The role of environmental hazard in mothers’ beliefs about appropriate supervision

Abstract

Understanding factors that influence mothers’ beliefs about appropriate levels of supervision for their children may assist in efforts to reduce child injury rates. This study examined the interaction of child (i.e. age, gender, and injury risk behavior) and maternal perception of environmental hazard (i.e. hazard level, injury likelihood, and potential injury severity) variables in predicting mothers’ beliefs about appropriate levels of supervision for their children. Participants were 58 mothers of 2- to 8-year-old community children who were interviewed about their beliefs regarding child injury risk and appropriate supervision. Results indicated that perceived environmental hazard level interacted with child age, gender, and injury risk behavior to predict mothers’ beliefs about supervision. Perceived injury likelihood also interacted with child injury risk behavior to predict beliefs about supervision. Findings underscore the complexity of factors influencing mothers’ beliefs about appropriate supervision and indicate the importance of environmental hazard level in such beliefs.

Keywords

child injury prevention maternal supervision

Unintentional childhood injuries pose a serious threat to the welfare of children in the United States. Indeed, unintentional injuries to children less than 19 years of age account for approximately 9.2 million emergency room visits and more than 12,000 deaths each year in the United States (Borse et al., 2008). Even when such injuries are nonfatal, they can result in permanent disability as well as long-term psychological distress (Le Brocque et al., 2010; Leech et al., 2006).

Given the negative impact of child unintentional injury in the United States, much research has focused on the correlates of child hood injury risk. Several recent studies have indicated that closer caregiver supervision is related to lower risk and frequency of child hood unintentional injury (Damashek et al., 2009; Morrongiello et al., 2009a; Morrongiello and House, 2004; Pless et al., 1989; Wills et al., 1997). Moreover, some studies have found that caregiver supervision may mitigate the effect of child behavior problems on risk of injury (Morrongiello et al., 2008; Schwebel et al., 2004).

Given that research has highlighted the potentially protective role of supervision in preventing children’s injuries, it is important to understand which factors influence mothers’ beliefs about appropriate levels of supervision for their children. If mothers have inaccurate beliefs about child injury risk and the need for supervision, perhaps interventions can be designed to target such beliefs. To date, there is a small body of literature examining factors (i.e. child and environmental variables) influencing mothers’ beliefs about supervision; however, there are still several gaps in the knowledge base, such as how such factors might interact to predict maternal supervision beliefs. Researchers have suggested that injuries occur as the result of the interaction of a number of factors (Morrongiello and Schwebel, 2008), suggesting that decisions about how to prevent injuries (i.e. supervising) should take these interactions into account. However, researchers have not examined how such factors interact to predict mothers’ supervision beliefs.

A small body of research (e.g. Garling and Garling, 1993; Peterson et al., 1991, 1993; Soori and Bhopal, 2002) has indicated that child characteristics (i.e. age, gender, and risky behavior) are associated with mothers’ supervision beliefs and behaviors. In particular, mothers report that more supervision is needed for younger children (Peterson et al., 1993) and have been found to provide closer supervision to younger children (2- to 3-year-olds) compared to older children (i.e. 4- to 5-year-olds; Morrongiello et al., 2006a). With regard to gender, research has found that caregivers believe that more supervision is needed for girls than boys and that mothers tend to promote less independence among girls than boys (Basow, 1986; Block, 1983; Morrongiello et al., 2009b; Soori and Bhopal, 2002). Other research has demonstrated that mothers supervise children who exhibit greater injury risk behavior more closely than those with greater behavioral control (Morrongiello et al., 2006b). However, research on the relation between children’s risky behavior and mothers’ beliefs about appropriate supervision is lacking.

Research has also been conducted on the role of the environment in mothers’ beliefs about supervision of their young children. To date, research has found that mothers report that young children are at greatest risk of injury in situations with dangerous environmental hazards (e.g. in a kitchen; Garling and Garling, 1993; Peterson et al., 1991, 1993) and that a higher level of supervision is needed for children in such situations (Morrongiello et al., 2009b; Peterson et al., 1991). Moreover, research examining mothers’ perceptions of children’s injury susceptibility and potential injury severity found that such factors are related to mothers’ general injury prevention behaviors (Morrongiello and Kiriakou, 2004) as well as supervision (Brown et al., 2005).

