Journal of Pediatric Psychology, Vol. 26, No. 3, 2001, pp. 175-184
© 2001 Society of Pediatric Psychology
Maternal and Child Reports of Behavioral Compensation in Response to Safety Equipment Usage
University of Missouri-Columbia
All correspondence should be sent to David DiLillo, Department of Psychology, 238 Burnett Hall, University of Nebraska-Lincoln, Lincoln, NE 68588-0308. E-mail: ddilillo{at}unl.edu .
 |
Abstract |
|---|
 Top Abstract IntroductionMethodResultsDiscussionReferences |
|---|
Objective: To assess maternal and child risk compensation behaviors in response to several commonly used safety measures.
Methods: We administered a previously validated self-report measure of risk tolerance to a total of 151 mothers and their children in grades 3-7. Mothers indicated the level of risk they would permit their child to assume; children were questioned regarding the degree of physical risk they would typically assume while unsupervised by an adult. Participating families were randomly assigned to conditions in which safety equipment either was or was not present during assessments of risk tolerance.
Results: Mothers who viewed the stimulus materials depicting the use of safety precautions reported significantly higher levels of tolerance for risky behavior on the part of their children than did mothers who viewed identical materials without the safety precautions. No significant differences in estimated risk taking emerged between children in the two experimental conditions.
Conclusions: These data may reveal a compensatory mechanism by which parents escalate their threshold for acceptable risk behavior in the presence of safety precautions for their children. Such tendencies have the potential to offset some of the protection provided by the use of safety equipment.
Key words: risk behavior; risk taking; risk tolerance; safety measures.
|
Introduction |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Unintentional injuries have supplanted illness as the major source of death and disability for youths over the age of one in the United States (Centers for Disease Control [CDC], 2000
; Rice & Mackenzie,
1989; Rodriguez,
1990). In 1996 alone, 13,000 children died from unintentional
injuries in this country (CDC,
2000). Furthermore, for every injury-related fatality, an
estimated 233 additional children require emergency medical treatment for
injuries (either unintentional or intentional) each year
(Burt & Fingerhut, 1998).
The financial consequences of these injuries are also disturbing. It has been
reported, for instance, that 15% of all medical spending for children from
ages 1 to 19 is directed toward treatment and rehabilitation following
unintentional injuries (Miller, Romano,
& Spicer, 2000). Clearly, the personal and economic impact of
unintentional childhood injuries makes prevention of these unfortunate events
an important societal issue.
Unfortunately, many of the activities that children most enjoy frequently
lead to serious injury. Bicycling, a ubiquitous activity for many youths, is
associated with 400 deaths and nearly 400,000 emergency room visits each year
for children under the age of 15 (Wilson,
Baker, Teret, Shock, & Garbarino, 1991). Water-related
recreation results in over 2,000 child drownings annually, a figure exceeded
only by the number of deaths involving automobiles
(CDC, 1990;
National Safety Council,
1992). Although rates of pedestrian injuries have decreased
slightly in recent years, children who walk and play near roadways are still
in danger, as 1,100 youths under the age of 14 are killed annually in road
environments (Wilson et al.,
1991). In addition to these more conventional sources of morbidity
and mortality, new threats to children's physical well-being are continually
emerging. In-line skating, for example, which has gained tremendous popularity
in the short time since its inception, is now recognized as a major source of
childhood injury, particularly in the form of bone fractures occurring to
those under 20 years of age (Adams, Wyte,
Paradise, & del Castillo, 1996;
Pudpud & Linares,
1997).
In response to the alarming rates of unintentional injuries suffered by
children in the course of routine play and recreational activities, various
controlled studies have examined the real-world effectiveness of safety
equipment usage in preventing injuries. These investigations have documented
clear improvements in safety across various injury categories. One of the most
impressive effects has been found for bicycle helmet usage, which may reduce
the risk of head injury by more than 80%
(Thompson, Rivara, & Thompson,
1996; Thompson, Rivara, &
Thompson, 1989). A similarly high degree of protection has been
attributed to wrist guards and elbow pads worn by children as protection while
roller blading (Schieber et al.,
1996). The use of personal flotation devices (a.k.a. life
jackets), which effectively keep swimmers' heads above water, is recommended
in the United States by the Coast Guard, the National Transportation Safety
Board, and the American Academy of Pediatrics
(Hermann & Stormer, 1985;
U.S. Coast Guard, 1995). There
is also an indication that painted crosswalks may decrease the risk of
pedestrian injury for young children
(Young & Lee, 1987).
