Improving the effectiveness of road safety campaigns: Current and new practices

The evaluation of campaigns aimed at improving road safety is still the exception rather than the rule. Because of this, ineffective campaigns and campaign techniques are allowed to continue to be utilised without question, while new methods of behaviour modification are often ignored. Therefore, the necessity and advantages of formally evaluating road safety campaign efforts are discussed.

See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/251590277
Improving the effectiveness of road safety campaigns: Current and new
practices
Article in IATSS Research · March 2011
DOI: 10.1016/j.iatssr.2011.01.003
CITATIONS
45
READS
561
2 authors:
Some of the authors of this publication are also working on these related projects:
Sustainable SafetyView project
Tamara Hoekstra
4 PUBLICATIONS46CITATIONS
SEE PROFILE
Fred Wegman
Delft University Of Technology, Faculty Civil Engineering and Geosciences
63PUBLICATIONS630CITATIONS
SEE PROFILE

Improving the effectiveness of road safety campaigns: Current and new practices
Tamara Hoekstraa, Fred Wegmana,b,
⁎
a SWOV Institute for Road Safety Research,The Netherlands
b Delft University of Technology,The Netherlands
a b s t r a c ta r t i c l e i n f o
Article history:
Received 1 September 2010
Received in revised form 22 November 2010
Accepted 22 November 2010
Keywords:
Road safety campaigns
Evaluation
Fear appeals
Automaticity
Priming
Framing
The evaluation of campaigns aimed at improving road safety is still the exception rather than the rule. Becaus
of this, ineffective campaigns and campaign techniques are allowed to continue to be utilised without
question,while new methods of behaviour modification are often ignored.Therefore,the necessity and
advantages of formally evaluating road safety campaign efforts are discussed.This article also describes the
pros and cons of some of the more common campaign strategies and introduces a number of new methods
that show a great deal of promise for the purpose of road safety campaigns. In order to infuse the field of road
safety campaigning with such new insights into road user behaviour and behaviouralmodification,one
should look beyond the confines of road safety campaign standards and learn from the knowledge gained in
other disciplines such as economics and social psychology. These new insights are discussed in terms of their
implications for the future of road safety campaigns.
© 2011 International Association of Traffic and Safety Sciences. Published by Elsevier Ltd. All rights reserved.
1. Road safety and road safety campaigns
Road traffic today is inherently dangerous.In fact, in contrast to
other modes of transport such as railways and air traffic,the road
traffic system was not designed with safety as a jumping-off point [1].
Consequently, in road traffic it is us humans who make the difference
between hazard and safety, with little keeping us from harm should we
make a mistake. Differently put, unlike other modes of transport that
have procedures, safeguards or fallbacks to limit both the occurrence
and impact of human error, road traffic relies more heavily on its users
to keep accidents from occurring.Given that humans are almost
inadvertently prone to make mistakes and commit violations, human
behaviour is of particular interest for most road safety professionals.
That is where road safety campaigns come in.Together with other
‘behavioural’measures (e.g.,law enforcement,education,training,
and even infrastructure to some extent) road safety campaigns are
used as a means of influencing the public to behave more safely in
traffic. Road safety campaigns can be defined as purposeful attempts to
inform, persuade,and motivate a population (or sub-group of a
population) to change its attitudes and/or behaviours to improve road
safety, using organised communicationsinvolving specific media
channels within a given time period [2]. It can have many and multiple
purposes,such as informing the public of new or little known traffic
rules,increasing problem awareness or convincing people to refrain
from hazardous behaviours and adopting safe ones instead.
This article is based on a keynote speech delivered by Fred Wegman
at the PRI (Prévention Routière Internationale) World Congress on June
24th 2009.In this speech,Wegman examined issues pertinent to the
present and future of road safety campaigns by discussing evaluation
practices, common fallacies concerning the nature of human behaviour
and new insights into behaviour modification that may be of use for the
future of road safety campaigns.In this article,we elaborate on the
insights put forth in the speech with the aim of providing inspiration for
the future of road safety campaigns. In order to do so, we first discuss the
current practices in evaluating road safety campaign evaluations or the
lack of such. We will then examine some of the methods of behavioural
modification that are already widely implemented in road safety
campaigns. Finally, we will delve into the many promising methods of
behavioural modification that can be found in fields such as psychology
and economics.These methods and insights are discussed in terms of
the ways they might be used to positively influence road user behaviour
and by extension, road safety itself.
1.1.Are road safety campaigns effective?
Surprisingly, all the more so given the sheer number of road safety
campaigns,only a fraction of such campaigns are formally and
thoroughly evaluated [3].Despite this dearth ofevaluation results,
there are a number of reports on the subject that give some indication as
to the potential effects of road safety campaigns. For example, in 2004
the World Health Organisation concluded that road safety campaigns
were able to influence behaviourwhen used in conjunction with
legislation and law enforcement. However, the report also states that “…
when used in isolation education,information and publicity generally
IATSS Research 34 (2011) 80–86
⁎ Corresponding author.Delft University of Technology,The Netherlands.
E-mail address: [e-mail] (F.Wegman).
0386-1112/$ – see front matter © 2011 International Association of Traffic and Safety Sciences.Published by Elsevier Ltd.All rights reserved.
doi:10.1016/j.iatssr.2011.01.003
Contents lists available at ScienceDirect
IATSS Research

do not deliver tangible and sustained reductions in deaths and serious
injuries” [4].