Findings from these studies provide preliminary evidence regarding the factors that are related to mothers’ beliefs about appropriate levels of supervision for their children. However, deciding how closely to supervise one’s children is a complex process that requires mothers to simultaneously take into account several factors, including environmental (hazard level, potential injury severity, and injury likelihood) and child variables (age, gender, and risky behavior). Previous research has only examined these factors in isolation, without taking into account potential interactions between child and environmental variables in mothers’ beliefs about appropriate levels of supervision. It is likely that the effect of child variables (i.e. age, gender, and risky behavior) on mothers’ beliefs about appropriate supervision may be moderated by the hazard level of the environment. Indeed, in a highly hazardous environment (e.g. near a lake), age may become less salient than in a less hazardous environment (e.g. a living room).

This study will examine the effect of interactions between child (i.e. age, gender, and injury risk behavior) and maternal perception of environmental hazard (i.e. perceived hazard level, perceived likelihood of injury, and perceived potential injury severity) variables on mothers’ beliefs about appropriate levels of supervision for their children.

Method

Participants

Participants were 58 mothers of 2-year-old (n = 17; 10 girls and 7 boys), 5-year-old (n = 22; 11 girls and 11 boys), and 8-year-old (n = 19; 10 girls and 9 boys) children. Mothers ranged in age from 19 to 51 years (M = 35.5, standard deviation (SD) = 6.8). Most (82.8%) of the mothers were Caucasian, and the rest were African American (12.0%), Asian American (1.7%), or of other ethnicity (3.5%). Mothers were primarily (82%) from middle- to upper-middle-class backgrounds (classes III and IV; Hollingshead, 1975). The majority (75.9%) of the mothers were married, 12.1 percent were divorced, 10.3 percent were never married, and 1.7 percent were living with a boyfriend. Most (70.7%) of the mothers had more than one child.

Mothers were excluded from the study if they (a) were not proficient in English; (b) reported that their child had a developmental (e.g. autism) or physical (e.g. blindness) disability; or (c) reported that their child had been hospitalized overnight as the result of an injury. The first and second criteria were used to exclude mothers who might have had difficulty understanding interview questions or whose supervision beliefs might have been biased by their child’s disability status, respectively. The third criterion was used to exclude mothers whose beliefs about supervision might have been unduly influenced by experience with a severe injury that occurred to their child.

Mothers were recruited from local daycare centers (n = 14), elementary schools (n = 41), and by word of mouth (n = 3). Daycare mothers were approached in person and then contacted by telephone, whereas the remaining participants were originally contacted via telephone. Of the mothers (n = 104) who were initially contacted, 24 (23%) declined to participate and 7 (6.7%) were found to be ineligible (e.g. child had been hospitalized for an injury). In addition, 15 (14.4%) participants withdrew or were dropped from the study following several consecutive failures to attend their appointments. Mothers received US$15 as compensation for participation in the study.

Procedure

Mothers participated in a 1-hour interview in their homes. Interviewers were three undergraduate research assistants in psychology. Each interviewer received approximately 15 hours of training prior to contact with the mothers. Interviews were conducted in families’ homes and took approximately 1 hour to complete. At the beginning of the session, the interviewer reviewed the general purpose and procedure of the study, explained confidentiality, and obtained written consent from the mother. The interviewer then administered a demographic questionnaire and a measure assessing the mother’s supervision beliefs, as well as perceptions of perceived hazard level in several situations that are depicted in photographs (see the section “Measures”). All procedures were approved by the Institutional Review Board of the University of Missouri

Measures

Demographics

A demographic questionnaire was used to assess children’s age and gender.

Child injury risk behavior (Injury Behavior Checklist)

The Injury Behavior Checklist (IBC) was used to assess children’s risky behavior. The IBC is a 24-item measure assessing the amount of risky behaviors in which young children engage. Participants rate the frequency of several child risk behaviors (e.g. “runs out into the street”) on a 0–4 scale. The measure has high internal reliability (α = .87) and has been found to be a valid predictor of child injury for preschool and school-aged children (Potts et al., 1997; Speltz et al., 1989). This measure is frequently used in studies examining children’s unintentional injuries (see Damashek et al., 2005; Morrongiello and Sedore, 2005; Schwebel et al., 2002). A total scale score was used for the present analyses. Coefficient α for the present sample was .88.