Supported by data from these and other intervention studies, preventionists have long touted the regular use of safety equipment as one of the best front-line defenses against unintentional pediatric injury. Rarely considered in the child injury literature, however, is the possibility that safety equipment usage may itself elicit some systematic behavioral change with the potential to affect injury outcomes. If, for example, the protective value of safety equipment brings with it a perception of invulnerability to injury, children or the adults who supervise them may respond with an increased tolerance for risky or reckless behavior. More specifically, the sense of security imparted by safety equipment may lead to increased risk taking by children—or greater tolerance for risk on the part of supervising parents—either of which might increase the likelihood of injury. The potential impact of these processes on injury rates is easily understood by considering the familiar example of football players, who would be much less inclined to hurl themselves at one another in such an unrestrained manner without the feeling of protection provided by pads and a helmet.
This premise, that individuals alter their behavior in response to changes
in perceived risk, forms the basis of a formal theory of risk behavior known
as "risk compensation" or "risk homeostasis."
According to this theory, developed largely by Wilde
(1982,
1994,
1998), individuals have an
optimal or "target" level of risk with which they are most
comfortable and which they strive to maintain in a given risk situation.
People are said to respond to decreases in perceived risk in a given situation
by increasing risk behaviors in an attempt to hold relatively constant their
individual target level of risk (Wilde,
1994). The net result of these compensatory behaviors, according
to Wilde (1982,
1998), is that overall injury
rates will remain essentially unchanged, even in the presence of interventions
intended to improve safety.
Although risk compensation is believed to occur in response to a variety of
risky situations (Wilde,
1982), empirical investigation of the theory has been limited
primarily to studies of adult driving behavior. Several investigations
utilizing both actual and analogue driving tasks have reported evidence of
compensatory behaviors (e.g., faster driving) among participants randomly
assigned to a seat belt condition
(Aschenbrenner & Biehl,
1994; Jackson & Blackman,
1994; Streff & Geller,
1988). On the other hand, risk homeostasis theory has been the
subject of considerable debate in the literature, with several writers
criticizing the conceptual underpinnings of the theory, including the
contention that overall injury rates will remain unaffected by the
implementation of common safety measures such as automobile safety belts and
airbags (see, for example, McKenna,
1988; O'Neill & Williams,
1998; Slovic & Fischhoff,
1982). In the realm of childhood injury, the many studies
documenting net reductions in injury rates resulting from the use safety
equipment suggest that whatever compensatory processes may occur probably do
not involve a strict homeostatic mechanism operating to counteract completely
the protective value of safety equipment used by children. It is nevertheless
quite possible that some offsetting behaviors may occur in response to safety
equipment usage, resulting in increased risk tolerance among children or the
caregivers who supervise them. If this is the case, then the effectiveness of
common safety equipment used by children may, to some degree, be diminished.
Although other child injury specialists (e.g.,
Peterson, 1984) have alluded
to risk compensation processes by suggesting that perceptions of increased
physical safety could lead to less vigilant supervision on the part of
parents, we are unaware of any controlled experimental efforts to examine the
phenomenon of behavioral compensation related to pediatric injury.
This study seeks to assess self-reported compensatory behaviors in a sample of 8- to 13-year-old children and their mothers. During the preadolescent years, youths spend much of their time under the watchful eye of adult supervisors who are expected to carefully monitor and regulate children's risk-taking behaviors to protect them from injury. Yet, with advancing age, children spend increasing amounts of time in unsupervised settings, where they alone must make decisions regarding physical risks. Because of the dual influences on youths' injury-relevant behaviors, we chose to assess compensatory tendencies in both children and their maternal caregivers. Using a between-subjects design, we randomly assigned child participants and mothers to view several risk-taking scenarios depicting children either using or not using common safety equipment. Child participants in each condition were asked to report the maximum level of risk in which they would engage; mothers reported the maximum risk level they would permit their child to take. Responses from both groups were used to evaluate the prediction that safety equipment usage results in increased tolerance for physically risky behaviors among children and their maternal caregivers.
