Journal of Pediatric Psychology Advance Access originally published online on February 23, 2005
Journal of Pediatric Psychology 2005 30(7):562-570; doi:10.1093/jpepsy/jsi043
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Children’s Risk Taking Behaviors: The Role of Child-Based Perceptions of Vulnerability and Temperament
1 Clinical Child Psychology Program, The University of Kansas, and 2 Columbus Children’s Research Institute, Columbus Children’s Hospital
All correspondence concerning this article should be addressed to Richard E. Boles, University of Kansas, Clinical Child Psychology Program, 2010 Dole Human Development Center, 1000 Sunnyside Avenue, Lawrence, Kansas 66045. E-mail: rboles{at}ku.edu.
Received December 15, 2003; revisions received May 28, 2004 and August 20, 2004; accepted August 20, 2004
 |
Abstract |
|---|
 Top Abstract MethodResultsDiscussionAcknowledgmentReferences |
|---|
Objective To examine the relationship between perceptions of vulnerability, temperament, and children’s risk taking behavior in a simulated home environment. Methods Children and their primary caregivers were interviewed regarding temperament and perceptions of vulnerability to injury. In addition, children’s interactions with simulated hazards were observed in an environment representing a typical home. Results Children whose caregivers reported higher levels of activity were significantly more likely to report lower perceptions of vulnerability to injury and show increased risky behavior. After controlling for gender differences, children’s risky behaviors were predicted from child-based perceptions of vulnerability. Conclusion Perceptions of vulnerability and active temperaments represent significant risk factors for potential injuries in the home. Modifying perceptions of vulnerability as well as identifying at-risk temperaments for injuries is important to consider when developing effective interventions.
Key words: home injuries; children; injury prevention; temperament; activity level; vulnerability.
Unintentional injuries are the leading cause of death in children and adolescents from age 1–19 years (Guyer et al., 1999
) and are identified as the greatest threat to the health and well-being of the nation’s children (Centers for Disease Control and Prevention, 2000; Finney et al., 1993; Roberts & Brooks, 1987). In the United States in 1998, home injuries that resulted in an emergency department visit occurred to more than 10 million people. Such injuries include fire-related burns, falls, poisonings, and scalds (National Center for Injury and Prevention and Control, 2001, 2002). Consequently, the prevention and minimization of injuries in the home and surrounding community have been identified as the prime foci of research for the future (National Center for Injury and Prevention and Control, 2001, 2002).
Children and adolescents are at greater risk of unintentional injuries for many reasons. Owing to their cognitive immaturity, children are often impulsive and may frequently be unable to judge accurately the level of safety in situations (National Center for Injury and Prevention and Control, 2002; Schwebel & Plumert, 1999). Child age and gender have also been identified as key variables in understanding child injury rates. For instance, injuries typically increase during grade school years (Scheidt et al., 1994). Also, boys have identified situations as being less risky when compared with girls (Hillier & Morrongiello, 1998). Moreover, boys engage in more risky behavior than girls during both naturalistic observations and analog studies (Coppens & Gentry, 1991; Rosen & Peterson, 1990).
Child temperament has also been previously identified as a risk factor for increased rates of injury (Bijur, Stewart-Brown, & Butler, 1986; Matheny, 1987; Schwebel & Plumert, 1999). Specifically, parents who described their children as highly active, impulsive, and under-controlled reported more daily injuries of greater severity (Plumert & Schwebel, 1997; Schwebel & Plumert, 1999). More specifically, temperament traits place children at a higher risk of injury, which may be mediated by ability overestimation that leads to injuries (Schwebel & Plumert, 1999). Few studies presently exist, however, on other possible mediating variables for the relation found between temperament and injuries.
Unintentional injuries are considered to be understandable, preventable, nonrandom processes (Peterson, Farmer, & Mori, 1987; Roberts, 1986; Roberts, Elkins, & Royal, 1984), in which explication through scientific theory and investigations can reveal etiology and parameters for prevention. Despite renewed research efforts from a variety of disciplines and methodological approaches, significant gaps in knowledge remain to be fully explored. One important area of exploration continues to be children’s perceptions regarding unintentional injuries. Relatively few studies have dealt with this focus, although a growing literature base is emerging. In an effort to reveal children’s self-reports about their behavior in risky situations, Gable and Peterson (1998) found that 8-year-old children most frequently identified fate as the principal reason for the occurrence of minor injuries. In fact, 30% of the interviewed children indicated that the injury occurred because of "reasons out of their control" (p. 330).