Mothers’ beliefs about appropriate levels of supervision

The Situational Supervision Questionnaire (SSQ) was developed for this study to assess mothers’ beliefs about appropriate levels of supervision and perceived environmental hazard in potential injury situations (see Appendix 1). The SSQ comprises eight 21.59 cm × 27.94 cm color photographs, each depicting a different scene in which young children are likely to spend time. The SSQ photographs were identified based on ratings from two experts on childhood injury. Each expert initially rated 27 photographs in terms of hazard level (i.e. low, high, or undecided) and also indicated whether a child would likely be present in each scene. Four scenes that both experts agreed were high risk and four scenes that they agreed were low risk were selected for this study. Each of the eight selected scenes was identified by both injury experts as one in which children would typically be present. The pictures identified as high-risk scenes included a living room with an open fireplace, an outdoor swimming pool, a laundry room with a plugged-in iron, and a stairway. The pictures identified as low-risk scenes included a kitchen table and chairs (with no hot or sharp objects present), a children’s area in a library, a child’s playroom, and a living room with two couches and a lamp.

The SSQ was administered by asking mothers questions about each picture. Mothers first answered questions regarding their beliefs about appropriate supervision. For each photograph, mothers were asked the following question: “How much supervision does your child require when he/she is in this situation?” The data collectors then read a 7-point Likert scale (1 = “very little” to 7 = “constant”) and asked the mothers to rate the amount of supervision that their child would require in that situation (see Appendix 1 for a more detailed description of the scale anchors). Mothers were instructed to base their ratings only on the objects that were present in each photograph. Internal consistency (α) for mothers’ supervision ratings was .91 across all photographs.

Maternal perception of environmental hazard level

Perception of environmental hazard level was measured in three ways using the SSQ. Based on previous research indicating that mothers’ supervision is related to both perception of injury likelihood and potential injury severity (Brown et al., 2005), we used the SSQ to assess both of these factors. We also gathered information on mothers’ overall perception of hazard level for each SSQ photograph (i.e. low vs high).

After reporting on the level of supervision mothers believed was needed for all eight photographs, mothers then answered questions about injury likelihood and potential injury severity. In order to help mothers envision what injury might occur, they were first asked, “what is the most likely injury to happen to your child in this situation?” Then, to determine mothers’ perception of injury likelihood, they were asked, “how likely is that injury to occur?” and were provided with a 1–7 Likert scale (1 = “not at all likely,” 3 = “somewhat likely,” 5 = “pretty likely,” and 7 = “very likely”). The value of α for perceived injury likelihood across all photographs was .77.

Afterward, mothers were also asked about their perceptions of potential injury severity. The question was worded as follows: “How serious is that injury?” Mothers responded using a 7-point Likert scale (1 = “slightly serious” to 7 = “very serious;” see Appendix 1 for a detailed description of the scale anchors). The value of α for perceived potential injury severity across all photographs was .78.

Finally, at the end of each interview, mothers were asked to sort the pictures into piles to indicate the level of environmental hazard that the pictures portrayed. They sorted the pictures into “low hazard,” “high hazard,” and “uncertain” categories. Mothers’ hazard ratings agreed with expert ratings 78.7 percent of the time. For the following analyses, we omitted instances in which mothers rated hazard level as “uncertain.”

Results

Descriptive statistics

Means for continuous variables can be seen in Table 1. Mean IBC scores were similar to the reported means in studies of similarly middle–upper income families (Damashek et al., 2005; Morrongiello et al., 2006b; Speltz et al., 1989). Mothers’ mean supervision, perceived injury likelihood, and potential injury severity scores were significantly higher for situations that mothers categorized as high-hazard versus low-hazard.

Table 1.

Means of child, perceived environmental hazard variables, and maternal beliefs about supervision.