|
Method |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Participants
Participants were recruited by means of letters sent to the parents of a randomly selected subsample of all children enrolled in grades 3 through 7 in the local public school system, over a period of 6 months. Follow-up phone calls were made approximately 1 week after the letters were mailed, to solicit the participation of the identified child and his or her mother in the study. Because mothers typically bear a disproportionate responsibility (relative to fathers) for supervising their children's activities, only mothers were invited to contribute data, along with the identified child. Incentives for participating families included a substantial discount toward the purchase of a bicycle helmet provided to all participants, as well as a lottery with two bicycles and three smaller gift certificates offered as prizes. Of the 346 parents who were contacted, 151 (44%) agreed to participate in the project, kept scheduled interview appointments, and provided complete data.
Measures
Risk-taking Measure. The risk-taking stimulus materials used to
assess mothers' and children's self-reported tolerance for physical risk
taking were adapted slightly from a measure developed by Potts and his
colleagues (Potts, Doppler, &
Hernandez, 1994; Potts,
Martinez, & Dedmon, 1995). The measure consisted of seven 12
x 18-inch pictorial illustrations of school-age children engaged in
common play activities involving some level of physical risk. In each scenario
a five-interval gradation of risk was represented. That is, the picture showed
different degrees of risk, which reflected proximity to a potential injury
source. For each illustration mothers reported the maximum acceptable level of
risk they would allow their children to engage in while in their presence;
children reported the maximum amount of physical risk they would take if left
unsupervised.
The seven risk items were chosen to represent a range of common injury risks in children's environments. These items (with corresponding questions asked of mothers) included (1) riding a bicycle down hills of variable steepness and height ("How high a hill would you allow your child to ride down?"); (2) jumping a bicycle various distances off of a small ramp ("How far would you allow your child to jump?"); (3) swimming varying distances out into a body of water to retrieve an errant frisbee ("How far out would you allow your child to swim?"); (4) crossing a street to retrieve a ball, with a car approaching from varying distances ("When would you allow your child to cross the street?"); (5) riding in-line skates down hills of varying steepness ("How high a hill would you allow your child to ride down?"); (6) climbing up a tree to retrieve a kite at varying heights ("How high would you allow your child to climb?"); and (7) approaching an unknown, chained dog to retrieve a frisbee ("How close would you allow your child to get to the dog?"). As noted, each child participant was asked to report the maximum level of risk he or she would take in each risk scenario.
During administration of the measure, a small illustrated figure of a child of the same sex and appearing approximately the same age as the child respondent was initially placed at the lowest risk level in each illustration. If a participant reported tolerance for this initial level of risk, the figure was then placed at the next higher level of risk and the participant was queried again about the acceptability of engaging in that behavior. This process was repeated until the respondent indicated a maximum level of acceptable risk, at which point a score ranging from 1 (lowest risk) to 5 (greatest risk) was assigned for the item. In this manner, Likert-type ratings ranging from 1 to 5 could be assigned to mother and child reports. On occasions when even the lowest level of risk was unacceptable, a score of 0 was assigned. Scores were averaged across items to obtain an overall indicator of acceptable risk.
Potts et al. (1995) have
reported psychometric properties for the risk-taking measure, including
moderate internal consistency and convergent validity with self-report
measures of sensation seeking, as well as with peer, teacher, and parent
ratings of risk-taking behavior, and actual injury history. This instrument
thus serves as a safe and apparently accurate proxy measure for actual
physical risk-taking behaviors among children.
Injury Frequency History. Parents completed a child injury history questionnaire that assessed life-time frequencies of injuries experienced by their child, including broken bones, muscle strains, serious cuts, concussions, burns, poisoning, animal bite, insect stings/bites, water inhalation, shock, and other (miscellaneous). Injuries were classified as either medically treated or not medically treated. Each of these was summed for two overall injury scores.