Similar findings were also reported by Morrongiello and Rennie (1998) when they interviewed children, ages 6, 8, and 10 years. Children viewed stimuli depicting children in various degrees of risk and were asked to first sort the pictures according to level of risk, with varying levels of facial expressions, ranging from confident to wary. They then reported how much they engaged in each activity and answered questions regarding the attributions of injuries during such activities. Results showed that children’s self-reported risk taking behaviors were related to beliefs that injuries are attributed to bad luck, that they were less likely than peers to incur injury, and that they more often minimized their own level of risk of injury. Additionally, boys, as compared with girls, were found to be more frequent risk takers, more often identified injuries as due to bad luck, rated lower levels of vulnerability to injury, and rated themselves less likely than their peers to be injured. Although these studies provide evidence of important child-based cognitive factors related to children’s actual risky behavior, much remains to be understood about children’s vulnerability to injury, particularly within the home environment. Research on injury control and prevention may also benefit from utilizing nontraditional methodologies toward advancing current knowledge.
An often neglected methodological approach to child injury prevention has been the use of observational techniques to gather data on child behaviors in various situations. Although a small portion of the injury literature has included child observations, most studies have relied on self-report, caregiver report, or medical records both retrospectively and prospectively. Observational data not only provide important steps toward identifying reliable relationships among injury and behavior, such data also have the advantage of revealing behavioral actions and environmental contingencies. When observational data are connected to past reported experiences of risky behavior and injury history, a much larger and perhaps more comprehensive picture may emerge regarding the etiology of child injury (Cataldo, Finney, Richman, & Riley, 1992; Cook, Peterson, & DiLillo, 1999).
In this study, children’s perceptions of vulnerability to risky home situations are examined to determine possible links to their observed behaviors in a simulated home hazard environment. Children who indicate lower perceptions of vulnerability to injury are expected to display greater rates of actual risky behaviors during observations. Mediation effects were conducted to further clarify the nature of the relationship between child-based perceptions of vulnerability, temperament, and observed risky behavior. In particular, perceptions of vulnerability were tested as possible partial or full mediators for the hypothesized relationship between temperament and observed risky behavior. Given that ability overestimation is considered a likely mediator of the relationship between temperament and injuries (Schwebel & Plumert, 1999), this study assesses whether perceptions of vulnerability play a similar role as ability overestimation.
|
Method |
|---|
|
TopAbstractMethodResultsDiscussionAcknowledgmentReferences |
|---|
Participants
Recruitment of children took place at several daycare facilities as well as at local elementary schools during after-school programs within a midwestern city. One hundred and twelve caregivers provided contact information and were called by a study researcher to ascertain interest. Seventy-two caregivers agreed to participate in the study and arranged to complete measures both at home and at a university clinic. Forty-seven participants (64%) completed all measures for this study in multiple locations for the purposes of additional projects unrelated to this study. All measures were counterbalanced in presentation. Forty-six children ages 4–7 and their caregivers comprised the final. Eighty percent of the caregivers who completed the questionnaires were the child’s mothers, 15% were fathers, and 5% were other family members. No differences were found in temperament scores between reporters. Children with a physical disability or families who reported a member having been significantly injured (i.e., requiring hospitalization during the previous 3 months) were considered ineligible to participate. Demographic information as well as additional measures not related to this study were collected at a home visit before the clinic visit. In addition, other measures were collected during the visit to the clinic as well as during an additional home visit, but they were unrelated to this investigation. Because of the overall research design (i.e., multiple phases), many participants were not included in this project (e.g., they were part of the control group for another study). Assessments for sampling bias using ANOVAs on the completers and noncompleters revealed no significant differences, p > .05, on education, income, family size, temperament, or ethnicity. This project was accepted by the institutional review board.
Measures
Family Information Form
Demographic information was obtained from a caregiver regarding family size, ages, ethnicity, income level, education, and gender of all household members. Socioeconomic status was derived from the information provided in this form. Additionally, family history of injury and child’s health were collected from the participating caregivers.
Temperament Assessment Battery for Children—Revised (Martin & Bridger, 1988)
The Temperament Assessment Battery for Children—Revised (TABC-R) is a 37-item normative-based questionnaire designed to determine temperamental type based on score profiles of Impulsivity and Inhibition dimensions. The dimensions are theoretically derived from the neuropsychological theory on the behavioral inhibition system (Inhibition dimension) and the behavioral activation system (Impulsivity dimension) (Gray, 1991). The Inhibition dimension was linked to introversion characteristics (e.g., highly unsociable or shy). The Impulsivity dimension has been linked to extraversion behaviors (e.g., management problems) during test validation and is conceptualized as an aggregate of Negative Emotionality, Activity Level, and Lack of Task Persistence scales. In addition to the aforementioned scales, 10 impulsivity augmentation items are included in the Impulsivity dimension t score to increase reliability (Martin & Bridger, 1988). Items from the Activity Level scale include "My child runs up and down stairs" and "When my child moves about in the house or outdoors, he/she runs rather than walks." A sample item from the Impulsivity scale is "My child is easy to manage" (with reversed scoring on this item). Caregivers indicate how often their child engages in each behavior on a Likert-type scale ranging from 1 ("hardly ever") to 7 ("almost always"). t scores are generated in the following areas: Inhibition (8 items; tendency to be cautious or hesitant in social or novel situations), Negative Emotionality (8 items; tendency to become emotionally distressed), Activity Level (6 items; tendency to show and control gross motor behavior), Lack of Task Persistence (5 items; level of continued engagement during tasks over time), and Impulsivity (29 items; an aggregate of the degree to which the child can control behavior, emotion, and attention). Martin and Bridger (1988) reported internal consistency estimates ranging from .71 to .90 for the various subscales. Test–retest reliability was reported with 1-year stabilities ranging from .53 to .76.
The Negative Emotionality and Lack of Task Persistence scales were not used in this study because of preliminary data analysis which indicated that they were not meaningfully related to risky child behavior or perceptions of child vulnerability, p > .05, offered no prior conceptual basis for inclusion, and ultimately reduced the power to detect significant relationships. It was also observed that TABC-Impulsivity and TABC-Activity Level (TABC-AL) scales were significantly positively correlated as might be expected, although multicolinearity diagnostics were found to be acceptable. TABC-Impulsivity and TABC-Inhibition were dropped from further analyses, however, for several reasons: First, the inclusion of both TABC-Impulsivity and TABC-AL variables produced nonsignificant models (indicative of multicolinearity), and TABC-Impulsivity was dropped. More important, TABC-Impulsivity is an aggregate of not only Activity Level but also Lack of Task Persistence and Negative Emotionality scales; these scales have not been theoretically linked to child injuries and were not hypothesized for the current study. Preliminary analyses also failed to reveal an association with Risky Behavior. Therefore, the primary contribution of linking Impulsivity with Risky Behavior and Perceptions of Vulnerability appears to be from the inclusion of Activity Level items within its scoring. Finally, TABC-Inhibition was not included because it was not significantly related to any other study variables of interest, thus increasing the predictive power of the retained variables.
Injury Vulnerability Assessment of Children
This measure was created by the authors of this study. The measure consists of 20 single photographs of a single child (age 5) interacting within a home environment. Two sets of pictures were made: one for each gender. Both sets of pictures were made using the same home props, camera, film, lighting, and body positions. Each picture depicts a situation in which a child is in either a low- or high-risk environment, depending on environmental conditions and child behavior. Seven low-risk and 13 high-risk situations were constructed based on common child activities in the home. A low-risk situation depicted a child sitting on a carpeted living room floor while watching television. A high-risk situation involved the child holding a pot of boiling water by the handle on top of a hot stove. "Low-risk for injury" was defined as a situation in which the child depicted in the picture would likely be safe from injury if allowed to continue without any additional human intervention. Conversely, "high-risk for injury" was defined as a situation where the child would have a high chance of being hurt either minimally (e.g., small cut or bruise) or severely (e.g., injuries requiring hospitalization) even when it is believed that the child may not get hurt every time in that situation and no other human intervention occurred during the activity.
Specifically, children were asked to view each Injury Vulnerability Assessment of Children (IVAC) picture one at a time and indicate the likelihood that they would be injured if they were the child depicted in the picture using a Visual Analog Scale (VAS). Before viewing the pictures, each child completed a brief training exercise to ensure reliable results from the VAS, although no direct reliability estimates were conducted. Participants were told that they would be viewing pictures of a child "about your age" in the home. They were told to pretend that no one was around to see or hear them at the time. Then, the children were told to consider how safe or unsafe they would be if they were in the depicted situation. "Safe" and "unsafe" were operationally defined the same as low and high risk. Children would then make a mark on a horizontal line (i.e., the VAS), indicating the level of safety for that situation. The VAS was arranged to show extreme "safe" on the left and extreme "unsafe" on the right. Pictures were also placed as visual cues underneath the scale to remind children which way meant high or low safety. Specifically, going left to right (or from safe to unsafe), the pictures were (1) a child being held by the mother, (2) a happy face, (3) a person with a question mark over the head indicating being unsure or undecided, (4) a Band-Aid® bandage, and (5) a child in an arm cast.
Pilot testing with three nonexpert graduate students in clinical child psychology doctoral program indicated that they were each able to categorize pictures into two groups based on low- and high-risk definitions with 100% agreement. That is, each adult was asked to place a picture into a category labeled either high or low potential risk for injury. This indicated that each picture elicited the intended response (i.e., a risky situation was actually considered risky, which helped assess face validity). Data from this sample of children yielded Cronbach’s alpha estimates of .84 for scores across all 20 items. Dichotomizing pictures into low- and high-risk categories yielded internal reliability estimates (Cronbach’s alpha) for VAS scores of .60 and .81, respectively. Thus, only high-risk items were retained for future analyses involving child vulnerability, given the low-risk items showed poor internal reliability.
Child Observations
Observations of child behavior took place at a local university clinic in a room containing simulated home hazards and typical toys. The six "simulated" hazards included a pocket knife, pill container, soldering iron, cigarette lighter, cleaning spray bottle, and a hand gun. These hazards were not pointed out by the researcher. All objects were modified to present no risk of injury above that found for common toys. For example, the knife’s sharp edge and point were dulled with a grinder and the hand gun was a CO2 powered BB gun pistol with no ammunition in the room. The child was prompted about the presence of several typical toys in the room, which included a doll, Play-Doh® clay, a puzzle, Matchbox® cars, an Etch A Sketch® drawing toy, and paper and colored markers.
Procedure
Caregivers and their child arrived at the university clinic and were told that two interviewers would be working with them. Consent to participate had been obtained from the caregiver during an initial home visit in which demographic information was collected and additional unrelated measures were administered. The caregiver was first introduced to the simulated home hazard room and informed that videotaping would occur from a wall-mounted camera in view, while the child assented to the project in the clinic lobby. Each child then entered the simulated hazard room with the researcher, and the caregiver was left to complete the TABC-R in a separate room. The child was told to remain alone briefly in the room while the researcher left to retrieve more papers. The child was continuously monitored by the researcher through a two-way mirror and was videotaped for 15 min. Next, the researcher entered the simulated hazard room with the child to complete the IVAC and an additional unrelated measure. Once all measures were completed, caregivers were debriefed on their child’s behavior during observations and given specific strategies toward reducing potentially risky behaviors. Lastly, caregivers were provided with a coupon for a free pizza or equivalent food item.
|
Results |
|---|
|
TopAbstractMethodResultsDiscussionAcknowledgmentReferences |
|---|
Preliminary Analyses
The various dependent measures were examined for multicolinearity, nonnormalcy, and univariate or multivariate outliers (via z scores and Mahalanobis distances, respectively). Following Tabachnick and Fidell’s (1989) guidelines, we recoded outliers (i.e., scores greater than ±3 SD away from the mean of each variable) to the most extreme score within ±3 SD of the mean.
