Environmental Influences on Fighting Versus Nonviolent Behavior in Peer Situations: A Qualitative Study with Urban African American Adolescents

Abstract

This qualitative study explored environmental factors that influence adolescents’ responses to problem situations involving peers. Interviews were conducted with 106 middle school students (97% African American) from an urban school system. Participants were asked to describe factors that would make it easier and those that would make it more difficult for adolescents to make specific responses to problem situations. Two types of responses were presented: nonviolent responses identified as effective in a previous study, and fighting responses. Qualitative analysis identified 24 themes representing family, peer, school, and neighborhood and broader social factors that were related to both nonviolent behavior and fighting. The identification of environmental influences on fighting and nonviolent responses has important implications for efforts to reduce aggression and promote effective nonviolent responses to problem situations encountered by adolescents.

Introduction

Efforts to improve the effectiveness of violence prevention programs could benefit from a clearer understanding of the factors that promote effective nonviolent behavior, particularly in the context of environments that may support aggressive responses. Making a successful transition to adulthood requires more than simply avoiding behaviors such as violence (Catalano et al. 2002). It also requires mastering skills that enhance an individual’s overall adjustment (Greenberg et al. 2003; Weissberg et al. 2003). Reviews of school-based violence prevention programs indicate that the majority include components that focus on addressing social-cognitive skills (Boxer and Dubow 2001). Although individual-level factors are an important target of prevention efforts, interventions may have a limited effect if they do not also address relevant environmental influences within the family, peer, school, and community domains. These factors can play a critical role in determining the extent to which the individual-level skills taught in prevention programs are adopted. For example, adolescents may not consider ignoring to be a viable response to a peer behavior such as teasing if they believe it will lead to them losing status or being targeted by their peers (Farrell et al. 2008).

This study is one in a series designed to provide a foundation to improve the relevance and effectiveness of programs to promote the positive development of urban minority middle school students. For many adolescents the transition from elementary school to relatively larger and less structured middle schools is challenging (Eccles et al. 1993). This transition co-occurs with many social and physical changes (Peterson and Hamburg 1986). In particular, the increased importance of peers may increase adolescents’ susceptibility to the influence of negative peer models (Dishion and Andrews 1995). As a consequence, many middle school environments contain social norms that support violence as an appropriate path to achieve social status or seek revenge for perceived injustices (Fagan and Wilkinson 1998). Not surprisingly, this stage of development is marked by increased risk for negative social and behavioral outcomes (USDHHS 2001). This is particularly true for minority youth growing up in poor urban areas (Attar et al. 1994). The neighborhoods in which they live are often marked by significant problems with crime, housing, school and neighborhood resources, and economic strain (Hawkins et al. 2004; Smith and Hasbrouck 2006). By the time they reach adolescence, many poor urban youth have been exposed to violence, drugs, poverty, family disruption, abuse, and neglect (Evans 2004). Detrimental neighborhood influences and inadequate community resources may interact with environmental factors in peer, school, and family contexts to promote aggressive strategies and reduce the effectiveness of nonviolent responses in conflict situations. These factors underscore the need for a clearer understanding of contextual factors that may influence the effectiveness of programs designed to promote positive outcomes for these youth.

An important goal of these studies was not only to identify factors associated with violent behavior, but those that promote effective nonviolent behavior. Although we have a wealth of information about factors associated with aggression, we know far less about factors that influence effective nonviolent behavior. Much of the research base for violence prevention programs draws from studies that compare aggressive and non-aggressive youth (e.g., Crick and Dodge 1994; Hawkins et al. 1998; Lipsey and Derzon 1998). One drawback to this approach is that non-aggressive youth in such studies are a heterogeneous group, not all of whom use effective strategies in problem situations. This is supported by Hanish and Guerra (2002), who identified different clusters of children who all had relatively low levels of aggression. Clusters included a well-adjusted group and other groups that while low on aggression, exhibited other adjustment problems such as internalizing symptoms, peer rejection, and low achievement. Not all nonviolent responses are effective. If the focus of prevention efforts is to go beyond reducing violence, further work is needed to improve our understanding not only of the factors that reduce violent behavior, but those that influence the use of effective nonviolent behavior.

These studies were guided by Goldfried and D’Zurilla’s (1969) behavior-analytic model. The first step was to identify the problem situations most often encountered by urban African American adolescents. In the first study, qualitative analysis of interviews with 60 adolescents and 50 adults from the community identified a diverse array of problem situations categorized into five domains: child, family, peer, school, and neighborhood (Farrell et al. 2007). The relevance of these situations was confirmed in a second study (Farrell et al. 2006) in which 176 students from three urban middle schools rated the frequency and difficulty of 61 problem situations identified in the qualitative study. The extent to which experiencing these problem situations was related to negative outcomes was also confirmed by correlations between the number of different problem situations experienced and outcomes including aggression, delinquency, depression, anxiety, and low self-worth. A subsequent study (Farrell et al. 2007) identified potential responses to these problem situations. Interviews with 122 middle school students identified an average of 14 responses to each of the 25 problem situations that emerged as most prevalent and difficult in the Farrell et al. (2006) study. The effectiveness of these responses for the specific situations with which they were linked was then rated by three groups: (a) 61 adolescents from schools and community centers; (b) 27 teachers, community center staff, and family interventionists; and (c) 9 researchers. This process identified nonviolent responses to each situation considered effective by all three groups.

The next step in this line of research was to identify factors that influence adolescents’ use of violent versus nonviolent responses to problem situations involving peers. The focus was restricted to peer interactions to provide a more in-depth examination of contextual factors that influence responses. Interactions with and approval from peers become increasingly important to youth during this time (Baumrind 1987). Not surprisingly, situations that occur within the peer context can be particularly problematic for adolescents and contribute to violence and victimization (Crean 2004; Farrell et al. 1998; Kochenderfer-Ladd and Skinner 2002). Interviews were conducted with 106 sixth and seventh grade students at three urban middle schools to identify factors that would encourage and those that would discourage the use of both violent responses and nonviolent responses that had been identified as effective in the Farrell et al. (2007) study. A previous report (Farrell et al. 2008) focused on the impact of individual-level factors and identified four domains (personal resources, beliefs and values, perceived consequences, and appraisal of the situation) that influenced the use of effective nonviolent and aggressive responses. A major finding was that responses deemed effective by previous research were not necessarily considered effective when presented to youth as possible solutions to peer conflict situations. Other factors, such as individual goals, aspects of the situation, and structural components of the environment, appeared to influence whether youth felt they were likely to enact a specific response. This finding emphasized the need for further analyses to examine how environmental factors impact the perceived efficacy of behaviors that prevention programs often promote.

The current study conducted qualitative analyses of these interviews to explore factors within the family, peer, school, and neighborhood and societal domains that influence how adolescents responded to difficult peer situations. In contrast to most previous studies of risk and protective factors, the present study used a qualitative approach. Naturalistic inquiry (Lincoln and Guba 1985) was used to give voice to adolescents’ perspectives on the factors influencing them to engage in particular responses when faced with challenging situations. The findings of previous studies (Farrell et al. 1998, 2007, 2008) confirm the value of using a qualitative approach to illuminate the challenges faced by adolescents and the complexities of their lived experience growing up in an urban environment.

Method

Setting

This study was conducted with adolescents in Richmond, Virginia, a community with a large population of low-income African American adolescents living in neighborhoods with high rates of crime and violence. Census data from 2000 indicate that 61% of 15–24 year olds in Richmond were African American, and 61% of children lived in neighborhoods high in poverty (Kids Count 2004). Richmond was ranked the ninth most dangerous city among all US cities with populations of over 75,000 based on 2003 FBI violent crime statistics (Nolan 2004). In 2001, the violent crime rate for youth under 18 was almost four times the state average. In 2002, intake cases involving juvenile delinquency were more than twice the state average (Kids Count 2004).

Sample

Participants were sixth and seventh graders from three public middle schools (ages 11–15) serving a predominantly African American student population (87, 99, and 100%) with high percentages of students eligible for the federal free or subsidized lunch program (72%).All procedures were reviewed and approved by the university’s institutional review board. Efforts were made to recruit all students in six classrooms selected to achieve a balance across grade levels (3 sixth grade and 3 seventh grade), schools (2 per school), and to include both general classrooms (four classrooms), and specialized classrooms including one classroom serving students with disabilities and one advanced curriculum classroom. Students received a $5 gift card for showing consent forms to their parents and returning them to research staff, whether or not they chose to participate. Participants received a second $5 gift card for completing the interview. Signed consents from parents and assents from students were obtained from 118 of the 139 total students in the six classrooms. Of these, 109 (78% of those eligible) completed interviews. Three interviews could not be transcribed due to audiotape malfunctions. The final sample of 106 included 65 girls, 41 boys, 48 sixth graders and 58 seventh graders. Ten percent were from classrooms serving students with disabilities and 19% from advanced curriculum classrooms. All but three participants were African American.

Interviews

Interviews were conducted during February and March of 2004. Interviewers were eight women (6 White, 2 Black) and two men (both Black); all of whom had at least a bachelor’s degree. The majority of interviews were completed by Black interviewers (36% by men and 23% by women). Interviewers completed 8 h of training in protection of research participants, developmental and cultural considerations in interviewing, and procedures for reporting suspected abuse and handling emotional upset. Role playing was used to give interviewers feedback, increase interviewer comfort and skill, and promote consistency across interviews. Two of each interviewer’s initial interviews were reviewed and oral feedback was given to individual interviewers before they conducted additional interviews.

The primary goal of the interview was to identify barriers and supports for fighting and nonviolent responses to peer situations. Interviews lasted 20–60 min and were conducted at school. Students were given a written description of a problem situation and response, were read the description, and asked to imagine the problem was happening to them and that they were going to make the specified response. Once participants were oriented to the problem situation and response, they were asked a series of questions about the response, including how easy it would be to make, how likely they would be to make it, and why they would or would not make that response if they were in that situation. Interviewers then explored participants’ perceptions of factors that would make it easier (i.e., supports) and those that would make it more difficult (i.e., barriers) to make the response in that particular situation. To encourage participants to think about a range of factors, they were specifically asked to identify any factors within each of the following domains: (a) personal characteristics (“things about yourself”), (b) family, (c) school, (d) neighborhood and community, (e) friends, and (f) other people. Students were encouraged to elaborate upon their responses and to identify factors that did not fit within these domains. Interviewers verbally summarized individual student responses periodically to check for accuracy and understanding; and were trained to avoid leading questions. At the conclusion of the interview, participants were given a briefer and more structured task in which they were asked to respond spontaneously to eight fill-in-the-blank statements about the domains (e.g., “Kids’ parents need to be [blank] to be able to help them do this”). Interviewers followed these with prompts for clarification to ensure they understood the student’s perspectives (Lincoln and Guba 1985).

The situation-response pairs that served as stimuli for the interviews were derived from earlier studies. Twelve problem situations identified in the Farrell et al. (2007) qualitative study that involved interactions with peers and that were found to be prevalent and difficult (Farrell et al. 2006) were selected. A total of 13 scenarios were formed by pairing each situation with 1 or 2 nonviolent responses that received the highest effectiveness ratings in the Farrell et al. (2007) study (see “Appendix”). Effectiveness ratings for these responses based on a 5-point scale (1 = really bad, 2 = bad, 3 = okay, 4 = good, 5 = really good) ranged from 3.6 to 4.4 (M = 4.0) for students’ ratings, 3.7 to 4.6 (M = 4.2) for community representatives’ ratings, and 3.7 to 5.0 (M = 4.4) for researchers’ ratings. Effective responses included assertive limit setting; communicating positively; seeking adult support; seeking adult intervention; problem solving and implementing a solution; offering support; behavioral change and adaptation; and disengaging, avoiding, or ignoring the problem.

Scenarios were organized into six sets with each set including two or three peer situations paired with an effective nonviolent response. Four of the sets also included one of four peer situations that was paired with a physically aggressive response (e.g. “I’d fight him/her”). Each participant was randomly assigned to respond to one of six sets of scenarios (15–19 students responded to each set).

Qualitative Analyses

Audiotapes of the interviews were transcribed and imported into N6, a software program designed for qualitative analysis (QSR International 2002). Two levels of coding occurred over 2 years. A team of six investigators was trained in methods of open coding, and decision rules were developed and discussed to increase consistency. Initially, two investigators read a subset of 20 male and female interviews, and marked text that met preliminary definitions of barriers and supports. Team members wrote memos about the coding and began developing definitions and procedures for first-level coding. Other team members triangulated with existing literature about factors influencing adolescent problem solving and barriers and supports for violent and nonviolent coping. Themes with definitions generated from open coding of the initial subset of interviews and the triangulation procedure were listed in a table and used to guide first-level coding.

During first-level coding, transcripts were coded for themes and for the domains they represented (i.e., individual, family, peer, school, neighborhood, and society). Codes for specific interview segments were not mutually exclusive because interview segments often represented multiple themes. Transcript segments were coded as a barrier or support for either a nonviolent or violent response. First-level coding occurred over a 1 year period during which researchers coded individually and in small groups, and held team meetings to discuss codes and develop decision rules to increase consistency. Throughout coding, memos were written, and team members engaged in constant comparisons between the transcripts and emergent findings. This process was used to ensure transparent results and to document bias reduction activities (e.g., searching for negative cases) (Huberman and Miles 2002). This resulted in an initial set of 51 themes. Intercoder agreement on a random sample of 10% of the transcripts was 79%.

Second-level coding was an iterative process that involved examining and recoding transcripts by a team of five investigators until there was 100% agreement in coding and all areas of non-agreement were either resolved and integrated into the thematic structure or articulated as unanswered questions. In many instances themes that emerged as supports for fighting were also discussed as barriers for making a nonviolent response, and vice versa. In order to clarify the role different factors may have played in influencing violent and nonviolent behavior, responses were coded according to both the theme content and the context (i.e., barrier or support for fighting or for nonviolence). For example, whereas participants might describe the presence of friends’ support for nonviolence as a factor that would encourage them to make a nonviolent response, they might also describe the absence of friends’ support for nonviolence as a factor that would make it easier for them to fight. This coding took ~9 months. During this process new codes were identified and defined, similar themes were merged, and definitions were refined. A table listing definitions and representative quotes for each theme was developed and refined by constant comparison between the emerging themes and transcripts (Strauss and Corbin 1998). Integrating relevant theory and literature identified themes consistent with the literature and those that represented new themes. The focus of the present study was on themes representing the peer, family, school, neighborhood, and society domains.

Results

Overview

The qualitative analysis identified 24 themes representing environmental factors related to either nonviolent responses or fighting. These themes were generally well represented in the data (see Table 1). Of the 20 themes identified as factors influencing nonviolent behavior, all but 5 themes were mentioned by 10 or more students, 13 were mentioned by 25 or more students, and 3 by over half the students. Of the 19 themes related to fighting, all but 3 themes were mentioned by 10 or more students, 15 were identified by 25 or more, and 6 by at least half. The following sections describe the themes within each domain and include representative quotes to illustrate their relation to fighting and nonviolent behavior and the influence of various contextual issues on responses. Domains are ordered from more to less proximal based on Bronfenbrener’s (1979) ecological model. Because of the larger number of themes in the family and peer domain, themes in those domains are organized according to whether their presence tended to support fighting and discourage nonviolent responses; or whether their presence tended to support nonviolent responses and discourage fighting.

Table 1.

Number of interviews with adolescents (total N = 106) during which specific themes were identified as barriers or supports for nonviolent responses or for fighting

Domain/Theme	Nonviolent	Fighting
Family influences supporting fighting
Parental endorsement of fighting	6	28
Mixed parental messages about     fighting	5	16
Parental modeling of violence     and antisocial behavior	1	8
Family influences supporting nonviolent responses
Parental values against fighting	27	45
Parental modeling of prosocial     behavior	16	4
Parental disciplinary practices	14	27
Proximal support from adult     authority figure at home	99	58
Peer influences supporting fighting
Friends’ support for fighting	27	59
Peer pressure for fighting	48	53
Peer instigation	20	26
Direct verbal victimization	46	28
Bystander pressure to fight	47	63
Concern over ‘‘tough’’ image     and reputation	36	38
Peer influences supporting nonviolent responses
Friends’ support for nonviolent     behavior	94	42
Peer pressure for nonviolence	4	0
Concern over prosocial image     and reputation	2	1
School influences
Proximal support or supervision     at school	93	65
School consequences	37	50
Neighborhood influences
Exposure to violence/trauma,     delinquency, and drug or alcohol     abuse	35	62
Proximal support, supervision,     or monitoring	33	20

Family Influences Supporting Fighting

Three themes within the family domain were identified whose presence supported fighting and discouraged nonviolent responses. Specific themes included parental endorsement of fighting, mixed messages about fighting, and parental modeling of violence and antisocial behavior.

Parental endorsement of fighting, representing implicit or explicit messsages that supported aggression, were mentioned in over one-quarter of the interviews. One boy indicated that his parents would encourage him to fight if other kids were teasing or picking on him: “Cause you know your family say, ‘Somebody hit you, hit them back.’…So you gotta do what they say even if you do get in trouble.” By the same token, some students reported that their parents’ endorsement of violence made them less likely to respond nonviolently to peer conflict. A boy explained what would make it hard for him to tell the teacher or principal when other kids are teasing him: “My parents, they’ll force me to fight them. Well, they don’t force me, they’re like, like if a teacher don’t do nothing I might as well do it on my own, like fight.”

Mixed messages about fighting, although by definition somewhat ambiguous, were most often viewed as encouraging rather than discouraging fighting. One girl said it would be easier for her to fight if “one person in my family’s like, ‘You don’t need to fight or you get suspended.’ But then somebody else in my family be like, ‘Well, she said this…you need to get in her face.’” Conversely, several students indicated that mixed messages would support them in responding nonviolently to conflict up to a point. For example, a girl explained how her mother’s mixed messages would help her talk to an adult at school if other students were teasing her:

My momma always influence me to try not to fight unless somebody puts their hands on you, ‘cause she be like, “Words don’t hurt.” And she say that unless they push you, shove you, or put their hands on you…that’s when you do what you have to do.

Parental modeling of violence and antisocial behavior was mentioned in a small number of interviews. Students noted that adolescents whose parents model aggression and poor emotion regulation are more likely to use violence to solve problems. One boy stated that kids who fight have parents who are “always fighting all the time or trying to put fighting into their souls.”

Family Influences Supporting Nonviolent Responses

Four themes within the family domain were identified whose presence supported nonviolence and discouraged fighting. These included themes specifically related to nonviolence including parental values against fighting, and parental modeling of prosocial behavior; and broader themes related to disciplinary practices and parental support.

Parental values against fighting were considered important determinants of adolescents’ behavior. Many students stated that their parents teach them that violence won’t solve problems and explain the consequences of violence. This factor was most often mentioned as deterrent to fighting. For example, when asked what parents should do to help their children say “I’m not going to fight,” even if other kids are boosting up the argument, one boy suggested: “They need to be prepared to teach them the right thing. Tell them to say, ‘I’m not going to fight’.” Another boy noted that knowledge of the consequences of fighting would prevent him from fighting: “They’d [my parents] tell me like it ain’t worth it for fighting, because that could take you to a lot of places that you wouldn’t want to be, such as jail or six feet under.” Students also explained how parental disapproval of aggression supported nonviolent responses. More specifically, encouragement to avoid fighting and messages about the appropriate way to handle conflict supported youth in behaving nonviolently. One girl explained how her mother’s influence would help her apologize for saying something that crossed the line: “She talks to me about treating people how you would want to be treated and if you didn’t mean to, say sorry, ‘cause it’ll get worse and worse with each day.” Similarly, a boy explained how his mother’s messages would help him ignore a student saying things about his family: “My mother always tell me that if anyone talks about her, why bother to fight the person. Because the person don’t even know my mom….and she tells me, just ignore what people say about my family.”

Parental modeling of prosocial behavior was viewed as an important influence on adolescents’ behavior, and was generally seen as a support for nonviolent responses. