Accident Analysis and Prevention 43 (2011) 1341–1347
Contents lists available at ScienceDirect
Accident Analysis and Prevention
journal homepage: www.elsevier.com/locate/aap
Understanding risk compensation in children: Experience with the
activity and level of sensation seeking play a role
Jennifer Lasenby-Lessard, Barbara A. Morrongiello ∗
Department of Psychology, University of Guelph, MacKinnon Building, Guelph, Ontario, Canada N1G 2W1
a r t i c l e
i n f o
Article history:
Received 9 August 2010
Received in revised form
30 November 2010
Accepted 5 February 2011
Keywords:
Risk taking
Risk compensation
Children
Sensation seeking
Sports activities
a b s t r a c t
The aims of this study were to determine if children 7–12 years show risk compensation when engaging in ecologically valid recreational sports tasks, and to explore how experience with the activity and
extent of sensation seeking influence this. Children were positioned up on a platform, on a bike or wearing rollerblades, and they were presented varying heights and inclines from which they selected the
greatest one they go down when wearing and not wearing safety gear appropriate to the activity; when
making their ratings they anticipated actually doing the task. Results revealed that children engaged in
significantly more risk taking when wearing safety gear, thereby demonstrating risk compensation, and
this was significantly greater for the activity with which they had greater experience. However, children
high in sensation seeking demonstrated significantly more risk compensation in both the high and low
experience activities, although the injury risk appraisals that predicted risk compensation varied with
experience level. Implications for the design of injury prevention programs and directions for future
research are discussed.
© 2011 Elsevier Ltd. All rights reserved.
1. Influences on children’s risk taking
Unintentional injury in childhood has been cited as one of the
leading causes of death and disability for children in most developed nations (Canadian Institute of Child Health [CICH], 2000;
National Center for Injury Prevention and Control [NCIPC], 2010;
World Health Organization [WHO], 2005). For school-age children,
many injuries occur when they are away from home and making their own decisions about injury-risk activities (Shanon et al.,
1992). Research has shown that children engage in behaviors that
put them at risk of injury for a variety of reasons (Morrongiello
and Lasenby-Lessard, 2007). Some of the key determinants of risk
taking are extent of experience with an activity, injury-related cognitions, anticipated emotions in reaction to risk taking, gender, and
children’s behavioral attributes.
Greater risk taking is more likely among children who: have
more experience with an activity (Lasenby-Lessard et al., in press;
Morrongiello and Dawber, 2004); think there is little danger, that
they are not vulnerable to injury, and that they would not be seriously injured (Morrongiello, 1997; Morrongiello and Rennie, 1998);
and feel less fear and more excitement when engaging in risky
activities (Morrongiello and Matheis, 2004); and boys (Coppens
and Gentry, 1991; Ginsburg and Miller, 1982; Morrongiello and
∗ Corresponding author.
E-mail address: [email protected] (B.A. Morrongiello).
0001-4575/$ – see front matter © 2011 Elsevier Ltd. All rights reserved.
doi:10.1016/j.aap.2011.02.006
Rennie, 1998; Rosen and Peterson, 1990). A variety of personality
and behavioral characteristics also influence risk taking decisions
(see Schwebel and Gaines, 2007, for extensive review). In particular,
sensation seeking (i.e., “the need for varied, novel and complex situations and experiences and the willingness to take physical and
social risks for the sake of such experiences”, Zuckerman, 1979,
p. 10) strongly relates to risk taking behaviors in adolescence and
adulthood (Bouter et al., 1988; Brown et al., 1974; Heyman and
Rose, 1980; Hymbaugh and Garrett, 1974; Rowland et al., 1986;
Straub, 1982; Zuckerman, 1979, 2007), as well as in childhood
(Morrongiello and Lasenby, 2006; Morrongiello et al., 2004a,b;
Morrongiello and Sedore, 2005). In a recent study examining the
impact of accumulated experience with an activity on risk taking decisions, for example, children high in sensation seeking were
especially likely to react with increased risk taking once they accumulated some experience with an activity (Lasenby-Lessard et al.,
in press). Thus, high sensation seekers emerged as a unique high
risk group who needed very little prompting to increase their level
of risk taking, presumably because doing so increases their level
of arousal (Zuckerman, 1979) and satisfies a need for novel and
intense stimulation (Zuckerman, 2007).
2. Risk compensation
One influence on risk taking that has recruited much attention
and about which there is some controversy is the use of safety
gear. Unquestionably safety gear reduces children’s risk of serious
1342
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
injury. However, research has shown too that individuals sometimes react with increased risk taking when they wear safety gear,
a phenomenon commonly known as risk compensation. While risk
compensation theorists agree that safety initiatives (e.g., safety
gear, environmental modifications to roads or products) produce
positive effects by reducing injury risk, they argue that the full benefit of such effects will not be realized in individuals who show risk
compensation. Although there is debate about the nature and scope
of this phenomenon (Adams and Hillman, 2001; Hedlund, 2000;
Thompson et al., 2001), and not all research has found support for
this phenomenon (see Thompson et al.,2001, for review), risk compensation has been observed in athletes (Biasca et al., 2002; Braun
and Fouts, 1998), adults when driving (Potvin et al., 1988; Simonet
and Wilde, 1997; Stanton and Pinto, 2000), parents of young children (Morrongiello and Major, 2002; Viscusi, 1984, 1985; Viscusi
and Cavallo, 1996), and even in school-age children (DiLillo and
Tremblay, 2001; Morrongiello et al., 2007a,b).