Likewise,a meta-analysis showed that the effects of mass media
campaigns alone are small, especially when compared to the effects of
campaigns that were combined with other measures [5].Without
enforcement and/or education a mass media campaign has virtually
no effect in terms of reducing the number of road accidents,while
adding either of both those measures ensures a reduction of over ten
percent (see Table 1).Interestingly enough, it is the local,personally
directed campaigns thatshow by far the biggest effect on road
accidents. However, it should be noted that the confidence interval for
this result is quite large,meaning that there is a large uncertainty
about the true value of this parameter. More importantly, however, is
that the meta-analysis only contained a few studies on this type of
campaign,and as such this result is based on rather few studies [6],
which provide another alternative explanation for the fact that local
individualised campaigns seem so much more effective than other
types of campaigns.
1.2.The business of evaluating road safety campaigns
The above discussion is based on the results of a meta-analysis. For
such analyses, the results of multiple evaluation studies are combined,
which allows for comparisonsbetween different types of media
campaigns alone and media comparisons in conjunction with other
measures.For the evaluation of individual road safety campaigns,
however, it is often quite difficult to isolate the effects of the campaign
component from the effects of the measures the campaign is combined
with. That is,it is hard to tell if obtained results are the result of the
campaign itself,or whether they are the result of the campaign
combined with these other measures, or even whether the same effects
might have been reached by using those accompanying measures alone.
One way to deal with some of the difficulties thatcome with
evaluating the implementation of a campaign, is by pretesting (parts of)
the method and of the campaign in a controlled environment. Pretesting
can take on various forms depending on the specific characteristics of
the intended campaign,but in general it refers to a small-scale study
where the campaign conceptor individual aspects ofit are tested
experimentally (i.e., comparing the behaviour or behavioural intentions
of people who were presented with the campaign to the behaviour of
people who have not been presented with the campaign; comparing the
behaviour of people who have been presented with different concepts of
the same campaign) to determine if the campaign strategy has any hope
of influencing people's behaviour. If the chosen method of influencing
behaviour proves successfulin an experimentalsetting,this may be
taken as an indication that it's worth trying on a larger scale.If the
method does not prove to be effective experimentally, the results of the
pretest may prove useful in figuring out how to improve the campaign at
a point when it is still possible to make such an overhaul.
When it comes to evaluating the results of a road safety campaign
that is or has actually been implemented in the field,a good research
design can do much to counteract many such difficulties in the
evaluation of road safety campaigns.
At the very least, setting up an evaluation study in a way that will
generate credible resultsrequires some basic knowledge of the
principles involved in such studies.Such knowledge can be gleaned
from the reports engendered by the European CAST (Campaigns and
Awareness-raising Strategies in Traffic safety) project. This project was
aimed at making clear guidelines and tools to encourage the proper
design and evaluation of road safety campaigns. In doing so, a number
tools were created that can be helpful in setting up a sound evaluation
study, such as a manual [2], an evaluation tool [7] and a reporting tool
[8]. For the purpose of this article, it is not our intention to outline the
necessary steps in the evaluation process.Indeed,the above publica-
tions acquit themselves ofthat purpose more thoroughly than one
article ever could.Our purpose in discussing the evaluation ofroad
safety campaigns is merely to point out the necessity of it.
1.3.The importance of evaluating road safety campaigns
Because reports on the evaluation results of road safety campaigns
are few and far between,there is still little insight available into the
effectiveness of campaigns in general, let alone which ingredients have
proven to be successful, and which have not. This in turn makes it hard
to determine if and how the practice and effectiveness of road safety
campaigns might be improved,thereby depriving the organisations
behind road safety campaigns of the opportunity to learn from their
successes and their mistakes and make a bigger difference. Evaluations
of road safety campaigns may, for example, shed some light on the mor
controversial of current practices (such as the use of fear appeals) and
help determine if and when these practices are really effective.
Another manner in which the current lack offormal evaluation
practices is to be lamented, is that it means that ineffective campaigns
and campaign techniques are allowed to continue unchallenged.The
continued use of funds for campaigns whose effectiveness is unclear
means that available funds are notspent on other, possibly more
effective road safety interventions.This may even keep the field from
looking to employ other new and less traditional approaches. Hopefully,
as evaluations become more common in the field of road safety
campaigns,it will also create some headway for newer methods of
behavioural modification to come into play in the stead of techniques
that have yet to be proven effective.
In furthering our aim to enhance the effectiveness of future road
safety campaigns,we will now proceed to discuss some insights into
current practices and beliefs within the realm of road safety campaigns,
such as different implications of the targeted behaviour type for the
campaign method,the notion that media campaigns can be used to
reach the whole population and the controversy surrounding the use of
fear appeals.
2. Current practices within the realm of road safety campaigns
2.1. Target behaviour: differences between different types of behaviour
targets
One of the things that are reflected by the meta-analyses into road
safety campaigns over the years,is that the effectiveness ofsuch
campaigns vary considerably depending on the type of behaviour that
is targeted [5,9].Campaigns aimed at increasing seat belt use,for
example, have been very effective in promoting its usage (see [10], for
this and other examples ofeffective road safety campaigns).To a
certain point that is, because as behaviours become more pervasive, it
becomes increasingly hard for campaigns to have any further effect on
them. More specifically,as the base levelof the measure ofeffect
increases, the impact and expected improvement as a result of a mass
media campaign is reduced [11].Thus,the ‘baseline’of a behaviour
must be taken into account in determining beforehand if a road safety
campaign is to be of any use.