Direct Experience and Safety Equipment Measures. To assess past direct experience with the seven risk-taking situations, we asked child participants to designate the frequency with which they had actually engaged in the basic activity represented in each of the seven scenarios, using a 5-point scale (1 = never, 2 = once in a while, 3 = every other day, 4 = most days, and 5 = every day). Thus, children reported how often they rode a bicycle, did tricks on a bike, used rollerblades, swam in a pool or lake, climbed trees, and crossed streets alone "in the summertime or when the weather is nice outside." Scores were averaged across items to obtain an overall indicator of direct experience with the risk activities.
A 3-point scale (1 = never, 2 = sometimes, and 3 = every time) was then used to assess participants' use of safety equipment appropriate to each situation. Children were asked to report their use of helmets for bicycling; elbow pads, knee pads, wrist guards, and helmets for rollerblading; and life preservers for water activities.
Demographic Questionnaire. Finally, a questionnaire was completed by parents to assess demographic information such as child's age, gender, and ethnicity, as well as parental marital status, education, and occupation.
Experimental Design and Procedures
Data were collected during individual testing sessions, prior to which
participants had been randomly assigned to either an experimental or control
condition. Each data collection session lasted approximately 30 minutes and
occurred in either the family's home or on campus, according to their
preference. For participants in the experimental condition, safety equipment
or features were depicted in five of the seven the risk scenarios (1-5); these
included the child figure wearing a helmet in the bicycle scenes; elbow pads,
knee pads, wrist guards, and a helmet in the rollerblading scene; a life
preserver in the water scenario; and using a crosswalk in the pedestrian
scenario. Participants in the control condition viewed the same five
scenarios, but with all safety equipment and protective features deleted from
the pictures. For the remaining two risk scenarios (6 and 7), participants in
both conditions viewed identical stimulus materials. These last two items thus
served as a manipulation check: the presence of group differences in ratings
of items 1-5, but not items 6 and 7, would support the strength of
manipulating the presence of safety equipment. The order of scenario
presentation was systematically varied using seven random series of the risk
items. Experimenters rotated though these series during data collection to
balance possible order effects.
After discussion and completion of parental consent and child assent forms
(and with children out of earshot), experimenters completed the risk-taking
measure with each mother. Mothers then completed the demographic and injury
history questionnaires. With mothers out of the room, children were then
administered the risk-taking measure, as well as the direct experience and
safety equipment measures, which the experimenter read aloud to them.
Following both interviews, participants were debriefed as to the purpose of
the study, and children were engaged in a discussion about the potential
dangers of several of the depicted activities. All experimental procedures
were in compliance with the American Psychological Association's
(1992) ethical guidelines for
research and were approved by the University of Missouri Human Participants
Review Committee.
|
Results |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Group Comparability
Participants in the experimental (Safety Equipment) and control (No Safety Equipment) conditions were compared with respect to several injury-relevant variables (i.e., injury history, direct experience, and safety equipment questionnaires; see Table I), as well as demographic indices (child sex, ethnicity, age, and socioeconomic status [SES]). T tests and chi-squares revealed no significant differences between conditions on any of these variables. Our sample of 151 mother-child dyads was evenly divided by gender (75 boys and 76 girls), predominantly white (92% Caucasian, 5% African American, and 3% other ethnic minority), and reported a mean child age of 10.8 years (range: 8.4-13.6, SD = 1.4). SES, computed using the Hollingshead Two Factor Index (Hollingshead, 1957
), ranged
from 14 to 66, with a mean of 49.1 (SD = 13.2), reflecting a broad
range of SES in our sample, with the average participant being of middle to
upper-middle SES.
|
Effect of Safety Equipment Manipulation
Mother and child ratings of acceptable risk were averaged across the five
manipulated scenarios and two constant scenarios, yielding two risk summary
scores for each informant. Mothers' summary scores for the manipulated
scenarios demonstrated adequate reliability (Cronbach's 
=.74),
although the two constant scenarios were less so ( =.35), probably
because one of them yielded a floor effect for mothers, who most often
disapproved of their children approaching the dog at all. The child summary
scores for manipulated and constant scenarios achieved alphas of.80 and.70,
respectively. These summary scores are depicted for mothers and children, by
condition, in Table II.