Variable	Overall, M (SD)	Possible range	Low-hazard situations, M (SD)	High-hazard situations, M (SD)	t statistic
Child age	5.10 (2.38)	2–8	–	–	–
IBC score	22.68 (11.23)	0–96	–	–	–
Perceived injury likelihood	2.58 (1.29)	1–7	2.26 (1.17)	2.88 (1.40)	−4.73^*** (n = 405)
Perceived potential injury severity	2.25 (1.83)	1–7	1.65 (1.32)	3.05 (2.15)	−7.67^*** (n = 405)
Maternal supervision ratings	4.79 (2.06)	1–7	3.94 (1.95)	6.01 (1.60)	−11.77^*** (n = 406)

M: mean; SD: standard deviation; IBC: Injury Behavior Checklist.

p < .05. **p < .01. ***p < .001.

Bivariate analyses

Bivariate analyses were used to assess the relation between child variables, variables indicating maternal perception of hazard, and mothers’ beliefs about supervision. T-tests were used to examine the relation of child gender to injury risk behavior, maternal perceived injury likelihood and potential severity, and maternal beliefs about supervision. Average injury likelihood, potential injury severity, and supervision scores were used for each mother because there were repeated measures of these variables for the eight photographs. Perceived potential injury severity was the only variable that differed significantly for boys versus girls. Mothers reported higher average potential injury severity scores for boys (M = 2.58, SD = 1.06) than for girls (M = 1.95, SD = 0.90; t(56) = 2.4, p = .02; Cohen’s d = .64).

Correlations were used to examine relations between continuous child variables (i.e. child age and child injury risk behavior), maternal perceptions of environmental hazard (i.e. injury likelihood and potential injury severity), and maternal beliefs about supervision. Average scores for perceived injury likelihood and potential severity were used because mothers had repeated measures of these variables for each photograph. Average supervision scores were negatively correlated with children’s age (r = −.64, p = .0001), but not with children’s injury risk behavior. There was a negative association between injury risk behavior and age (r = −.29, p = .03) such that younger children reportedly displayed more injury risk taking behavior. In addition, mothers reported a higher likelihood of injury for children who reportedly engaged in greater risk taking behavior (r = .49, p < .0001). Finally, there was a trend for mothers to report greater injury likelihood for younger children (r = −.23, p = .09).

Predicting maternal supervision beliefs

When predicting supervision, multilevel modeling was used rather than ordinary least squares regression because of the nested structure of the data. Due to the fact that mothers reported on beliefs about supervision, perceived environmental hazard level, injury likelihood, and potential injury severity for each photograph, we had eight data points for each mother for each of these variables. The presence of repeated measurements breaks the ordinary least squares assumption of independence among observations; however, multilevel modeling was designed to handle this type of data structure (Raudenbush and Bryk, 2002).

We first examined the main effects of the child and three perceived hazard variables. As can be seen in Table 2, younger child age predicted higher maternal supervision scores. In addition, each of the three perceived hazard variables significantly predicted beliefs about appropriate levels of supervision; however, maternal perception of overall hazard level was the stronger predictor, followed by maternal perception of potential injury severity and then injury likelihood.

Table 2.

Predicting maternal supervision beliefs from child variables and maternal perception of hazard.

Variable	b (SE)	SE
Child age	−0.33^***	0.07
Child gender	0.30	0.30
IBC score	0.00	0.01
Perceived hazard level (low, high)	0.62^***	0.11
Perceived injury likelihood	0.17**	0.07
Perceived potential injury severity	0.22^***	0.04
Random components/variance
Intercept (level 2)	1.04^***	0.25
Residual (level 1)	2.17^***	0.15

SE: standard error; IBC: Injury Behavior Checklist.

N for number of subjects = 58; N for number of observations = 463; for maternal perceived hazard level, 0 = low and 1 = high; for child gender, 0 = male, 1 = female.

p < .05. **p < .01. ***p < .001.

Next, we examined interactions between environmental variables (perceived hazard level, perceived injury likelihood, and perceived potential injury severity) and child variables. As can been seen in Table 3, there were significant interactions between perceived hazard level and all three child variables (i.e. child age, child gender, and child injury risk behavior). Each interaction will be discussed below.

