Research activities 1987-1989

(1)

Research activities

1987-1989

Sl.lP

O

I.P

1990

(2)

Photos: Studio Verkoren

SWOV Institute for Road Safety Research

(3)

SWOV's objects,

methods and organisation

How SWOV is organised

The SWOV Institute for Road Safety Research is governed by a board of governors and a director. SWay's board of governors consists of a minimum of seven and a maximum of nine members. Six members are appointed: the board is free to elect the remaining members, including the Chairman. The following each appoint one member:

- the Minister of Transport

- the Royal Dutch Touring Club ANWB - the Netherlands Association for Automobile

Insurance NVVA

- the Netherlands Association of Bicycle and Automobile Industry RAI

- the Director General of the Public Works Department

- the SWay works council.

The management is responsible for the day-to-day running of the organisation. The director represents the board of governors, both I'nternally and externally.

Director of the Institute is Prof. MJ. Koornstra.

Functions

Three main functions have been hid down for SWOV in close cons Utation with the Ministry of Transport, SWOV's main client and so It'ce of subsidy.

I. Information control function

SWOV is responsible for ascertaining the information needs of the government, keeping records of the information available form Dutch and foreign research and making it appnGlble to the formulation and implementation of road safety policy.

2. Archl'tect of contract research

SWOV contracts out road safety research projects to specialised research institutes, designs and supervises them, integrates the results, and compiles the reports.

3 . Research function

SWay will carry out interdisciplinary and basic research. Where contracting out a special proje Q

or part of a project, WIll be possible, SWOV WIll

4 not do the research itself.

SWOV's objects

The SWOV Institute for Road Safety Research was founded in 1962 on the initiative of the Minister of Transport, the Royal Dutch Touring Club ANWB and the Netherlands Association for Automobile Insurance NVVA. The reason was the constant increase in the number of road casualties and the realisation that scientific research was indispensable for an effective approach to road safety problems. SWOV's objects is to contribute to road safety by means of scientific research and dissemination of the results. Its activities cover all aspects and areas of road safety. SWOV also contracts research to third parties. In this way it has a co-ordinating function in planning road safety research in the Netherlands.

The research results and know-how are spread: - among policy-making bodies which can put the research results into practice;

-among scientists, in order to exchange research results and methods, and

-among institutions and persons concerned with road safety.

Much of SWOV's research is focused on

obtaining information for policy measures. Mos t

assignments for this come from the Ministry of Transport. In recent years, however, more and more assignments have been received from provincial and municipal authorities and private institutes.

(5)

SWOV's philosophy

About a million traffic accidents occur in the Netherlands each year, killing about 1,500 and injuring more than 50,000 persons. The total economic loss caused by traffic unsafety amounts yearly to about 6 milliard Dutch guilders. There is every reason for tackling this gigantic problem with united strength. In the first place because it involves a great number of human lives, but also because of financial-economic considerations. Somehow we accepted traffic unsafety in the course of time as a kind of natural phenomenon, which cannot be modified essentially. SWay has always vigorously contested this opinion and also indicated in which way traffic unsafety can be reduced. In the first place we must get rid of the idea that the total problem can be finally solved by continually looking for isolated solutions for various problems. We have to look for multi-causal instead of mono-multi-causal explanations. In order to reduce traffic unsafety to an acceptable level, the functioning of the entire traffic and transport system has to be improved. This requires a deep insight into the interplay of traffic participants and other factors of the system, i.e. roads, vehicles and surroundings.

More part'~ularly the knowledge of various

branches 0 f science has to be combined and more

thorough integration of research, policy and control in necessary as well. For this reason, SWOV seeks for ways and means to realise these aims in an effective and efficient manner in co-operation with foreign associated institutes. And to all appearance, not unsuccessfully. In the interplay between traffic participants, vehicles, road and surroundings (so-called "critical

situations") may arise, which, in turn, may lead to traffic accidents. What matters is the detection of such critical situations in the traffic. This requires a close co-operation between scientists of various disciplines, e.g. psychologists, sociologists, physicists and engineers. However, this is not enough. A comprehensive approach is only possible with the aid of systematically collected data concerning the traffic and accident process, and highly advanced analysis techniques. The application of computers to such investigations is of vital importance. The computer also plays a significant part in developing mathematical models of the traffic and accident process. Such form of simulation of real situations is relatively cheap in comparison with other simulations methods.

(6)

6

International SWOV symposium

10 celebrate its 25th anniversary, SWOY

organized an 'tt ~rnational symposium on the

theoretical and methodological background to

road safety research from 26 to April 1988·

The symposium, entitled Traffic Safety Theo ry and Research Methods, was held at the RAI Conference Centre in Amsterdam and attended by just under 200 people from 24 different countries.

The aim of many conferences

is

to present the

data obtained from research and enable an

exchange of knowledge to take place. ~ is not

only the findings that are of interest. however; tie way in which the research is carried out is also important. In other words, research methods and techniques are also a valid subject for discussion. The last conference devoted to the methodo-logical background to road safety research was

held in 1969, and much progress has been mal~

in this field since then, to which SWOY has

made major contribution -yet another reason for

organizing the symposium.

The aim SWOY had in mind was to take a look into the future and help mark out the framework for research. A select group of promi Jent scientists of various nationalities were invited to present recent developments in research methods,

technologies and theories· The five subject areas

chosen covered the main fields of research.

t

Context and scope of traffic safety theory

This session dealt with traffic safety as a totality:

models that give a valuable description of the developments in traffic safety, the data needed for this purpose and the quality standards that need to

be met. The discussJ'on was devoted particularly

to the role of research in underpinning road safety policy and the relationships between the safety of the traffic system and the development of the total traffic and transport system as such, economic aspects of road safety and the health

lervice aspect. Those who presented papers

J"ncIuded guest speakers professor Frank A .

HaJ'ght of the University of Cah'fomia, USA and

Professor Niels 0 "J~rgensen of the Technical University of Denmark.

2. Models for evaluation

Professor Ezra Hauer of the University of Toronto was invited to address this session as guest speaker. He explained how we should

learn form the experience we gain from the imp lementation of safety measures, as regards theireffects on both traffic and safety. Evaluation research in particular suffers from a number of methodological problems which are in need of attention. By improving the design of evaluation research we can achieve a major improvement in the efficiency with which measures are applied. Other papers were delivered by Ekkehard

Briihning of the Bundesanstalt fUr Strassenwesen and Heather Ward of University College, London, among others.

3. Theoretical analysis and models

In recent years both researchers and government have displayed increasing interest in the part road users themselves play in the mistakes which occur in the traffic system. The main emphasis here is on actual road user behaviour and the conditions in which certain types of behaviour occur. More heed is being paid in particular to the cognitive aspects: not just for instance, to the performance of certain manoeuvres in traffic but also to the strategies underlying them" The value of various cognitive models and behaviour models was examined and a number of new developments presented by guest speakers Professor John A. Michon of the University of

Groningen, Professor A.R. Hale of Delft

Technical University, Wiel Janssen of the TNO

Institute for Perception, Soesterberg and Terje

Assum of the Institute of Transport Economics,

Oslo, among others.

4. Statistical analysis and models

Research into road safety is made more difficult by the fact that accidents are relatively rare

occurrences, especially if we wish to study them

within a limited area and a limited period of time.

This is why, right from the start of road safety research, a good deal of consideration was given to statistical methods of assessing reliults . New techniques have also been developed to describe the relationships between the many factors

involved in accidents. These were presented by

(7)

Maycock of the British Transport and Road

Research Laboratory, and Heinz Hautzl·nger of

the Institut fUr angewandte Verkehrs -und

Tourismusforschung, Heil bronn.

5. Time-dependent models

For some years now, to a hrge extent at SWOV's instigation, road safety has been defined as a

quality aspect of the dynam·E traffic system,

which develops over time. Progress has been made recently with the development of analytical model which describe traffic as a dynamic

system. Other ways of describing time-dependent

road safety processes have a so been considered.

These models were presented by Professor A.C. Harvey of the University of London and Siem Oppe of SWOY.

At all the sessions ample use was made of the opportunity to ask questions and make comments

on the papers, often resulting in an animated dis

-cussion. The participants' general impression

following the symposium was favorable. It was

apparent from the many letters SWOV

subsequently received that they regarded it as a valuable step along the stony path of traffic safety research.

(8)

8

Research plan for 1989-1992

SWOV has d lawn up a Provisiona I Medium-'lerm Road Safety Research Plan for 1989-92 which describes the framework the Institute considers approp ftite for road safety research during this period. The priorities it sets out are based on a combination of considerations:

relevance to government policy, social feasibility, potential for improving road safety and scientific value.

SWOV used to have an annual research

programme comprising over a hundred projects and part-projects based on requests by govern-ment departgovern-ments for research or information. Research into ways of improving road safety was carried out at a variety of specialist institutes in the Netherlands; consequently the problems were approached from too many different angles, and there was the risk of projects overlapping or work being duplicated. Such a fragmented system, lacking any clear research strategy, cannot provide the information needed to optimalize (a) the transport & traffic process and (b) measures to improve road safety.

If the gaps in our knowledge are to be filled, it is essential to concentrate the quality of research and plan it carefully: this will provide a better basis upon which to decide the priorities in traffic and transport policy and road safety policy. The Provisional Medium -'Ierm Road Safety Research Plan for 1989 -92 sets out both short -term and longer--term ob~ctives. The main short

-term aims are to support and evaluate specific policies, with particular reference to the Ministry of Transport's Med'l!lm-Term Road Safety Plan for 1987 -91. The prior'ly in the longer-term

is

to expand and J'mprove ways of controlling traffic and road safety.

The Research Plan compnose" a total of eight projects, two 0 fwhich are dJorectIy connected with the JOmplementation of the Medium -Term Road Safety Plan. Another five are concerned more WJOth longer-term objectives, while incor -porating certain elements from the Ministry'S Piano

The MedIum -Term Road Safety Plan SUppl~t Pro/ect J'nvolves support for, and p tt bdJic

evaluation of, measures connected with the Ministry's Plan, with the emphasis on drink-driving, speeds and use of seat belts and crash helmets.

The other project connected with the Ministry's Plan is the Conditions for Participation in Traffic

Project, which consists of a review of the system of driving instruction and driving tests, including the development of a provisional driving licence, a course of instruction for moped-riders and the drafting of a highway code.

Yet another project, the Basic Data for Research

and Policy Project, is concerned with the accessibility by computer of linked data banks. This project has no objectives of its own but is designed more to serve the other projects in the Plan and other SWOV projects and activities. Traffic and road safety problems have so far developed more or less independently. It is not entirely without significance in this context that it was following the formulation of a road safety target that an unexpectedly sharp rise in mobility resulted in a rise in road accident casualties, thus hampering the achievement of the target in the short term. The current target for the reduction in road fatalities has been raised from 200 to just under 300 and the timescale has been reduced by two years. It is necessary, therefore, to gain some

o~ea of the interrelationships between these

developments, and it is with this in mind that the

Mobility and Road Safety Project focuses on models for forecasting mobility and the effects on road safety of alternative mobility scenarios and traffic and transport plans.

A basic factor in the control of traffic and road safety is the design of traffic situations -not only their individual design but how consistent they are· Another major factor is how clear or unclear the situations are in combination with the rules on expected and desirable behaviour ° Research into

these factors

is

contained in the Designing the

Infrastructure Project, which enta ils classifying and systematizing the information available.

The Interactlon between Road Users and Traffi1c EnVIronment ProJect is concerned with

(9)

information for various projects. There are two

main components·.

I. A diagnosis of differences in individual road user behaviour, which has consequences in

particular for education, training and selection;

and

2. An examination of situation -related behaviour,

in particular the interactions between the various road users. This has consequences particularly for the design of traffic situations so as to prevent potential conflicts.

Emergency aid to road accident casualties is a subject bordering on traumatological research. The predominance of road casualties among patients suffering from serious, life-threatening and disabling traumas was one of the reasons behind the decision to institute the Injury Prevention Project, which includes an exam-ination of the effectiveness of measures in this field in reducing fatalities and permanent injuries from road accidents.

A subject that has received re btively little attention in the past is the commercial road

haulage sector, which

is

likely to exert an

increasing in-fluence on road safety problems. The Heavy Traffic Project examines the interrelationships between vehicle collision properties, vehicle dynamics and control, ergonomics, regulations and the training and selection of drivers.

Various areas in which SWOV has until recently conducted research are not included in the Provisional Medium-Term Road Safety

Research Plan as specific projects. Research into the collision phase, particularly its practical applications, is included in the Designing the Infrastructure Project. Similarly, there are no projects specifically concerned with the vehicle aspect or electronics in traffic, but this research is included in the various projects to which it is relevant.