|
Of the four summary scores (mother and child across both manipulated and constant scenes), only the mother ratings of manipulated scenes demonstrated an effect of condition; t(149) = 4.82, p =.001. Figure 1, which depicts maternal ratings for each scenario, by condition, shows a pattern of greater risk tolerance reported by mothers in the safety condition across all five risk items. The constant scenes yielded virtually identical risk ratings across conditions for both mothers and children. Although this general pattern appears to confirm the utility of the constant items (6 and 7) as a check on the manipulation of safety equipment across conditions, support for this conclusion is limited somewhat by the restricted response range for mothers on the "approach a strange dog" item. Subsequent analyses focused on the manipulated scenes only.
|
Comparing Mother and Child Risk Ratings
To characterize the effect of experimental condition and child sex on risk
ratings, a 2 (condition, safety vs. no-safety, between subjects) x 2
(child sex, male vs. female, between subjects) x 2 (respondent, mother
vs. child, within subjects) repeated measures ANOVA was conducted, with
ratings of acceptable risk (the first and third rows of
Table II) as the dependent
variable. Of particular interest were two potential interaction effects on
risk ratings: Condition x Respondent, indicating a difference in how the
experimental manipulation affected the risk tolerance of mothers versus
children, and Condition x Child Sex, indicating a difference in how the
experimental ratings affected risk ratings for boys versus girls. The former
was confirmed, F(1, 147) = 6.99, p =.009 (see
Table III). Visual examination
of the means presented in Table
II reveals that mothers' ratings were lowest—and discrepant
from child ratings—in the no-safety condition but approached child
ratings in the presence of safety equipment.
|
The interaction between condition and child sex fell short of statistical significance, F(1, 147) = 2.73, p =.10, indicating that the effect of the safety equipment manipulation on both mothers' and children's ratings did not show a strong pattern of difference by child gender (see Table IV). Further inspection of the mother ratings only, however, suggests a pattern in which the upward drift of maternal risk ratings in the presence of safety equipment (reported above) appears more substantial for sons than for daughters. Mothers' ratings approach children's ratings in the safety condition, for both boys and girls. Nevertheless, because the risk ratings offered by girls are relatively low in comparison with the boys', the increment required for maternal ratings to approach child ratings in the Safety condition is smaller for girls.
|
Finally, although we have demonstrated a convergence of the average mother's and child's risk ratings in the safety condition, some index of covariation at the level of individual mother-child dyads would be helpful. Pearson correlation coefficients, reflecting correspondence between risk ratings offered by mother and child dyads (for manipulated scenarios only) yielded values of.51 (p <.001) for the safety condition and.20 (p =.10) for the no-safety condition.
|
Discussion |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
To our knowledge, this is the first study to specifically investigate the phenomenon of behavioral compensation related to childhood injury. At issue here was whether participants' tolerance for physical risk taking, across several popular but potentially hazardous childhood activities, would vary as a function of safety equipment usage. Theories of risk compensation suggest that, all other factors being equal, individuals who use protective equipment will engage in greater levels of physical risk than those who do not. Here, in comparison to mothers in the no-safety condition, those viewing safety stimuli reported significantly greater tolerances for children's risk-taking behaviors across all of five manipulated risk scenarios. Assuming equivalence of groups (which is likely given our sample size, random assignment of participants, and a lack of group differences on a number of potentially confounding variables), it is reasonable to conclude that the increased acceptance of risk by mothers in the safety condition was in fact due to the experimental manipulation of safety equipment between conditions. In contrast to mothers, child participants demonstrated no significant group differences in risk taking in response to either manipulated or constant scenarios. These results and their implications for injury prevention will be discussed in the context of risk compensation processes.
Maternal Risk-Taking Reports
As noted, mothers in the safety condition were significantly more lenient
in their standards for risky behavior on the part of their children. These
loosened restrictions were evident for children of both sexes (though somewhat
more so for sons than daughters) across all five scenarios including the
bicycle, water, in-line skating, and pedestrian risk situations. The greater
acceptance of risk taking reported by mothers in the safety equipment
condition may have resulted from an increased confidence in their children's
physical well-being, engendered by the presence of protective equipment in the
risk scenarios. This finding may constitute cause for concern, because of the
possibility that supervising adults who permit greater risk taking when
children use protective equipment may unwittingly undermine the intended
function of that equipment, exposing children to a greater likelihood of
injury than might exist under conditions of constant supervisory
vigilance.
