In Their Own Words

Abstract

Injury is a significant public health problem among youth. A primary cause of adolescent injury is risk-taking behavior, including alcohol use, interpersonal violence and road-related risks. A novel approach to prevention is building on friendships by encouraging adolescents to intervene into their friends’ risk taking. Fifty-one early adolescents (13-14 years) and 44 older adolescents (16-17 years) from two Australian schools participated in focus groups, aiming to explore stories of intervening. Findings showed preference for talking to friends; however, participants also spoke to adults, monitored friends’ behavior and planned ahead. Close friendships, perceived harm, and self-efficacy influenced the likelihood of intervening. These findings have implications for the design of risk and injury prevention programs, by suggesting strategies to promote adolescents’ communicative ability for risk reduction. The findings also highlight the language and dialogue of adolescents and suggest that methods for increasing intervening behavior should focus on building social connectedness and increasing self-efficacy.

Keywords

adolescents health promotion communication risk behaviors Australia

Injuries are the leading cause of fatalities among young people. High rates of injury are typically attributed to the fact that adolescents have high rates of involvement in risky behaviors (Buckley, Chapman, & Sheehan, 2012; Pickett et al., 2002). Although adolescent peers are commonly acknowledged as risk factors for such engagement, friends might also provide a protective influence by intervening and trying to stop risky behavior (Syversten, Flanagan, & Stout, 2009). There is little research, however, that examines intervening behavior and how this is experienced for young people. The purpose of this exploratory study is to understand the strategies that adolescents use to protect their friends in the context of risk-taking behaviors, and to understand contextual factors that facilitate intervening behavior as described by the early adolescents.

Adolescent Injury and Risk-Taking Behavior

In 2004-2005, almost a quarter of young Australians reported having sustained an injury in the previous 4 weeks (Australian Bureau of Statistics [ABS], 2008a). There is substantial evidence that adolescent engagement in violence, alcohol use, and road-related activities contributes significantly to their injury. A U.S. national survey found that 4% of 14- to 15-year-old students (9th grade) and 3% of 17- to 18-year-old students (12th grade) reported having been injured in a physical fight (Centers for Disease Control and Prevention, 2010). Australian research has shown 13% of early adolescents reported an injury experience associated with alcohol, 20% reported an injury experience with a motorcycle or quadbike, and 13% as vehicle passenger over a 6-month period (Chapman, Buckley, & Sheehan, 2011). Furthermore, adolescents identify such risk behaviors as injury-causing. Buckley and Sheehan (2010) conducted qualitative research regarding adolescents’ definitions of risk-taking behaviors and found that young people aged 14 to 17 years were likely to report violence risks, for example, getting into fights and using weapons as causes of their, and other young person’s injuries. They also reported a number of risk behaviors around the road, for example, being a passenger of a drink driver, riding a motorcycle, and cycling. In addition, students noted alcohol use was risky and likely to cause injury including in violence and transport scenarios. There can be considerable injury consequences, as well as educational and criminal consequences for engaging in risk taking, suggesting a significant need to understand and prevent such health compromising behaviors. Furthermore, these consequences are recognized by young people.

Adolescent Protective Behavior

Research has shown the risk-taking behavior of adolescents is a social activity and is influenced by peers (Gerrard, Gibbons, Vande Lune, Pexa, & Gana, 2002). Early adolescents, who are more susceptible to peer pressure, have friends who engage in risk-taking behaviors and believe their friends expect them to engage in such behaviors (Zakrajsek & Shope, 2006). A systematic review by Brechwald and Prinstein (2011) suggested frequent interactions, social roles, motivations (identity), and feedback play a role in influence and that conforming is associated with higher status, matched norms, external reinforcement, and intrinsic rewards. Thus, there is opportunity for a friend to be physically present and have an awareness that an adolescent is about to engage in risk-taking behavior. For example, Chen, Elliott, Durbin, and Winston (2005) and Preusser, Ferguson, and Williams (1998) found that adolescents took more risks on the road and were more likely to have a greater number of motor vehicle crashes with more than two people present in the car. Furthermore, Rigby and Johnson (2006) found that early adolescents indicated that they would intervene in bullying situations. These findings suggest that developing peer intervening skills might be an effective way to reduce risk-taking behavior and consequently reduce injuries and fatalities among young people.

Little, however, is known about adolescent intervening behavior; that is, adolescents’ ability and context of taking direct action to reduce their friends’ engagement in risk taking. Syversten et al. (2009) examined students’ intended responses to knowledge of a peer with a plan to “do something dangerous.” The researchers found that positive perceptions of, and connectedness to students and teachers, compared with feeling alienated, predicted a greater likelihood of talking to teachers and fellow students about the friend’s hypothetical plan. This finding suggests that students’ sense of connectedness to school (i.e., their feelings of support and acceptance within the school social environment) may potentially impact on their intervening behavior, particularly within the school context. Furthermore, research involving young adults suggests that relationship factors play an important role in understanding intervening behavior. Monto, Newcomb, Rabow, and Hernandez (1992) found that social support and similarity between the individual intervening and the individual performing the risky behavior were important predictors of college students’ self-reported drink driving interventions. Thomas and Seibold (1995) too found that it was the relationship and perceived potential consequences of harming that relationship that were important in predicting likelihood to intervene. The researchers found that those more likely to intervene had a strong relationship with the risk-taker and perceived greater negative consequences of failing to protect. The threat to close friendships potentially represents a greater social cost of failing to protect within the college student groups. While these studies provide correlate factors of those who report intentions to intervene, they do not provide the young person’s description of such factors and thus insight into how young people see these operating in the intervening context.

Smith, Kennison, Gamble and Loudin (2004) conducted one of the few qualitative studies regarding intervening behavior, specifically in the context of passenger behavior. They found that in all intervening scenarios described by participants, there was an indication of potential harm. A large quantitative study examining adolescent friends’ intervening in cigarette, drug, and alcohol use similarly found perceptions of harm to be important. Flanagan, Elek-Risk, and Gallay (2004) reported three conditions that impact on friends’ likelihood of protection, including age, sex, and perceived health-risk implications. The researchers reported that with greater implied seriousness of risk, the likelihood was greater that adolescents would intervene by talking to a friend or adult over ignoring the behavior.

This limited research into adolescents’ intervening combined with the findings from college student samples does indicate that intervening behavior is more likely when young people identify negative consequences in failing to protect, such as the threat of harm to close friends. However, an additional factor is evident from the young adult literature and that is a need to be confident in being able before intervening occurs. Ulleberg’s (2004) study found 16- to 25-year-old drivers’ likelihood of addressing unsafe driving of same-age peers was predicted by capability and fewer perceived costs. Similarly, Rabow, Newcomb, Monto, and Hernandez (1990) found that 73% of college students reported intervening when able. Although these studies provide important information about the individuals and their beliefs, they fail to elicit the detail of the adolescent’s perception of the context of intervening and how young people describe the factors that facilitate intervening.

Exploring Intervening Through Narratives

The research on intervening in bullying scenarios provides some additional insights into intervening behavior. O’Connell, Pepler, and Craig (1999) conducted naturalistic observations of elementary school student’s bullying and identified that around a quarter of the time students intervened when there were opportunities to intervene (these were children across elementary school, Grades 1 to 6, ages 5 to 12 years). However, many more students self-reported that they would try and help when a fellow student was being bullied. Such discrepancy highlights a need to understand the experiences and narratives of the individuals who might be involved to better bridge the gap between intention and behavior and understand the individuals who perform the behavior. A more in-depth understanding of the behavior and the individuals who intervene might assist in the development of strategies that promote intervening.

In addition, while some of the quantitative studies described earlier provide insight into factors that strengthen intention and inquiry into the stories of young people, a qualitative design will provide researchers with a complex textual description of adolescents’ unique perspectives into the strategies, context, and intrapersonal qualities associated with intervening. Focus groups will also provide a rich description of how adolescents perceive intervening into their friends’ risk taking, and will allow an interactive process for adolescents to generate ideas and brainstorm with one another. Furthermore, identifying the factors that increase peer intervention behaviors in the language used by adolescents will assist in directing and informing the design of programs aiming to develop adolescents’ likelihood to intervene. The purpose of the current study was therefore to identify, from young peoples’ perspective, the nature of intervening behavior and the reasons young people might intervene in particular situations. Early adolescents were the focus of the current study as this represents the period in which risk-taking behaviors are largely initiated and therefore harm minimization and prevention behavior change programs typically target this age group. The older adolescents were recruited to provide insight into injury prevention for early adolescents and reflect on their risk-taking experiences over time. It is anticipated that utilizing an exploratory qualitative design will help to identify intangible factors which have not been made readily apparent in previous quantitative research.

Method

Participants

Participants were 95 students: 51 early adolescents (13-14 years old; 54% female), and 44 older adolescents (16-17 years old; 48% females) from two state-funded schools in the outer suburbs of South-East Queensland. The schools were recruited as part of a larger study involving evaluation of an injury prevention program; to occur in the following year with a different cohort as recruited from a conveniently located school district. The early adolescents represented seven randomly selected 9th-grade classes. This age group was the focus of the current study as risk-taking harm minimization and prevention behavior change programs typically target this age group; approximating when risk-taking behaviors (e.g., alcohol use, violence etc. are largely initiated). The older adolescents represented those who were selected by the school as leaders (i.e., they were prefects and school captains) and one randomly selected additional senior class from one school. The older adolescents were recruited to provide a context on how adolescents may develop and reflect on their experiences over time. The older adolescents were expected to articulate with greater detail their experiences and provide insight into intervening behavior as it develops.

All participants had written approval from their parents. The active consent rate was 98% in obtaining parental permission, and all eligible students present on the day agreed to participate. Focus groups were conducted anonymously to the researchers to encourage students to speak openly; therefore, the only identifying information collected was age and gender. While no personal questions related to socioeconomic status were included, school-level data are available using the Index of Relative Socioeconomic Advantage/Disadvantage, as derived from the 2006 Australian Census (see ABS, 2008b). The Index is a rating constructed from attributes of the population in the area (by postcode), such as educational attainment, income, employment, and occupation. Index rating scores range from 1 to 10, with low values indicating disadvantage and high values indicating advantage. One of the schools was located in a somewhat advantaged area, with an Index score of 7, and the other school was located in a disadvantaged area, with an Index score of 1 (ABS, 2008b). Each of the schools are coeducational high schools, located approximately 30 kilometers from the nearest central business district (CBD).

Prompts

Students were invited to take part in a focus group discussion about how adolescents can reduce injury among early teens. As part of this discussion, they were asked about how they have previously and how they might look out for their friends in the future. Older adolescents were particularly asked about injury prevention for early teens and to reflect on their own experiences from earlier years in high school. All participants were asked about what motivates teens to look out for their friends and the characteristics of those who do. Throughout the focus groups, initial prompts incorporating broad questions were used, followed by more specific questions used to seek further clarification. Some example initial broad prompts included in the discussions were as follows: “If friends were around to protect you from risky stuff what kinds of things could they do?” “I’d like to know about how friends can protect you, if friends could be protected what kind of things could you do?” “So tell me the kinds of young people who do that?” and “For any of the things we’ve been talking about, are they the same for younger ones?” Some of the more specific prompts were “Okay, so how do you decide what to do?” “What affects your choice?” “How would they interact?” “Do you have anything else that you want to add to that one, about stopping people doing that?” and “Can you provide us with an example?”

To maintain the exploratory nature of this study, the focus group prompts were not framed in terms of theoretical constructs. Instead, broad open-ended questions were used. The purpose of the broad framework was twofold. First, studies examining student intervening behavior are rare and do not provide a clear theoretical framework to test. Second, an aim of this study was to explore and understand the language used by young people to describe intervening behaviors, and to serve this purpose, questioning was required to be less directed.