(10)

10

Results of SWOV research 1987 -1989

S,Ubjects for road safety education

The Ministry of Transport's Medium-Term Road Safety Plan for 1987-1991 sets out the policy keynotes for the next four years. Some of these relate to target groups among road users (young people and the elderly), some to factors in road safety (drink-driving and speeds), some to measures designed to reduce injuries (seat belts and moped riders' crash helmets), and some to infrastructural measures. SWOV has made a survey of subjects for road safety education on which the Ministry could base its public educa-tion policy as part of its Plan.

Elderly road users

The risk of injury per travellers ki bmetre is higher for the 65+ age group than for other adults. This is due to three interrelated factor: declining mental and physical faculties, decreas-ing mobility and greater physical vulnerability. Educational measures could take account of this, by providing information on the decline in faculties and the problems that elderly people can encounter in traffic when they get out of practice.

It needs to be made clear, not only to the elderly

but also to the social and medical world in which they live, that it is essential that they remain active as road users, since this delays the decline in performance. Information could be provided to highways authorities on safe routes for elderly cyclists and pedestrians and special measures to cater for the elderly. It is also im-portant for the elderly to be informed of changes

-including temporary ones -in the infrastructure,

e g . roadworks, and new traffic rules· Lastly, it is important that other road users understand the need for appropriate behaviour, especially as regards speed and giving way, so as to take the

problems and behavl'our of the elderly into

account·

Young road users

The development of young people up to the age of 24 as road users is gradual 0 They have b gain

experience of each new mode of transport and

new situauoon, and each tlOme there is the danger

that they may be overtaxed by their rol e IOn traffic 0

The infrastructure is only partly geared to this phased learning process, as in the division into residential and traffic zones. Children need to gain experience at the correct rate, they need education and training and supervision. In

gene ral, parents and other carers should be more

involved in their children's road safety education. Their attention could be drawn to educational programmes, e.g. on television for schools. Parents should also be made aware of the need to set a good example.

Seat belts

The seat-belt use rate in the Netherlands in 1988 was about 77% on non-urban roads and 63% on urban roads. Better enforcement of the rule that makes wearing of seat belts compulsory could make a major contribution to road safety. A campaign in Friesland demonstrated that a combination of publicity and police surveillance was effective in increasing use of seat belts. The publicity could be aimed, for example, at people who wear their seat belts on long journeys but not on short journeys, or at those who believe that seat belts are a hazard in particular types of accident.

Moped-riders' crash helmets

Every year dozens of moped-riders lose their lives on the roads as a result of incorrectly fastened chin straps or helmets which are too

large or in poor condition 0 Hundreds more are

admitted to hospital with serious head injuries 0

Only one in six moped -riders wears a helmet which is in good condition and correctly secured. More stringent standards are needed, but, in addition, educational work could make for better

and more effective use 0 A publicity campaign,

possibly combine with tighter police surveillance, could induce more moped -riders to wear helmets which are in good condition and correctly secured. Another kind of publicity could provide prospective purchasers of helmets with

information on the quality of the various types.

Lastly, it is important that good clear instructions

(11)

Drink-driving

At least a quarter of all road deaths occur in accidents in which one of the parties has been drinking. About a quarter of all motorists admit to driving occasionally after having consumed alcohol. On weekend nights one in eight

motorists drinks and drives. Even among cyclists drinking is a problem. It is because drink-driving has become a habit for many people, and is even accepted to a certain extent, that, as in the case of

anti-smoking campaigns, a strategy is needed. A

combination of education and police surveillance is called for, with the emphasis on the latter in the short term. The measures could be linked up with campaigns against excessive drinking, or these campaigns could be used as a basis for educating the public about the effects of alcohol on safety. Speed

Speed is a major factor in road accidents. The ril;k of an accident and the seriousness of the

outcome i~' "trongly correlated to speed. The collision speed affects the outcome of an accident as regards the risk of injury and the seriousness of the injury. The relationship between speed and risk is by no means simple, however: it is not the case that the higher the speed, the more accidents there are. The greater the differences in speed

between vehicle~~ on the other hand, the greater

the risk of accidents. It is these differences that

cau~e dangerous and unpredictable manoeuvres

to be carried out, thus increasing the accident risk.

Government policy is to control and regulate speeds on Dutch roads by means of speed limits and their enforcement, physical speed restrictions, recommended speeds and limits on the engine capacity of mopeds. In addition to this, road users could be educated to improve their knowledge -i.e. their understanding of the risks - and their skill in avoiding risks, and to make them change

their habits. (R-88-6)

(12)

12

Measures needed to reach Medium-Term Road Safety Plan target

The Medium-Term Road Safety Plan for 1987-91

has a clearly-defined target: a 25% reduction in casualties by the year 2000. The target for 1990 is 200 fewer deaths and 1,500 fewer hospital

patients than in 1985. Whether the trend is going in the right direction can be judged from a SWOV report which gives an overview of the road accident rates for recent years. SWOV comes to the conclusion that the general trend is good, but we must not slacken our efforts to make traffic safer, especially as mobility increases.

It should be noted that 1985 was a particularly

good year, owing partly to the severe winter in the early part of the year. Making due allowance for this, the targets for both 1990 and 2000 should be achievable, although the 1986 and 1987 figures show that we need to continue our efforts. The "fatalities" target for 1990 is likely to be beyond our reach, though it should be possible to achieve the "serious injuries" target. The "other injuries" target is not likely to be reached either in 1990 or 2000.

Given that the targets are essentially quantitative, it is important that the quality of the accident records system remains unchanged, or else that

the effects of any changes be known. SWOV

proposes that the quality of the system be examined every year.

Another important element in the Plan is the

selection of

six policy keynotes with good

chances of success in relation to the target. A

survey of developments under these headings

yields the following conclusions.

(a) According to the accident statistics the trends were no better in those provinces where a Regional Road Safety Council operated than in the other regions.

(b) The campaign for a 25% reduction was very popular with the municipal authorities, as shown by the number of municipalities participating. (c) It is not possible to establish how effective the Education as Catalyst scheme has been on the basis of the information currently available; this also applies to the What is the Government doing? keynote.

(d) Drink-driving is on the decline. This success shows that we should continue with the same approach.

(e) There is a slight increase in seat-belt wearing. Crash helmets are being worn in almost 100% of cases but the way in which they are used is no better than it was a few years ago.