An additional point of interest emerged from comparisons of mothers' and children's mean risk levels across the safety and no-safety conditions. When viewing unprotected figures, mothers were more conservative than children in the risk levels they endorsed. That is, caregivers permitted less risk taking than their sons or daughters reported they would take. This probably reflects greater safety awareness and more accurate judgments of physical risks on the part of adult caregivers. Surprisingly, however, these differences were not maintained across conditions, as mothers in the safety condition reported risk levels that approached those provided by children. Thus, in the presence of safety equipment, caregivers reported a relaxed tolerance for risky behaviors, allowing both boys and girls to engage in similar levels of risk that children themselves would take while unsupervised and without the aid of protective equipment. To the degree that these tendencies reflect real-world behaviors, mothers of children who use protective equipment may be undercutting the crucial role they as caregivers play in limiting risky child behaviors that could lead to injury.
Additional information regarding the correspondence between risk ratings can be found in our correlational data showing a strong positive relationship between maternal and child risk tolerances in the safety condition, but not in the no-safety condition. These findings supplement the group data, again suggesting a convergence of mother and child risk ratings in the presence of safety equipment. These results indicate the possibility of some transmission of attitudes or tolerances toward risk between mother and child. Such influences could conceivably be either unidirectional or bidirectional (i.e., mother to child, child to mother, or both).
Child Risk-Taking Reports
Analyses of gender differences in risk taking showed boys selecting greater
levels of risk than did girls, in both the safety and no-safety conditions.
These data are congruent with previous research on childhood risk taking and
injury (e.g., Kafry, 1982;
Matheny, 1991;
Potts et al., 1995;
Zuckerman, 1979), finding that
boys consistently demonstrate higher levels of risk taking (and suffer more
injuries) than girls. More germane to this study was a lack of support for the
hypothesis that children viewing safety figures would evince greater levels of
risk taking than their no-safety counterparts. Although the overall pattern of
means suggested higher risk tolerance among safety participants on four of
five risk scenarios, these differences were not statistically significant.
This finding may in part reflect children's lack of knowledge about the
protective function of safety equipment (e.g., perhaps helmets are just
another type of hat to them). It is also possible that children who are
distracted by the excitement of play activities simply become oblivious to the
presence of safety equipment, rendering it inconsequential to their
risk-taking decisions. In such cases, dispositional variables (e.g., sensation
seeking, impulsivity, activity level), which have been empirically linked to
physical risk taking and injury (DiScala,
Lescohier, Barthel, & Li, 1998;
Kafry, 1982), may supersede
whatever impact safety equipment might have on children's willingness to
engage in risky behaviors.
Implications for Injury Prevention
Here, parents reporting as though their children were using protective
equipment permitted significantly increased levels of child risk taking.
Although we cannot be certain whether a shift in risk tolerances of this
magnitude would translate in an increased risk of actual injuries, one might
reasonably predict that significant increases in risk taking would be
accompanied by a greater likelihood of injury-producing events (e.g., falls
while rollerblading, bicycle crashes, etc.) of a severity beyond the ability
of safety equipment to contain. In this manner, the protective value of
equipment may be undermined to some extent by compensatory behaviors. Thus,
interventions that promote safety solely by encouraging increased utilization
of safety equipment may be missing an important piece of the injury prevention
puzzle: the need to consider behavioral responses to safety equipment usage. A
twofold strategy of promoting safety equipment usage and discouraging
the loosening of standards for safe behavior in the presence of such
precautions may therefore have the potential for improved effectiveness over
equipment-only interventions. Reminding caregivers who may overestimate the
protective value of safety equipment that such measures neither confer
invulnerability to injury nor reduce the need for vigilant supervision of
children seems a logical first step in this process. Parents could be
instructed not only to insist on the use of proper safety equipment but also
to maintain the enforcement of safe standards of behavior in the presence of
such measures.
Our failure to find evidence of compensatory behaviors among child participants may be an encouraging sign regarding children's risk behaviors in unsupervised settings. If children's real-world risk behaviors parallel their self-reports here (i.e., if children are not inclined to take greater risks when they use safety equipment), this suggests that youths who are unsupervised, but using safety equipment, may be receiving the maximum benefit from that equipment.
Limitations and Future Directions
Several methodological features of this study shed light on directions for
future research in the area. The first issue concerns the between-subjects
nature of the design, which compared risk reports of different groups of
participants under two separate conditions. Because risk compensation is
believed to operate at an individual as well as an aggregate level
(Wilde, 1988), a
within-subjects design assessing the same individuals under two conditions
(one "safety," one "no-safety") could provide unique
information not available from a between-subjects design. Such a study could
evaluate the extent to which risk compensation may result from a
"contrast effect" produced when an individual switches from nonuse
to use of some type of safety equipment
(Streff & Geller, 1988). A
second issue concerns our use of a self-report measure to assess child and
maternal risk-taking tendencies. Although this instrument has demonstrated
some degree of validity, the possibility remains that mothers' or children's
real-world tolerances for children's risk behaviors differ from the reports
provided here. An observational measure would allow for greater confidence in
the assessment of participants' risk-taking behaviors. Third, a more thorough
knowledge of risk compensation processes could be gained from attempts to
explore the specific mechanism by which risk compensation is said to occur
(i.e., the role that risk perceptions play in risk behaviors following the
introduction of safety equipment). Though shifts in risk perceptions could be
inferred from the present data, they were not measured here directly. Finally,
in further establishing the parameters within which risk compensation occurs,
it will be crucial to consider the relevance of developmental changes in risk
compensation tendencies. The question of whether compensatory shifts in
parental tolerance for children's risk taking vary as a function of child age
may have implications for the tailoring of interventions strategies to
different parental and child cohorts.
|
Acknowledgments |
|---|
David DiLillo is now at the Department of Psychology, University of Nebraska-Lincoln. George Tremblay is now at the Department of Clinical Psychology, Antioch New England Graduate School. This research was sponsored by grant #URC-99-010 from the University of Missouri Research Council.
Received July 30, 1999; revision received January 24, 2000; accepted September 22, 2000
|
References |
|---|
|
TopAbstractIntroductionMethodResultsDiscussionReferences |
|---|
Adams, S. L., Wyte, C. D., Paradise, M. S., & del Castillo, J. (1996). A prospective study of in-line skating: Observational series and survey of active in-line skaters— Injuries, protective equipment, and training. Academic Emergency Medicine, 3, 304-311.[ISI][Medline]
American Psychological Association. (1992). Ethical principles of psychologists. American Psychologist, 47, 1597-1611.
Aschenbrenner, M., & Biehl, B. (1994). Improved safety through improved technical measures? Empirical studies regarding risk compensation processes in relation to anti-lock brake systems. In R. M. Trimpop & G. J. S. Wilde, (Eds.), Changes in accident prevention. The issue of risk compensation. Groningen, The Netherlands: Styx.
Burt, C. W., & Fingerhut, L. A. (1998). Injury-related visits to hospital emergency departments: United States. Vital Health Statistics, 13, 1-76.
Centers for Disease Control and Prevention (CDC). (1990). Fatal injuries to children—United States, 1986. Morbidity and Mortality Weekly Report, 39, 443-451.
Centers for Disease Control and Prevention, National Center for Injury Prevention and Control (CDC). (2000). U.S. Injury Mortality Statistics. Atlanta, GA: Author. Retrieved September 12, 2000, from the World Wide Web: http://www.cdc.gov/ncipc/osp/usmort.htm.
DiScala, C., Lescohier, I., Barthel, M., & Li, G.
(1998). Injuries to children with attention deficit hyperactivity
disorder. Pediatrics, 102,
1415-1421.
Hermann, R., & Stormer, A. (1985). Results of recent research on life-jackets. Journal of Royal Naval Medical Services, 71, 161-166.
Hollingshead, A. B. (1957). Two Factor Index of Social Position. New Haven, CT: Unpublished manuscript.
Jackson, J. S. H., & Blackman, R. (1994). A driving-simulator test of Wilde's risk homeostasis theory. Journal of Applied Psychology, 79, 950-958.
Kafry, D. (1982). Sensation seeking of young children. Personality and Individual Differences, 3, 161-166.
Matheny, A. P., Jr. (1991). Children's unintentional injuries and gender: Differentiation by environmental and psychosocial aspects. Children's Environments Quarterly, 8, 51-61.