In organized sports, for example, wearing safety gear has been
associated with increased aggressive and injurious behavior by
hockey players, as well as resistance to calling penalties by coaches
and referees (Biasca et al., 2002). Adult drivers have been shown
to behave more recklessly (e.g., speeding) when wearing seat belts
(Stanton and Pinto, 2000). Product safety interventions directed
towards children also have been found to lead to risk compensation.
For example, with the invention of childproof caps on medication
there was an increase of 3500 cases of poisoning in children under
the age of five, presumably because parents were less vigilant about
the handling and storage of these ‘safer’ bottles (Viscusi, 1984).
The introduction of safety mechanisms on cigarette lighters also
has been linked to reduced parental caution about these products
(Viscusi and Cavallo, 1996). Similarly, parents have been shown
to react with increased tolerance for risk taking when their children are wearing safety gear during recreational sports and play
activities (Morrongiello and Major, 2002). Even during childhood
wearing safety gear has been shown to result in more reckless risk
taking behavior (DiLillo and Tremblay, 2001; Morrongiello et al.,
2007). Thus, despite debate about how robust and generalized risk
compensation is, research demonstrates unequivocally that this
phenomenon occurs across a broad age range and across numerous
activities.
3. Current study
Building on past research, the current study examined if the
extent of risk compensation shown by school-age children varies
as a function of the extent of experience with an activity, and what
role individual-difference factors (sensation seeking, injury-related
appraisals, emotional reactions) play in this process. Children participated in both high experience (bicycling) and low experience
(rollerblading) activities, and to assess for risk compensation a
within-participant design was used (cf. Streff and Geller, 1988) in
which each child’s risk taking when not wearing safety gear was
compared to their risk taking when wearing gear. Bicycling and
rollerblading were used in this study because both carry a potential for serious injury risk, children are aware of this risk (Adams
et al., 1996; Pudpud and Linares, 1997), children in this region
are more likely to bike (high experience) than rollerblade (low
experience), and both activities have designated safety gear that is
encouraged or mandated for use during these activities. To assess
sensation seeking, children completed a standardized questionnaire (Morrongiello and Lasenby, 2006). Appraisals of injury-risk
cognitions (danger, vulnerability, potential injury severity) and
emotions (excitement, fear) in response to risk taking also were
measured to explore if these provide insight into the mechanisms
by which risk compensation operates.
4. Methods
4.1. Participants
A sample of 99 children 7–12 years of age was divided into four
groups: young girls (N = 27, range: 7–9 years , M = 8.76, SD = 0.93),
young boys (N = 23, range: 7–9 years , M = 8.97, SD = 0.64), older girls
(N = 25, range: 10–12 years , M = 11.47, SD = 0.85), and older boys
(N = 24, range: 10–12 years , M = 11.673, SD = 0.81). Participants
were randomly selected from the Child Development Research
Unit Database at the University of Guelph; a database comprised
of over 13,000 families who have previously indicated an interest
in participating in research on child development. Mothers were
predominantly Caucasian (98%) and generally well educated, with
79% having completed a university or college degree, 16% having completed some university or college courses, and 4% having
completed high school. Family income fell within the mid- to highsocioeconomic status range, with 63% earning $80,000 and above,
25% earning between $60,000 and $80,000, 12% earning between
$40,000 and $60,000, and 1% earning under $40,000 per year. Children were English speaking, normally developing, and no child had
experienced an injury resulting in hospitalization. Children were
pre-screened over the phone and were asked to rate their level of
experience bicycling and rollerblading on a scale of 1–5. Only children with high experience bicycling (ratings of 4 = do it regularly
and often when it is nice weather or 5 = do it very often when it
is nice weather) and low experience with rollerblading (ratings of
1 = have never done it or 2 = have done it a little but not more than
3×) were selected for participation in the current study; participation rate = 74%. The Institutional Review Board at the University of
Guelph provided approval for this research and parents and children granted consent prior to participating.
4.2. Measures
Risk taking tasks: Each child participated in a task that was
designed to allow for risk compensation to occur. Specifically, children were randomly assigned to participate in either an incline
(speed) or a height task, with the constraint that the groups be
balanced by age and sex; two different types of tasks (judgements
based on speed and height) were used to ensure variation in risk
taking and because pilot tests revealed both were interesting for
children at these ages. For both of these tasks an apparatus was custom built that would allow the children to choose varying heights
and inclines according to the level of risk they were willing to take
with and without safety gear on.
The platform on which the child stood was on the top of the
apparatus and was three feet wide and six feet long; this size
accommodated a bicycle on the platform. Wood boxes were built to
go under the platform and to act as steps for the child to go up and
down so s/he could survey the different height conditions (i.e., 6, 12,
18, 24, 30, 36, and 42 in.) or incline conditions before making his/her
risk taking selection (i.e., what is the greatest height or steepest
incline condition the child would bike or rollerblade down). For the
incline condition, three ramps were designed for each of the seven
step heights (6–42 in.) and these were constructed such that at each
height there was a ramp that would lay at a 20 degree angle, a 32.5
degree angle and a 45 degree angle; these angles correspond to
those that are rated as novice, intermediate, and advanced angles,
respectively, for slope (speed) based recreational activities such as
rollerblading and skiing (Beaudry, 1994). The selection of higher
step heights and greater incline angles, therefore, indicated greater
risk taking.
Ratings: Once children made their risk taking decision (height
or incline), they positioned themselves at that location, sitting on
the bike or standing and wearing rollerblades, and were asked to
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
use a 5 point Likert scale (1 = not at all, 2 = slightly, 3 = somewhat,
4 = fairly, 5 = very) to rate their cognitions and emotions; children
completed separate ratings for each of the four test conditions
(bike, rollerblade × gear on, off), with ratings randomized each
time. Specifically, children were asked to rate their perceptions of:
danger (e.g., How dangerous do you think it is for you to go down the
ramp?), vulnerability (e.g., How likely do you think it is that you could
fall doing this?), and injury severity (e.g., How hurt do you think you
could get if you were to fall?), excitement (e.g., How excited do you
feel now that you are getting ready to go down the ramp?) and fear
(e.g., How scared or nervous do you feel now that you are getting ready
to go down the ramp?).
Questionnaires: Each child completed the Sensation Seeking
Scale for Children-Child version (SSSC-C, Morrongiello and Lasenby,
2006), a measure that assesses sensation seeking in children. The
scale consists of 27 items in which children are presented opposing statements and asked to choose the one from each pair that
best describes them (i.e., I would like to jump or dive off a diving
board versus I would not like the feeling you get when standing on a
diving board). Items are scored such that higher numbers indicate
greater sensation-seeking (range: 0–27); internal reliability was .86
for this sample. The parent also completed a Demographic Information Sheet that provided information about parental education and
family income.
4.3. Procedure
Regardless of whether the child was assigned to the height or
incline task, the length of the session was 45 min.
Height risk taking task: Children performed the height task four
times (bike, rollerblade × gear off, gear on), with task order randomized with the four groups of participants (2 ages × 2 sexes). Children
walked up each of the seven steps and looked down at a 20 degree
angle ramp attached to the front. Children were told they were
going to ride a bike (or rollerblade) down the ramp and to choose
the highest step height at which they would feel comfortable doing
so. The children were then allowed to walk up and down the steps
in order to choose the height at which they felt comfortable biking (or rollerblading) down on the 20 degree ramp. For the safety
gear condition, they were wearing the necessary safety gear for
either biking (helmet) or rollerblading (helmet, wrist guards, elbow
and knee pads). In order to elicit a more valid response, children
were told that they would actually be performing these activities
after they had made their decisions about how high they wanted
the platform. After the children had chosen the height at which
they felt comfortable biking (or rollerblading), the 20 degree ramp
was set to that height and the children were seated on an actual
bike or wore rollerblades up on the platform, and they then rated
their cognitions and emotions (as outlined above under Ratings),
either with or without safety gear on, depending on condition. Thus,
when children made their risk selection for height they expected
they would perform the activity (rollerblade or bike down the
ramp) at that height, and when they gave their ratings, they were
fully equipped depending on condition (rollerblades on, seated
on bike × safety gear on, off) and positioned at the height they
selected.
Incline risk taking task: Before beginning the incline task each
child’s height was measured and 50% of their overall height was
calculated and recorded; this ensured that each child’s experience
of height was the same when making their risk taking decisions
about the steepest incline they would go down. Each child performed the incline task (i.e., selecting the steepest ramp they would
go down) at this height several times (bike, rollerblade × gear
off, on), with order randomized. Once children had chosen the
ramp they wanted, they would get on top of the bike (or stand
with rollerblades on) and rated their cognitions and emotions
1343
Table 1
Risk taking scores (range: 1–7) as a function of gear condition (safety gear on or
off) and level of experience with the activity (high = bike; low = rollerblade). Overall
(N = 99) risk taking and risk compensation scores (RC; risk taking with gear on minus
risk taking with gear off) are also indicated.
Experience
High
Low
Overall RT
Gear
Overall
Off
On
RC
3.66 (1.50)
2.77 (1.18)
3.21
4.34 (1.68)
3.33 (1.32)
3.84
.68
.46
(see above), either with or without safety gear on, depending on
condition.
Debriefing: Designing real-life activities that are safe and ethical in order to measure risk taking and risk compensation is very
difficult. Because these tasks are potentially dangerous, particularly when not wearing safety gear, children were not required
to actually perform these activities. After the children had made
their decisions about height or incline for all conditions they were
debriefed as to why they were told they would have to do the
activities but were not required to do so.
5. Results
Although participants were randomly assigned to the height and
incline condition, preliminary analyses were conducted to assess
for age and sex differences in risk taking across conditions. Results
revealed comparability in risk taking scores across conditions for
gear and no-gear scores (p > .05 for all effects). Because results
revealed no significant differences for test condition, the effects
of primary interest were based on within-participant and withincondition comparisons, and there was no interest in examining
results as a function of risk taking condition per se, we collapsed
across incline and height conditions in all analyses reported herein.
For risk taking, scores could range between 1 and 7, with higher
scores indicating greater risk taking.
Did children show risk compensation and did it vary as a function
of extent of experience with the activity?
An Age group (2) × Sex (2) × Experience (2: high, low) × Safety
Gear (2: on, off): Analysis of Variance (ANOVA) was conducted to
assess for risk compensation (i.e., higher risk taking scores for gear
on than gear off trials), with experience and safety gear as withinparticipant factors and age and sex as between-participant factors.
Results revealed main effects of sex, experience, and safety gear,
along with an experience × safety gear interaction. Specifically,
boys engaged in more risk taking than girls overall (M = 3.90 and
3.21, SD = 1.49 and 1.26, respectively), F(1, 95) = 8.83, p < .01, 2p =
.29. Children engaged in significantly more risk taking for the high
than the low experience activity overall (M = 4.03 and 3.05, SD = 1.59
and 1.25, respectively), F(1, 95) = 68.33, p < .01, 2p = .42. Children
engaged in more risk taking when wearing safety gear (M = 3.84,
SD = 1.50) than when not wearing safety gear (M = 3.11, SD = 1.34),
confirming that they showed risk compensation, F(1, 95) = 56.73,
p < .01, 2p = .37. An interaction between experience and gear also
emerged, F(1, 95) = 4.86, p < .05, 2p = .25, which reflected the fact
that the extent of risk compensation (i.e., difference in risk taking
scores across gear conditions) was greater for the high (bicycling)
than low (rollerblading) experience activities (see Table 1).