Procedure

All procedures adhered to approved protocols of the university and education board ethics committees (institutional review boards). Only students who had returned signed consent forms from their parents were invited to participate.

Two focus group facilitators experienced in working with young people led discussions of approximately six to eight students per group (groups were either all Year 9 students or all Year 12 students). The facilitators were trained in psychology and had conducted previous focus groups and had undergone training in ways in which to elicit comment from many students and be supportive of differences of opinion (e.g., encouraged interaction and verbalization of opinions through prompts such as “Do you all agree? Or does anyone have any different ideas?”). Groups were conducted over a 1 month period. Students were provided with a description of the research and completed consent forms. Following this, 30 min of class time remained for discussions, which were audio-recorded and later transcribed. Refreshments were provided for students. A teacher remained in sight of the group but did not take part in discussions.

To increase the rigor, a number of steps were undertaken throughout the process. An introductory paragraph was read to emphasize that every individual’s thoughts were appreciated and valued and that differences of opinion were respected. In addition, throughout the discussion the facilitator invited participants to share their thoughts or reflect on the discussions to that point. For example, an individual might have been asked, “Would you like to share how this fits with your experience?” This process was in attempt to encourage responses from a variety of participants and reiterate to the participants that all opinions were valued. In addition, checks occurred through the discussion with experienced facilitators paraphrasing and determining accuracy in situ, for example, “so let me make sure I got this straight” or “what I hear you say is.”

Analyses

Discussions were transcribed verbatim by the first author to begin the process of familiarization with the data. Following from this, the first author continued the process of familiarization by reading and rereading the transcripts. Themes were generated through a process of familiarization and ongoing interpretation by the first author. Data were managed using the software, NVIVO, QSR International (2009) for qualitative data analysis. Qualitative analysis enables the conduct of an exploration of relationships between identified themes. It involves a process of managing, summarizing and finding meaning in large semistructured quantities of data.

From this process, index categories of information were developed to produce codes. The number of codes produced was not predetermined. Two overall areas of discussion were explored: the intervening behaviors used and the context of intervening behaviors. Through repeated reading of the transcripts, categories or content were firmed in each of these two conceptual areas. For example, each time an intervening behavior was mentioned a theme was highlighted (i.e., for “say something to make them stop”—“say something” was coded as a particular behavior theme). The themes and subthemes were checked across focus groups to assess dependability and conformability, and thus the recurrence of themes was explored. This was especially important in the current research with two participant groups taking part (i.e., early and older adolescents).

When the initial coding was complete, a second researcher checked the coding accuracy by going through a proportion of the transcripts independently (three early adolescent and three older adolescent randomly selected group transcripts). To improve the reliability and validity of the coding, frequent consultations with the research team were held to discuss the selection of themes from both coders; thus, with the addition of another two researchers, it provided “fresh eyes” to the material. The team then determined the final set of codes.

Findings

Young people were able to provide detailed descriptions of their intervening experiences, including the strategies they used to protect their friends. Furthermore, they were able to describe the characteristics of young people who intervene, as well as instances that might increase or decrease the likelihood that they would intervene or had previously intervened.

Intervention Strategies

Talking to friends

All participants described saying something to their friend as a key method by which they intervene in their friends’ risky behaviors. Such comments were applied to the overall concept of risk taking and some specific, such as around the road (e.g., driving after drinking) and getting into fights. For example, one early adolescent suggested, “you could tell them.” While some participants suggested simple statements that they could use with their friends, many elaborated. The elaborations included descriptions of the consequences of such behavior in terms of physical harm or social consequences as well as describing the need to take responsibility for one’s actions.

Describing consequences: Harm

The most common addition to simply telling a friend not to do a particular injury-risk behavior related to talking to their friend about the consequences. For instance: “you tell them before they go down there (on the motorbike track) there’s a chance you’re gonna come off (and get hurt)” (early adolescent boy) and “tell the repercussions of it” (early adolescent girl). Others, who thought that the risks were not worth the consequences and that it would be unwise to undertake a particular behavior, suggested that they would remind their friend of such, “tell them that’s stupid” (older adolescent boy). The inclusion of describing the consequences was considered an important part of telling a friend to engage in a less risky behavior.

Describing responsibilities

Many students also spoke about personal responsibility to choosing particular behaviors and thus sought to remind friends of their responsibility or sense of individual character, “you could say, ‘just be yourself mate’” (older adolescent boy) and “you could say, ‘look in the mirror it’s just stupid to be doing it’s not worth anything’” (older adolescent boy). Consideration of the consequences, whether it be regarding the specific nature of the behavior or to the individual’s sense of self might thus also reflect salient issues that influence young people to intervene.

Describing disapproval

The data revealed that a number of participants disapproved of particular behaviors, and this was reflected in the way in which they would respond to their friends’ risk taking. This was particularly true for behaviors occurring around the road. For example, “hooning behavior,” which includes risky driving behaviors (e.g., illegal street racing, burnouts, drifting, speeding and unnecessary acceleration), was reported as unpopular by a number of participants, and these participants suggested intervening by expressing this dislike. One early adolescent boy remarked, “I hate hoons. I don’t like that.” An older adolescent boy reflected on another individual’s intervening behavior who had also indicated dislike for hooning, “you’re in the car and your mates decided to hoon, then one of your mates would go, hey, stop doing that. That’s not good.” A small number of other students emphasized more strongly their dislike for a particular behavior and suggested that they would emphasize such disapproval more strongly. This was again highlighted in risky driving situations; for example, one early adolescent suggested, “if I was in the car with you and you’re doing like a hundred down a little side street, I’d tell you, ‘I’ll hit you in the head’ and I would.”