Table 3.

Predicting maternal supervision beliefs from interactions between child variables and maternal perception of hazard variables (maternal perceived hazard level (PHL), perceived injury likelihood (PIL), and perceived potential injury severity (PPIS)).

Variable	b (SE)	Variable	b (SE)	Variable	b (SE)
Child age	−0.47 (0.07)^***	Child age	−0.34 (0.06)^***	Child age	−0.32 (0.07)^***
Child gender	0.39 (0.31)	Child gender	0.13 (0.29)	Child gender	0.24 (0.30)
IBC score	−0.01 (0.01)	IBC score	0.00 (0.01)	IBC score	0.02 (0.01)
PHL (low, high)	−0.03 (0.47)	PIL	0.41 (0.08)^***	PPIS	0.32 (0.05)^***
Child age* PHL	0.32 (0.06)^***	Child age^* PIL	0.03 (0.03)	Child age* PPIS	0.04 (0.02)⁺
Child gender^* PHL	−0.67 (0.26)^**	Child gender^* PIL	0.04 (0.14)	Child gender^* PPIS	−0.05 (0.10)
IBC score* PHL	0.04 (0.01)^**	IBC score* PIL	−0.01 (0.01)^*	IBC score* PIL	0.00 (0.00)
Random components/variance
Intercept (level 2)	1.01 (0.24)^***	Intercept (level 2)	0.85 (0.24)^***	Intercept (level 2)	0.91 (0.25)^***
Residual (level 1)	1.51 (0.11)^***	Residual (level 1)	2.56 (0.20)^***	Residual (level 1)	2.44 (0.19)^***

SE: standard error; IBC: Injury Behavior Checklist.

N for number of subjects = 58; N for number of observations = 464; for perceived hazard level, 0 = low and 1 = high; for child gender, 0 = male, 1 = female.

p < .10; *p < .05; **p < .01; ***p < .001.

See Figure 1 for a graphic depiction of the interaction between perceived hazard level and child age. Examination of the interaction indicates that in both low- and high-hazard situations, mothers reported that they would supervise younger children more closely. Overall supervision scores were higher in high-hazard situations; however, the effect of age was stronger for low-hazard situations (b = −0.47) than for high-hazard situations (b = −0.15). Visual inspection of Figure 1 shows a steeper slope for the low-hazard line.

Figure 1.

Effect of child age by mothers’ perception of hazard level in predicting mothers’ reports of appropriate levels of supervision.

As shown in Figure 2, examination of the simple slopes for the interaction between child gender and perceived maternal hazard level indicates that higher levels of appropriate supervision were reported for both boys and girls in high-hazard situations; however, there was a stronger relationship between hazard level and supervision for boys (b = 2.4) than for girls (b = 1.7). Thus, the effect of hazard level appeared to have a much greater effect on mothers’ beliefs about appropriate levels of supervision for boys than for girls.

Figure 2.

Effect of child gender by mothers’ perceived hazard level in predicting mothers’ reports of appropriate levels of supervision.

Examination of the simple slopes for the interaction between injury risk behavior and perceived hazard level indicates that mothers reported a positive relation between injury risk behavior and supervision in both low (b = .02) and high (b = .04) hazard situations; however, the relation between injury risk behavior and supervision beliefs was slightly stronger in high-hazard situations (see Figure 3). In other words, mothers reported that more supervision was needed for children with higher levels of injury risk behavior, particularly in high-hazard situations.

Figure 3.

Effect of child injury risk behavior (Injury Behavior Checklist (IBC)) by mothers’ perception of hazard level in predicting mothers’ reports of appropriate levels of supervision.

As Table 3 shows, additional analyses were conducted for interactions between child variables and the two other variables representing mothers’ perception of environmental hazard (i.e. injury likelihood and potential injury severity). Injury risk behavior interacted with perceived injury likelihood, and there was a trend toward a significant interaction between child age and potential injury severity. We graphed the interaction of injury risk behavior by perceived injury likelihood using recommendations by Aiken and West (1991) for examining interactions between continuous variables. As can be seen in Figure 4, supervision ratings were highest when perceived injury likelihood was high, regardless of IBC level. However, when injury likelihood was low, mothers reported that children with high IBC scores needed more supervision than those with mean or low IBC scores.

