be particularly effective in preventing young driver drinkdrive crashes.
Driving at night
Driving at night requires specific skills, extra concentration – and extreme care. Young drivers may be under
the impression that because roads are quieter at night it is safer for them to speed or pay less attention to
the road. In fact, you can’t see as far. Driving at night requires specific skills, extra concentration and
extreme care. Drivers are more likely to be tired, drunk or drugged and so are any other road users who are
out and about, such as drunk pedestrians rolling out of pubs and clubs in the early hours.
Evidence published by the Association of British Insurers and the US Insurance Institute for Highway Safety
shows that young drivers in this country have a higher proportion of crashes in the evenings and early
mornings than older drivers. Between the hours of 2am and 5am, male drivers are 17 times more likely to be
killed or seriously injured in a road crash than male drivers of all ages. 18 The proportion of crashes that occur
at night reduces as the age of the driver increases. 19
Carrying passengers
Research shows that peer pressure can encourage bad driving and result in drivers ‘showing off’ to their
passengers. Behavioural research shows that young male drivers in particular, are much more likely to take
risks on the road if there are young male passengers in their vehicle.20 All too often when serious crashes
12
Young Drivers: Road Safety and the Cost of Motoring, Interim report and consultation paper, (Association of British Insurers 2005)
Young and Novice Driver Policy Statements (Royal Society for the Prevention of Accidents 2002)
14
Young Driver Attitudes, S. Stradling, M. Meadows (DfT, 2001)
15
Young Driver Attitudes, S. Stradling, M. Meadows (DfT, 2001)
16
Road Casualties Great Britain: 2005 (DfT, 2006)
17
Road Casualties Great Britain: 2005 (DfT, 2006)
18
Night-time Accidents, H. Ward, (Centre for Transport Studies, University College London, 2005)
19
See Annex 3: “The components of GDL – analysing the US experience”, Susan Ferguson, Former Senior Vice President, Research,
Insurance Institute for Highway Safety, USA
20
Understanding Passenger Influences on Driver Behaviour: Implications for Road Safety and Recommendations for Countermeasure
Development, Regan & Mitsopoulos, Monash University Accident Research Centre, Report 180 (2001)
13
4
involve a young driver, it is not only the driver who has been killed or seriously injured, but their young
passengers as well.
US research confirms that the already high crash rate for 16-19-year-olds driving alone is greatly increased
when passengers are present. The more passengers, the more risk and the risk is higher when the drivers
are aged 16 and 17 rather than 18 and 19. With two or more passengers, the fatal crash risk for 16-19 yearold drivers is more than five times what it is when driving alone. 21 22
Due to the substantial additional risks for young drivers, restrictions on driving at night and while carrying
passengers are key components of GDL systems in many countries. Brake urges the Government to
introduce similar restrictions as part of a GDL system to minimise drivers’ exposure to these risks until they
have more driving experience under their belt. Many GDL systems set a lower drink-drive limit for drivers at
the ‘learner’ and ‘restricted’ licence stages: Brake is urging the Government to reduce the current
dangerously high drink-drive limit for all drivers. Restrictions should also be considered for any other
substantial additional risks identified by analysis of insurance and crash data – anecdotal evidence suggests
that inexperienced drivers are more likely to crash when driving more powerful vehicles than they are used
to, which would suggest a restriction on the size of engine would be a key additional measure.
Q2. Do you have any comments about the contents of the Partial Impact Assessment published alongside
this Paper?
Brake notes the high level of assumptions being made about costs in the Partial Impact Assessment, which
make it difficult to comment on the cost-benefit estimates for each of the options outlined. For example, the
annual value given to holding a full driving licence for new learner drivers is estimated to be between £800
and £1,500. The footnotes state that these figures are based on:
 an analysis of Norwegian drivers’ willingness to pay for not losing their licence
 a DSA estimate of the average cost of learning to drive
Removing the privilege of holding a licence once drivers have become used to it is by no means the same
scenario as delaying access to a full licence for drivers who do not already hold one. It is similarly unclear
how the average cost of learning to drive relates to the annual value of holding a full driving licence – it would
seem more likely that 17-18 year-olds would save money by not having a driving licence.
Government impact assessments tend to assign low values to the benefits for society of preventing death
and injury on the roads. We therefore think it likely that the paper underestimates the potential benefits of
improving safety on the roads, in particular when assessing the likely impact of introducing combinations of
the different options, which could prevent larger numbers of deaths and serious injuries.
The Partial Impact Assessment seems to ignore any likely reductions in insurance premiums due to fewer
deaths, serious injuries and overall numbers of road crashes. Lower average insurance costs would provide
less incentive for drivers to risk driving uninsured and the impact of having fewer uninsured drivers on the
roads would help to bring insurance costs down even further.
Q3. What are your views about our analysis that improved training and testing is the best way of improving
the safety of newly-qualified drivers?
While effective training and testing is extremely important in setting a high minimum standard of safety on
the roads, there are clear benefits that could be introduced by both of the other main options identified by the
Government. Brake does not agree that the third option is ‘the only one which can definitely improve driving
standards’; nor is there any evidence that either of the first two options would ‘disadvantage or
inconvenience safe, law abiding driver, or encourage people to drive outside of the official system’.
The existing theory test, hazard perception test and single test of practical skills are clearly inadequate in
identifying whether someone is a safe driver. There is currently no minimum learning period or requirement
for professional tuition, which means learner drivers may obtain a full driver licence within weeks of turning
17, without adequate tuition and with very little experience. The existing tests do not test a driver’s attitude to
safe driving, or their ability to drive in different levels of traffic, in the dark, in different weathers, or on the
motorway. Researchers have also voiced concerns that the current hazard perception test only provides a
very crude measure of hazard perception and may in fact be putting some drivers with good hazard
perception skills at a disadvantage. 23
21
Doherty, ST; Andrey, JC; and MacGregor, C, 1998. The situational risks of young drivers: the influence of passengers, time of day,
and day of week on accident rates.
22
See Annex 3: “The components of GDL – analysing the US experience”, Susan Ferguson, Former Senior Vice President, Research,
Insurance Institute for Highway Safety, USA
23
See Annex 3: ‘The Road Ahead: Hazard perception skills among young and novice drivers’, Dr. David Crundall, Accident Research
Unit, School of Psychology, University of Nottingham
5
Brake urges the Government to adopt a combination of all approaches described in Chapter 4 of the
consultation paper. The three options outlined are by no means mutually exclusive and if applied together
would be likely to save more lives.
In Brake’s view, the priority for Government action should be to change the emphasis of the process of
learning to drive, so it focuses more on the attitudes and experience needed to be a safe driver, while
ensuring that drivers have the necessary technical skills. Adopting a GDL system which incorporates a
minimum learning period, gradual exposure to more risky driving situations, clear communication of the
standards needed for safe driving and robust testing would bring benefits in making sure drivers have
substantial driving experience under their belt before acquiring a full licence.
Brake urges the Government to adopt a three-stage GDL system (learner, novice and full licence holder),
structured as follows:
Stage one – learner (displaying red ‘L’ plates)
Drivers should hold a learner’s licence for a minimum of at least 12 months, before taking the practical
driving test, theory test and hazard awareness test. A minimum learning period of at least 12 months
ensures that drivers are older and would help to ensure that they are more experienced before being able to
take their test. It also allows learners to gain experience in all seasons and in all types of weather. The
Partial Impact Assessment suggests that a 12-month minimum learning period would prevent 548-840
deaths and serious injuries – a significant number.
During this time, learner drivers should not drive unsupervised. They should undergo a minimum of 10 hours
professional tuition in a car with dual controls, but should be encouraged to sign up for much more. All
candidates should complete a standard syllabus which is monitored through a mandatory logbook and
‘signed off’ by an ADI to state that a candidate is ready to take the test to gain a ‘novice’ driver licence.
Accompanying drivers
Brake is not opposed to learners and novice drivers having additional practice with non-professional
accompanying drivers, as this can be an important way of getting on-road experience. However, we would
like to see additional controls put in place to ensure that this role is taken seriously, by responsible
individuals. Brake recommends that the minimum age of non-professional accompanying drivers should be
raised to at least 25 to ensure that they do not fall into the high-risk age group of 17-24 year-old drivers.
Accompanying drivers should have held a full driving licence for at least five years and should have a clean
driving licence. This would help to prevent drivers who take dangerous risks on the road from passing their
dangerous habits and attitudes on to learner drivers. Drivers wishing to accompany a learner driver should
register themselves as ‘approved accompanying drivers’, by completing a questionnaire to prove their
suitability, which could be checked by their insurer.
Drink-drive limit
Brake supports a lower drink-drive limit of a maximum of 20mg alcohol per 100ml blood for learner and
novice drivers (and all drivers).
Stage two – novice/provisional (displaying green ‘N’ or ‘P’ plates)
Drivers should hold a ‘novice’ licence for a minimum of two years after passing a theory, hazard perception
and practical driving test.
Novice drivers should be allowed to drive unsupervised, but with certain restrictions to limit their exposure to
high-risk situations. Brake recommends the following restrictions on unsupervised driving:
 carrying young passengers (e.g. under-21s, or between the ages of 14-25). Parents who are novice
drivers and need to carry their own children or other dependants could be exempt from this
restriction;
 driving between 11pm and 5am;
 driving on motorways;
 driving vehicles with powerful engines. The Government should seek specialist advice on what size
engines should be restricted.
 other risky driving situations identified by crash, casualty and insurance data.
During the novice driver period, drivers should be required to take a minimum of a further 10 hours of
professional tuition, but should be encouraged to sign up for much more. As with the learner stage,
candidates should complete a standard syllabus (including a requirement to gain tuition and supervised
experience driving on motorways and at night) which is monitored through a mandatory logbook and ‘signed
off’ by an ADI to state that a candidate is ready to take the test to gain a full driver licence.
6
Stage three – full licence
Drivers should only be able to apply for a full licence after holding a novice licence for a minimum of two
years. They should be required to pass a second theory, hazard perception and practical driving test to help
ensure their level of safe driving is consistent on all types of roads, including motorways.
Q4. Which do you think would be most helpful to improve the educational value of the theory test?
Q5. Do you agree or disagree that case studies could have a role in helping assess whether learners have
understood driving theory better?
Q6. What other methods could be used to assess whether learners understand driving theory?
Brake agrees that the theory test should seek to assess a candidate’s understanding of the subject matter
and not just their knowledge. We do not have a strong view on whether the Government should continue to
publish the theory test question bank and answers. If the question bank continues to be published, perhaps a
greater number of questions would help to ensure that candidates cannot just learn the answers by rote?
Brake agrees that case studies that assess learners’ understanding of driving scenarios would be of value.
We also think there is scope within the theory test to include more questions on the risks drivers face, and
pose to others, on the road. This could include questions on, for example, how the speed a vehicle is
travelling affects the survival chances of a pedestrian it hits, or the numbers of people who are killed and
seriously injured on the road.
Q7. How can we improve road safety using the hazard perception test?
Although there is evidence that the introduction of hazard perception testing has been beneficial,24
researchers and practitioners have identified problems with the current button-press test, which cannot tell
whether someone is pressing the button in response to the hazard that the test wants them to identify, or in
response to something else on the screen.25
More sophisticated ways of testing driver hazard perception could include tracking eye movements in driving
simulators – this would enable examiners to see not only when the driver spots hazards, but how effectively
they are scanning the scene. 26
Brake is not convinced that 3D animation would be appropriate – there is a danger it could seem one step
even further removed from reality than film clips – but agrees that it would be useful to research the potential
of using animation.
Q8. Do you agree or disagree that the marking system for the practical test should focus more on evidence
of competence than on evidence of weakness?
Q9. Do you agree or disagree with the introduction of independent driving into the practical test?
Q10. Do you agree or disagree with the introduction of Situational Judgement exercises into the practical
test?
Q11. Do you have any comments on the way in which we test specific manoeuvres in the practical test?
Brake agrees that the proposed changes to include elements of independent driving and ‘situational
judgement’ into the practical driving test would be beneficial and it would seem sensible to require a
candidate to display a consistently good standard of driving throughout the driving test.
We also think a longer driving test would be beneficial, opening up possibilities for ensuring that the test
covers a wide range of different types of road and requiring the candidate to display consistently safe driving
over a longer period of time.
Q12. Do you agree or disagree that:
(a) The theory test should be uncoupled from the hazard perception test?
(b) The specified driving manoeuvres should be tested separately from the general driving part of the
practical test?
Brake has no strong views on whether it would be beneficial or not to split the driving tests into a number of
different modules. However it would be important to trial any changes to ensure that modular testing was a
24
Cohort II: A Study of Learner and New Drivers Vol 1: main report, (TRL, 2008)
See Annex 3: ‘The Road Ahead: Hazard perception skills among young and novice drivers’, Dr. David Crundall, Accident Research
Unit, School of Psychology, University of Nottingham
26
See Annex 3, “Assessing the road ahead – hazard perception and risk management among young and novice drivers”, Dr Robert
Isler, Director of Traffic And Road Safety research group (TARS), University of Waikato, Hamilton, New Zealand; ‘The Road Ahead:
Hazard perception skills among young and novice drivers’, Dr. David Crundall, Accident Research Unit, School of Psychology,
University of Nottingham; ‘Going the extra mile: holistic learning based on international research’, Gary Austin, Managing director,
A2OM Academy
25
7
good indicator of safe driving. It would also be important to ensure that a modular format did not encourage
drivers who had passed some, but not all, of the modules to drive unsupervised, as if they had already
gained their licence.
Q13. What are your views about providing more comprehensive feedback to all candidates at the end of
each assessment, regardless of the result?
Brake would welcome examiners providing constructive feedback at the end of assessments to candidates
and supervising drivers.
Q14. What are your views about the proposed student workbook? How useful would a voluntary document
be for all learners when they start learning to drive?
Q15. Do you support the idea of progress being recorded in a student workbook?
Brake would welcome the introduction of a driver syllabus to set out the minimum standards required to pass
the test. It should provide a clear framework of the skills, understanding, experience and attitudes needed to
be a safe driver.
However the proposed student workbook to accompany the syllabus should be mandatory. While voluntary
schemes may be of benefit to more responsible, safety-conscious young drivers, because they are voluntary,
they are unlikely to be taken up by risk-taking young drivers, who believe themselves invincible and see the
driving licence as their ticket to independence and status among their peers. Previous voluntary schemes
introduced by Government, such as the logbook introduced in March 2003 and ‘Pass Plus’ post-test training
have resulted in extremely low take-up and it is unlikely that any voluntary workbook would be widely and
systematically used.
A mandatory workbook could require learners and novice drivers to demonstrate that they had gained
minimum levels of experience of driving in different situations prior to taking the test. Learners should be
required to complete the workbook and get it ‘signed off’ by an ADI prior to sitting the driving test.
Q16. What sort of information should be considered in creating a star rating system to help learners in
choosing their instructor?
Q17. What are your views on the usefulness of publishing the pass rates for different instructors?
Brake supports the idea of providing learners with extra information, help and advice to enable them to
choose a suitable instructor and understand the qualifications they hold and the services they provide.
However it is not clear from the consultation paper how such a system might work.
Publishing pass rates on their own would be of limited use to learner drivers, as they are difficult to interpret
without additional information on the typical ‘types’ of driver taught by the instructor and their risk of crashing
once they had passed their test. This would mean criteria for any star rating system would have to be
developed extremely carefully. It would need to be designed to enable new instructors to earn a living, while
encouraging them to gain additional qualifications and develop their skills.
Q18. Do you agree or disagree that learners should be required to have a ‘test readiness’ certificate signed
by a supervising driver or driving instructor, before they can take a practical test?
Brake strongly agrees that learners should have a ‘test readiness’ certificate, signed by a driving instructor
(not a supervising driver) before they can take a practical test. This should only be signed on completion of
the mandatory student workbook.
Q19. Do you agree or disagree that practical test candidates and their supervising drivers would benefit if the
supervising driver were to sit in on the test?
Q20. Do you agree or disagree that practical test candidates and their supervising drivers would benefit from
the supervising driver sitting in on the debrief at the end of the test?
Brake does not have any strong views on whether supervising drivers should be allowed or required to sit in
on the practical test. However, it would be important to ensure that this did not put extra pressure on the
examiner. We agree that having the supervising driver sitting in on the debrief at the end of the test would be
extremely beneficial, particularly when the candidate has not passed the test, as this will provide useful
information for further training.
Q21. Do you think an Attitude Advisor is likely to offer benefit by:
(a) providing useful guidance to students to help their learning programmes?
(b) providing useful guidance to their supervising drivers?
Brake agrees that the development of an Attitude Advisor could provide useful guidance to students and to
their supervising drivers. We would also like to see the Government carrying out research to examine
8
whether this type of attitude assessment, or other useful psychometric testing, could be incorporated into the
driving test to help reduce crashes among newly-qualified drivers.
Psychometric testing is already used by a number of fleet operators as a way of assessing driver risk and
tailoring training and other fleet safety measures accordingly. There are already a number of providers of risk
assessments, including:
 DriverMetrics – a company which offers psychometric testing for fleet, bus and emergency service
drivers. Developed by Cranfield University, the testing system used by driver metrics is called the
Driver Risk Index. It assesses drivers’ level of risk on the road, which takes into account behavioural
and attitudinal factors, including those linked to stress. 27
 Interactive Driving Systems – a research-led provider of global fleet risk management solutions, has
developed an online driver risk assessment called RoadRISK that helps companies identify which
drivers are most at risk of collisions, and tailor interventions accordingly. 28
 The Royal Society for the Prevention of Accidents’ (RoSPA) – RoSPA has developed a Driver Risk
Assessment Software program, which provides organisations with an assessment of their drivers’
strengths and weaknesses in relation to safe driving. It helps managers to decide an appropriate
level of training for each individual as required, to target any weaknesses. 29
Psychometric testing has also recently been used in rehabilitation programmes for driving offenders. For
example, Thames Valley Police runs a speed awareness workshop for people caught speeding, developed
by Reading University psychology professor, Frank McKenna. It uses a computer-based program and
questionnaire to assess driver risk. Participants are shown video sequences where they are required to spot
hazards, or state at which speed they would drive in different scenarios. A risk profile for the participant is
built up, including their likelihood of speeding, tailgating, using driving as an emotional outlet and
susceptibility to fatigue. Feedback on the risk profile is given to the participants, with advice tailored to
reducing risk. The programme has also been adapted for use in companies.30
One key failing of the current training and testing system is that there is no examination of a driver’s attitudes
towards safe driving. If a test could be developed that could identify attitudes and behaviours that mean a
driver is more likely to take risks on the road, Brake would like to see this forming part of the driver licensing
process.
Q22. How much do you think that learners would benefit from attending driver discussion groups?
While Brake thinks that learner drivers could potentially benefit a lot from attending driver discussion groups,
much would depend on the format and content of these groups. It is difficult to see how the over-confident
drivers who would most benefit from having their attitudes and beliefs challenged by their peers would be
persuaded to attend driver discussion groups, unless there were some incentive to do so.
Discussion of safe road use, to encourage safer attitudes, would perhaps be better delivered in schools, as a
compulsory part of the National Curriculum, through the PSHE or Citizenship syllabus. There is much
anecdotal evidence that attitudes towards safe driving are formed well before young people are of the age
where they can gain a provisional licence and holding discussion groups in schools would allow these to take
place well before this age.
Q23. Are you aware of any evaluated road safety education programmes which could inform our work with
pre-drivers, and that you would like to make us aware of?
Brake has been running a road safety education programme for 15-25 year-olds (and younger youth
offenders) through its Road Safety Academy for five years. Trained volunteers show Brake’s DVD Too
Young To Die and run a discussion group based on the road safety issues it raises, which include speed,
drink, drugs, seat belts and the possible consequences of bad driving. The young people are encouraged,
through discussion, to come up with their own strategies for keeping safe in different scenarios where peer
pressure may be exerted on them to take risks as passengers or drivers.
Feedback from the young people taking part in this education programme suggests that the vast majority of
them engage with the presentation and consider the advice to be memorable, with many pledging to use
what they have learnt by driving safely.
27
See www.drivermetrics.co.uk
See www.virtualriskmanager.net
29
See www.rospa.co.uk/drivertraining/software/index.htm
30
See www.perceptionandperformance.com
28
9
A pilot of the programme was evaluated by Professor Steve Stradling and a report was published in 2005. 31
A further evaluation, as yet unpublished, has just been completed by Kirklees Metropolitan Borough Council
as part of a project grant-funded by the Department for Transport.
Q24. As well as the subjects mentioned in this Paper, what else should be covered in the pre-driver
qualification in safe road use?
Some important topics are mentioned in the proposed content for a pre-driver qualification in safe road use,
but it is also important to include:
 understanding risks relating to road use, including: the risk of death and injury on different modes of
transport and for different age groups when using roads; the risk of crashing, death and injury
associated with different types of dangerous behaviour on roads; and additional risk factors that are
associated with increased crash and casualty risk for young drivers and passengers;
 understanding and emotionally engaging with the consequences of serious crashes on roads,
including: the effects of death and serious injury on families, friends and communities; and how
causing a serious crash can impact on a driver’s life (emotional and financial costs, as well as
possible costs to freedom) and the life of their family;
 elements of ‘planned behaviour’, such as planning possible strategies for dealing with difficult
situations and ensuring that they do not arise in the first place;
 ‘brain-training’, using film clips to help develop advanced hazard perception skills (as described in
Annex 3).
Q25. How can we make this qualification appeal to as wide a range of people as possible? How can it be
made engaging, and where should it be made available?
Q26. What are your views on a pre-driver qualification in safe road use? Do you think young people would
benefit from participating in it?
It is unclear from the consultation paper how the proposed pre-driver qualification in safe road use will be
examined and whether it would count towards gaining a driving licence, which could provide some incentive
for people to take it. It is obviously important to ensure as far as possible that a pass mark does reflect
whether someone is a safe road user, in particular if it is envisaged that the qualification could give
candidates credit towards gaining their driving licence.
If the qualification is to be voluntary, it is likely to suffer from similar problems as previous voluntary
schemes, such as the current DSA driver logbook and Pass Plus: low take-up; and a tendency to be taken
up only by people who already have some of the safest attitudes towards learning to drive. In order to ensure
better take-up of a voluntary scheme, there would have to be better incentives. These could include:
 lower insurance costs for learners who have already gained the qualification – but discounts would
have to beat the lowest insurance quotes significantly to provide a real incentive to take the
qualification (and the qualification would need to deliver tangible crash and casualty reductions in
order to justify lower insurance costs)
 credit towards passing the driving theory test – although it would be important to ensure that the
qualification was not an ‘easy’ alternative to taking the theory test and would not act to speed up the
process of gaining a driving licence.
Brake strongly recommends that making road safety a compulsory part of the National Curriculum taught in
schools is prioritised above developing a voluntary pre-driver qualification. It is obviously beneficial for
children to be educated about the risks they face on the roads, given comprehensive information about
safety on and around the roads, and informed about what they can do to minimise risks on the road, at all
stages of their development. Starting at an early age would help to address dangerous young driver attitudes
before they develop, and replace them with safer attitudes to driving.
Government is currently failing to ensure that children are given life-saving road safety lessons. Road safety
is not a compulsory part of the national curriculum and although local authority road safety officers do
fantastic work at local level, they do not have the necessary resources to deliver road safety education to all
pupils in all schools. While the Driving Standards Agency has stepped up delivery of its ‘Arrive Alive’
programme, in which driving examiners give road safety presentations to teenagers in schools and colleges,
information provided by DSA to Brake in 2006 confirmed that it is still reaching less than 1% of 15-25 year
olds in the UK.32
31
Stradling S, Kinnear N, Mann H, Evaluation of Brake Young Driver Education Scheme: A Pilot Project funded by the Bridge House
Trust, Transport Research Institute, Napier University (2005) – available on www.brake.org.uk
32
DSA examiners deliver approximately 5,000 ‘Arrive Alive’ presentations each year, to an average of 50,000 teenagers. (DSA, 2006)
According to the National Cenus 2001, there are 6,962,447 15-25 year olds in the UK.
10
By making anti-drug education compulsory in schools in 2000, the Government clearly demonstrated that it
accepts the importance of compulsory education on issues which lead to dangerous, anti-social behaviour.
Brake urges the Government to extend this approach to road safety education.
Q27. How do you think we can use additional qualifications to encourage a culture of lifelong learning?
As can be seen from the low take-up of additional driver training, such as Pass Plus and advanced driver
training, drivers do not currently feel that the incentives to refresh and develop their driving skills outweigh
the inconvenience and expense of doing so. In a Brake and Green Flag survey of 789 drivers and
motorcycle riders, published in 2005, although more than half of respondents (54%) thought their driving
would benefit from additional driver training, less than a quarter (23%) had received any training whatsoever
since passing their test.33 To develop a culture of ‘lifelong learning’ when it comes to driving, the incentives to
take further training (such as the possible insurance incentives described in the consultation paper) would
have to be much greater.
While Brake supports the development of a replacement scheme for Pass Plus and a common brand and
assessment system for advanced driver training, we think it is crucial that drivers should pass an
assessment to prove they have successfully completed any additional training or qualifications before being
eligible for insurance discounts.
Brake is pleased to see the Government is planning additional measures to support employers who train
their drivers and develop new qualifications based on specific relevant driving competences. It is important to
ensure that people who drive for work, particularly those who drive high mileages for work purposes, are
encouraged to adopt a culture of lifelong learning.
The best way of encouraging a culture of lifelong learning would be to require regular re-tests or driving
assessments for all drivers, for example every five or ten years. This measure has a substantial level of
public support as, despite the inconvenience factor, the benefits are obvious. In a Brake and Green Flag
survey of 789 drivers and motorcycle riders, published in 2005, nearly half of respondents (47%) said that
drivers should have to re-take their driving test at least every 10 years, while one in eight (13%) thought it
should be more often.34
Q28. How can motorway driving be taught more effectively?
Brake agrees that it is right to continue to prohibit learner drivers from driving on motorways. The best way of
ensuring that motorway driving is taught more effectively would be by incorporating mandatory instruction in
a dual-controlled car, supervised by an ADI, and followed by additional supervised driving on motorways,
into the ‘novice’ phase of a graduated driving licence.
Q29. How can we best apply our reforms for learning to drive to those who want to ride a motorcycle?
Brake has no specific recommendations for applying the measures outlined in this consultation response to
learner and newly-qualified motorcyclists. However it is important to note that the principle of restricting the
riskiest riding situations has already been adopted for motorcycles and graduated access to riding larger
motorcycles is due to be phased in, under EU legislation. This would provide an excellent opportunity for the
Government to introduce a graduated driver/rider licensing scheme for all drivers and riders.
33
34
Are you ready to drive? The Green Flag report on safe driving part 5 (Brake, 2005)
Are you ready to drive? The Green Flag report on safe driving part 5 (Brake, 2005)
11
ANNEX 1: Graduated Driver Licensing – evidence that it works
The UK Government has stated that research indicates that extending the learning period would lead to a
reduction in the number of casualties on our roads. In 2002, it calculated that a six-month minimum learning
period would prevent about 900 casualties each year, including 120 would be deaths and serious injuries.
The longer the minimum learning period, the greater the casualty savings would be. 35
Brake is calling for a minimum learning period as part of a Graduated Driver Licensing (GDL) system. A GDL
system would enable drivers to gain experience on ALL types of roads, and over a period of time to ensure
that they are skilled before being allowed to drive in the higher risk situations. The GDL system outlined by
Brake in this consultation response would also mean that drivers were at least 20 years old and had a
minimum of three years’ driving experience, including at least two years’ solo driving experience on a
restricted licence, before they were able to hold a full driving licence.
GDL systems have been successfully adopted in a number of other countries with the effect of reducing
young driver casualties and evidence from these countries suggests that they could lead to a dramatic
reduction in the number of deaths and serious injuries of young people on our roads. Some examples are
given below.
New Zealand
New Zealand’s multi-stage GDL, the first in the world, was introduced in 1987. Prior to that date, a full car
licence could be obtained at age 15 after passing written, oral and practical tests. Before May 1999 GDL
applied only to people under 25 but it now covers novice drivers of all ages.
Learner’s permit
 Minimum age for holding learner’s permit is 15
 Must be held for 6 months minimum
 Learner must always be accompanied by supervisor (a licence holder aged 20 or older who has held
a licence for 2 years or more) in front seat
 A maximum blood alcohol content (BAC) of 0.03 applies
 Learners must carry their learner licence when driving
 Violations result in extensions to learner period of up to 6 months
 Must display L-plates
Restricted licence
 Drivers under 25 must hold a restricted licence for 18 months, but this can be reduced to 12 months
if the driver attends an approved driving course
 Drivers 25 or over must hold the restricted licence for at least 6 months (reduced to 3 months on
completion of approved course)
 Restricted licence holders must not drive between 10pm-5am unless accompanied by a supervisor
 Restricted licence holders must not carry passengers unless accompanied by a fully licensed adult
driver
 A maximum BAC of 0.03 applies
 The licence must be carried in the vehicle when driving
 Any violation of restrictions may lead to an extension of the restricted period of up to 6 months (can
be imposed by Chief Traffic Officer)
 Drivers must pass an exit test at the end of the restricted stage, focussing on higher order driving
skills such as hazard perception. This is a road test and involves stopping the vehicle and reporting
hazards to the examiner and describing hazards while driving.36
An analysis of the system in New Zealand by TRL showed that following its introduction, there was a
reduction in car crash injuries of 23% for 15-19 year olds. Importantly, given that drivers in the UK cannot
obtain a learner’s licence until the age of 17, the drop in casualties was also significant –12% – for 20-24
year olds.37
A 2002 study, by the Injury Prevention Research Unit in New Zealand, found that young people were
reasonably accepting of the restrictions, with the passenger restriction being the least acceptable. 38
Introducing a More Structured Approach to Learning to Drive – Consultation, (DfT, 2002)
Graduated Driver Licensing – A Review of Some of the Current Systems, (TRL Report 529, 2001)
37
Graduated Driver Licensing – A Review of Some of the Current Systems, (TRL Report 529, 2001)
38
Graduated Driver Licensing: The New Zealand Experience, (Dorothy Begg & Shaun Stephenson, Journal of Safety Research, 342,
2002)
35
36
12
California, USA
California is just one of several US states to have introduced some form of GDL. California introduced a new
licensing system in 1983, but made further changes in 1998, which strengthened the graduated licensing
provisions.
Provisional permit phase:
 Provisional permit issued from age 15 on passing knowledge test (traffic law an signs) and vision
screening
 Must be supervised by fully-licensed driver aged 25 or more
 Permit must be held for at least 6 months before applying for a provisional licence
 Driver education course to be completed
 Six hours of professional driver training required
 At least 50 hours of practice certified by supervising adult. At least 10 hours of this must be at night.
 Must pass driving test to obtain provisional licence.
Intermediate phase (provisional licence):
 During first year, must not drive between midnight and 5am unless accompanied by a driver aged 25
or more
 During first 6 months, must not carry passengers under the age of 20 unless accompanied by a
driver aged 25 or more
 During second 6 months, must not carry passengers under the age of 20 in the car between
midnight and 5am unless accompanied by a driver aged 25 or over
 Full licence available from age 18 years
 Maximum BAC for drivers under 21 is 0.01
 There is a youthful driver improvement programme in place (warning letter after 1 st offence, a one
month licence revocation after 2nd offence)39
 An investigation of the 1998 changes to the system in California reported a 20% reduction in at-fault
fatal and injury crashes for 16-year-old drivers, and a 21% reduction in deaths and injuries among
teenage passengers of 16-year-old drivers. 40
Washington State, USA
A GDL system was introduced 1991, including:
 a driving curfew between 1-5am in the first year of driving
 a ban on carrying teenage passengers in the first six months of driving
 an automatic licence suspension for under-18s of up to six months for committing two traffic offences
Annual fatal and serious crashes involving 16 and 17 year old drivers decreased by 59% comparing the two
years before and after introducing GDL. Surveys of the public have found substantial support for the
measures introduced:
 69% support restrictions on novice drivers (including 43% of 17-24 yr-olds surveyed)
 67% support a minimum learning to drive period
 83% think learners should have to take lessons with a qualified instructor
39
40
Graduated Driver Licensing – A Review of Some of the Current Systems, (TRL Report 529, 2001)
Graduated Driver Licensing – A Review of Some of the Current Systems, (TRL Report 529, 2001)
13
ANNEX 2: Analysis of the evidence collated by Brake about the attitudes and behaviour of young
drivers and passengers
Evidence gathered prior to Road Safety Academy presentations for young drivers
Updated October 2008 (initial analysis of statistics, not previously published)
Levels of risk-taking from a survey of 9,349 young drivers and passengers (average age 16-17), of whom
2,126 (23%) say they have a full or provisional driving licence
 1,553 (17%) say they have crashed at least once while driving
o 427 (5%) say they have crashed more than once
 446 (5%) say they have driven a stolen vehicle
 402 (4%) admit having driven on illegal drugs
o Drugs they admitted driving on included: amphetamines, acid, cannabis, crack, crystal meth,
cocaine, ecstasy, heroin, LSD, magic mushrooms, speed, and others
 779 (8%) admit having drunk alcohol before driving
 2,095 (22%) say they have been a passenger in a vehicle when the driver was driving under the
influence of drugs or alcohol
 1,064 (11%) say they have driven at more than 70mph on a rural road
 1,967 (21%) say they have raced another driver (or been a passenger in a vehicle being raced
against another one)
 5,397 (58%) say they have been a passenger in a vehicle being driven at more than 40mph in a
30mph zone or at more than 70mph on a rural road
o Of these, 2,177 (40%) have asked the driver to slow down, 3,162 (59%) have not
 7,248 (78%) say they always belt up in the front, 141 (2%) admit they never belt up in the front
 6,266 (67%) say they always belt up in the back, 440 (5%) admit they never belt up in the back
Evidence gathered through surveys of drivers 2004-7
Published in The Green Flag Report on Safe Driving Part Six: Young drivers (Brake, 2007)
(Survey of 4,640 drivers, 533 of whom were aged between 17 and 24)
Copies of the full report are available on request from ckeeler@brake.org.uk or on www.brake.org.uk.
Levels of risk-taking of drivers aged 17-24, compared to older drivers
 Almost half of young drivers (45%) admit driving the morning after drinking a lot of alcohol,
compared to a quarter of older drivers (26%)
 One in 11 young drivers (9%) admits driving on illegal drugs, compared to just 3% of older drivers
 Six in ten young drivers (60%) say they drive at or above 35mph in a 30mph zone once a week or
more frequently, compared with less than four in ten older drivers (37%)
 Four in ten young drivers (40%) say they speed on derestricted rural roads, compared to two in ten
older drivers (20%)
 One in 3 young drivers (33%) admits overtaking blind, compared to one in 11 older drivers (9%)
o One in 6 young drivers (17%) admits doing this once a week or more
 Half of young drivers (50%) admit using a hand-held mobile phone while driving, compared to 35%
of older drivers
o One in 5 young drivers (20%) admits doing this once a week or more
Drivers’ views on young driver safety
 More than half of young drivers (55%) believe they are safer than most drivers, compared to just
under half (49%) of older drivers
 The majority of drivers of all ages (58%) agree that there should be restrictions on young drivers
o More than four in ten young drivers (43%) agree that there should be restrictions on young
drivers, while almost three-quarters of older drivers (73%) agree there should be restrictions
on young drivers
Published in The Green Flag Report on Safe Driving Part Three: are you ready to drive? (Brake, 2005)
(Survey of 789 drivers and riders, 95 of whom were aged between 17 and 24)
Copies of the full report are available on request from ckeeler@brake.org.uk or on www.brake.org.uk.
Drivers’ views on the current procedures for teaching and testing novice drivers
 One in five drivers (20%) said they think the current driving test is too easy
14


More than two-thirds of drivers (67%) said they are in favour of novice drivers completing a minimum
number of hours of supervised driving (e.g. 50 hours) over a minimum period (e.g. six months)
before they can take their driving test
More than four-fifths of drivers (83%) said novice drivers should be required to take lessons with a
qualified driving instructor
Drivers’ views on restrictions on young drivers
 Seven in ten drivers of all ages (69%) said there should be restrictions on young drivers
o Nearly half of 17-24 year-old drivers (43%) said there should be restrictions on young drivers
 Of the drivers who said there should be restrictions on young drivers
o more than four-fifths of drivers (84%) support a requirement for young drivers to display ‘P’
plates after passing their test, to show they are on probation
o more than a third of drivers (34%) support a curfew preventing young drivers from driving
late at night
o a quarter of drivers (25%) support a limit of one passenger
o more than seven in ten (71%) thought these restrictions should not be lifted until drivers
reached the age of 21
Drivers’ views on further training and testing
 Four-fifths of drivers (79%) said Pass Plus should be compulsory
 Two-thirds of drivers (66%) said drivers should have to pass an eyesight test at least once every five
years
 Nearly half of drivers (47%) said drivers should have to re-take their driving test at least every 10
years
 Three-quarters of drivers (75%) had not received any additional training since passing their test, but
more than half (54%) thought their driving would benefit from additional training
 Almost six in ten drivers (57%) have not read the Highway Code since passing their driving test
NB: Brake is currently in the process of coordinating a similar survey – results will be forwarded as soon as
they are available.
Evidence gathered through a 2006 survey of 15-25 year olds across the UK
Published November 2006
A third of young people (32%) have driven or been a passenger in a road race, according to the results
of a survey of nearly 4,500 people aged 15-25 across the UK. The survey was released today by road safety
charity Brake at a Westminster press conference for National Road Safety Week 2006.
Other results from Brake’s survey of 4,486 young people in the UK include:
 A third (30%) have driven without a licence
 One in 14 (7%) has driven a stolen vehicle
 One in five (18%) has been a passenger with a driver who they knew was unlicensed, uninsured, or
who had stolen the car
 Two in five (38%) don’t always belt up in the back
 One in three young people (30%) has been a passenger in a car driven by a driver on drink or drugs
Out of the young people surveyed who drive:
 Almost half (45%) break 30mph limits by 10mph or more. Of these, three-fifths (59%) are male
 One in five (19%) young drivers has overtaken at speed when they are unsure of what is coming in
the other direction. Of these, seven out of 10 (71%) are male
 A quarter of young drivers (25%) have driven at more than 70mph on a rural road. Of these, twothirds (64%) are male
 One in five (21%) drive after drinking alcohol
 Almost one in ten (9%) drive on illegal drugs
15
ANNEX 3: Summary of information presented at Brake conferences on young driver safety
The minutes below represent a summary only of evidence presented by some of the key speakers from two
conferences run by Brake in May 2007 and January 2008. Full minutes and copies of speakers’ powerpoint
presentations are available on request from ckeeler@brake.org.uk.
The road to young driver safety – research and recommendations for developed countries
Divera Twisk, Chair of the Young Drivers Working Group, Organisation for Economic Co-operation and
Development, the Netherlands
Divera spoke about the findings of an Organisation for Economic Cooperation and Development (OECD)
study, ‘Young drivers: the road to safety’.
Magnitude and forecast
 Young drivers: 27% of all driver fatalities but only 10% of the population in OECD countries
 For each young driver killed, about 1.3 other road users die
 Safer countries have safer novice drivers; if older, expert drivers are more at risk in a given country,
so too are young, novice drivers. This means there are high benefits to young drivers from general
road safety. Reducing young driver deaths is not only about specific measures for young drivers, but
also good general road safety measures such as:
o strict drink and drug driving laws and enforcement, e.g. random breath testing
o seat belt use, e.g. reminders in vehicles
o speed management, e.g. widespread use of speed cameras
o vehicle design – old cars offer less protection
o safe infrastructure, e.g. roads that allow drivers to recover from errors safely by having
roadsides that reduce the effects of a crash (e.g. soft verges); are predictable, so that
drivers know from the type of road what they can expect – e.g. whether there are likely to be
cyclists and pedestrians; and have low speed limits (30kph) in residential areas
Specific young driver risk factors
There are several factors specific to young drivers that need to be addressed:
 Young age: biological immaturity
 Lack of experience
 High exposure to dangerous traffic conditions
 The highest risk of crashing for young drivers across all countries is when they are driving on
weekend nights
Research concludes that age is less important in determining risk than experience.
NB: The number of female young drivers who crash tends to follow general road safety trends, but young
male driver safety has not improved in line with older male driver safety between 1994 and 2001. The
number of young male drivers who crash has not decreased as much as the number of older male drivers
who crash.
Effective countermeasures specific to young driver risk factors
 Minimum age of 18 years for solo driving
 Delay licensing if possible (e.g. free Public Transport Pass in the Netherlands)
 Training should focus on ‘higher-order’ skills (e.g. self-control)
 Ensure that experience is gained before solo driving (e.g. through accompanied driving)
 Impose protective measures during the first period of solo driving, when risk is the highest
 Ensure compliance during these first years
 Avoid migration to more unsafe motorised two-wheelers (10 times more risky)
 Accompanied driving
o In Austria, gaining accompanied driving experience before taking their test significantly
reduced the crash risk for young drivers – particularly young males.
o Accompanied driving novices have lower crash rates than traditionally trained novices, even
if controlled for socio-economic factors
o Swedish experience – crash reduction of 30% for young drivers who had taken an
accompanied driving scheme
o Debate: how many hours is effective?
 Sweden has most hours and highest effect
Conclusions
16




Implement general safety measures
Introduce accompanied driving as an instrument to boost experience
Impose restrictive measures: only zero alcohol has wide acceptance in the short term, but we
shouldn’t be put off by resistance – there is always resistance to any new safety measure
Develop specific measures to target young male drivers
Assessing the road ahead – hazard perception and risk management among young and novice
drivers
Dr Robert Isler, Director of Traffic And Road Safety research group (TARS), University of Waikato, Hamilton,
New Zealand
Why are young drivers over-represented in crash statistics world-wide?
 Because they lack maturity: biology and lifestyle factors
 Because they lack experience: haven’t learned enough
Training young drivers
Traditional methods of driver education and driver training have not delivered demonstrable safety benefits.
In fact, some types of car-handling training have often been cited as leading to increased risk-taking due to
learners’ inflated self-rated driving skills and subsequent show-off behaviour (Gregersen, 1996). A growing
consensus among driver training and road safety researchers is that greater emphasis should be placed on
higher-level cognitive functions underlying driving skills. Some researchers have argued further that there is
an urgent need for a holistic and structured plan of education and training (Engstroem, Gregersen,
Hernetkoski, Keskinen, & Nyberg, 2003).
A study of four Graduate Driver Licence (GDL) systems (US, Canada, Australia, & New Zealand) found a
31% median decrease in 16-year-olds’ crash rates during the first year, with a longer-term 7% crash
reduction.
Driver training programme evaluation findings show:
 driver education and training contributes little to reducing crash risk/ involvement for road users (prelicence, defensive, advanced, driver improvement)
 off-road skid control training may increase crash risk, particularly among males
 there is no evidence that professional driver training is effective in reducing crash risk.
However, there was a lack of methodological soundness in previous evaluation studies of driver training
programmes. Most of the pre-2000 evaluations were not published in peer review journals and often used:
1) no control group
2) hypothesis testing procedure, inappropriately
3) crude outcome measures such as number of collisions/deaths
Are we teaching the wrong things?
Traditional driver training involves teaching basic control skills. However, these are not the problem for
young drivers. A common crash risk factor for teenagers seems to involve a lack of higher ‘executive’
cognitive functions that could be associated with the frontal lobe of the brain – such as emotion regulation,
impulse control and decision processes, but also functions that are considered as higher-level driving skills
such as eye scanning, hazard anticipation, and risk management. Interestingly, brain imaging studies have
shown that the frontal lobes are not fully developed until young people reach the age of 25 years (Dahl,
2004), the same time when age disappears as a risk factor for crashes, even after driving experience is
taken into account. (Mayhew, Simpson & Pak, 2003).
Immature executive functioning may lie behind the poor hazard anticipation and detection, decision-making
and risk management skills that characterise many adolescent drivers. But like any other brain function, they
can be enhanced with appropriate training interventions. Recently, McKenna, Horswill and Alexander (2006)
found that hazard anticipation training using video simulation techniques decreased hazard detection times
in novice drivers and were associated with a substantial decrease in their risk-taking behaviour.
Eye tracking research
Eye movement tracking has shown that novice drivers (compared to more experienced drivers):
 concentrate their search in a smaller area, closer to the front of the car
 fix their eyes in one place for longer in hazardous situations
 do not look for hazards over such a wide area when driving on dual carriageways
More effective eye scanning can be taught via video simulations.
17
Hazard detection
 Hazard anticipation skills in novice drivers can be significantly improved by training in the laboratory
using video simulation techniques – novice drivers can be improved to the level of experienced
drivers within four hours of training
 Hazard anticipation training reduced risk-taking behaviour (speed choice, following distance, gap
acceptance).
Two educational projects in New Zealand designed for training high-level cognitive skills:
1. ‘Cd-Drives’ CD-ROM (Isler & Cockerton, 2003), developed at the Psychology Department of the University
of Waikato for the ‘Practice’ intervention programme in New Zealand (available to all novice drivers at no
cost – www.practice.co.nz). The Practice programme is a joint initiative between Land Transport New
Zealand and the Accident Compensation Corporation, promoting the benefit of extensive supervised driving
experience for novice drivers. The ‘Cd-Drives’ program provides more than 90 video-based traffic
simulations including a fully functional 3D-dashboard and a near 360 degrees of vision around the virtual car
through the inclusion of side and rear-view mirrors. The ‘Cd-Drives’ program consists of five training
modules: eye scanning, hazard detection, risk management, road commentary and the ‘final challenge’. All
modules (except the road commentary module) consist of 18 live-action video simulations. The program
received a Telecommunications Users Association of New Zealand (TUANZ) award for best educational
software in New Zealand, 2003.
2. The a2om mind® DVD series is based on the ‘Cd-Drives’ programme and was developed by the
University of Waikato for a2om® Ltd (www.a2om.com), the first University-affiliated Driving Academy in the
UK. The a2om mind program provides more than 100 high-definition video-based traffic simulations, filmed
in the UK, including many night driving scenarios and different weather conditions. The program uses the
latest computer based high-definition video technology, enabling a2om students to learn anticipate and
respond to hazards as far ahead as 600 feet. It also has an extensive road commentary module which allows
students to develop ‘situation awareness’ and improve their hazard detection times, which are typically 30%
longer in novice drivers than in more experienced drivers and are directly related to crash risk. A post-test
module for newly-licensed drivers exposes them to motorway-related hazards that were not encountered
during the learner phase.
Cognitive neuroscience: the science of behaviour and mental processes and the role of age
(maturity)
Young people’s vision, strength, reaction time, health, and resilience are at their peak, but the frontal lobe of
the brain and its executive functions are not fully developed until the age of about 25.
Executive functions of the frontal lobe of the brain are:
 inhibition
 planning ahead
 impulse control
 risk management
 reasoning
 self monitoring
 verbal self-regulation
 emotion regulation
 motivation
 hazard perception
 eye movements
The frontal lobe project 2006
New Zealand research into the benefits of car control skills (traditional) training (i.e. steering, manoeuvring,
parking) as opposed to higher level driving skills training (i.e. road commentary, self-assessment, group
work, video-based hazard perception training)
Baseline tests were conducted which included measures of driving performance, personality, driver attitude
and risk-taking. In addition, each participant’s executive frontal lobe function was psychometrically assessed.
 Questionnaires included: driver behaviour/attitude, sensation-seeking, self-reported risk-taking
behaviour, confidence in driving skills.
 Extensive psychometric testing included: Delis-Kaplan Executive Function System (D-KEFS) a
psychometric test which assesses executive frontal lobe functions; and psychometric tests for
general ability, depressive and anxiety tendencies.
 Driving skills assessment: search, speed choice, directional control.
18
Baseline and post-training driving performance (higher level skills and in-car control skills) were assessed by
professional driver assessors. The driving assessors did not know which participants had received training.
Results showed:
 increased frontal lobe function improved driving performance
 training in higher level driving skills improved real driving performance without increasing selfreported driving confidence levels
Reducing young driver deaths – an insurance perspective
Justin Jacobs, Assistant Director of Motor and Liability, Association of British Insurers
The problem
 Average young male motor claims are significantly higher than average claims for older drivers
 Statistics show that young male drivers are disproportionably represented in the number of car
drivers killed, injured and seriously injured per 1,000 licence holders
 Inexperience is a factor in between 35% and 40% of crashes involving drivers aged 17-18. The
proportion of crashes in which inexperience is a factor diminishes rapidly with the age of the driver.
 Excessive speed is a contributory factor in more than twice as many male driver crashes as female
driver crashes. This factor diminishes with the age of the driver, particularly for male drivers.
 Highway Code violations are higher for males, but decrease significantly with the age of the driver
for males and females.
 Driver fatalities per 10 million trips for young drivers increase with the number of passengers in the
vehicle. (Source: Chen, L.H., Baker, S.P., et al. Carrying passengers as a risk factor for accidents
fatal to 16- and 17-year-old drivers. Jama 283(12): 1578-82)
ABI recommendations for reducing young driver crashes
 Minimum learning period enabling drivers to undertake supervised practice without an incentive to
rush to take the practical test.
 Structured learning programme for all new drivers, covering the different road and traffic conditions
learner drivers should experience before taking the test. This could be implemented via a logbook in
which the driver would record the number of hours they had spent driving, and their experience.
 Graduated driving licences with, for example, a requirement not to carry more than a certain number
of passengers during the first few months of driving.
 Encouragement for young drivers to avoid driving at night. Some insurers are already offering lower
premiums to young drivers who only drive during the day. A targeted advertising campaign to remind
young drivers of the particular risks they face when driving at night should also be launched.
The components of GDL – analysing the US experience
Susan Ferguson, Former Senior Vice President, Research, Insurance Institute for Highway Safety, USA
Causes of the young driver problem
 Driving inexperience
 Youthful age (immaturity, brain development)
 Young driver crashes peak at night time (between 9pm – 6pm). The proportion of crashes that occur
at night reduces as the age of the driver increases.
Fatal crash characteristics
Percent by driver age, FARS 2005
driver error
speeding
single vehicle
3+ occupants
positive BACs
16
74
34
49
29
15
17
73
32
47
24
23
18
71
33
44
23
30
19
68
33
46
24
32
20-25
64
30
45
19
53
26-49
51
19
38
17
48
Driver death rates increase for drivers aged 16 and 17 according in proportion to the number of passenger
they are carrying. The number of passengers does not increase the crash rate for drivers aged 30-59.
19
The percentage of drivers killed with a positive Blood Alcohol Count (BAC) increases with the driver age until
drivers reach age 20-29, after which the percentage of drivers killed with positive BACs decreases.
Licensing approach prior to graduated licensing
 Teach young beginners rudiments of driving.
 Turn them loose on the highways.
 Try to pick out and deal with youthful “problem drivers”. This is problematic because many young
drivers who crash do not have points on their licence and are not so-called ‘problem drivers’.
Principles of graduated licensing
 Find appropriate trade-off between safety and mobility
 Encourage low-risk on-road practice driving
 Keep young beginners out of high-risk driving situations
 Delay full driving privileges until teenagers are older and starting to mature out of risky driving
practices
Graduated licensing
The Insurance Institute for Highway Safety recommends the following three-stage licensing system as a
minimum:
Learner
 Maintain the starting age at 16, or raise it to 16
 Establish a 6-month learner’s phase
 Require adult supervision
 Require 30-50 hours of certified driving, some at night
Intermediate
 No unsupervised driving before age 16 ½
 Restrict unsupervised night driving
 Limit teenage passengers
Full
 Minimum age 18
Additional GDL components might include:
 Extended learner period
 Parent certification requirements
 Night time restrictions
 Passenger restrictions
 Cell phone restrictions
Results of some of the evaluations of graduated licensing programs in North America
British Columbia
California
Georgia
Florida
Maryland
Michigan
North Carolina
Nova Scotia
Ohio
Pennsylvania
age groups
crash reductions
16-18
16
16
15-17
16
16
16
16-17
16-17
16
16%
23%
30%
9%
21%
38%
39%
29%
23%
28%
Is graduated licensing effective?
 Consistent, positive findings
 Greater reductions with stronger GDL programmes – i.e. those with more restrictions and longer
learner and intermediate periods
 No increase in crash risk for 17 or 18 yr-olds once they obtain their full licence
 No male/female differences in reductions
Licence delay
20


In 13 jurisdictions that have introduced GDL, the extension of the learner permit period delayed
young drivers from being able to gain a licence by two-six months
Studies in Kentucky, Connecticut, and Nova Scotia indicate substantial crash reductions ranging
from 22-50%
Extended practice driving
 Does GDL increase the quantity of extended practice?
o Very little evidence, but in California about 21% more parents indicated they spent 50 hours
or more driving with their teen after GDL was enacted
 Does more experience produce better drivers?
o In Nova Scotia, collision rates for 16-17 year old drivers was 11% lower during first 6 months
of unsupervised driving after GDL
Restricted night driving
Crash reductions, Nighttime vs. daytime*
percent reduction
jurisdiction
restricted hours
night
day
Florida
11-6
16
9
Michigan
12-5
59
32
North Carolina
9-5
47
22
Nova Scotia
12-5
49
5
*Data are for 16 year-olds in Florida, Michigan, and North Carolina and for 16-17 year olds in Nova Scotia
Passenger restrictions
Maximum number of passengers allowed*
number of passengers
number of jurisdictions
None
15
1
18
2
2
3
2
No restriction
14
*Ten states relax their restrictions over time (e.g., allowing none the first 6 months then up to 3); the table
includes the restriction that applies immediately after licensure




New Zealand reported moderately positive effects
Four studies of California GDL have indicated positive effects
o Zwicker et al., found a 38 percent reduction in 16 year-old per capita crashes involving other
injured teenagers in the vehicle
Positive effects found in forthcoming study for California, Massachusetts, and Virginia
In North Carolina, crashes with multiple passengers declined by 32 percent among 16 year-old
drivers
Cell phone restrictions
More information is needed about cell phone restrictions, but it is known that:
 Distraction is thought to be a particular problem for novice drivers, especially when others are in the
vehicle.
 Cell phone use results in a fourfold increase in injury crashes for drivers of all ages. Risk is
increased for both handheld and hands-free phones.
 16-24 year olds are more likely to talk on their cell phone while driving than other age groups.
Numbers of drivers talking on their cell phone while driving increased for all age groups between
2000 and 2005
 As of April 2007, 15 US jurisdictions ban cell phone use for teens in graduated licensing systems,
and more are considering bans.
Fatal crash rate trends in the US
Since GDL has been introduced across the US, there has been a significant reduction in fatal young driver
crashes.
Fatal crashes per 100,000 population
21
By driver age, 1996 vs. 2005
age
1996
2005
16
17
18
19
30-59
33
39
47
43
20
19
30
40
40
17
percent
reduction
42
23
15
7
15
Conclusions
 Number of teenagers killed in passenger vehicles in the U.S. in 2005 is at its lowest since 1992
despite highest population of teenagers since 1977
 Dramatic reductions in 16 year-old driver fatal and police-reported crashes per population during last
decade
 Gains somewhat greater at night than during the day
 Fatal crashes among 16 year-olds less likely to involve multiple teens in the vehicle
 Fatal crashes involving alcohol declined more among 16 year-olds than among older teens
 GDL has proven effective
o Reasonably certain that delayed licensing, nighttime and passenger restrictions are effective
in reducing crashes
 Still many questions
o Role of holding periods, supervised driving, timing of permit and licence
o The effect of cell phones, and other distractions
o Does GDL produce safer drivers?
The Road Ahead: Hazard perception skills among young and novice drivers
Dr. David Crundall, Accident Research Unit, School of Psychology, University of Nottingham
The hazard perception test currently being used by the DSA is one in which you sit down, watch a video from
the perspective of a driver and press a button when you spot a hazard. Hazard perception testing tries to
encourage explicit training of the hazard perception skills in the driver curriculum. This training can be
difficulty as hazard perception skills are usually learnt gradually in the years after passing your driving test.
Hazard perception research:
 Early Pioneers: Pelz & Krupat (1974), Watts & Quimby (1979) – researched the idea that there was
a correlation between drivers who had more crashes and drivers who had poor hazard perception
skills
 The Modern Age: McKenna & Crick (1991); Horswill & McKenna (2004) – researched the differences
between experienced and inexperienced drovers to see if there was the potential to teach and test
hazard perception skills
 The Doubters: Crundall et al., (1999); Chapman & Underwood, (1998); Sagberg & Björnskau, (2006)
– Concerned about whether the existing hazard perception test is an effective diagnostic of good
and bad driving. Not against hazard perception testing per se, but have some problems with the
efficacy of the current model
Are there real underlying differences in HP tests between good and bad drivers and what difference
in behaviour does this represent?
 More experienced drivers’ eye movements are different from less experienced drivers’.
 More experienced drivers spend less time focusing on hazards (fixation duration decreases) but their
peripheral attention is better and recovers more quickly after having spotted a hazard.
 In contrast, less experienced drivers zoom in to focus on a hazard, spend a long while fixating on the
hazard, and then it takes a long while for their peripheral attention to return.
Problems with hazard perception testing – research findings
 The current button-press test cannot tell whether someone is pressing the button in response to the
hazard that the test wants them to identify, or in response to something else on the screen. This could
mean people are able to get good results without actually identifying hazards in the way they are
supposed to
 Stopping the clip to ask “What happens next?” and getting the person being tested to describe what they
thought was about to happen, based on their observation of the impending hazard allowed us to judge
22

how well they had spotted the hazard in advance, and to be sure they were observing the hazard that we
had intended for them to. Their answers were scored for accuracy on a scale of 0-6
More experienced drivers demonstrated a much higher accuracy. So, although the existing push-button
test does not measure accuracy, there are ways and means of testing accuracy
What makes a good hazard perception clip?
 Different hazard perception clips favour different drivers. Abrupt onset hazards favour younger people
rather than experienced drivers. This is because young people are biologically faster, so do better in
simple response time tests.
 Slow onset hazards favour inexperienced drivers rather than experienced drivers, because they are
more skilled at pre-empting hazards early. However, by pre-empting the hazard and clicking immediately
you can sometimes fail to get points because the system does not recognise that the hazard is there yet.
 Instead we need clips that:
o Contain a foreshadowing element – some sort of predictive element
o Have clearly defined onsets that allow pre-emptive responses from experienced drivers
(providing that they are accurate responses!)
o Vary along real-world dimensions that provide a typology of hazard clips, whose varying
demands are understood through experimentation
We have concluded that there are three types of external hazard that ought to be tested:
1. Predicting the behaviour of other road users (anticipation). Ability to pre-empt a hazard at the early
stages – spot the ‘pre-cursor’ to the hazard.
2. Developing a mental model of the situation (knowledge). Ability to use your knowledge about the
environment to predict when hazards are likely to occur.
3. Dividing and focusing attention (scanning). Ability to prioritise which hazards pose the most risk, and
monitor sources of hazards.
Conclusions so far
 Safer and more experienced drivers do look at hazards differently, even if a simple push button
response does not distinguish them from inexperienced drivers
 Accuracy judgements can be introduced into hazard perception testing in a meaningful way
 We have evidence for a typology of hazard perception clips which we suggest should be used for the
development of any future clips used by the DSA (i.e. there are three types of hazards)
 Pressing a button cannot tell us how well people react to hazards – e.g. whether they brake, or steer
away. Research has shown that when confronted with a hazard, expert drivers brake earlier than less
experienced drivers. Ideally, a hazard perception test would monitor driver behaviour in response to
hazards.
 Simulator studies can allow us to identify how drivers respond to these hazards in a more detailed
manner than a simple button response can allow
Going the extra mile: holistic learning based on international research
Gary Austin, Managing director, A2OM Academy
Our driving test is skills based – attitudes are not addressed. A2OM has created a new educational model
 On-line assessment through psychometric profiling
 On-line neuroscientific training: frontal lobe
 Lectures and group sessions
 In car lessons, including ‘buddy’ system
 E-learning platform and lesson plan/diary
 All weather advanced driver training
 Telematics monitoring to see how drivers apply the training
It is essential to consider the psychology of the driver – we are all different. This means looking at individual
personalities. Personality is composed of: intellect; emotions; body impulses; intuition and imagination. All
these elements require activation if learning is to be more effective.
Young people interpret their experiences differently – some are more susceptible to risk taking than others.
We therefore need to find out more about the individual, background influences and approach to risk. The
best way to do this is through Psychometric profiling.
Brain training
23
The frontal lobes of the brain, which controls our executive functions, only become fully developed by the
age of 25. They control executive functions, such as:
 Eye Movements
 Hazard Perception
 Inhibition
 Planning ahead
 Impulse control
 Risk Management
 Reasoning
 Self Monitoring
 Verbal self-regulation
 Emotion regulation
 Motivation
Chapman, P., Underwood, G. and Roberts, K. (2002).
Less developed executive functions mean that novice drivers (compared to more experienced drivers):
 concentrate their search in a smaller area, closer to the front of the car
 have longer fixation times in hazardous situations
 have a smaller spread of search when driving on dual-carriageways
This can be combated by teaching more effective eye scanning via video simulations
Does hazard anticipation training affect drivers’ risk taking?
McKenna et al., (2006) showed that:
 Hazard Anticipation skills in novice drivers could be significantly improved by training in the
laboratory using video simulation techniques.
 Novice drivers could be improved to the level of experienced drivers within 4 hrs of training.
 Hazard Anticipation training reduced risk taking behaviour (speed choice, following distance, gap
acceptance).
Does ‘higher-level’ skills training affect drivers’ risk taking?
New Zealand research by Dr Robert Isler showed that statistically significant improvement in novice drivers
was seen only after higher level skills training.
 (over-)confidence levels in their driving skills were significantly lowered
 attitudes to speeding, overtaking and close following improved significantly
 larger-scale research is being done on the long-term effects of higher-level skills training on crash
rates.
END/
24