Perceptions of vulnerability scores for the IVAC were calculated by summing measurements across all items. That is, for each photo scene, the child provided a mark on the VAS, resulting in an individual item score ranging from 0 to 150 mm. Scores across all 13 risky photographs were then summed to create the final total vulnerability score. Higher scores indicate that the child felt more vulnerable toward receiving an injury; lower scores reflected a tendency to rate less vulnerability to risky situations, with a range of scores from 0 to 1950 mm (150 mm x 13 photo scenes).
Child observation scores were recorded during live child interactions within the simulated home environment. Risky behavior frequency was recorded for the entire 15 min the child was in the room. Risky behavior was defined as having occurred when the child made physical contact with any of the previously identified hazards for this study (e.g., the gun, knife, or lighter). The intent of the touch was not differentiated for this study. A participant’s risky behavior was calculated into a total frequency score by summing the occurrences across risky items. This study utilized the frequency of risky behavior as the criterion variable and caregiver-reported temperament and self-reported vulnerability to injury as predictor variables. Interobserver agreement was calculated by an independent observer reviewing a sample of videotaped observations (15%), dividing the number of agreements (i.e., total number of touched risky objects) by the number of agreements plus disagreements and multiplied by 100%. Agreement was 100% for all participants.
Sample Characteristics
As can be seen in Table I, this sample was predominantly Caucasian and belonged to the upper middle class, with nearly half of the sample (44%) reporting an income at or above $50,000. Most of the children lived in households with married caregivers. Caregivers’ average age was 35 years, and they had an average of two children in each family. Both caregivers were high-school graduates on average with some college credit. The average age of participating children was nearly 6 years, with approximately equal distribution groups of gender. During observations, the average number of hazards contacted was 1.78, with the following frequencies of interaction: (a) pocket knife, 29.9%; (b) hand gun, 28.6%; (c) medication, 14.3%; (d) cigarette lighter, 11.7%; (e) soldering iron, 13.0%; and (f) cleaning solution, 2.5%.
|
Associations Between Child Characteristics and Observed Risky Behavior
Examination of the Pearson correlation coefficients in Table II revealed that children were significantly more likely to indicate lower perceptions of vulnerability to injury as their actual risky behavior increased. As expected, boys were significantly more likely to engage in risky behavior. As child age increased, their perceptions of vulnerability ratings increased, indicating that older children tended to report greater feelings of vulnerability to injury compared to younger children. Children whose caregivers reported higher levels of activity on the TABC-R were significantly more likely to show increased risky behavior and report lower perceptions of vulnerability to injury.
|
Primary Analyses
It was hypothesized that children who indicated lower perceptions of vulnerability to injury were expected to display greater rates of actual risky behaviors during observations. A linear regression analysis was conducted using risky behavior as the criterion variable and child perceptions of vulnerability to injury as the predictor variable. As can be seen in Table III, the regression equation (Analysis 1) revealed that children’s perceptions of vulnerability scores significantly predicted risky behavior, R2 = .11, F(1, 39) = 4.56, p < .05. This finding indicates that lower IVAC scores (feelings of being invulnerable to injury) significantly predicted higher occurrences of risky behavior. Although gender was considered to be a possible interactive variable with our predictor variables (IVAC and TABC-AL), no significant findings were revealed during examinations. Thus, gender was utilized as a control variable during mediational analysis.
|
Mediational Analysis
Mediation analysis tested the hypothesis that child-based perceptions of vulnerability may partially or fully explain the revealed relationship between temperament and risky behavior. To test this mediation hypothesis, preliminary steps (Baron & Kenny, 1986; Holmbeck, 1997) were conducted to determine which variables met criteria as possible mediators. These steps included the traditional four steps of identifying that (a) the independent variable (IV) is significantly associated with the mediator, (b) the IV is significantly associated with the dependent variable (DV), (c) the mediator is significantly associated with the DV, and (d) the impact of the IV on the DV is reduced after controlling for the mediator. Additionally, child gender, but not age, was entered into the model as a control variable, given the significant relationship between gender and risky behavior. Although child age has been shown in the previous studies to be related to risk perceptions (e.g., Hillier & Morrongiello, 1998), it was not found to be a predictor of or associated with risky behavior and was thus not included in the mediation models as a control variable.