(f) There is not much to report on the Appropriate Speeds scheme, owing partly to lack of

information and absence of a clear-cut policy. Another problem is the continuing absence of a proper speed monitoring network.

(g) Improving dangerous locations has been part of government policy for some years now. Whether the measures taken have had any effect is impossIble to say as there have been no recent

surveys.

(h) The elderly and the young present the biggest road safety problem. Although relatively their position has improved in recent years we must continue our efforts on behalf of these two

(13)

Young motorists have been recognized as one of the main road safety problem areas in the

Netherlands: more of them are involved in accidents than the members of any other age group. SWOV has carried out research into the causes of this problem and possible solutions, including such ideas as prohibiting alcohol consumption, prohibiting night driving at weekends and raising the minimum age for car drivers. It should be noted that data collection tenninated at the end of 1985, thus the survey does not include developments since that date. Demographic factors

Young people aged 18 to 24 make up

approximately 12% of the total population. Males

are slightly more numerous than females in this age group. The proportion of the population represented by young people is expected in the long run to drop to some 8-9%. This should in itself have a markedly beneficial effect on road safety.

Driving licences and vehicle ownership Young people account for an increasing proportion of driving licence applicants. The number of young people taking the test, however, is declining sharply, as is the case in the other age groups. The number of young people holding licences will rise very slightly and then fall from

1990 as the size of this age group shrinks· Vehicle

ownership irrespeqive of licence-holding among

young people has declined very sharply in recent

years. Of 18 and 19-year-old males 12% own

vehicles; this compares with 52% of the 20-24

age group· More young men than women own

vehicles, and thiS is alSo true of older men: 83%

as against 20% of women·

Travel patterns

The travel pattern of young people differs substantially from that of the' 'middle group" of 35 to 54 -year-old road users· The reasons for travel by young people are spread more or less equally among work, visits/stays, shopping. sport/leisure and education. In the 35 -54 age

group the distribution is fairly uneven, with shopping and work the principal rea&bns. Young people make far more night trips than older people, particularly at weekends but al .. o during the week.

Accidents

Accidents involving young motorists differ noticeably from those involving the 35-54 age group: more accidents at weekends; more accidents at night; higher pre-collision speeds; more accidents in normal straight-ahead driving, fewer in stationary vehicles; more overturned

vehicles; more collisions w'th obstacles and one

-sided accidents; more men than women a.. front

-seat passengers; less use of -seat belts; lighter and older cars; head rests less common.

Traffic risk

The traffic r i5k is the name given to the ratio of the number of motorists direc dy involved in collisions to passenger-kilometre&' covered by the age group in que .. lion. Motorists aged 18 to 20 have a risk thatis tw'ce as high as that of the 21

-24 age group, which Le; already higher than that of the 35-54 age group. The risk at night, especially at weekends, is much higher than at other times of day for all age groups.

Road user behaviour in theory and in practice

Young motorists are unable initJ'ally to come up

to the 'itandard of skills reqUl'red ' They need some time to learn how to handle their vehicles and behave in traffic. After a certain period their skIlls are adequate but they then enter into a situation where excessive self -confidence and deliberate risk -taking contribute to a very high

traffe risk. It has been established that young

male motorists use their seat belts les .. than thel'r

more experienced counterparts. They also have

more convictions for a van'ety of traffic offences.

The data on alcohol con'iumption point not so mUch to higher consumption among young

people as to the exposure problem at weekend'i.

when more young people are out and about and there is a greater likelihood of thel'r drinkI"ng.

(14)

14

Crash and post-crash factors

Theoretically collision speed, weight and strength of vehicle, type of collision and use of safety devices are the main factors determining the outcome of collisions. In practIce motorists have control over only a limited number of these factors, collision speed and seat-belt wearing. Data on responsibility for and cause of accidents suggest that young people are no more frequently responsible for accidents involving other road users than experienced drivers. There are indications, however, that there is a connection between accidents and young people's speed behaviour.

Problem analysis

If we put the above factors together, we find that young motorists present the following pattern. They suffer from a substantial struct U'al

disadvantage in comparison with expert drivers, resulting from a combination of lack of

experience and age-related effects. The structural disadvantage diminishes with increas'ng age (and experience), sooner for women than for men. The problem is that virtually the only way of

overcoming the lack of experience

is

through

practice, but it is this inexperienced practIce that causes the safety problem.

The lack of expe:rience demands a much more

cautious style 0 fdriving than that of the experts,

but it is here that young people display

undesirable behaviou uul traits such as high speeds

and frequent n ght journeys. It is evidently in the

nature of young people that they lack the caution needed to compensate for the lack of experience in the broadest sense when performing their role

in traffic· They take greater risks than the experts

in hazardous conditions when precisely the

opposite is what is required. If we are to do

something about the lack of safety of young motorists we need to offer, or even impose, a considerable quantity of compensatory elements.

Compensatory system

The period from passing the driving test to achieving expert status can be seen as an apprenticeship, a period of probation. Suitable elements in a compensatory system might be a change in the minimum age for driving a car and associated changes in the requirements for driving courses and tests. Raising the age would result in a lower risk, at least for motorists; but it would force young people to travel using such modes of transport as cycles and mopeds, which

in many cases have a higher risk than cars· It

would be worthwhile, therefore, also to consider lowering the age: perhaps on balance this might

have a beneficial effect on road safety. In

addi-tion, a radical approach is needed to achieve a fundamental change in behaviour, not only speed

behaviour but also showing -off, recklessness and

rejection of authority and rules· Changes in the

driving course could help young people to gain more experience. Other measures which could be considered are a ban on night driving combined with a total ban on driving by young people who

(15)

Driving lessons for moped-riders could improve safety

There were two million mopeds in the

Netherlands in 1970. Since then the number has declined considerably, to 560,000 in 1986. The

records indicate an annual average of 11,500

moped-riders injured on the roads in 1985 and 1986. Of these just under 2,900 were admitted to hospital and 125 died. Moped-riders accounted for 22.7% of all recorded road casualties, 8% of road fatalities and 19.7% of all hospital

admissions in these two years. The data for 1987 indicate a slight increase in the number of deaths among moped-riders in comparison with the average for 1985 and 1986.

Compared with other modes of transport, the moped is associated with a higher risk of injury per kilometre travelled. The highest risk of fatal injury per passenger-kilometre in 1986 was to

moped-riders and their passengen;, even higher than the risk to motorcyclists. It was twice as high for male as for female moped-riders.