What individual-difference factors relate to risk compensation and
do these vary for high and low experience activities?
To address this question we computed a single index of risk
compensation for each participant for the high and low experience
activities separately (cf. Morrongiello and Major, 2002) and these
scores were then related to other variables of interest. Specifically,
each ‘risk compensation’ score was determined by subtracting each
participant’s risk taking ‘gear off’ score from their risk taking ‘gear
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
Table 2
Correlations showing variables that related to extent of risk compensation shown
for the high and low experience activity for the overall sample (N = 99).
Variable
Age
Sex
Change in danger
Change in vulnerability
Change in severity
Change in excitement
Change in fear
SS
*
**
Experience
High
Low
.07
−.02
−.19*
−.06
−.04
.10
−.09
.28*
−.09
−.14
−.05
−.19* .
−.29**
.07
−.16
.27*
p < .05.
p < .01.
on’ score so that higher scores reflected greater risk compensation
(i.e., increased risk taking when wearing gear). Similar difference
scores were computed for ratings of danger, vulnerability, severity, and fear; for the excitement ratings the reverse was done.
This mathematical approach ensured that all ratings resulted in
a positive score and reflected the extent of change in rating that
corresponded to the extent of risk compensation (i.e., extent of
increase in risk taking when wearing safety gear). The risk compensation scores for high and low experience activities were then
each correlated with sensation seeking, age, sex, and the change
in ratings for danger, perceived vulnerability for injury, perceived
injury severity, excitement, and fear.
As can be seen in Table 2, sensation seeking correlated with risk
compensation for both high and low experience activities. Regardless of their level of experience with the activity, children who
scored higher in sensation seeking showed greater risk compensation. For the high experience activity, extent of decrease in rating of
danger also was associated with risk compensation. In contrast, for
the low experience activity, risk compensation related to decreases
in ratings of injury vulnerability and potential injury severity in
reaction to wearing safety gear. Specifically, children who showed
greater decreases in ratings of vulnerability and injury severity in
reaction to wearing safety gear, showed greater risk compensation for the low experience activity. Interestingly, there was only
a modest relation found between risk compensation scores for the
high and low experience activities, [r(96) = .49, p < .05], which suggests that it is not necessarily the same children across high and
low experience activities who showed risk compensation. Examining the correspondence in these risk compensation scores as
a function of extent of sensation seeking, however, revealed an
interesting difference: for children high in sensation seeking (i.e.,
top 1/3 of sensation seeking scores) there was a much higher
degree of correspondence [r(32) = .69, p < .01] compared with those
low in sensation seeking (bottom 1/3 of sensation seeking scores;
r(32) = .49, p < .01). Thus, high sensation seekers represented a
much more ‘high risk group’ in that they were more likely to show
risk compensation across high and low experience activities than
were children scoring lower in sensation seeking.
To explore the mechanisms by which sensation seeking influences risk compensation and determine if this attribute moderates
the impact of other variables on risk compensation, a regression
was conducted to predict risk compensation, separately for the
low and high experience activities, from interaction terms based
on crossing sensation seeking with changes in risk appraisals (danger, vulnerability, severity) and rating of emotions (excitement,
fear) in response to putting on safety gear when risk taking; main
effect variables were centered prior to computing interaction terms
and an examination of inter-correlations revealed no evidence of
multi-collinearity concerns (Tabachnick and Fidell, 1989). Results
revealed that although sensation seeking was positively associated
Predicting RC for the High Experience Activity: Sensation Seeking x Danger
2.00
Extent of risk compensation
1344
Sensation
Seeking
High
1.00
Medium
Low
0.00
Low
Medium
High
Change in danger rating
Fig. 1. Predicting risk compensation for the high experience activity: examining the
change in danger rating × sensation seeking interaction.
with risk compensation for both activities, it impacted risk compensation via different mechanisms for the high and low experience
activities.
For the high experience activity, risk compensation was predicted
by an interaction of sensation seeking × change in danger rating
[F(5, 95) = 2.10, p < .05; t = 2.51, p < .05; B = .19, SE = .07], and this
accounted for 22% of the variance in risk compensation. Follow-up
simple slopes analyses (cf. Aiken and West, 1991) were conducted
to test for sensation seeking as a moderator and to determine if
change in danger rating was a significant predictor of risk compensation at each of three levels of sensation seeking (low = 1
SD below mean; mid = mean; high = 1 SD above mean). Results
indicated that sensation seeking moderated the relation between
change in danger rating and risk compensation. Specifically, sensation seeking had a differential impact on risk compensation when
children reacted with large decreases in danger ratings in reaction to wearing safety gear: then high levels of sensation seeking
predicted significantly greater risk compensation, F(3, 95) = 3.38,
p < .05; t = 2.50, p < .01; B = .54, SE = .21. As the data in Fig. 1 illustrates, when engaged in an activity with which they have a
high level of experience, children react to wearing safety gear by
decreasing their rating of danger, and for those who react with a
large decrease in this rating, being high in sensation seeking further
exaggerates injury risk and predicts significantly greater levels of
risk compensation than that shown by those not high in sensation
seeking.