Only TABC-AL satisfied the above necessary relationships among the predictor (IV), mediator, and criterion variables (DV). TABC-AL was tested to be mediated by IVAC; although children with higher TABC-AL scores were observed to show greater risky behavior, this relationship may be mediated by the perception of vulnerability to injury. Perception of vulnerability was tested as a mediator of the relationship between temperament characteristics (TABC-AL) and risky behavior.
To test mediation effects of IVAC on TABC-AL scores, two hierarchical regression analyses were conducted alternating Blocks 2 and 3; Block 1 entered gender as a control variable. As can be seen in Table III, IVAC scores significantly predicted risky behavior over and above gender, 
R2 = .11, F(2, 36) = 4.67, p < .05, but when TABC-AL was added, after controlling for gender and IVAC, the TABC-AL scores remained a statistically significant predictor, R2 = .08, F(3, 35) = 4.67, p < .05, suggesting that children’s perceptions of vulnerability (IVAC scores) do not mediate the relationship between TABC-AL and risky behavior (Table III, Analysis 2). A follow-up Sobel test of the magnitude of change on path coefficients (recommended by Sobel, 1982; Holmbeck, 2002), utilizing the Goodman (II) version, further supported that no significant mediating effect of IVAC scores occurred on TABC-AL when predicting risky behavior, z = 1.26, p = .21.
|
Discussion |
|---|
|
TopAbstractMethodResultsDiscussionAcknowledgmentReferences |
|---|
This study provides some insight into children’s perceptions of vulnerability and actual behaviors of risk taking. A strength of this study lies in the use of real-time observation data and child-based perceptions of vulnerability to injury. Children’s perceptions of vulnerability to risky home situations were linked to their observed behaviors in a simulated home hazard environment. Children were significantly more likely to engage in observed risky behavior when they reported less vulnerability to becoming injured at home. Although a logical finding, such an empirical link between child self-report and observed behavior has been scarce in the literature.
Several limitations are present in this study. First, the sample size may have limited the ability to detect modest relationships that have been found in past injury-prevention investigations on cognitive factors related to risk appraisals (Hillier & Morrongiello, 1998). Indeed, a retrospective power analysis with this sample size and alpha set at .05 indicated that power was .70, below the suggested .80 mark (Cohen, 1977), when calculating multiple regressions using three predictor variables. Additionally, the transformation of outlying scores may have artificially increased the magnitude of the reported relationships, although comparisons between original and recoded scores showed very similar trends. Second, the observed risky behavior occurred in a simulated home setting as opposed to a natural setting. Although the room was designed to reflect a typical home environment (upholstered chairs, a table, toys, as well as the risky items), children may have responded differently after entering a university clinic as opposed to entering their own home. However, this setting did allow for control of the hazards and permitted direct observation and taping. Third, risky behavior was conservatively defined as the touch of an identified risky object during the elapsed time. Several children clearly displayed even more risky behavior by running their fingers across the dulled edge of the pocket knife or repeatedly pulling the trigger of the handgun. Although these behaviors are qualitatively different from a brief touch, the distinctions of safety may not be significant. The proximity of touch brings the child to a point where injury can occur. Thus, a conservative approach to categorizing risky behavior as any touch was deemed most appropriate. Differences in the level of the risky behaviors or intentions of the child may be important to ascertain in future research (although even well-intended hazard touching, such as moving it to a safer place, might result in injury).
The measure of child vulnerability to injury (IVAC) was constructed for this study. This initial study indicates acceptable internal estimates of reliability and face validity, given the mean scores on safe versus risky situations. This finding also supports past research that indicated that children can reliably show an ability to distinguish between safe and risky situations (Coppens, 1985; Hillier & Morrongiello, 1998).