Mopeds are used mainly by young people and, as

we would logically expect, the casualty rate is

highest among them. If we look at the accident

rate among cyclists, moped-riders and pedestrians we see a marked rise in the accident rate at the age of 16, when the moped becomes available'. the number of accidents among 16-year- old cyclists, moped-riders and pedestrians combined

is over four times as high as among their 15-year

-old counterparts. There is no comparable increase in the accident rate at any other age. The number of moped-rider casualties is highest among the

16-year-olds; at the age of 17 the rate has already fallen by a quarter. Experience could well play a major part here.

(16)

16

As regards collisions with other vehicles, the majority of casualties occur in collisions with cars. In the fatal accidents, however, almost as many of the casualties result from collisions with heavy vehicles as with cars. Most of the casualties in collisions between other modes of transport and mopeds are cyclists and pedestrians. Over half the fatalities among moped Tiders occurred in accidents in non-urban areas, and such accidents more often have a more serious outcome.

Driving lessons

The minimum age for driving a moped in the Netherlands is 16, and there are no requirements as to driving lessons or state of health such as apply to other motor vehicles. Mopeds are counted as "slow traffic" and thus treated as equivalent to cycles and pedestrians as regards the training required. In practice, however, moped-riders differ from the other "slow" road users: children receive road safety education at primary school and are usually given some supervision by their parents, both on foot and when cycling, whereas they receive little if any supervision on when riding mopeds. While the current system of driving lessons and road safety education may not be as appropriate and effective as might be wished, the moped differs from all other modes of transport in having virtually no training provided in its use.

Factors affecting the safety of moped-riders are to

be found both in the riders themselves and in the other road users, the traffic system and traffic facilities. It has been established that other road users do not make sufficient allowances for the problems of moped -riders, but it is partly as a result of the nders' own behaviour that the other road users are prevented from making such allowances. Lack of expen'ence plays a maJ'or part, affecting vehicle control and understanding of hazards, among other things. Combined with high speeds this is likely to increase the risk. Attitudes are risk factors, as expressed, for example, in macho behaviour and "expen·menting".

Measures could be sought in various areas· One solution might be to introduce a classifica lion of two-wheeled motor vehicles with, say, increas-ingly stringent requirements imposed on the rider the more highly-powered the vehicle. Courses of driving lessons could be geared to the new classi-fication: the course for the low-power category could serve as a basis for all two-wheeled motor vehicles, and issue of a licence for a higher-power category could be made contingent on experience with a lower category and an addi-tional course of lessons.

Alternatively, a driving course could be introduced irrespective of any classification. It should be designed to overcome the need for riders to learn "by trial and error" and to enable riders to acquire the practical experience they need once they have completed the course in order to become "experts" in safer conditions. The subjects that the course should cover can be determined from accident studies and behavioura I

observations. Speed is a major foundation; this is also the aspect that differs most in comparison with cycling, of which riders already have some experience. High speed is a new factor for riders to learn to handle. Beginners in particular might well be surprised by its consequences. It causes a deterioration in riders' own perception and in other road users' perception of them; in many cases they do not realize this and overestimate their ability to stop in time. They al so overrate their ability to manoeu vre: mistakes in estimation can in crease the aoc iden t

ns

k, particularl y at intersec tions wi th other traffic present. Ri ders should in any event be taught the following aspects: the rules of the road, basic veh·.cle control skills, ability to recognIZe hazards and modes of behaviour designed to antiCipate danger. Theoretical and practl'callessons should be integrated. The course shoul d al so inc lude information on such matters as the dangers of drink -driving and the importance of a good, correctly fastened crash helmet.

(17)

Cyclists particularly underrepresented in official traffic statisUes

Every year I in 34 of the Dutch population is injured in a road accident - an annual total of 430,000 casualties. Of this number, however, only 50,000 find their way '\J1to the official statistics - a meagre 12 %.

The discrepancy between the actual numbers of road casualties and the official statistics is caused by two factors. Firstly, not an accidents are reported to the police (who supply the official figures). Secondly, the police are not required to record an accident unless it corresponds to the international definition: it must take place on a public road, at least one moving vehicle must be involved and there must be an injury (defined in the Netherlands as more than just a scratch). The SWOV survey showed that, of the 430,000 casualties, about 210,000 (49%) were covered by the definition; the police attended 45% of these cases (95,000 casualties), and of these, 50,000 ultimately found their way into the official statistics.

The largest category of road casualties in the official statistics is car occupants. The most common type of accident is the collision between a car and another vehicle. The incompleteness of the records is not in itself an obstacle to a proper understanding of the situation, provided they give a representative picture; this is not the case, however. The survey found that cycl ists were the largest category, in particular cyclists aged under 15. One-sided cycling accidents (i.e. not involving another party) result in far more casualties than any other category of accident.

Road accident casualties

Total from survey In-patients 18,000 Out-patients 117,000 No hospital treatment 292,000 Total 430,000 2

The official figures were more complete the more serious the injuries. A quarter of those injured seriously enough to require out-patient treatment were recorded. The statistics on those admitted to hospital as in -patients were reasonably complete and gave a reasonably good cross-section; the figures on fatal accidents were complete. Also, the police records include a good deal of information on the time, location, nature and circumstances of accidents, information which is not available from the hospitals. Another point in favour of the official statistics is that through the efforts of the police, the Road Accident Records Department and the Central Bureau of Statistics they are quickly made available. This makes them indispensable for the kind of road safety research SWOV carries out.

The Netherlands is not exceptional in keeping incomplete records: this is the case in every country. SWOV does see some scope for

improving accident records, however. In addition to the police records, use could be made, for instance, of other records of accidents and injuries, particularly to cover the less serious casualties. An experiment is taking place in collaboration with the Ministry of Transport which involves extending the Consumer Safety Foundation's personal accident records system (PORS). Victims of accidents in or near the home or connected with leisure or sporting activities treated as out-patients are recorded from a sample of hospitals; during the experiment road

casualties are also being included. (R -89 -13)

3 3

InJ'ured InJ'ured as percentage ( international (official of2

definition) statistics)

18,000 14,262 79

93,000 24,232 26

99DOO 11254 Jl

(18)

18

The safety of cycle traffic

There has been increasing I'nterest in recent years

in the vulnerable position of cycl'1sts in traffic. Most cyclist casualties occur in conflicts with motorized traffic ("fast traffic'), in particular with cars. During the years 1981 -83 inclusive a total of 13,344 cyclists were killed or sufficiently injured to warrant hospitalization as a result of road accidents. There had been a collision with a car in the case of 7,151 of them.