However, not only the baseline of the targeted behaviour matters
when considering what type of road safety campaign to implement.
As was stated before, all aspects of the target behaviour itself should
be considered in determining the best approach. That is, what works
Table 1
Effects of road safety campaigns on road accidents [5].
Best estimate 95% confidence interval
General effect −9% (−13; −5)
Mass media alone +1% (−9;+12)
Mass media + enforcement −13% (−19; −6)
Mass media + enforcement + education−14% (−22; −5)
Local individualised campaigns −39% (−56; −17)
81T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86

for a campaign aimed at drink driving may not work for speeding, and
vice versa. There are several examples of research illustrating this very
phenomenon. The effectiveness of road safety advertising aimed both
at fatigue [12] and drink driving [13], for example, was most
determined by whether or not suitable alternatives were provided
(i.e., having a designated driver,and alternative forms of transpor-
tation). However, providing people with alternatives for their
behaviour is not effective in influencing all traffic related behaviours.
In the case of speeding behaviours,for example, such alternative
behaviours are not readily available. That is, it is not really possible to
suggestalternative ways to deal with the dangers that speeding
presents,other than to just not engage in this particular behaviour
[13]. It is worth noting that speeding in generalis a notoriously
difficult behaviour to influence through campaigning.Not only is it
difficult to provide people with viable alternatives for this behaviour,
but it also seems that even enforcement does little to sway people in
their propensity for speeding.Indeed,many young drivers hold the
belief that speeding is socially acceptable and that avoiding detection
by law enforcement is a breeze.More importantly,however,is that
reportedly not even increases in the chance of getting caught would
sway these youngsters from speeding [14]. The lack of effect garnered
by enforcement in relation to speeding may be owed in part to the
public's conviction that enforcementis in place not because the
behaviour is truly dangerous,but because the government sees it as
an expedientway to rake in cash [13]. Thus, in contrast to drink
driving and fatigue,speeding behaviours are far less receptive to the
provision of behaviouralalternatives and even enforcement or the
publicity thereof.For road safety campaigns in general,this implies
that the nature of the behaviour determines whether or not a certain
tactic will be effective, something which can be determined by
pretesting a campaign.For speeding behaviours in particular,this
means that different means should be sought to try and influence this
behaviour for the better.Since it has been suggested that the lack of
effect garnered by enforcement is due to the belief that speeding is
socially acceptable,that detection and the subsequent penalties can
be easily avoided and that the imposed fines represent nothing but
governmentalgreed,the first step may be to try and tackle these
beliefs.
Not only do the individual characteristics of a publicity campaign
have different implications on different types of outcome behaviours,
but the combination of publicity with other measures(such as
enforcement, education and personal communication) also has differing
effects depending on the type of targeted behaviour. In a study on both
the individual and combined effects of publicity campaigns and
enforcement aimed at drink driving on the one hand and or speeding
on the other, it was shown that the combination of publicity campaigns
with enforcement did indeed have a different effect depending on the
type of target behaviour.With drink driving campaigns both enforce-
ment and publicity campaigns had significant independent effects, but
no combined reinforcing effect.The reverse was true for speeding
campaigns,where only the combined effect of enforcementand
publicity campaigns was significant and not the two measures
independently [13].This last finding may have something to do with
the finding mentioned earlier that speeding is not easily dissuaded
through enforcement.The fact that in spite of the lack of individual
effects of either enforcement or publicity,a combined effect was still
obtained, is somewhat more surprising. The author suggests that since
the focus of publicity campaign was on the severe consequences of the
crashes,this may have counteracted some of the prevalent cynicism
concerning the actual motives of the government to enforce the laws on
speeding [13].
Taken together, these results suggest a number of things that should
be taken into account. First, it is important that the individual
characteristics ofthe targeted behaviour be taken into account,for
what works for one type of behaviour may not work for another. Second,
it is important to realise that not only does this have implications for the
measures you wish to employ individually, but that even the
combination of different measures may also interact differently given
a different set of targeted behaviours. Furthermore, to determine what
possible effects any given approach might have on the behaviour at
hand, carefulpretesting may be usefulin order to determine what
particular approach or combination ofapproaches is in order.Andfinally, the fact that different road user behaviours seem to react
differently to various types and combinations of road safety interven-
tions means that specialcare should be taken when conducting or
interpreting meta-analyses on road safety campaigns.These types of
analyses tend to bundle evaluation results using different methods,
focusing on different behaviours and groups within various settings. The
research discussed in this paragraph clearly illustrates that this may
prove problematic in the interpretation of the results.
2.2.Target audience: using mass media campaigns to reach the entire
population
Concerning the prospective audience of a publicity campaign, there
seems to be an implicit belief that when you need to reach a bigger range
of people or perhaps the whole of a population, mass media campaigns
are the way to go.The idea seems to be that almost everyone either
watches television,listens to the radio or reads the paper or perhaps
even does all of these things at some time or other. Therefore, the greate
the reach of the media channel you use to put your message out there,
the greater number of people who will receive that message. However,
research consistently shows that any type of campaign, including mass
media types, is more effective when it is focused on specific groups [2].