Figure 4.

Effect of child injury risk behavior (Injury Behavior Checklist (IBC)) by mothers’ perception of injury likelihood in predicting mothers’ reports of appropriate levels of supervision.

Discussion

Although maternal supervision has been found to protect young children from unintentional injuries, our understanding of factors related to mothers’ beliefs about appropriate levels of supervision is still somewhat limited. In particular, little is known about the way in which child variables interact with mothers’ perceptions of environmental hazard to influence mothers’ beliefs about appropriate levels of supervision for their children. This study addressed this gap in the literature by examining how child (i.e. age, gender, and injury risk behavior) and perceived environmental hazard level interact to predict mothers’ beliefs about supervision. We found that mothers’ perceptions of environmental hazard level moderated the effect of three child variables (i.e. age, gender, and injury risk behavior) on mothers’ beliefs about supervision. Mothers’ perceptions of injury likelihood also interacted with child injury risk behavior to predict beliefs about supervision.

When examining these interactions, we found that the effect of child age was weaker in high-hazard situations, indicating that when mothers believe that the environment is hazardous, even older children need greater levels of supervision. Conversely, in low-hazard situations, age is a more salient predictor of beliefs about appropriate levels of supervision. These findings may be due to mothers’ accurate understanding that low-hazard environments may pose greater risks for younger than older children because younger children have lower levels of physical coordination and greater impulsivity than do older children (Hill et al., 2000; Zeedyk et al., 2002). However, highly hazardous situations may pose a risk to older children as well.

With regard to child gender and environment, we found that mothers endorsed higher levels of supervision for both boys and girls in high-hazard situations; however, the effect of hazard level was stronger for boys than it was for girls. Thus, mothers’ beliefs about appropriate levels of supervision for boys were influenced more strongly by the environmental hazard level than were mothers’ beliefs about appropriate supervision levels for girls. Perhaps this reflects a belief that boys are more likely to approach hazards than girls. Indeed, previous research has found that mothers expected sons to engage in more risky behavior than daughters (Morrongiello and Hogg, 2004). Moreover, such a belief would be consistent with literature indicating that boys are more likely to approach and touch hazards than girls (Morrongiello and Dawber, 1998).

Finally, mothers reported the need for higher levels of supervision for children with higher levels of injury risk behavior, particularly in situations viewed as more hazardous. Perhaps mothers believe that children with increased risky behavior are also more likely to approach and interact with hazards than children with less injury risk behavior. Interestingly, the opposite pattern was found when examining the interaction between injury risk behavior and perceived injury likelihood. This latter interaction indicated that mothers believed that in situations in which injury likelihood was high, all children needed similar levels of supervision and that in low injury likelihood situations, children with high injury risk behavior needed the most supervision. Perhaps these disparate findings emerged because injury likelihood may not be a pure measure of environmental hazard level. In making ratings about injury likelihood, mothers might take into account several factors, including environmental hazard level, children’s injury risk behavior, and children’s age.

The present findings provide more information about the complexity that may underlie mothers’ beliefs about appropriate levels of supervision for their young children. The beliefs of mothers in the present sample appear to be consistent with research indicating that factors leading to child injury are multifaceted and complex (Morrongiello and Schwebel, 2008). Further research examining the role of interactions between child and environmental hazard level in predicting injury risk would help us to better understand how mothers should make decisions about how to supervise their children. Indeed, research examining such interactions is limited.

Some limitations of this study should be noted. First, we only examined mothers’ beliefs about supervision, rather than their actual supervisory behaviors. It is possible that mothers may not actually engage in the same levels of supervision that they endorsed in this study, due to a range of factors such as time constraints and distractions. Examining whether child and environmental variables interact to influence mothers’ actual supervisory behaviors would be an important next step. Moreover, research that includes multiple measures of maternal behavior (i.e. self reports and direct observation) may indicate whether mothers’ beliefs about injury risk are actually related to their supervision behaviors.