Active intervention

Intervening behavior was primarily focused around the immediate relationship of the potentially risk-taking friend and the intervener. As discussed, this was most commonly described in terms of something the intervener could say about the risk-taker, including the consequences of risk taking or their disapproval. There were, however, a number of participants who provided behavioral examples. Across both drinking and fighting behaviors, monitoring was suggested as an important part of looking out for friends. One early adolescent suggested “watching before a fight happens,” and in relation to amount of alcohol consumed, “kind of monitoring, watching” (older adolescent girl) was suggested. Among the older female students, this was elaborated on and the participants suggested that if issues could be “fixed” earlier on then there would likely be “less chance of stuff going wrong (and/or) someone getting hurt.”

An additional example of managing the situation behaviorally was described among three groups and involved removing the friend from the risky situation; that is, “take them away from the situation” (early adolescent girl). A notable difference that was evident among the older adolescents was the consideration of planning to prevent the friend from being in the situation in the first instance. Older adolescent participants suggested a number of alternative activities, such as, “we’d be going to the movies,” “going to Glorias (coffee shop),” which would have lessened the likelihood that they and their friend would be in a situation requiring intervention, in particular, parties involving alcohol. One older adolescent boy articulated that this would be “kind of like lead(ing) by example.” Thus, older adolescents were cognizant of their ability to plan and prevent involvement in risk taking when not prompted to consider the timings of intervening behavior.

While as stated, participants primarily focused on the immediate context in which the risk behavior was occurring, two groups of students described how they could intervene by talking to a trusted adult. Among older adolescents, it was acknowledged that school staff might play a role in helping when they were at school. Early adolescents also acknowledged that they could, “tell their father and mother” (girl). Thus, participants also reported being able to seek support from adults. Additional descriptions also concerned potential risks of talking to adults, however. As one early adolescent girl stated, “you get in trouble if you talk.” Interpersonally, therefore, adolescents did sometimes perceive trade-offs for speaking out and saying something.

In addition to active intervention strategies, older adolescents also recognized that more passive behaviors could potentially change their friends’ behavior. One older adolescent boy described this as, “sometimes if you ignore it as well because they might be doing it for attention as well; if you don’t give them the attention they’re after then there’s no point in doing it anymore.” A different rationale for ignoring the friend or behavior provided by an older adolescent potentially represents more self-preservation motives, “Like, he’d not be my mate anymore . . . if he was having races along this big road. And I’d say straight after that, I’m not coming out with you again.”

Facilitators to Intervening

Friendship

All participants indicated that they would be most likely to intervene with a good friend. Many participants indicated that this was necessary for taking action, “it depends how well you know the person. If they’re really good friends then I would (intervene)” (early adolescent), and “I worry about the others (in my group)” (early adolescent). Others were more decisive regarding trying to stop a good friend from risk taking, “you have to talk to your good friends” (early adolescent girl). While most described the necessity of looking out for good friends, others noted that this would make it easier to intervene, “I just know my friend and how to talk (to them).” A similar number were not sure how this would help or hinder, “depends who are your friends are” followed by, “(they) expect it.”

Weighting physical harm and efficacy

Along with intervention among good friends, intervention was also reported as being more likely in the context of more serious circumstances and scenarios. Such seriousness, several individuals noted, was among the more important of factors that led them to intervene in their friends’ risk-taking behaviors. If it’s, “something serious, people getting hurt, then (I would)” (early adolescent boy). Another noted, “before they’re hurt” (older adolescent boy). While one group of older adolescents identified that a more serious situation led them to suggest care, “get that caution (to them), I’ve seen the effects of what can happen.”

Participants recognized that trying to stop the risk-taking behavior of their friends was complicated by others in the scenario. Some older students noted that intervening might cause an additional stress for a driver that would outweigh the current potential harm. As one older adolescent boy noted, “with um, like encouragement with people in the back, if it’s just like one person that’s different, and if there was like a big group, music playing, there’s all those distractions there; makes it harder.” A group of males conversed about the challenge, “’cos driving on emotions can cause accidents as well,” “like if you had any arguments before, like if they’re, if someone had an argument before they got into the car,” “if you’re fighting you’ll just speed up,” “’cos they’re in control,” and “yeah (it’s) a bit claustrophobic(in the car).” Finally, they concluded, “anything could be happening, ’cos the emotions in the car might not be that good.”

Additional complications related to individuals’ strengths. Primarily, students of both age groups identified that having confidence or a stronger sense of self made it easier to intervene. One early adolescent boy suggested that to do something, “you’ve just got to have a bit of confidence about yourself.” Another older adolescent girl suggested that “having a strong sense of personality and who you are” facilitated the likelihood of intervening. A greater confidence and sense of self was also recognized as a potential outcome resulting from feedback after intervening. Such feedback was not necessarily viewed as immediately positive; however, it reinforced a greater sense of self or of being a “good” person for looking out for their friends. Such self-awareness was, however, only articulated by the older students. For instance, one older adolescent boy noted, “you can get that feeling that you do get a bit more respect from the group because you are willing to put that forward.” In another group, an older adolescent suggested that “I think you get a lot of respect as well” (girl), while another indicated, “but it’s not (direct) feedback, they don’t like give you anything” (boy). The early adolescents, however, noted that the relationship would be strained as a result of commenting publically on the risks. One early adolescent girl noted that, “Sometimes you have to be the bad guy as well. I get really nervous (doing that).” Similarly, another group of early adolescents noted this challenge, “I’m the one person kind of in my group that will be like I don’t think you should do that.” “Yeah, I find that as well” (boy). As evident through dialogue within one group of older adolescents, this was noted as being especially challenging when others were present and encouraging the risky behavior.

If there’s more, if there’s some people that are actually pushing the person who’s driving to do it. (boy)

Yeah, peer pressure (boy). They’re telling him to car faster and someone else is telling him to slow down. (boy)

They’d probably try to say something and then one of them will say, don’t be a wus. (boy)

That’d be hard. (girl)

Discussion

The goal of this study was to describe the strategies used by adolescents to intervene in their friend’s risk-taking behavior and the perceived facilitators to intervening among early adolescents. Understanding intervening behavior provides an insight into strategies that adolescents are able to put into practice to reduce harm to their friends, and also potentially provides detail on the kinds of strategies that may be encouraged in health and injury prevention programs.