Furthermore, there is some evidence that people with low education are
less likely to be reached with information through mass media
campaigns. This is not due to a lack of comprehension of the message;
rather, people with a lower level of education seem less prone to even
pay attention to the message conveyed by such campaigns [15]. Instead,
people with lower degrees of education are far more likely to be reached
by personally contacting them than by ads placed in a local newspaper.
Personally contacting people with lowerdegrees of education (by
phone, mail or personal visitations) seems to have a greater chance of
getting the message across. This supposition is lent even more credence
by the results of the meta-analysis mentioned earlier,which also
suggested that local,personally directed campaigns were particularly
effective in diminishing road accidents [5]. Taken together, these results
indicate that campaigns using more personalforms of delivery and
communication are indeed worth the time and effort.
2.3.The pros and cons of fear appeals
Judging by the number of road safety campaigns that make use of
fear appeals,there is a firm belief in the ability to “scare people
straight”.The idea is that when fear is aroused,people will become
more motivated to accept the message and recommendations
presented in a campaign [16].Implicitly, the way people sometimes
react to these types of campaigns (shock,horror, or even tears) is
taken as a sign that the message got through to people, with the firm
believers in these types of campaigns saying that they know it works
“because you can see the kids leave the classroom with tears in their
eyes”. The belief that this is somehow indicative of the effectiveness of
this approach supposedly is that when people are thus affected by the
campaign,they are certain to comply with the dispensed advice.
However, it is important to remember that the amount of tears shed is
not the ultimate goalof these campaigns.The ultimate goalis the
effect the campaign has on actual behaviour and, by extension, on the
number of road accidents,and on that count the jury is still out.
Because the fact is that although fear can motivate people, it can also
have the opposite effect. It may in fact lead people to employ so-called
defensive responses.Such responses may take many forms, for
example with people discounting the veracity ofthe claims in the
82 T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86

campaign,by them saying that the campaign bears no personal
relevance to oneself,or even by avoiding exposure to the campaign
altogether [16,17]. Indeed, from a scientific point of view, fear appeals
are rather controversial, in the sense that research into this approach
shows a mixed bag of results.In some cases,fear appeals seem to
generate promising results in terms of perceived severity,suscepti-
bility and message acceptance, while in others the approach seems to
engender the aforementioned defensive response [16].
These diverging results may have something to do with perceived
self-efficacy on the one hand, and the perceived threat and
susceptibility on the other [16].Perceived self-efficacy refers to the
extent to which people think they are able to do something to prevent
whatever the fear appeal is portraying. Thus, when self-efficacy is low
(i.e., one does not believe there is anything one can do to stop the
fearful image from happening),people are more likely to show the
defensive reactions to fear appeals which render them ineffective
[16,17]. Perceived threat refers to the extent to which people believe
themselves to be in any danger ofthe consequences shown.Even
when people are sympathetic to the plight shown by the fear appeal
and feel that the recommended precautions are both sensible and
doable, the fear-appeal will not have the intended effect if people do
not believe that the consequences shown will ever happen to them.
Only if people feelthat the portrayed consequences are relevant to
themselves and feel they are able to take the preventive measures the
campaign proposes, does the fear-appeal have a chance to work.
Another possible explanation for the diverging results is that the
effect of fear appeals may be dependent on cultural differences. In the
Netherlands,for example,there is a long tradition of road safety
advertising with an emphasis on humour rather than fear.This in
contrast to countries such as Australia, New Zealand, the United States,
and Great Britain that often show explicit pictures of crashes, casualties,
injuries and blood, and the related emotions of pain, sorrow and grief of
traffic victims and relatives [18]. Using such methods in countries where
the people are not used to them, may cause a controversy which might
detract from the impact of the campaign.
Yet another factor that determines whether or not fear-appeals
have the desired effect,is the gender of the group the campaign is
aimed at.Women tend to respond more favourably to fear appeals
than men,as evidenced by greater recall,more effect on behavioural
intentions and more positive attitudes towards the central message
(e.g.[19,20]).Young males,especially,seem to have little suscepti-
bility to fear appeals. Lewis, Tay and Watson (2007) [20], for example,
posit that fear appeals have less effect on young males,because they
tend to discount and avoid them.So too did Tay & Ozanne [39] find
that a fear-evoking drink driving campaign resulted in a reduction in
fatal crashes in the group of women of all age categories and in older
men (35–54 years old), but not in the target group of young men ages
18 to 24. As young males in generalhave a higher crash rate than
others, the implications of the differing effects of gender on the
effectiveness of fear appeals seem highly poignant.
The above discussion on the possible pros and cons of fear appeals
illustrates that the wielding of this method is fraught with difficulty.
For one thing, one must limit the possibility that people will respond
defensively to the campaign,meaning that the fear the campaign
evokes should not be so overwhelming that people feel they can do
nothing to prevent it. One possible way to do this is by supplying the
audience with specific actions that can be taken as preventive
measures.It is also important that this action is not only something
people think they will be able to do, but that they feel is both credible
and suitable to be considered a preventive measure. Furthermore, the
audience should be made to feel that the problem is relevant to them.
Finally,there are the cultural backdrop and the characteristics of the
target audience to consider.Other than the age and gender ofthe
prospective audience, this might also mean focusing on consequences
that the target audience will care about.Young people,for example,
may be far more impressed by the repercussions of a road accident on
their looks or on their social standing than on their general health and
longevity.