The implications of this study enhance the important roles of both children and caregivers in preventing injuries in the home. Children whose caregivers reported them as having an active temperament demonstrated an inability to recognize the level of risk in typical home environments. Therefore, children should be taught, for example, that kitchens are dangerous places, which would potentially increase risk perception. In addition, because older children revealed higher levels of vulnerability but not a reduced level of risky behavior, future investigations are needed to identify why such knowledge may not translate to behavior change. Although modifying perceptions of vulnerability, as well as identifying at-risk temperaments for injuries, such as high activity levels, is important to consider when developing effective interventions, these strategies may be ineffective on their own given the relatively stable nature of temperament and limitations of cognitive therapy for younger children (Schwebel & Plumert, 1999). Thus, additional interventions are likely needed to supplement the above recommendations, one of the most important strategies being environmental modifications.
Modifying the environment has been effective at reducing injuries, sometimes requiring little active effort by individuals to attain safety benefits (Walton, 1982; Wilson, Baker, Teret, Shock, & Garbarino, 1991), and where pertinent, it is considered the preferred method of preventing injuries (DiLillo & Peterson, 2001). For example, gun hazards can be controlled by caregivers through the modification of the environment, such as placing guns in locked cabinets and using gun locks. These environmental changes will remain necessary parts of effective interventions for reducing child injuries as additional research continues to further identify possible psychological variables that mediate the link between temperament and risky behaviors.
|
Acknowledgment |
|---|
|
TopAbstractMethodResultsDiscussionAcknowledgmentReferences |
|---|
This research was funded in part by the Rebecca Routh Coon Injury Prevention Grant (now entitled The Lizette Peterson-Homer Injury Prevention Grant) by The Society of Pediatric Psychology.
|
References |
|---|
|
TopAbstractMethodResultsDiscussionAcknowledgmentReferences |
|---|
Baron, R. M., & Kenny, D. A. (1986). The mediator–moderator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173–1182.[CrossRef][Web of Science][Medline]
Bijur, P. E., Stewart-Brown, S., & Butler, N. (1986). Child behavior and accidental injury in 11,966 preschool children. American Journal of Diseases of Children, 140, 487–492.
Cataldo, M. F., Finney, J. W., Richman, G. S., & Riley, A. W. (1992). Behavior of injured and uninjured children and their caregivers in a simulated hazardous setting. Journal of Pediatric Psychology, 17, 73–80.
Centers for Disease Control and Prevention. (2000). Healthy People 2010. Retrieved August 10, 2003, from http://www.health.gov/healthypeople/document
Cohen, J. (1977). Statistical power analysis for the behavioral sciences (Rev. ed.). New York: Academic Press.
Cook, S., Peterson, L., & DiLillo, D. (1999). Fear and exhilaration in response to risk: An extension of a model of injury risk in a real world context. Behavior Therapy, 30, 5–15.
Coppens, N. M. (1985). Cognitive development and locus of control as predictors of preschoolers’ understanding of safety and prevention. Journal of Applied Developmental Psychology, 6, 43–55.
Coppens, N., & Gentry, L. (1991). Video analysis of playground injury risk situations. Research in Nursing and Health, 14, 128–136.
DiLillo, D., & Peterson, L. (2001). Prevention of childhood disorders. In C. E. Walker & M. C. Roberts (Eds.), Handbook of clinical child psychology (3rd ed., pp. 974–988). New York: Wiley.
Finney, J. W., Christophersen, E. R., Friman, P. C., Kalnins, I. V., Maddux, J. E., Peterson, L., et al. (1993). Society of Pediatric Psychology Task Force: Pediatric psychology and injury control. Journal of Pediatric Psychology, 18, 499–526.
Gable, S., & Peterson, L. (1998). School-age children’s attributions about their own naturally occurring minor injuries: A process analysis. Journal of Pediatric Psychology, 23, 323–332.
Gray, J. (1991). The neuropsychology of temperament. In J. Strelau & A. Angleitner (Eds.), Explorations in temperament: International perspectives on theory and measurement (pp. 105–128). New York: Plenum.
Guyer, B., Hoyert, D. L., Martin, J. A., Ventura, M. A., MacDorman, M. F., & Stobino, D. M. (1999). Annual summary of vital statistics 1998. Pediatrics, 104, 1229–1246.