McKenna, F. P. (1988). What role should the concept of risk play in theories of accident involvement? Ergonomics, 31, 469-484.
Miller, T. R., Romano, E. O., & Spicer, R. S. (2000). The cost of childhood unintentional injury and the value of prevention. Future of Children, 10, 137-163.[ISI][Medline]
National Safety Council. (1992). Accident facts—1992 edition. Itasca, IL: National Safety Council.
O'Neill, B., & Williams, A. (1998). Risk
homeostasis hypothesis: A rebuttal. Injury Prevention,
4, 92-93.
Peterson, L. (1984). The "Safe at Home" game: Training comprehensive prevention skills in latchkey children. Behavior Modification, 8, 474-494.[Abstract]
Potts, R., Doppler, M., & Hernandez, M. (1994). Effects of television content on physical risk-taking in children. Journal of Experimental Child Psychology, 58, 321-331.[ISI][Medline]
Potts, R., Martinez, I. G., & Dedmon, A. (1995).
Childhood injury and risk-taking: Self-report and informant measures.
Journal of Pediatric Psychology,
20, 661-668.
Pudpud, A. A., & Linares, M. Y. (1997). In-line skating: A deadly pediatric activity? Pediatric Emergency Care, 13, 376-379.[ISI][Medline]
Rice, D. B., & Mackenzie, E. J. (1989). Cost of injury in the United States: A report to Congress 1989. San Francisco: University of California Institute for Health and Aging.
Rodriguez, J. G. (1990). Childhood injuries in the United States: A priority issue. American Journal of Diseases of Children, 144, 625-626.[Medline]
Schieber, R. A., Branche-Dorsey, C. M., Ryan, G. W., Rutherford, G.
W., Jr., Stevens, J. A., & O'Neil, J. (1996). Risk factors
for injuries from in-line skating and the effectiveness of safety gear.
New England Journal of Medicine,
335,
1630-1635.
Slovic, P., & Fischhoff, B. (1982). Targeting risks. Risk Analysis, 2, 227-234.
Streff, F. M., & Geller, E. S. (1988). An experimental test of risk compensation: Between-subject versus within-subject analyses. Accident Analysis and Prevention, 20, 277-287.
Thompson, D. C., Rivara, F. P., & Thompson, R. S. (1996). Effectiveness of bicycle safety helmets in preventing head injuries: A case-control study. Journal of the American Medical Association, 276, 1968-1973.[Abstract]
Thompson, R. S., Rivara, F. P., & Thompson, D. C. (1989). A case-control study on the effectiveness of bicycle safety helmets. New England Journal of Medicine, 320, 1361-1367.[Abstract]
U.S. Coast Guard. (1995). Draft regulatory evaluation: Recreational inflatable personal flotation device standards. Washington, DC: Author.
Wilde, G. J. S. (1982). The theory of risk homeostasis: Implications for safety and health. Risk Analysis, 2, 209-225.
Wilde, G. J. S. (1988). Risk homeostasis and traffic accidents: Propositions, deductions, and discussion of dissension in recent reactions. Ergonomics, 31, 441-468.
Wilde, G. J. S. (1994). Target risk: Dealing with the danger of death, disease, and damage in everyday decisions. Toronto: PDE.
Wilde, G. J. S. (1998). Risk homeostasis theory: An
overview. Injury Prevention, 4,
89-91.
Wilson, M. H., Baker, S. P., Teret, S. P., Shock, S., & Garbarino, J. (1991). Saving children: A guide to injury prevention. New York: Oxford University Press.
Young, D. S., & Lee, D. N. (1987). Training children in road crossing skills using a roadside simulation. Accident Analysis & Prevention, 19, 327-341.[ISI][Medline]
Zuckerman, M. (1979). Sensation seeking and risk taking. In C. E. Izard (Ed.), Emotions in personality and psychopathology (pp. 163-197). New York: Plenum.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  I B Pless Is this journal really needed? Inj. Prev., December 1, 2003; 9(4): 289 - 291. [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Landen, U. Bauer, and M. Kohn Inadequate Supervision as a Cause of Injury Deaths Among Young Children in Alaska and Louisiana Pediatrics, February 1, 2003; 111(2): 328 - 331. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||