All in all, the mantra “if you scare them, they will change” is not as
easy as it might at first glance appear to be. Even when all the pointers
above are taken into account,people may react differently to a fear-
based campaign than expected.Therefore,careful pretesting is in
order,not just in terms of how people experienced the imagery,but
rather of what most road safety campaigns are actually trying to
accomplish, namely, a change for the better in terms of behaviour or
behavioural intentions.
3. New approaches on the horizon
Thus far, the discussion of road safety campaigns has centred around
the status quo: what has been done, how effective has it been and what
steps can be taken to improve upon some of the current practices both
the method and evaluation of road safety campaign. In the remainder o
this article,we turn our eyes to the future,to discuss some promising
insights into behaviour and behaviour modification that can be gained
from social psychology and economics alike. These insights, though not
necessarily new or unheard of in their own right, are promising in the
sense that they may prove to be useful when adapted to the field of roa
safety campaigns. Furthermore, all are centred on the same supposition
namely that we humans are by no means the rational decision makers
that we think we are [21,22].
Many road safety campaigns seem to be based on the implicit tenet
that if people knew better, they would do better. Usually, such
campaigns seek to rectify the current situation by providing the right
amount and quality of information (not to mention the reliability of its
sources), expecting that as a result, people will come around to the righ
way of thinking and doing.Granted,some campaigns also employ
elements such as attitude and the socialnorms that play a role in
people's chosen behaviour. Yet even when such elements are
implemented in a campaign,the strategy often seems to veer right
into presenting people with information on either or both the
ramifications of the undesirable behaviour and the profits of the correct
behaviour.
Underneath it all, most methods seem to be implemented because
it is assumed that the behaviour people exhibit is the result of rational
thought and careful deliberation. That is, given a set of options people
will choose the option that makes the most sense and will ultimately
result in the biggest payoff.Logically,that would mean that telling
people that certain behaviours are dangerous,should result in those
people choosing to do the sensible thing and stop taking such
unnecessary risks. People are rational creatures, after all. Or are they?
Evidence that human decision making is decidedly not always the
result of rational thought,is bountiful.On the contrary,many of our
actions are the result of our habits, feelings, biases, circumstances, and
so on and so forth [23,24].In fact,many behaviours are so much the
result of such things that conscious rational thought does not play the
significant role that many seem to attribute to it. As such, many
behaviours have come to be considered as automatic: behaviour that
is performed unintentionally and without conscious knowledge and
control [25].One way to overcome this is to break a habit by either
timing your message to be delivered at the precise moment when
people are prone to revaluate their habits (e.g.,change of job or
address,having children, etc.) or by aiming the intervention at
stopping bad habits from forming [26,27].Another way to deal with
the fact that many behaviours are in factautomatic,is by simply
embracing that fact.That is, rather than ignoring this insight and
holding out hope that people must eventually succumb to our
carefully reasoned arguments,we propose the following: ‘if you
can't beat them, join them’. What we of course mean by that glib little
idiom, is that when we cannot overcome people's inbred habits and
inclinations with cold hard logic, why not try to make use of the very
83T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86

fact that many behaviours are,at least to some extent,automatic in
nature.
3.1.Making use of the automaticity of behaviour
Automaticity has been demonstrated to play a role in any number of
behaviours.For some types of behaviours,this fact has been well
established. The task of driving a car, for example, is widely accepted to
become automatic in that after a certain amount of practice it does not
require our express attention and we do it without really thinking about
it [24]. It is not a far leap to consider that a degree of automaticity might
also play a role in many traffic behaviours: people who violate speeding
limits may not necessarily do so because they are ignorant or
misinformed about the possible ramifications oftheir behaviour,or
make a conscious decision to put the pedal to the metal.
The qualification of more complex behaviours as being in some
part due to automatic rather than deliberate processes,however,is
something that many people seem uncomfortable with [23].Yet
research has shown that any number of behaviours and even traits
that are considered fairly stable can be influenced without us being
aware of it.For example,having people make sentences with words
generally associated with the elderly (e.g., bingo, begonias,and
Florida) causes them to cross a certain distance more slowly than the
people who were asked to do the same with neutral words [28].
Having them think and write about people generally thought to have
above average intelligence (college professors) makes people perform
better on intelligence tests [29].
3.1.1.Priming
The examples above describe something that is known as
“priming”: the activation of certain mental constructs or schemas by
presenting people with sensory input (e.g.,words, and images) that
are closely associated with those constructs [30].This activation in
turn influences not only thought processes but actions and behaviours
as well.Another example of this principle at work is by using visual
rather than verbal stimuli: presenting people with a bill that has the
logo of a credit card company on them, leads people to decide quicker
on giving bigger tips. The credit card logos are closely associated with
buying and spending,that it appears to lead people to do spend
money more easily than when these constructs are not activated [31].
The same principles ofpriming may be applied for road safety
campaigns, the advantage of such an approach being that it does not
require people to actively process the campaign message and
accompanying imagery.The research amassed on automatic behav-
iour and priming suggests that,when presented with certain images
and words, people may automatically exhibit the behaviour that those
stimuli invoke. This means that it is a matter of finding which images,
words or otherwise invoke the behaviours one would like to see. For
this purpose, carefully pretesting of possible stimuli may be the key.
The concept of priming is also of interest if one decides to go the
more traditional route of using a campaign to inform or convince with
information and reasoned arguments,if only to remind us that our
often well-intentioned words and images may have the opposite
effects of what we intend by them. For example, recent campaigns to
prevent people from getting driving while fatigued showed people
sleeping behind the wheel or depicted a car with a pillow where the
steering wheelshould be.These images are meant to visualize this
phenomenon and remind people that getting behind the wheel when
tired poses a real threat. It is possible, however, that such images may
unwittingly invoke the very concept one is trying to prevent,and as
such, may to a certain extent even induce the concept of sleepiness. To
our knowledge,the evaluations carried out on such campaigns [32]
have not included the measures necessary to ascertain whether such
images have any unintended side effects.Rather,such evaluations
include measures of knowledge,behavioural beliefs,intentions,risk
comprehension and self-reported behaviour rather than the associa-
tions one has with the images and words in the campaign,let alone
possible testing ofwhether people felt more or less fatigued after
viewing them.Therefore,we cannot now conclude whether or not
these images might have any such unintended side effects. However,
we use this example to illustrate that caution should be exercised in
choosing the images and words used in a campaign. As was suggested
earlier, to determine these potential side-effects,some form of
(experimental) pretesting should be in order.
3.1.2.Modelling and social influence
The above paragraph dealt with the power of words,images and
other such stimuli and the almost involuntary behaviour people may
exhibit as a result of them.In much the same way,the same can be
accomplished by presenting people with an example of the way we
would like them to act rather than showing them how they should
not. This is because we humans have a very strong tendency to mimic
the behaviours we see in others. For example, we “whisper to
someone who is whispering,we start to speak much louder when
others do so” (p. 3, [30]). Leading by the right rather than the wrong
example can therefore have the desired effect simply because we just
cannot help mimicking people.Furthermore,by demonstrating the
desired behaviour you are also playing on another important human
tendency: to conform. People are less prone to exhibit behaviour that
other people see as inappropriate.This is another good reason to
emphasize that the desired behaviour is something everybody does
and is perceived by the most people as normal and appropriate.
Many campaigns aimed at changing unwanted behaviour,how-
ever, feature people portraying the very behaviour that the campaign
is supposed to discourage. Often, this is meant to raise awareness of a
certain problem, or to hold up a “mirror” for the audience to look into.
This is based on the contention that if people see how stupid,ugly,
silly, etc., it looks when other people act a certain way, hopefully in the
future they will think twice before acting like that themselves.
However, “holding up a mirror” like that suggests that multiple people
will be able to recognize themselves in the reflection.This, then,
suggests thatthe behaviour portrayed is something thata lot of
people already do, and therefore may act to normalize the very
behaviour that it attempts to discourage.These types of campaigns
may thus be ineffectual or even inadvertently enforce the unwanted
behaviour.This once again underlines the importance not only of
providing the target audience with the right example, but to carefully
pretest these example to see ifpeople see them the way they are
meant to be seen.
3.1.3.Framing the message: losses or gains
Another way to subtly move people in the direction you want
them to take,is by using the implication of the prospect theory [33].
This influential theory describes how differently people react to
information based on whether they are framed as potential losses or
potential gains. To illustrate this difference, consider having to make a
choice of whether or not to undergo a medical procedure. If a doctor
told his patient that “out of a hundred patients, 90% will still be alivefive years after the operation”, that patient will probably be inclined to
go through with it. If, however, the doctor stated that “out of a
hundred patients,10% will die within five years after the operation”,
the patient will probably be a lot more reluctant to undergo what is
essentially the same procedure (example courtesy of [34]).
Another example of the above principle at work comes from the
research of Meyerowitz and Chaiken [35], who looked into ways to get
more women to check for changes in their breast tissues in order to
increase the likelihood of early detection of breast cancer.They did
this by spreading several versions of the same brochure on this
subject, one describing this procedure in terms of the potential gains
(“by checking your breast tissue regularly,you will be more familiar
with their texture which will make it easier to detect any eventual
changes”),potential losses (“by not checking your breast tissue
84 T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86

regularly, you will be less familiar with their texture which will make
it harder to detect any eventual changes”) or without any mention of
either the potential losses or gains. In this particular case, it turned out
that the brochure thatwas framed in terms of losses yielded the
biggest results: 57% of the women that had received this version had
started to check their breast regularly in contrast to either the gain
frame or neutral brochure that had only moved about 38% to do so.
Prospect theory may be applicable in the realm ofroad safety
campaigns as well. Depending on the type of behaviour one is trying
to encourage or discourage,it will be worthwhile to investigate
whether different frames make a difference in getting people to
behave as you would like them to. Again, the key to success would be
to take the time to pretest different ways in which to frame the
message you would like to get across before settling on any one tactic.
3.2.Informal education
One last thing to take into account is that the approaches
illustrated above can be put to good use beyond trying to influence
the behaviour of the target audience themselves. In order to maximize
their potential impact, one might look beyond the principal targets of
the message.Indeed,we do not just learn traffic behaviours from
formal education and campaigning.Rather,a lot of what we learn
comes from our own experiences and what we observe in others [1].
Therefore,it might be worthwhile to look not just at how we might
influence the initial recipients of a campaign,but also at how these
recipients might further influence others around them.