In addition, because the mothers included in this study were predominantly White and middle class, the results may not generalize to other populations (e.g. low-income mothers or mothers from other ethnic backgrounds). Given that lower socioeconomic status (SES) has been found to be a risk factor for injury (Faelker et al., 2000; Hippisley-Cox et al., 2002), it seems particularly important to examine factors related to supervision in low-income mothers. It would be interesting to compare beliefs about supervision across different social strata and ethnic groups. Such research may allow us to develop more targeted interventions for particular at-risk groups.

Findings from this study may have implications for educating mothers about appropriate levels of supervision. Indeed, although previous interventions have been successful in improving caregivers’ use of home safety devices (Gershater-Molko et al., 2002; Llewellyn et al., 2003), few systematic attempts have been made to help mothers modify their supervisory behavior. Findings such as the ones from this study, that clarify the relation of mothers’ beliefs about injury risk to their beliefs about appropriate supervision, can provide information about which cognitions should be targets for intervention. Mothers in this study had relatively realistic beliefs about the relation of child injury risk to appropriate levels of supervision for their young children. However, future studies examining mothers’ beliefs about injury risk and supervision in populations at higher risk of injury (e.g. families from a low SES background) would be particularly beneficial for developing interventions to prevent childhood injury.

Footnotes

Appendix 1 Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.

References

Aiken

West

(1991) Multiple Regression: Testing and Interpreting Interactions. Thousand Oaks, CA: SAGE.

Basow

(1986) Gender Stereotypes: Traditions and Alternatives. 2nd ed. Monterey, CA: Brooks/Cole.

Block

(1983) Sex Role Identity and Ego Development. San Francisco, CA: Jossey-Bass.

Borse

Gilchrist

Dellinger

. (2008) CDC Childhood Injury Report: Patterns of Unintentional Injuries among 0–19 Year Olds in the United States, 2000–2006. Atlanta, GA: Centers for Disease Control and Prevention, National Center for Injury Prevention and Control.

Brown

Roberts

Mayes

. (2005) Effects of parental viewing of children’s risk behavior on home safety practices. Journal of Pediatric Psychology 30: 571–580.

Damashek

Williams

Sher

. (2009) Relation of caregiver alcohol use to unintentional childhood injury. Journal of Pediatric Psychology 34: 344–353.

Damashek

Williams

Sher

. (2005) Risk for minor childhood injury: an investigation of maternal and child factors. Journal of Pediatric Psychology 30(6): 469–480.

Faelker

Pickett

Brison

(2000) Socioeconomic differences in childhood injury: a population based epidemiologic study in Ontario, Canada. Injury Prevention 6: 203–208.

Garling

(1993) Mothers’ supervision and perception of young children’s risk of unintentional injury in the home. Journal of Pediatric Psychology 18: 105–114.

10.

Gershater-Molko

Lutzker

Wesch

(2002) Project SafeCare: improving health, safety, and parenting skills in families reported for, and at-risk for child maltreatment. Journal of Family Violence 18: 377–386.

11.

Hill

Lewis

Dunbar

(2000) Young children’s concepts of danger. British Journal of Developmental Psychology 18: 103–119.

12.

Hippisley-Cox

Groom

Kendrick

. (2002) Cross sectional survey of socioeconomic variations in severity and mechanisms of childhood injuries in Trent 1992-7. British Medical Journal 324: 1132–1137.

13.

Hollingshead

(1975) The Four-Factor Index of Social Status (Unpublished manuscript). New Haven, CT: Yale University.

14.

Le Brocque

Hendrikz

Kenardy

(2010) The course of posttraumatic stress in children: examination of recovery trajectories following traumatic injury. Journal of Pediatric Psychology 35: 637–645.

15.

Leech

Larkby

Day

. (2006) Predictors and correlates of high levels of depression and anxiety symptoms among children at age 10. Journal of the American Academy of Child and Adolescent Psychiatry 45: 223–230.

16.

Llewellyn

McConnell

Honey

. (2003) Promoting health and home safety for children of parents with intellectual disability: a randomized controlled trial. Research in Developmental Disabilities 24: 405–431.

17.

Morrongiello

Dawber

(1998) Toddlers’ and mothers’ behaviors in an injury-risk situation: implications for sex differences in childhood injuries. Journal of Applied Developmental Psychology 19: 625–639.

18.

Morrongiello

Hogg

(2004) Mothers’ reactions to children misbehaving in ways that can lead to injury: implications for gender differences in children’s risk taking and injuries. Sex Roles 50: 103–118.

19.

Morrongiello

House

(2004) Measuring parent attributes and supervision behaviors relevant to child injury risk: examining the usefulness of questionnaire measures. Injury Prevention 10: 114–118.

20.

Morrongiello

Kiriakou

(2004) Mothers’ home-safety practices for preventing six types of childhood injuries: what do they do, and why? Journal of Pediatric Psychology 29: 285–297.

21.

Morrongiello

Schwebel

(2008) Gaps in childhood injury research and prevention: what can developmental scientists contribute? Child Development Perspectives 2: 78–84.

22.

Morrongiello

Sedore

(2005) The influence of child attributes and social-situational context on school-age children’s risk taking behaviors that can lead to injury. Journal of Applied Developmental Psychology 26: 347–361.

23.

Morrongiello

Corbett

Brison

(2009a) Identifying predictors of medically-attended injuries to young children: do child or behavioural attributes matter? Injury Prevention 15: 220–225.

24.

Morrongiello

Corbett

McCourt

. (2006a) Understanding unintentional injury risk in young children I. The nature and scope of caregiver supervision of children at home. Journal of Pediatric Psychology 31: 529–539.

25.

Morrongiello

Corbett

McCourt

. (2006b) Understanding unintentional injury risk in young children II. The contribution of caregiver supervision, child attributes, and parent attributes. Journal of Pediatric Psychology 31: 540–551.

26.

Morrongiello

Klemencic

Corbett

(2008) Interactions between child behavior patterns and parent supervision: implications for children’s risk of unintentional injury. Child Development 79: 627–638.

27.

Morrongiello

Walpole

McArthur

(2009b) Young children’s risk of unintentional injury: a comparison of mothers’ and fathers’ supervision beliefs and reported practices. Journal of Pediatric Psychology 34: 1063–1068.

28.

Peterson

Cook

Little

. (1991) “Mom lets me go there:” the role of environment and supervision in children’s minor injuries. Children’s Environments Quarterly 8: 15–23.

29.

Peterson

Ewigman

Kivlahan

(1993) Judgments regarding appropriate child supervision to prevent injury: the role of environmental risk and child age. Child Development 64: 934–950.

30.

Pless

Verreault

Tenina

(1989) A case-control study of pedestrian and bicyclist injuries in childhood. American Journal of Public Health 79: 995–998.

31.

Potts

Martinez

Dedmon

. (1997) Brief report: cross-validation of the Injury Behavior Checklist in a school-age sample. Journal of Pediatric Psychology 22: 533–540.

32.

Raudenbush

Bryk

(2002) Hierarchical Linear Models: Applications and Data Analysis Methods. 2nd ed. Thousand Oaks, CA: SAGE.

33.

Schwebel

Brezausek

Ramey

. (2004) Interactions between child behavior patterns and parenting: implications for children’s unintentional injury risk. Journal of Pediatric Psychology 29: 93–104.

34.

Schwebel

Speltz

Jones

. (2002) Unintentional injury in preschool boys with and without early onset of disruptive behavior. Journal of Pediatric Psychology 27(8): 727–737.

35.

Soori

Bhopal

(2002) Parental permission for children’s independent outdoor activities: implications for injury prevention. European Journal of Public Health 12: 104–109.

36.

Speltz

Gonzales

Sulzbacher

. (1989) Assessment of injury risk in young children: a preliminary study of the Injury Behavior Checklist. Journal of Pediatric Psychology 15: 373–383.

37.

Wills

Tanz

Christoffel

. (1997) Supervision in childhood injury cases: a reliable taxonomy. Accident Analysis and Prevention 29: 133–137.

38.

Zeedyk

Wallace

Spry

(2002) Stop, look, listen and think? What young children really do when crossing the road. Accident Analysis and Prevention 34: 43–50.