For the low experience activity, the model also significantly predicted risk compensation [F(5, 95) = 2.65, p < .01], accounting for
26% of the variance. Significant predictors included changes in
rating of severity (t = 3.02, p < .01; B = .29, SE = .10), sensation seeking (t = 2.38, p < .05; B = .10, SE = .04) and a marginally significant
interaction of sensation seeking × severity (t = 1.94, p = .055; B = .09,
SE = .02). Follow-up simple slopes analyses (cf. Aiken and West,
1991) were conducted to test for sensation seeking as a moderator and to determine if change in severity rating was a significant
predictor of risk compensation at each of three levels of sensation seeking (low = 1 SD below mean; mid = mean; high = 1 SD
above mean). Results revealed, as shown in Fig. 2, that sensation
seeking had no differential impact on extent of risk compensation when children made small decreases in severity ratings in
response to wearing safety gear. However, those high in sensation seeking engaged in significantly greater risk compensation
when medium and large decreases in severity ratings were made
in reaction to wearing safety gear, F(3, 95) = 7.21, p < .01; t = 3.77
and 3.18, p < .01; B = .44 and .26, SE = .12 and .08, respectively. Thus,
when children were engaged in an activity with which they had
limited experience, risk compensation was linked to judgements
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
Predicting RC for the Low Experience Activity: Sensation Seeking x Severity
Extent of risk compensation
2.00
Sensation
Seeking
High
1.00
Med
Low
0.00
Low
Medium
High
Change in injury severity rating
Fig. 2. Predicting risk compensation for the low experience activity: examining the
change in severity rating × sensation seeking interaction.
about potential injury severity. Children high in sensation seeking
were easily prompted to show risk compensation when they judged
there were moderate and high declines in potential injury severity
due to wearing safety gear.
6. Discussion
Using an ecologically valid task that tapped normative recreational activities (bicycling, rollerblading) the present findings add
to the accumulating literature that children show risk compensation when wearing safety gear. The extent of risk compensation,
however, was found to vary based on children’s level of experience with the task and their level of sensation seeking. Previous
research has demonstrated that children engage in greater risk
taking in tasks with which they have more experience (LasenbyLessard et al., in press; Morrongiello et al., 2004a,b; Morrongiello
and Sedore, 2005). However, the findings that children engage in
greater risk compensation when engaged in high than low experience activities, and that sensation seeking moderates this relation,
are unique to this study; each of these findings will be discussed in
turn.
Past research has demonstrated that individuals engage in
greater risk taking for activities with which they have greater
experience. Presumably, as individuals accumulate non-injurious
experience with an activity they perceive this activity as more
familiar and risk appraisals change to reflect a belief that the activity
is less risky than thought initially (Zuckerman, 2007). The fact that
risk compensation is greater for high than low experience activities, however, suggests that not only do children show greater risk
taking for high vs low experience activities but they also respond
with significantly greater levels of risk taking when wearing safety
gear during high than low experience activities. Hence, wearing
safety gear has a much more deleterious impact on risk taking
as children become increasingly familiar with an activity. At the
same time, therefore, that parents are likely to be becoming more
lax in monitoring because they assume children’s familiarity with
the activity indicates expertise (Morrongiello and Major, 2002),
children’s behavior is likely to be becoming increasingly risky, particularly if they are wearing safety gear. Furthermore, the child’s
level of sensation seeking, moderates the extent of impact of wearing safety gear on risk compensation, with the most significant
impact of injury risk being evident for children high in sensation
seeking.
Lasenby-Lessard et al. (in press) found that high sensation
seekers are particularly likely to engage in risk taking when the
contextual demands or their experiences support them doing so.
The current findings support this notion: independent of their level
1345
of experience with the activity, when high sensation seekers were
wearing safety gear they engaged in significantly greater risk taking
than when not wearing safety gear. Zuckerman (2007) has noted
similar findings among high sensation seeking adults, namely –
that they assume risks are lower, and respond accordingly, even for
activities with which they have limited experience. Thus, individuals high in sensation seeking appear to form a unique ‘high risk’
group who will take risks and engage in risk compensation even
for activities with which they have limited experience. Zuckerman
(1979, 2007) argues that individuals who are high in sensation
seeking seek out intense activities in order to increase their level
of arousal. It may be, therefore, that repeated experience with an
activity (Lasenby-Lessard et al., in press) or wearing safety gear (as
in the current study) results in a reduced level of arousal to which
sensation seekers then react by increasing their level of risk taking
and, consequently, level of arousal.
Past research has shown that children often consider personal
vulnerability for injury, potential injury severity and extent of danger when making decisions about risk activities (Morrongiello,
1997; Morrongiello and Rennie, 1998). The present findings
demonstrate, however, that children actually considered different
injury appraisals when deciding on activities with which they have
high vs low experience. For high experience activities, it appears
that children considered how well they could do the activity based
on their past experiences, and focused on how dangerous they
thought it was for them. In contrast, for the activity with which
they had limited experience, they focused on if they could get
injured (vulnerability) and how hurt they could be if they were
injured (severity); having limited experiences by which to judge
how dangerous the activity was for them, they instead focused on
the potential for injury to occur and the severity of injury that could
occur if they were not successful in doing the activity. Interestingly,
sensation seeking moderated relations between risk appraisals and
level of risk compensation. Specifically, for high experience activities children in the low and middle sensation seeking groups
showed lower levels of risk compensation regardless of how much
they decreased their danger rating in response to wearing safety
gear, whereas children high in sensation seeking showed increasing levels of risk compensation the greater their decline in danger
ratings. Thus, children high in sensation seeking were more easily provoked to respond with greater risk compensation for high
experience activities as their danger ratings declined in reaction to
wearing safety gear. Similarly, for the low experience activity, low
sensation seekers showed reduced risk compensation regardless of
how much they decreased their appraisal of injury severity when
wearing safety gear. However, children who fell in the mid and high
sensation seeking groups showed increasing risk compensation as
their appraisal of injury severity systematically decreased. Overall,
then, the findings indicate that for both high and low experience
activities, scoring higher in sensation seeking predicts greater risk
compensation corresponding to increasing changes in extent of risk
appraisal (i.e., danger appraisal for high experience activity; injury
severity for low experience activity).