Of course, looking beyond the principal target of a campaign
should not be to the detriment of the main objective: it is important to
keep one's eye on the prize.However,sometimes it might even be
expedient to look into such possibilities. For example, if young males
seem elusive in their susceptibility to road safety campaigns, why not
aim for their peers or other related groups who hold some sway over
them, such as older,more experienced drivers who are less prone to
risk taking and who might be convinced to lead by example.Or
perhaps by persuading young women that it would be in their
own best interest to withhold their admiration from the vehicular
antics young males sometimes display when trying to impress the
girls. Admittedly,whether these examples prove effective is conjec-
ture at this point,although the former example has been implemen-
ted in another form in a number of countries,namely in the form
of accompanied driving.In some countries novice drivers are only
allowed to drive when they are accompanied by an experienced
driver. Depending on the amount ofkilometres driven,the super-
visor's skills and the variation in routes, this approach has been
known to garner some measure of success [36].This does in no way
provide any indication, however, that a road safety campaign
encouraging older, more experienced drivers to share that experience
and mentor younger drivers would have any effect at all. Yet again we
will stipulate that it would take some form of pretesting to see if it
is possible to obtain such indirect results by means of a road safety
campaign.
Another example of this approach pertains to parents and children,
who are uniquely positioned to mutually influence each other's
behaviour.Indeed,this very fact proved very usefulin a campaign
aimed at encouraging seat belt use among kids, using toys that can be
attached to the seat belt.Not only did seat belt use among children
increase as a result of the campaign, but the children also encouraged
their parents to wear their seat belt [37]. In their turn, parents
influence their children because those children are young and
impressionable and look to their parents to see how to behave.In a
recent report on this last approach, many of the methods mentioned
in this article are outlined in relation to how they might be of use for
encouraging parents to take a more active role in their children's
traffic participation ([38]; report in Dutch, with English summary).
4. Conclusion
Hopefully, with this article we accomplished what we set out to do,
and that is to provide some inspiration for the future of road safety
campaigns, whether this is ultimately achieved by having a good hard
look at methods that are already widely implemented or by turning to
new and alternative ways to influence road user behaviour.For this
we should pay special attention to other disciplines studying human
behaviour such as social psychology and economics. Here, interesting
insights into human behaviour and behavioural modification can be
found which may prove to be of use within the practice of road safety
campaigns.One of the major current challenges in terms of research
into influencing road user behaviour is to figure out whether the
implication of these and other insights into human behaviour in
general can be adapted to influence road user behaviour in particular,
more specifically when designing road safety campaigns. Most
importantly,and we cannot stress this enough,is that any and all
forays into setting up a road safety campaign are formally and
thoroughly evaluated. Because as they say, the proof is in the pudding,
and the careful pretesting and evaluating of these and other
approaches is the only way to determine whether the pudding is
worth having.
References
[1] F.C.M. Wegman, L.T. Aarts (Eds.), Advancing Sustainable Safety: National Road Safe
Outlook for 2005–2020,SWOV Institute for Road Safety Research,Leidschendam,
2006.
[2] P. Delhomme,W. De Dobbeleer, S. Forward, A. Simões (Eds.), Manual for
Designing,Implementing,and Evaluating Road Safety Communication Cam-
paigns,Directorate-Generalfor Energy and Transport,European Commission,
Brussels,2009.
[3] A. Boulanger, M. Divjak, I. Orozova-Bekkevold, V. Zabukovec, Typology of Evaluation
Methods: Current Practices and Campaign Evaluation; Deliverable 2.1 of Campaigns
and Awareness Raising Strategies in Traffic Safety (CAST),Directorate-General for
Transport and Energy, European Commission, Brussels, 2007.
[4] M. Peden,R. Scurfield,D. Sleet,D. Mohan, A.A. Hyder, E. Jarawan,C. Mathers
(Eds.),World Report on Road Traffic Injury Prevention,World Health Organiza-
tion, Geneva, 2004.
[5] R. Elvik, T. Vaa, A. Hoye, A. Erke, M. Sorensen (Eds.), The Handbook of Road Safety
Measures,2nd revised edition,Elsevier,Amsterdam,2009.
[6] T. Vaa, T. Assum,P. Ulleberg,K. Veisten,Effects ofInformation Campaigns on
Behaviour and Road Accidents:Conditions,Evaluation and Cost Effectiveness,
Institute of Transport Economics,Oslo,Norway,2004.
[7] A. Boulanger(Ed.), Evaluation Tool for Road Safety Campaigns,Directorate-
General for Energy and Transport,European Commission,Brussels,2009.
[8] A. Boulanger,A. Meng (Eds.),Reporting Toolfor Effects ofa Single Campaign,
Directorate-Generalfor Energy and Transport,European Commission,Brussels,
2009.
[9] T. Vaa, R. Philips, Effects of Road Safety Campaigns; Deliverable 1.3 of Campaigns
and Awareness Raising Strategies in Traffic Safety (CAST), Directorate-General for
Transport and Energy,European Commission,Brussels,2009.
[10] SUPREME,Best Practices in Road Safety.Handbook for Measures at the Country
Level,European Commission,Brussels,2007.
[11] B. Elliott, Road safety campaigns: a meta-analysis,Report CR118,Department of
Transport and Communications,Federal Office of Road Safety,1993.
[12] R. Tay,B. Watson,Changing drivers'intentions and behaviours using fear-based
driver fatigue advertisements,Health Marketing Quarterly 19 (4) (2002) 55–68.
[13] R. Tay, The effectiveness ofenforcementand publicity campaigns on serious
crashes involving young male drivers: are drink driving and speeding similar?
Accident Analysis & Prevention 37 (5) (2005) 922–929.