The risk of accidents is not the same to all cyclists, however: it is highest for young people and the elderly. The government is trying to reduce the risk as far as possible by means of regulations to protect cyclists, but existl'ng traffic rules and regulations are not always observed. The question is to what extent do protective regulations achieve their objective, and do they meet with an adequate Iesponse form cyclists themselves? Conversely, there are signs of a malaise affecting the cycle: it is said to have been degraded to the status of an everyday object which no longer receives proper care and attention. SWOY carried out a census of the observation of certain behavioural rules and regulations concerning cyclists and cycles: this involved 2,400 cyclists in the municipality of Assen in January, February and March 1987. Some of them were questioned. The points examined were whether the cycles were fitted with lights and their general state of repair .. also, whether they were fitted with wheel reflectors, and whether

lights were used in the hours of dusk and dark

-ness.

Cycle tracks

The Netherlands is a country of cyclists. Total

cycle ownership is 11 5 million to a population of

14.5 million, in other words 8 out of 10 inha

-bitants possess a cycle. This compares with some

600 DOO mopeds. The level of cycle and moped

The main conclusions from the survey - which reflects only the situation at the monitoring points in Assen and should not necessarily be regarded

as representing a national average -were as

follows.

1. Cyclists were more inclined to use their lights on a dark or over- castday than on a sunny day.

2. The average proportion of cycles with wheel

reflectors was 65%.

3. The principal reason given - by over 50% of respondents - for cycling with lights was "in order to be seen". Of cyclists aged over 45,300/0 gave the legal requirement as their reason. Twelve percent cycled with lights in order to see better themselves.

4. Fifty-four percent of those cycling without lights did so because they thought it was still light

enough. Twenty percent of the young people (in

the 12025 age group) were on cycles with broken lights or none at all.

5, Ninety percent of the respondents used their cycles every day, Young people used them to go to school and for sport or to keep fit. Older people used them mainly to go to work and for visits and shopping,

6, As regards state of repair, the condition of brakes and tyres was examined; this tells us something about the care the cyclist takes of his

cycle in terms of safety precautions, Rust was

also looked at, since this tells us something about

the general care taken of the cycle, Ninety -seven

percent of the cycles had good brakes and 73% had good tyres, The better its state of repair, the more likely it was that the cycle had both lights and wheel reflectors and that the cyclist actually

used his lights, (R -88 -5)

ownership I'n the Netherlands has not risen I'n

recent years. Twenty four percent of the

population cycle for recreational and sporting

purposes; 18% use thel'r cycles to travel to work,

(19)

The road accident rate in the Netherlands is relatively low, 10 fatalities a year per 100,000 of population. About 22% of these casualties are cyclists and 8% moped-riders. Comparing the accident rates for cyclists and moped-riders with that for car occupants we find that the risk (per kilometre travelled) is four times higher for cyclists and as much as eleven times higher for moped-riders. Seventy percent of moped-riders killed on the roads are in the 16-24 age group. There are two overrepresented groups among cyclist fatalities: 20% in the 10-20 age group and 40% in the 65+ group.

Of the 97,189 kilometers of road in the Netherlands, 13,422 are provided with cycle paths and 1,064 with cycle lanes. In other words, about 15% of the road network has facilities for cyclists, also moped -riders in many cases.

Three types of cycle facility can be identified'.

(a) Cycle paths, separated from the carriageway by a shoulder or verge.

(b) Cycle lanes, where a white line on the road marks off the area

allotted to cycles and mopeds.

(c) Cycle routes, marked out specially for cyclists' use.

It is often asked which type is the best from the road safety point of view, and SWOV has carried out extensive research into the matter, the

conclusions from which as are follows.

The construction of cycle paths along urban roads (as opposed to intersections) results in an

improvement in road safety for cyclists. If there is a cycle path along a particular section of road, it is safer for cyclists if there is none at the

terminating intersection, i.e. the cycle path should be "truncated" before it reaches the intersection. More research is needed into the safety of cycle lanes. It is asserted that these do not have a beneficial effect on moped-riders; cycle lanes along roads are even claimed to have a harmful effect on cyclists. This might well be due to their narrowness, as a result of which cyclists and moped-r'lders fail to keep within the markings as motorists expect them to. More research is also needed into accidents at and near intersections

and the behaviour of road users at these locations·

On non -urban roads the safety of cyclists and moped 1'iders is improved by the construction of cycle paths. The information available on the various types of cycle facilities at non -urban intersections is unfortunately inadequate.

(20)

20

Urban cycle tracks and the safety of moped-riders

SWOV has compiled a report on the road safety aspects of cycle facilities in urban areas, in particular the danger to moped-riders using them, The report is based on data from a general survey of the safety of urban cycle facilities,

Three types of facilities for cyclists and moped-riders are identified:

(a) Cycle paths, with a physical separation between (i) cycles and mopeds and (ii) motorized vehicles,

(b) Cycle lanes, reserved for cycles and mopeds and indicated by a white line, There are no physical obstacles to prevent cars etc, using this area,

(c) No special facilities for cycles and mopeds; all road users except pedestrians use the same traffic area,

The question is, what effect do these three types of facility have on the safety of cyclists and moped-riders and their collision partners? Data on sections of road with the three types of facl1ity in urban areas of fourteen towns and cities of over 50,000 inhabitants were analyzed, with particular reference to acddents involving

moped-riders, During the 1980-82 period almost

85% of the casualties (killed or admitted to hosp'J.. tal) in these accidents were moped-riders, just under 9% cyclists, 6% pedestrians and only 0,7%

occupants of motorized vehicles,

The accident rates per kilometre travelled for

moped-nders on sectl'ons of road with each of the

three types of facility were almost the same as

regards collisions with motor vehldes, The

acddent rate per km fOr moped -riders wl'th moton'zed traffic at terminating intersections of

sections of road with cycle paths was two or three

times higher than with other types of facility,

As for the general effect on moped

-riders

in

collisions with motorized traffic, cycle paths had

the highest weighted accident rate per km, The

accident rate per km for moped-riders on both sections of road and terminating intersections with cycle paths was highest as regards collisions with other moped-riders, cyclists and pedestrians, We may therefore conclude that the danger to moped-riders, in terms of the risk of injury accidents, is greater on roads with cycle paths than on roads without cycle facilities or with cycle lanes,