6.1. Implications
The current findings are important for both theoretical and
practical reasons. On a theoretical level this study confirms that
risk compensation operates in children when engaged in sports
and recreation activities. Moreover, the results highlight that the
level of sensation seeking should be taken into account when
conducting research on risk compensation. Previous research has
indicated that individuals who have greater experience with an
activity take greater risks (Horvath and Zuckerman, 1993; LasenbyLessard et al., in press; Morrongiello and Dawber, 2004; Zuckerman,
2007), individuals wearing safety gear engage in greater risk tak-
1346
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
ing than if they were not wearing gear (Biasca et al., 2002; DiLillo
and Tremblay, 2001; Morrongiello et al., 2007a,b), and that various injury appraisals have a significant impact on risk decisions
(Morrongiello, 1997; Morrongiello and Rennie, 1998). However, the
current study is the first to examine these three factors together,
and the findings illustrate important ways these factors interrelate, and highlight that children scoring high in sensation seeking
constitute a particular ‘high risk’ group. Overall, in comparison to
other children, those who are high in sensation seeking generally
engage in greater risk taking, show greater risk compensation in
reaction to wearing safety gear, and they show elevated risk compensation in response to decreases in perceived danger and injury
severity when wearing safety gear.
The current results also have important implications for intervention. The findings highlight the need for parents, coaches, and
other supervisors to be particularly vigilant when children are
engaged in sports and recreation tasks for which they have a base of
experience. It is essential that supervisors not assume children can
accurately appraise injury risk in these activities or that they need
less supervision because they are familiar with the activity or are
wearing safety gear (Morrongiello and Major, 2002). Raising awareness among supervisors that children are likely to show high levels
of risk taking and risk compensation during activities with which
they have experience may help motivate them to remain vigilant.
These issues are particularly relevant to the domain of organized
sports, where questions about who should be responsible for children’s safety and how they behave (e.g., parents, coaches, children
themselves) continue to be debated (Emery et al., 2007).
Tailoring of interventions based on unique individual and or
contextual factors has gained in popularity (Gielen et al., 2006;
Rosenbloom, 2003; Zuckerman, 2007) and the current results are
relevant to this approach. Previous research has indicated the
need for intervention programs to target high sensation seekers
(Lasenby-Lessard et al., in press; Zuckerman, 2007) and the current research not only confirms this but highlights the importance
of also considering extent of experience with the risk activity. In
fact, the findings indicate that the content of intervention programs targeted at high sensation seekers needs to differ for high
and low experience activities. If a safety program seeking to reduce
risk compensation is targeting high sensation seekers, then the
content needs to focus on perceived reduction in danger for individuals with high experience with the activity, and on perceived
reduction in injury severity for individuals with low experience
with the activity. Given that tailoring of interventions has been
shown to maximize effectiveness (Gentilello et al., 2005; Gielen
et al., 2006; Ebel et al., 2003; McDonald et al., 2005; Nansel et al.,
2002; Zuckerman, 2007), considering sensation seeking and extent
of experience with the activity may produce the best strategies for
counteracting excessive risk taking and risk compensation among
children in sports activities.
6.2. Limitations and future research
Although the present study provides important insights regarding the impact of safety gear on children’s risk taking a few
limitations should be noted. Children were preselected for participation based on their level of experience with bicycling and
rollerblading. Future research on risk compensation may want to
examine risk compensation in a task such as the one utilized by
Lasenby-Lessard et al. (in press) in which experience was accumulated on site and carefully measured to ensure uniformity across
participants. Second, the present study examined individual sports
and recreation tasks. Different results may be found for team sports
or when peers are present. Research indicates that children engage
in greater risk taking when peers are present (Christensen and
Morrongiello, 1997; Horvath and Zuckerman, 1993; Morrongiello
and Dawber, 2004; Sandels, 1977; Wilson et al., 1991). Future
research should examine whether or not children engage in greater
risk compensation when peers are present during individual sports
and recreation tasks (bicycling, rollerblading, etc.) or during team
sports activities; children may show off and demonstrate even
greater risk compensation when peers are present. Finally, in future
research it would be informative to explore ways to counteract
children showing risk compensation in response to wearing safety
gear. Morrongiello and her colleagues have developed a number of
behavioral strategies that effectively reduce risk taking in children
(Morrongiello and Matheis, 2007; Morrongiello and Marks, 2008),
but whether these can be extended to address risk compensation
remains to be determined and is an important consideration for
future research on child injury prevention.
References
Adams, J., Hillman, M., 2001. The risk compensation theory and bicycle helmets.
Injury Prevent 7, 89–91.
Adams, S.L., Wyte, C.D., Paradise, M.S., del Castillo, J., 1996. A prospective study on inline skating: observational series and survey of active in-line skaters–-injuries,
protective equipment, and training. Academic Emergency Medicine 3, 304–311.