[14] R. Tay, B. Watson, S. Hart, Personal and social influences of speeding, Proceedings
of the Third International Conference on Traffic and Transportation Studies,
American Society of Civil Engineers,Reston,2002.
[15] W.H. Weenig, C.J.H. Midden,Mass media information campaigns and knowledge
gap effects,Journal of Applied Social Psychology 27 (1997) 945–958.
[16] K. Witte, M. Allen, A meta-analysis offear appeals:implications for effective
public health campaigns,Health Education & Behavior 27 (5) (2000) 608–632.
[17] R.A.C. Ruiter, C. Abraham, G. Kok, Scary warnings and rational precautions: a review
of the psychology of fear appeals, Psychology and Health 16 (2001) 613–630.
[18] SWOV, Fear-based information campaigns,SWOV-factsheet,April 2009,SWOV
Institute for Road Safety Research,Leidschendam, 2009.
[19] C. Goldenbeld,D. Twisk, S. Houwing, Effects of persuasive communication and
group discussions on acceptability of anti-speeding policies for male and female
drivers, Transportation Research Part F; Traffic Psychology and Behaviour, vol. 11,
nr. 3, 2008,pp. 207–220.
[20] I. Lewis, B. Watson, R. Tay, Examining the effectiveness of physical threats in road
safety advertising: the role of the third-person effect, gender, and age,
Transportation Research Part F.2007 /01.10(1), 2007,pp. 48–60.
85T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86

[21] D. Kahneman, New challenges to the rationality assumption, Journal of
Institutional and Theoretical Economics 150 (1994) 18–36.
[22] A. Tversky, D. Kahneman, Rational choice and the framing of decisions, Journal of
Business 59 (1986) S251–S278.
[23] J.A. Bargh,The Automaticity of Everyday Life,in: R.S.Wyer Jr. (Ed.), Advances in
Social Cognition,Vol. 10,Erlbaum,Mahwah,NJ, 1997,pp. 1–61.
[24] J.A. Bargh, E. Morsella, The unconscious mind,Perspectives on Psychological
Science 3 (2008) 73–79.
[25] J.A. Bargh,The four horsemen of automaticity: awareness,efficiency,intention,
and controlin social cognition,in: R.S.Wyer Jr., T.K. Srull (Eds.), Handbook of
Social Cognition,2nd ed.,Erlbaum,Hillsdale,NJ, 1994,pp. 1–40.
[26] G.R.Maio, B. Verplanken,A.S.R.Manstead,W. Stroebe,C.S.Abraham,P. Sheeran,
M. Conner,Social psychological factors in lifestyle change and their relevance to
social policy,Social Issues and Policy Review 1 (2007) 99–138.
[27] B. Verplanken,W. Wood, Interventions to break and create consumer habits,
Journal of Public Policy & Marketing 25 (2006) 90–103.
[28] J.A. Bargh,M. Chen,L. Burrows,Automaticity of social behavior: direct effects of
trait construct and stereotype activation on action,Journal of Personality and
Social Psychology 71 (2) (1997) 230–244.
[29] A. Dijksterhuis, A. Van Knippenberg, The Perception–Behavior Expressway: Automatic
Effects of SocialPerception on SocialBehaviour,in: M.P. Zanna (Ed.),Advances in
experimental social psychology, 33, Academic Press, San Diego, 1998, pp. 1–40.
[30] A. Dijksterhuis,J.A. Bargh, The Perception–BehaviorExpressway: Automatic
Effects of Social Perception on Social Behavior,in: M.P. Zanna (Ed.),Advances in
Experimental Social Psychology,33,Academic,San Diego, 2001,pp. 1–40.
[31] R.A. Feinberg, Credit cards as spending facilitating stimuli: a conditioning
interpretation,Journal of Consumer Research 13 (1986) 348–356.
[32] W. De Dobbeleer, T. Nathanail, G. Adamos, Pan-European Road Safety Campaign:
Campaign Evaluation;Deliverable 5.3 of Campaignsand Awareness Raising
Strategies in Traffic Safety (CAST), Directorate-General for Transport and Energy,
European Commission,Brussels,2009.
[33] A. Tversky, D. Kahneman, The Framing of Decisions and the Psychology of Choice,
Science,vol. 185,1981,pp. 453–458.
[34] R.H. Thaler, C.R. Sunstein, Nudge; Improving Decisions About Health, Wealth and
Happiness,Expanded editionPenguin Group,New York,2009.
[35] B.E. Meyerowitz, S.E. Chaiken,The effect of message framing on breast-self
examination: attitudes: attitudes, intentions, and behavior, Journal of Personality
and Social Psychology,52,3, 1987,pp. 500–510.
[36] SWOV, Accompanied driving, SWOV-Factsheet, January 2009, SWOV Institute for
Road Safety Research,Leidschendam,2009.
[37] P. Silverans,P. de Neve,SUPREME; Thematic Report Education and Campaigns,
Directorate-Generalfor Transport and Energy,European Commission,Brussels,
2007.
[38] A.T.G.Hoekstra,D. Twisk, Informele educatie SWOV Institute for Road Safety
Research,R-2010-19 Leidschendam,2010.
[39] R. Tay, L. Ozanne,Who are we scaring with high fear road safety advertising
campaigns? Asia Pacific Journal of Transport 4 (Summer 2002) 1–12.
86 T. Hoekstra,F. Wegman / IATSS Research 34 (2011) 80–86
View publication statsView publication stats