Context of Intervening

A study aim was to have young people identify the context in which they thought intervening behavior was more prevalent or likely to occur. Interestingly, young people were unlikely to describe the barriers to intervening and instead focused on facilitators (or opposite of facilitators, for example, if it was not a good friend). While a study limitation is perhaps the lack of detailed probes of such issues, it highlights challenges for young people in articulating failure to undertake a socially desirable behavior. The described context aligns with constructs in the Theory of Planned Behavior (Ajzen, 1991). Participants identified efficacy, identification of consequences (costs to not performing would be harm to the friend), and the importance of the relationship (looking out for close friends and their expectations) as being important in the context of intervening in friends’ risk taking. In addition, the perceived likelihood of intervening aligns with the model of bystander helping behavior in emergency contexts as described by Darley and Latané (1968). This theory suggests that there must first be a recognized need to intervene, social responsibility raised to decide intervening is appropriate and skills and efficacy to undertake intervening behavior.

One of the main impetuses for intervening behavior appeared to be a perception of harm in the situation. The Theory of Planned Behavior would suggest that action results from examining the consequences of the outcomes of such actions (or inaction). Therefore, if adolescents are able to identify a cost of inaction (harm to friends) they are likely, as indicated by the current results, to intervene. This initial perception of harm is the first step to intervening in the theory explaining bystander helping behavior. Thus, a critical element to enacting is identifying risk of harm, and in perceiving risk and harm, young people reported that they were more inclined to act. Clear and explicit social cues or environmental factors indicate intervention is required (Latané & Darley, 1970; Schroeder, 1995). Once the cues are interpreted (e.g., harm), the next stage of the process is enacted, which involves assuming responsibility and choosing a method by which they could intervene (or help in the bystander context).

A further facilitator to intervening noted in the current data centered around friendship. Adolescents reported that they were most inclined to intervene if a good friend was engaged in an injury-risk behavior and some described their friends’ expectations prompted intervening. Of note, this expectation prohibited speaking up to an adult. For early adolescents, there appears to be social norms around acceptable levels of harm as well as social norms associated with looking out for friends and this concept aligns with that of normative expectations, whereby perceived key referent groups’ expectations predict intended behavior (as outlined in the Theory of Planned Behavior). The findings also support the promotion of an ethic of social responsibility as a useful target for preventing adolescent risk taking evidenced in adolescents value of prosocial behavior (Bergin, Talley & Hamer, 2003).

Self-efficacy or confidence to effectively stop a friend’s risk-taking behavior was key to early adolescents intervening. Efficacy has been show to predict many teen health behaviors (see Armitage & Conner, 2001), as well as intervention in the bullying context and, for college-aged students and intervention as a passenger (Buckley & Davidson, 2012). Bystanders, for example, are more likely to act if they feel like they know what to do and feel like they possess the necessary resources to act (Latané & Darley, 1970). In contrast, they are less likely to act when they believe another bystander is more competent. The early adolescents in this study indicated that their knowing what to do was associated with their intervening and Latané and Darley’s major theory in which actions in emergency situations occur with the identification of a need to act and perceived social responsibility.

Intervention Strategies

A key step to promoting a positive behavior is to define that behavior, and this study sought to describe what adolescents do in practice to intervene in their friends’ risk taking. Once the necessary processes are in play to facilitate intervention, such as a recognized need, perceived responsibility (or social expectation), and efficacy, the kinds of strategies used can be considered. The experiences described by the adolescents in this study showed that, in many cases, they were willing to intervene and had previously intervened in friends’ risk-taking behavior. The most common strategy included talking to friends to persuade them to change; however, they also noted talking to an adult to interfere with the action, interfering directly with the friend, and planning ahead with alternative activities were options (the later with the older teens). Such descriptions are consistent with the work of Smith, Kennison, Gamble, & Loudin (2004), who found that persuasion, interference, planning, and threatening were common responses to intervening as passengers (16-18-year-) with potential drink driving scenarios.

The kinds of strategies used align with the stages of empathy development among adolescents described by Eisenberg and Strayer (1987). In understanding how the strategies fit with displays of empathy suggests that there may be developmental considerations in program design. In addition, understanding the particular competencies of considering others suggests that to promote some strategies additional support work will be needed so that the adolescent is best able to understand the context and choose a likely strategy. Eisenberg described three key behavioral responses that are displayed during late childhood and adolescence, including self-reflective empathy (responding in ways consistent with the one’s own perspective), empathy that reflects social values and empathy that reflects one’s own identity or values.

Although less frequently demonstrated in this study, some responses aligned with communication of empathy typically displayed in late childhood. That is, a self-reflective empathic orientation of active role-taking in the decision process and an emotional reaction to the situation, accordingly the child would be thinking about what it would mean for them and their affect if they were in that situation. As one early adolescent boy commented, he would describe his own reaction or focus on the negative consequences that would be expected if he were undertaking the behavior (it would be easy to get hurt if I went down that path on my motorbike). The next developmental stage of empathic communication development described by Eisenberg and Strayer (1987) was more apparent among the early adolescents in this study, whereby they were incorporating larger social values in their communication strategy, such as it being wrong to undertake a particular behavior. The final stage of the developmental process of empathy suggests a response that would be consistent with one’s own self-concept, which is said to develop through late high school and into adulthood. This skill is consistent with some of the older adolescent responses or their own behavior, such as modeling or describing personal responsibility as a strategy. The older teens also only identified strategies of planning ahead, in particular activities that best fit with who they were.