Hillier, L., & Morrongiello, B. A. (1998). Age and gender differences in school-age children’s appraisals of injury risk. Journal of Pediatric Psychology, 23, 229–238.
Holmbeck, G. N. (1997). Toward terminological, conceptual, and statistical clarity in the study of mediators and moderators: Examples from the clinical-child and pediatric psychology literatures. Journal of Consulting and Clinical Psychology, 65, 599–610.[CrossRef][Web of Science][Medline]
Holmbeck, G. N. (2002). Post-hoc probing of significant moderational and mediational effects in studies of pediatric populations. Journal of Pediatric Psychology, 27, 87–96.
Martin, R. P., & Bridger, R. C. (1988). Temperament Assessment Battery for Children—Revised: A tool for the assessment of temperamental traits and types of young children. Brandon, VT: Clinical Psychology Publishing Co.
Matheny, A. P. (1987). Psychological characteristics of childhood accidents. Journal of Social Issues, 43, 45–60.
Morrongiello, B. A., & Rennie, H. (1998). Why do boys engage in more risk taking than girls? The role of attributions, beliefs, and risk appraisals. Journal of Applied Developmental Psychology, 20, 227–251.[CrossRef]
National Center for Injury and Prevention and Control. (2001). Injury fact book 2001–2002. Atlanta, GA: Centers for Disease Control and Prevention.
National Center for Injury and Prevention and Control. (2002). Injury research agenda. Atlanta, GA: Centers for Disease Control and Prevention.
Peterson, L., Farmer, J., & Mori, L. (1987). Process analysis of injury situations: A complement to epidemiological methods. Journal of Social Issues, 43, 33–44.
Plumert, J. M., & Schwebel, D. C. (1997). Social and temperament influences on children’s overestimation of their physical abilities: Links to accidental injuries. Journal of Experimental Child Psychology, 67, 317–337.[CrossRef][Web of Science][Medline]
Roberts, M. C. (1986). Health promotion and problem prevention in pediatric psychology: An overview. Journal of Pediatric Psychology, 11, 146–161.
Roberts, M. C., & Brooks, P. (1987). Children’s injuries: Issues in prevention and public policy. Journal of Social Issues, 43, 1–12.
Roberts, M. C., Elkins, P. D., & Royal, G. P. (1984). Psychological applications to the prevention of accidents and illness. In M. C. Roberts & L. Peterson (Eds.), Prevention of problems in childhood: Psychological research and applications (pp. 173–199). New York: Wiley.
Rosen, B., & Peterson, L. (1990). Gender differences in children’s outdoor play injuries: A review and integration. Clinical Psychology Review, 10, 275–294.
Scheidt, P. C., Harel, Y., Trumble, A. C., Jones, D. H., Overpeck, M. D., & Bijur, P. E. (1994). The epidemiology of nonfatal injuries among US children and youth. American Journal of Public Health, 85, 932–938.
Schwebel, D. C., & Plumert, J. M. (1999). Longitudinal and concurrent relations among temperament, ability estimation, and inquiry proneness. Child Development, 70, 700–712.[CrossRef][Web of Science][Medline]
Sobel, M. E. (1982). Asymptotic intervals for indirect effects in structural equations models. In S. Leinhart (Ed.), Sociological methodology 1982 (pp. 290–312). San Francisco: Jossey-Bass.
Tabachnick, B. G., & Fidell, L. S. (1989). Using multivariate statistics. New York: Harper Collins.
Walton, W. W. (1982). An evaluation of the Poison Prevention Packaging Act. Pediatrics, 69, 363–370.
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.
CiteULike 
Connotea 
Del.icio.us What's this?
This article has been cited by other articles:
|
|
 |
|
  C. L Duncan and A. G Dempsey Using Basic Behavior Research on Children's Emotions to Inform Prevention Research: A Commentary on Pooley and Fiddick Social Referencing "Mr. Yuk" J. Pediatr. Psychol., November 11, 2009; (2009) jsp101v1. [Full Text] [PDF]
|
|
|
|
|
|
R. E. Boles and M. C. Roberts Supervising Children During Parental Distractions J. Pediatr. Psychol., September 1, 2008; 33(8): 833 - 841. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||