Aiken, L.S., West, S.G., 1991. Multiple Regression: Testing and Interpreting Interactions. Sage, Newbury Park, CA.
Biasca, N., Wirth, S., Tegner, Y., 2002. The avoidability of head and neck injuries
in ice hockey: a historical review. British Journal of Sports Medicine 36,
410–427.
Bouter, L.M., Knipschild, P.G., Feij, J.A., Volovics, A., 1988. Sensation seeking and
injury risk in downhill skiing. Personality and Individual Differences 9, 667–673.
Braun, C., Fouts, J., 1998. Behavioral response to the presence of personal protective
equipment. Human Factors and Ergonomics Society 2, 1058–1063.
Brown, L.T., Ruder, V.G., Ruder, J.H., Young, S.D., 1974. Stimulation seeking and the
change seeker index. Journal of Consulting and Clinical Psychology 42, 311.
Beaudry, G., 1994. Recueil de références sur la planification, l’aménagement et la
construction d’une station de ski alpin au Québec – Chapitre 1 La planification
de l’aménagement. Association des stations de ski du Québec, Montréal.
Canadian Institute of Child Health, 2000. The Health of Canada’s Children: A CICH
Profile, 3rd ed. Canadian Institute of Child Health, Ottawa, CA.
Christensen, S., Morrongiello, B.A., 1997. The influence of peers on children’s judgements about engaging in behavior that threatens their safety. Journal of Applied
Developmental Psychology 18, 547–562.
Coppens, N.M., Gentry, L.K., 1991. Video analysis of playground injury-risk situations. Research in Nursing and Health 14, 129–136.
DiLillo, D., Tremblay, G., 2001. Maternal and child reports of behavioral compensation in response to safety equipment usage. Journal of Pediatric Psychology 26,
175–184.
Ebel, B.E., Koepsell, T.D., Bennet, E.E., Rivara, F.P., 2003. Use of child booster seats
in motor vehicles following a community campaign. Journal of the American
Medical Association 289, 879–884.
Emery, C., Hagel, B., Morrongiello, B.A., 2007. Injury prevention in child and adolescent sport: whose responsibility is it? Canadian Journal of Sports Medicine 16,
514–521.
Gentilello, L.M., Ebel, B.E., Wickizer, T.M., Salkever, D.S., Rivera, F.P., 2005. Alcohol interventions for trauma patients treated in emergency departments and
hospitals: a cost benefit analysis. Annals of Surgery 241, 541–550.
Ginsburg, H.J., Miller, S.M., 1982. Sex differences in children’s risk-taking behavior.
Child Development S3, 426–428.
Gielen, A.C., Sleet, D.A., Diclemente, R.J., 2006. Injury and Violence Prevention:
Behavioral Science Theories, Methods and Applications. Jossey-Bass, San Francisco.
Hedlund, J., 2000. Risky business: safety regulations, risk compensation, and individual behavior. Injury Prevention 6, 82–90.
Heyman, S.R., Rose, K.G., 1980. Psychological variables affecting scuba performance.
In: Nadeau, C.H., Halliwell, W.R., Newell, K.L., Roberts, G.C. (Eds.), Psychology of
Motor Behavior and Sports. Human Kinetics Press, Champaign, IL.
Horvath, P., Zuckerman, M., 1993. Sensation seeking and risk appraisal and risky
behavior. Personality and Individual Difference 14, 41–52.
Hymbaugh, K., Garrett, J., 1974. Sensation seeking among skydivers. Perceptual and
Motor Skills 38, 118.
Lasenby-Lessard, J., Morrongiello, B.A., Barrie, D., in press. The Impact of Accumulated Experience on Children’s Appraisals of Riskand Risk Taking Decisions:
Implications for youth injury prevention. Health Psychology.
McDonald, E.M., Solomon, B., Shields, W., Serwint, J.R., Jacobsen, H., Weaver, N.L.,
Kreuter, M., Gielen, A.C., 2005. Evaluation of kiosk-based tailoring to promote
household safety behaviors in an urban pediatric primary care practice. Patient
Education and Counseling 58, 168–181.
Morrongiello, B.A., 1997. Children’s perspectives on injury and close call experiences.
Journal of Pediatric Psychology 22, 499–512.
Morrongiello, B.A., Dawber, T., 2004. Social and cognitive influences on schoolage children’s risk-taking decisions. Canadian Journal of Behavioral Science 36,
255–266.
J. Lasenby-Lessard, B.A. Morrongiello / Accident Analysis and Prevention 43 (2011) 1341–1347
Morrongiello, B.A., Lasenby, J., 2006. Finding the daredevils: development of a sensation seeking scale for children that is relevant to physical risk taking. Accident
Analysis and Prevention 38 (6), 1101–1106.
Morrongiello, B.A., Lasenby-Lessard, J., 2007. The psychology of risk taking by children: an integrative model implications for intervention. Injury Prevention 13,
20–25.
Morrongiello, B.A., Lasenby, J., Walpole, B., 2007a. Risk compensation in children:
why do children show it in reaction to wearing safety gear? Journal of Applied
Developmental Psychology 28, 56–63.
Morrongiello, B.A., Major, K., 2002. The influence of safety gear on parental perceptions of injury risk and tolerance for children’s risk taking: why a “Use the Gear”
message may not be sufficient to maximize child-injury preventions. Injury
Prevention 8, 27–31.
Morrongiello, B.A., Marks, L., 2008. “Practice what you preach”: induced hypocrisy as
an intervention strategy to reduce children’s risky playground behaviors. Journal
of Pediatric Psychology 33, 1117–1128.
Morrongiello, B.A., Matheis, S., 2004. Determinants of children’s risk taking in different social-situational contexts: the role of cognitions and emotions in predicting
children’s decisions. Journal of Applied Developmental Psychology 25, 303–326.
Morrongiello, B.A., Matheis, S., 2007. Addressing the issue of child-injury prevention:
a behavioral intervention to reduce fall-risk behaviors on playgrounds. Journal
of Pediatric Psychology 32, 819–830.
Morrongiello, B.A., Ondejko, L., Littlejohn, A., 2004a. Understanding toddlers’ inhome injuries I. Context, correlates and determinants. Journal of Pediatric
Psychology 29, 415–431.
Morrongiello, B.A., Ondejko, L., Littlejohn, A., 2004b. Understanding toddlers’
in-home injuries II. Examining parental strategies, and their efficacy, for moderating child-injury risk. Journal of Pediatric Psychology 29, 433–446.
Morrongiello, B.A., Rennie, H., 1998. Why do boys have more injuries than girls: the
role of attributions, beliefs, and risk appraisal. Journal of Pediatric Psychology
23, 33–44.
Morrongiello, B.A, Sedore, L., 2005. The influence of child attributes and socialsituational context on school-age children’s risk-taking behaviors that can lead
to injury. Journal of Applied Developmental Psychology 26, 347–361.
Morrongiello, B.A., Walpole, B., Lasenby, J., 2007b. Understanding children’s injuryrisk behavior: wearing safety gear can lead to increased risk taking. Accident
Analysis and Prevention 39, 618–623.
Nansel, T., Weaver, N., Donlin, M., Jacobsen, H., Kreuter, M., Simons-Morton, B., 2002.
Baby be safe: the effect of tailored communications for pediatric injury prevention provided in a primary care setting. Patient Education and Counseling 46,
175–190.
National Center for Injury Prevention and Control (NCIPC), 2010. WISQARS (WebBased Injury Statistics Query and Reporting System). Retrieved July 8, 2010,
<http://www.cdc.gov/ncipc/wisqars/>.
Potvin, L., Champagne, F., Laberge-Nadeau, C., 1988. Mandatory driver training
and road safety: the Quebec experience. American Journal of Public Health 78,
1206–1209.
1347
Pudpud, A., Linares, M., 1997. In-line skating: a deadly pediatric activity? Pediatric
Emergency Care 13, 376–379.
Rosen, B.N., Peterson, L., 1990. Gender differences in children’s outdoor play injuries:
a review and an integration. Clinical Psychology Review 10, 187–205.
Rosenbloom, T., 2003. Risk evaluation and risky behavior of high and low sensation
seekers. Social Behavior and Personality 31, 375–386.
Rowland, G.L., Franken, R.E., Harrison, K., 1986. Sensation seeking and participation
in sporting activities. Journal of Sport and Exercise Psychology 8, 212–220.
Sandels, S., 1977. An overall view of children in traffic. In: Jackson, R. (Ed.), Children,
the Environment, and Accidents. Pitman Medical, Kent, England.
Schwebel, D.C., Gaines, J., 2007. Pediatric unintentional injury: behavioural risk factors and implications for prevention. Journal of Developmental and Behavioral
Pediatrics 28, 245–254.
Shanon, A., Bashaw, B., Lewis, J., Feldman, W., 1992. Nonfatal childhood injuries: a
survey at the Children’s Hospital of Eastern Ontario. Canadian Medical Association Journal 146, 361–365.
Simonet, S., Wilde, G., 1997. Risk: perception, acceptance, and homeostasis. Journal
of Applied Psychology 46, 235–252.
Stanton, N., Pinto, M., 2000. Behavioral compensation by drivers of a simulator when
using a vision enhancement system. Ergonomics 43, 1359–1370.
Straub, W.F., 1982. Sensation seeking among high and low-risk male athletes. Journal
of Sport Psychology 4, 246–253.
Streff, F.M., Geller, E.S., 1988. An experimental test of risk compensation: betweensubject versus within-subject analyses. Accident Analysis and Prevention 20,
277–287.
Tabachnick, B.G., Fidell, L.S., 1989. Using Multivariate Statistics, 2nd ed. Harper and
Row, Philadelphia.
Thompson, D.C., Thompson, R.S., Rivara, F.P., 2001. Risk compensation theory should
be subject to systematic reviews of the scientific evidence. Injury Prevention 7,
86–88.
Viscusi, W.K., 1984. The lulling effect: the impact of child-resistant packaging on aspirin and analgesic ingestions. The American Economic Review 74,
324–327.
Viscusi, W.K., 1985. Consumer behavior and the safety effects of product safety
regulations. The Journal of Law and Economics 28, 527–554.
Viscusi, W., Cavallo, G., 1996. Safety behavior and consumer responses to
cigarette lighter safety mechanisms. Managerial and Decision Economics 17,
441–457.
Wilson, M., Baker, S., Teret, S., Shock, S., Garbarino, J., 1991. Saving Children: A Guide
to Injury Prevention. Oxford, New York.
World Health Organization, 2005. Child and Adolescent Injury Prevention: A Global
Call to Action. WHO, Geneva, Switzerland.
Zuckerman, M., 1979. Sensation Seeking: Beyond the Optimal Level of Arousal. Erlbaum, Hillsdale, NJ.
Zuckerman, M., 2007. Sensation Seeking and Risky Behavior. American Psychological
Association, Washington, DC.