There is much speculation that social competencies build from close friendships and an experience of comfort within friendship groups may predict disclosure (Buhrmester & Furman, 1986). While this is primarily conceptualized as an ability to share a problem, it is likely that it operates in a multidimensional manner. Such that experiencing acceptance and trust in close friendship fosters empathy and an ability to discuss sensitive topics regardless of whether it’s needed for advice or to discuss threatening information. The social skills reported by the young people as intervening, discussion of disapproval and social responsibility reflect those which, according to Glick and Rose, are evidenced in quality friendships. Their study highlighted helping behaviors coded by 12-year-olds as advice giving, reassurance giving, and distraction giving was associated with closeness in friendship. Similarly Rose and Asher (2004) found positive friendship qualities predict fewer negative strategies in helping situations (e.g., avoiding friends). The findings of this research suggest another element of quality friendship: conversation to protect. Buhrmester and Furman (1986) indicated that social competencies are allowed to foster with experienced comfort and ability to set aside insecurities. High-quality friendship might predict increased disclosure and offering advice. The discussion of responsibility as a rationale for ceasing the risky behavior coupled with the likelihood of intervention within close friendships thus suggests that there is quality within the friendship.

While the discussion prompts may have focused young people to consider the strategies that they may personally implement, fewer early adolescents discussed talking to an adult as a potential intervention strategy. It is thus likely that intervening behavior represents another example of adolescent’s increasing independence from adults and highlights the importance of developing strategies that acknowledge a preference to talk to friends. Daddis (2008), however, identified different scenarios in which adolescents considered issues and behaviors to be private or whether they could be regulated by parents. In particular among early adolescents, parents were seen to be legitimate sources of authority with regard to moral issues (e.g., those regarding the welfare of others). Options for talking to an adult should also be considered particularly in the context of providing appropriate support for friends. Similar to the results of this study, Syversten et al. (2009) found that adolescent peers reported more often taking their own action when intervening in a “dangerous plan” rather than talking to an adult. Building policy in schools to promote adult help should consider ways in which adolescents have confidence and allow (where possible) the close friendship to be supported. This may be promoted through positive communication and understanding of adults whether they be teachers or parents.

Given the increasingly complex social-cognitive skills that are developed through adolescence, complex ways to negotiate with friends about their safety and behavior may develop alongside such social-cognitive skills. The challenging social context of intervening may be such that despite a desire or willingness to ensure friends are safe, a corresponding action is required and such an action might not be automatic. Importantly, the qualitative nature of the current study allowed adolescents to report on the strategies that they can use in response to the risk-taking behavior of their peers. This is in contrast to previous studies, such as the Flanagan et al (2004) research, which have provided a list of option behaviors that assume adolescents are capable of producing such strategies. The qualitative research also allows descriptions of strategies that might not be easily documented in observational research.

A strength of the current study is its assessment of reported behavior compared with solely intended behavior. It is of interest to note though that some adolescents described intended responses suggesting they may not have had experiences with friends undertaking such behaviors where others in the group may have. Although it cannot be known without direct observation, the strategies and behaviors the adolescents describe are likely to be a more accurate representations rather than an assessment of intentions. O’Connell et al (1999) note that there is a discrepancy between intended intervening behavior in bullying situations compared with recorded observations later conducted in the playground. Thus, the descriptions from young people are likely to provide a solid basis for developing intervention strategies that promote intervening in their friends’ risk taking. The qualitative approach undertaken provides participant language and method of explaining social interactions. Tobler and Stratton’s (1997) meta-analysis of alcohol prevention programs demonstrated that a program’s inclusion of elements of classroom interaction and discussion was the only factor common to effective interventions. As such, the participants’ language can be used to develop discussion prompts and process activities for school-based intervention programs.

Discussion and Implications for Practice

The context in which the early adolescents were likely to intervene reflected the constructs of the Theory of Planned Behavior and as such the constructs may reflect appropriate targets for change in practice. Not surprisingly, a threat to harm was a key prompt to intervening. Many school-based health promotion programs include components that seek to raise awareness of the consequences of certain risk behaviors. Such programs typically include other components designed to promote safer behaviors, however, provide insight into the ways in which to operationalize perceptions of harm in such a way that leads to positive action. Further threats to harm are often considered as key to changing individual’s own risk behavior; the findings of this study suggest it may be extended to friends’ behavior. Efficacy was a particularly important factor identified in the current study, and rehearsal and safe spaces in which to build confidence as well as key adults supporting and reinforcing intervening behaviors may be appropriate strategies to promote intervening that build on this important factor.

The empathic stages described by Eisenberg and Strayer (1987) may be useful in the development of friends’ protective intervention programs, particularly through the selection of communication strategies that are likely to be applicable for early adolescents. For example, discussions and activities in school-based programs that promote responsibility may provide the intervening adolescent with content working toward larger identity issues. Such an approach of promoting personal responsibility is increasingly described in approaches to reducing bullying, and while the person likely to experience harm is different (the victim of bullying is not the perpetrator), some such programs have demonstrated efficacy (e.g., Kärnä et al., 2011). The findings of the current study suggest encouraging adolescents to describe responsibility. The decision discuss the intervening behavior between friends is consistent with research on communicative behavior during adolescence in that talk facilitates interaction and closeness in friendship is associated with being able to talk about challenging issues. Talk facilitates intimacy in friendship and there is greater talk within more intimate relationships (Daddis, 2008). Such experiences are increasingly important as individuals move into adolescence for both transmitting information and strengthening social interactions. Young people are increasingly more likely to spend time communicating with their friends through adolescence (Buhrmester & Furman, 1986).

Beyond the need to identify risk and protective factors for adolescent risk-taking behavior, there is an increasing need to develop effective prevention and/or harm minimization strategies that contribute to reducing the burden of adolescent injury. The findings of this study suggest that building on the communicative patterns existing in adolescent friendship as well as strengthening the potential to seek adult help, particularly among early adolescents, are key factors to target to reduce adolescent harm. The participants in this study also described planning ahead and ways in which they might interfere with risk-taking behavior, which represent potential strategies for change that can be enacted in a safe and effective manner. Notably, there is scope to build on current, more common approaches to reducing adolescent risk taking that focus on developing self-efficacy and responsibility, and to develop means of applying these skills in an intervening context rather than, as is more traditional, focusing on preventing adolescents’ own risk taking. While there are approaches that have demonstrated efficacy of peers intervening in everyday situations for smoking (ASSIST; Campbell et al., 2008), this study describes some of the strategies which may be facilitated and applied to other contexts indeed some of the strategies applied in ASSIST may be important to test in extensions to injury prevention. In summary, the available literature that frames friendship influences as a protective factor suggests that adolescents are willing and capable of intervening behavior in response to their friends’ risk taking, thus representing a critical area for injury prevention.