As regards collisions between cycll_{15ts and}

moped-riders, sections of road with cycle paths both along the road and at the terminating intersection had worse accident rates per km -much worse in some cases - than roads with cycle lanes or without cycle facilities,

The problems of pedestn'ans in conflic t W'IQ;} moped-riders, as far as injury aCCIdents are concerned, related mainly to roads (as opposed to intersections), The proportl'on of total injury accidents involving pedestrians accounted for by collisions with moped-riders was almost three times higher on sections of road with cycle paths

than on other sections of road, The accident rates

per km for moped -riders with pedestrians on

sectIons of road with cycle paths were also worse

than on sections of road without cy de faci 6ties or

with cycle lanes,

To sum up, the danger to moped -riders, in terms of injury accidents, is greater on roads with cycle paths than on roads without cycle facl1ities or

with cycle lanes, Colh'sions between pedestrians

or cychsts and moped -riders are more frequent on

roads with cycle paths than on the other types,

(21)

Wheel reflectors on cycles

Wheel reflectors have been a legal requirement on cycles since 1 January 1987. To assess the initial effect of the measure, and the subsequent effect on road safety, SWOV carried out a census in October and December 1986 and February, April and December 1987. This took place in five municipalities, at three types of locality in each: the cycle shed of a secondary school, the cycle shed at the railway station and - during the hours of dusk and darkness - among moving cycle traffic. This enabled a link to be established between the presence of wheel reflectors and the use of cycle lights. The December 1987 census included a poll to establish the link between cyclists' ages and the presence of wheel reflectors.

Schools

The proportion of secondary school pupils'

cycles with reflectors on both wheels rose from

11 % in October 1986 and 25% in December 1986 (i.e. before the legal requirement was introduced)

to 69% in April 1987. From February to April

1987 there was Ilttle further increase. The

different schools were not large, which leads us

to assume that these figures can be regarded as representative of secondary school pupils throughout the Netherlands.

Railway stations

The proportion of cycles with wheel reflectors at

railway station cycle sheds has also risen

considerably since October 1986, albeit reaching

a lower level than at the schools. The average

proportion in April 1987 was 54% . There were

large differences between the different stations,

the lowest proportion being found at Amsterdam

(37%) and the highest at Hoom (70%); thus these

figures cannot necessarily be taken as

representative of railway stations throughout the

Netherlands.

Moving traffic

Here again there were large differences in the proportion of cycles with wheel reflectors at the

different locations. The censuses taken from 4.00

of 56% over the February-April 1987 period,

with the lowest figure again in Amsterdam (35%)

and the highest 10 Hoom and Hi Iversum (65%).

Lights were more often in use on cycles with wheel reflectors than those without. An average of 64% of all cycles had their lights on at a census in February 1987, as opposed to 75% of cycles with wheel reflectors and 46% of those without. From the road safety point of view this is a highly unsatisfactory situation.

Wheel reflector types

There were three main types of approved wheel reflectors: tyre reflectors, spoke reflectors and rim reflectors. The first type is a vulcanized strip on both sides of the tyre, the second covers a wide variety of circular reflector systems which

are affixed to the spokes or in between them.

The incidence of rim reflectors was so low as to

be almost negligible. The proportion of tyre

reflectors found in April 1987 was about 33%, leaving spoke reflectors in the majority (67%). Most of the increase since October 1986 was accounted for by spoke reflectors.

Quality of reflecting materials

In its current form the census was not designed to

look at the quality of the reflecting materials.

It was clear from the experiences of the census

-takers, however, that there is some doubt as to

both the mechanical quality and the efficiency of

various types of reflectors. There are also

indications that cheap unapproved tyres with

reflectors are being sold and used.

Age factor

The differences in the proportion of cycles found with wheel reflectors at the schools and stations and in moving traffic can ascribed partly to the

age factor. The proportion of cycles with wheel

reflectors was markedly higher in the 12 -16 age group. It was partly for this reason that it

was decided to carry out a separate survey in

December 1987 to gain more information on

what was presumed to be the age factor.

(22)

22

December 1987 census

The results of the Dece mber 1987 census showed an increase in the proportion of cycles with wheel reflectors, by 10% acros~'the board, bringing the total proportion - insofar as this can be assessed from the limited number of census locations - to an average of 70%. The proportion found among the schoolchildren had risen to some 80%, again

~howing that secondary school pupils are the

most diligent in fulfilling the legal requirement. The survey revealed that the lowest proponion of

wheel reflectors was found in roughly the 20-35 age group. This i~' an area where improvements are called for. The distribution of the different

type~' of reflector had changed considerably by

December 1987: there had been a marked increase in the proportion of tyre reflectors and an equally marked decrease in that of spoke reflectors, with the result that the two types are now equally common. It is not unusual, in fact, for both types to be found on the same wheel.

(23)

Slight increase in proportion of cycles with wheel reflectors

For several years now SWOV has been

monitoring the fitting and use of safety features on cycles, including wheel reflectors and the use of lighting.

More cycles displayed good wheel reflectors in December 1988 (73%) than during a survey in December 1987 (70 %). The increase was smaller than during the previous period. The legal

requirement no longer seems to be having the direct effect of accelerating the growth in the proportion of cycles with wheel reflectors. In April 1987 about a third of these were tyre reflectors. This rose to some 600/0'n December

1988, with a marked decline in the proportion of cycles with spoke reflectors, to 42% in

December 1987 and only 34% twelve months later. The change has been due to the influx of new cycles and the replacemen t of old tyres with new ones bearing reflector strips. There is now a marked preference for reflector- tyres over spoke reflectors.

The monitoring of wheel reflectors is carried out at cycle sheds of secondary schools and railway stations, and among cycles on the move at various locations. In December 1988 the

proportion of cycles WIth wheel reflectors at the school cycle sheds was about 77%, 4% higher than that measured in December 1987. The

observations at the railway station cycle sheds showed a similar increase from December 1987 to December 1988 as those at the schools. Here the average proportion of cycles with wheel reflectors in December 1988 was about 66% -again 4% higher than in December 1987. There was scarcely any increase in the proportion of cycles monitored on the move with wheel reflectors: 65% in December 1987.

About 86% of cycles had some form of wheel reflector, including whimsical and defective types and reflectors on only one wheel. At least 15% of these dId not meet the statutory requirements. Lights were more often in use on cycles with wheel reflectors than on those without: on 70% and 36% respectively. The risk to those cyclists whose cycles are without wheel reflectors and who do not use their lights can be assumed to be higher than before wheel reflectors were made compUlsory.

The present design of the survey does not enab ~

the quality of the reflective materials to be

measured. It is clear from the observers' expe

-riences that the amount of reflection varies greatly. Even some tyres with approved reflectors provide hardly any reflection.

(24)

24

Motorcyclists wear helmets better than moped -riders

SWOV has carried out a survey to establish whether Dutch moped-riders and motorcyclists wear their crash helmets correctly and the state of the helmets worn .

The main conclusion from the survey of

motorcyclists was that more motorcyclists fasten their chinstraps, ·and fasten them better, than moped-riders: 81 % of motorcyclists had their chinstraps correctly fastened, as against only 32% of moped-riders. As in the case of the moped-riders it was mainly the younger ones who did not fasten their chinstraps, the 18 and 19-year-old males.

Thirteen percent of the helmets did not bear an official symbol of approval. The older the helmet, the more likely it was not to be approved: this was the case with 6% of the helmets that were under twelve months old and 50% of the helmets of 8 years and older.

The outer shell of the helmet is designed to protect the rider against sharp objects. The materials of which this is usually made are highly sensitive to chemicals, and helmets bearing stickers or paint are therefore more likely to crack in an accident than "undamaged" helmets.

Anyone wearing a helmet in this condition - all other things being equal - runs a n·sk of more serious head injuries on average. Of the helmets examined, 16% had been painted or had stickers attached. Here again this was more common among the younger wearers.

It is recommended that a helmet that has been stressed in an accident be replaced, sl·nce the

inner shell does not subsequently return to its original shape and thus loses some of its

cushioning capacity. Seven percent of the helmets had not been replaced after being stressed in an accident.

Various safety requirements can be laid down for helmets and the way in which they are worn, e.g.

correctly fastened chinstrap, helmet of approved type, no chin cup, surface of helmet undamaged and unpainted, not stressed in an accident. Fifty percent of motorcyclists' helmets met these requirements, as against on

0/

17% of moped

-riders' .

The report recommends working towards a standard press-stud fastener for chinstraps . This should cause great inconven·~nce to a wearer who fails to fasten it but be comfortable in use

-it should more or less force the wearer to use -it. Devices of this kind are already on the market. Medical practitioners are also urging the intro-duction of a standard fastener.

In the short term information should be supplied to helmet -wearers on the increased risk of head injunes to those who fail to use the chinstrap correctly or wear damaged or painted helmets. The origin of unapproved helmets should also be investigated. SWOV recommends reintroducing random testing of helmets on sale in the shops. In additIon, consumers should be provided with information on the points they should take into consideration when buying a helmet.

(25)

More people wearing seat belts

SWOV has carried out annual censuses and surveys of the presence and use of seat belts in the Netherlands since

1969.

Again in

1988

we examined whether the drivers and front-seat passengers of cars at light-controlled intersections and motorway exits were wearing their seat belts. A number of drivers were also questioned. Campaigns to encourage seat-belt wearing were held regionally and locally in

1988,

in particular in the provinces of Gelderland and North

Brabant. These resulted in higher use rates at almost all the monitoring points in the two

Seat belt use rates (drivers)

provinces. The figures in the other provinces presented a much less satisfactory picture. The general trend is upward. The rates of seat-belt use are rising in both urban and non-urban areas· In non-urban areas 77% of drivers wore a seat belt, as against

63%

in urban areas. The table below shows how the rates have changed over the last nine years. The

1981

level has been regained in both urban and non -urban areas, but this is still below the maximum of

1980. If

we break down the figures by type of road we find that the

1988

rates were higher than those in

1979,

the first census year.

Year Urban roads Non-urban Motorways Non-motor- Mixed Local roads

1979

51

69 1980

57

73 1981

52

70 1982

50

66 1983

46

65 1984

50

67 1985

49

66 1986

49

67 1987

55

70 1988

63

77

Another striking point was that there were marked differences in use rates between the four types of road during the entire period from

1979

to

1988

inclusive: over

80%

on motorways, about

70%

on non -motorway roads, over

65%

on roads with mixed traffic, and about

60%

on roads carrying local traffic. The rate was lower at the local traffic monitoring points in Amsterdam and The Hague than at other local traffic monitoring points. The Hague had the lowest figure -a mere

30%

l It was found that journeys were pre

-way traffic traffic traffic

70

76

73

70

69

73

72

73

77

83

67

58

44

68

62

52

65

57

45

61

56

43

58

54

39

58

57

44

60

59

40

60

58

41

63

48

72

67

59

dominantly long on non -urban roads and short on urban roads; the longer the journey, the higher the rate of seat -belt use.

The rate for &ont -seat passengers was virtually the same as for drivers. The percentage of women wearing thei r seat belts was higher than that of men, on both urban and non -urban roads: the rate fo'rmen on non -urban roads in

1988

was

76%

as opposed to

83%

for women; the rateS on urban roads were

61 %

and

69%

respectively.

(R

-87

-}7,'

R

-

87 -

18 ;

R -87 -

21 ;

R -88 -8

;

R

-89

-

I

)

(26)

26

Tw

e

n

w-tive percent of cars have rear seat bel

~ As of I January 1990 seat belts mu~t be fitted to

rear seats of cars in the Netherlands. At the end of 1988 SWOY carried out a trial survey to measure the proportion of cars fitted with rear seat belts and the use made of them before they became compulsory.

Not one of the ten highest-sel lng type~' of car in the Netherland~' was found to have rear seat belts fitted as standard, although more expensive models and American cars had this feature. A few manufacturers fit belts to rear seats as

standard on all their models. A small-scale survey in three towns in the Netherlands indicated that a

quarter of all car~' had rear seat belts at the end of 1988. Ofthe 2,172 cars looked at, 1627 (75%) had no rear seat belts. There was a great diversity in the types of belt found in those cars which had at least one fitted: most cars had three-point types, in some cases combined with a lap or three-point belt in the centre (16% of the total); 5% had only lap belts. More rear seat belts were found in newer cars than older ones. In those cars which did have rear seat belts only a quarter of the rear-seat passengers were using them. Children under 12 used their belts more often than older rear-seat passengers.

Research activities 1987-1989