While the findings of this research provide some key descriptions of intervention strategies and contextual factors relating to these, there are several limitations that must be considered when interpreting the results. For example, the focus group prompts used in this study did not seek for comments on any particular risk-taking behaviors. This was a deliberate strategy used to capture a broad range of responses; however, this strategy resulted in responses relating to a range of behaviors that are not well defined by the participating adolescents. In addition, just two schools participated in this exploratory research for 30-min discussions, and therefore results may not be generalizable to wider school-based populations. The goal of this study was, however, to obtain descriptions of behaviors and context rather than to be a large study of predictors. A further drawback is the length and depth of young people’s quotations. The quotes provided are generally brief and are indicative of the language used by adolescents, which more often involved young people building on each other’s stories rather than longer individual descriptions.

Minimal background information was collected from participants to ensure the anonymous nature of the focus groups. While focus groups are a rich source of gaining information through group interaction, emphasizing confidentiality within the group was one method used to encourage students to speak openly. A further limitation of note is that information relating to peer-group membership was not identified, as peer-group predictors were not a focus of the current study. Further research, however, should incorporate measures of peer-group to explore possible effects between group membership and intervening in risk-taking situations.

We also caution readers with respect to the self-report methods used in this study. The aim of the research was to understand intervening behavior from the perspective of young people; however, there is some discrepancy noted in observational studies showing that adolescents’ attitudes toward intervening differ from their actual intervening behavior, at least in bullying scenarios (O’Connell et al., 1999). This issue speaks to the potential for socially desirable responses from students. While we did have students comment on the challenges and the context of those challenges, there may be bias toward reporting more desirable behaviors, particularly to adults. There is a potentially implicit power imbalance in the adult/adolescent relationship.

The limitation described leads to suggestions of further research, first, to potential verification such that peer groups stories are matched to perhaps get a better indication of what might actually occur. Observation research would be a more ideal form of data collection, however, is perhaps limited to a school context and risky behaviors that may be able to be observed on school campus. Along such lines, follow-up quantitative surveys may better capture generalized intervening behavior and extension adolescent’s behavior of ignoring risk or joining in. The reasons in which young people choose not to intervene or choose to join in would be of particular interest in developing programs that encourage intervening to enable students to consider alternative behavioral responses to their friends’ risk taking. Such research might better explore a wider range of reasons for intervening that ideally would be prospectively linked to student’s behavior and link to more specific types of risk-taking behavior to distinguish between alcohol contexts and risky driving contexts, for example.

Specific sex differences were not examined and the focus groups were held with males and females. There was very little difference in the nature of the responses for males and females with regard to wanting to talk to friends and the active strategies they might use. There were, however, some differences evident in the way in which boys described their disapproval and the consequences of risk taking to their friends. Boys more often commented that they would indicate to their friends the potential for physical harms. Similarly, males described more of the physical consequences for harm as a consideration for intervening particularly weighted against their confidence that their intervening would make a difference and thus would prompt them to intervene. These and other potential sex differences could be further explored in future research that specifically examines differences between males and females in their stories of intervening.

An additional finding of the research is the understanding of some characteristics of the individual who is intervening, that may be strengthened through intervention programs. In this research, it was found that intervening behavior is more likely within close friendships, when the adolescent has confidence to intervene, and when they perceive that there is a real need to do something. As this article has described, intervention programs may target these factors and highlight consequences of behavior, build confidence and efficacy among adolescents, and also strengthen supportive relationships among adolescents and with others in the broader school context. The article thus provides some important contributions in (a) describing the factors that might be appropriate targets for behavior change programs, (b) understanding how young people describe the risk taking and intervening contexts in which they live, and (c) focusing on a behavior that is supported by young people and takes into account the social nature of risk taking in adolescence.

Footnotes

Acknowledgements

We acknowledge the contribution of the wider research team and participating students.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Health and Medical Research Council through Injury Prevention and Control Australia (219325).

Author Biographies

Lisa Buckley is a registered psychologist and a senior research fellow at Centre for Accident Research and Road Safety–Queensland (CARRS-Q). Her research interests are in the areas of adolescent injury and risk-taking prevention, adolescent peer relationships, educational curriculum, and designing, implementing, and evaluating of behavior change programs.

Rebekah L. Chapman is a PhD scholar and senior research officer at CARRS-Q and has worked for a number of years in adolescent health and injury prevention. Her area of interest is school connectedness and preventing and understanding adolescent risk-taking behavior and injury.

Mary C. Sheehan holds a professorial research appointment at CARRS-Q. Her specific area of research concentration is the application of attitudinal and behavior change strategies to community and population-based interventions. She has an established research profile in the public health area of adolescent injury prevention.

Bianca N. Reveruzzi is a research officer and PhD student at the CARRS-Q. Her specific research interests are in the areas of adolescent risk taking and injury prevention, adolescent first aid skill development, adolescent prosocial behaviors toward peers, curriculum-based interventions, and behavior change programs.

References

Ajzen

(1991). The theory of planned behavior. Organizational Behavior & Human Decision Processes, 50, 179-211.

Armitage

Conner

(2001). Efficacy of the theory of planned behaviour: A meta-analytic review. British Journal of Social Psychology, 40, 471-499.

Australian Bureau of Statistics. (2008a). Risk taking by young people. Canberra, Australia: Australian Government. Retrieved from http://www.abs.gov.au/AUSSTATS/abs@.nsf/Lookup/4102.0Chapter5002008

Australian Bureau of Statistics. (2008b). Socio-economic index for areas (SEIFA). Canberra, Australia: Author.

Bergin

Talley

Hamer

(2003). Prosocial behaviours of young adolescents: a focus group study. Journal of Adolescence, 26, 13-32.

Brechwald

W. A.

Prinstein

M. J.

(2011). Beyond homophily: A decade of advances in understand peer influence process. Journal of Research on Adolescence, 21, 166-179.

Buckley

Chapman

R. L.

Sheehan

(2012). Adolescent involvement in anti-social and delinquent behaviours: Predicting future injury risk. Accident Analysis & Prevention, 48, 518-522.

Buckley

Davidson

(2012). A psychosocial model of young adult passengers’ intervening in unsafe driving of their friends. Accident Analysis & Prevention, 53, 98-103.

Buckley

Sheehan

(2010). An adolescent injury intervention: Selecting targeted behaviours with implications for program design and evaluation. Australian Health Review, 34, 487-492.

10.

Buhrmester

Furman

(1986). The changing function of friends in childhood. A eo-Sullivanian perspective. In Derlega

Winstead

(Eds.). Friendship and social interaction (pp. 41-62). NY: Springer-Verlag.

11.

Campbell

Starkey

Holliday

Audrey

Bloor

Parry-Langdon

Moore

(2008). An informal school-based peer-led intervention for smoking prevention in adolescence (ASSIST): A cluster randomised trial. The Lancet, 371, 1595-1602.

12.

Centers for Disease Control and Prevention. (2010). Youth risk behavior surveillance—United States 2009. Atlanta, Georgia: Centers for Disease Control and Prevention. Retrieved from http://www.cdc.gov/mmwr/pdf/ss/ss5905.pdf

13.

Chapman

R. L.

Buckley

Sheehan

(2011). The development of the extended adolescent injury checklist (E-AIC): A measure for injury prevention program evaluation. Youth Studies Australia, 30, 49-58.

14.

Chen

Elliott

Durbin

Winston

(2005). Teen drivers and the risk to child passengers in motor vehicle crashes. Injury Prevention, 11, 12-17.

15.

Daddis

(2008). Similarity between early and middle adolescent close friends’ beliefs about personal jurisdiction. Social Development, 17, 1019-1038.

16.

Darley

J. M.

Latané

(1968). Bystander interventions in emergencies: Diffusion of responsibilities. Journal of Personality & Social Psychology, 18, 377-383.

17.

Eisenberg

Strayer

(1987). Empathy and its development. Cambridge: Cambridge University Press.

18.

Flanagan

C. A.

Elek-Risk

Gallay

L. S.

(2004). Friends don’t let friends . . . or do they? Developmental and gender differences in intervening in friends ATOD use. Journal of Drug Education, 34, 351-371.

19.

Gerrard

M. G.

Gibbons

F. X.

Vande Lune

L. S.

Pexa

N. A.

Gana

M. L.

(2002). Adolescents’ substance-related risk perceptions: Antecedents, mediators, and consequences. Risk Decision and Policy, 7, 175-191.

20.

Kärnä

Voeten

Little

Poskiparta

D. E.

Kaljonen

Samivalli

(2011). A large scale evaluation of KiVa Antibullying program: Grades 4-6. Child Development, 82, 311-330.

21.

Latané

Darley

(1970). The unresponsive bystander: Why doesn’t he help? New York, NY. Appleton-Century-Crofts.

22.

Monto

M. A.

Newcomb

M. D.

Rabow

Hernandez

(1992). Social status and drunk-driving intervention. Journal of Studies on Alcohol, 53, 63-68.

23.

O’Connell

Pepler

Craig

(1999). Peer involvement in bullying: Insights and challenges for intervention. Journal of Adolescence, 22, 437-452.

24.

Pickett

Schmidt

Boyce

W. F.

Simpson

Scheidt

P. C.

Mazur

Harel

(2002). Multiple risk behavior and injury: An international analysis of young people. Archives of Pediatric Adolescent Medicine, 156, 786-793.

25.

Preusser

D. F.

Ferguson

S. A.

Williams

(1998). The effect of teenage passengers on the fatal crash risk of teenage drivers. Accident Analysis & Prevention, 30, 217-222.

26.

QSR International Pty Ltd. NVivo qualitative data analysis software. Version 8. Victoria, Australia.

27.

Rabow

Newcomb

M. D.

Monto

M. A.

Hernandez

(1990). Altruism in drink driving situations: Personal and situational factors in intervention. Social Psychology Quarterly, 53, 199-213.

28.

Rigby

Johnson

(2006). Expressed readiness of Australian schoolchildren to act as bystanders in support of children who are being bullied. Educational Psycholist, 26, 425-440.

29.

Rose

Asher

(2004). Children’s strategies and goals in response to help-giving and help-seeking tasks within a friendship. Child Development, 75, 749-763.

30.

Schroeder

D. A.

(1995). Social dilemmas. Perspectives on individuals and groups. Westport. CT: Praeger.

31.

Smith

M. J.

Kennison

Gamble

Loudin

(2004). Intervening as a passenger in drink/driving situations. Applied Nursing Research, 17, 142-149.

32.

Syversten

A. K.

Flanagan

C. A.

Stout

M. D.

(2009). Code of silence: Students’ perceptions of school climate and willingness to intervene in a peer’s dangerous plan. Journal of Educational Psychology, 101, 219-232.

33.

Thomas

R. W.

Seibold

D. R.

(1995). College students’ decisions to intervene in alcohol-related situations. Journal of Studies on Alcohol, 56, 580-588.

34.

Tobler

Stratton

(1997). Effectiveness of school-based drug prevention programs: A meta-analysis of the research. Journal of Primary Prevention, 18, 71-128.

35.

Ulleberg

(2004). Social influence from the back-seat: Factors related to adolescent rear seat car passengers. Accident Analysis & Prevention, 36, 627-629.

36.

Zakrajsek

Shope

J. T.

(2006). Longitudinal examination of underage drinking and subsequent drinking and risky driving. Journal of Safety Research, 37, 443-451.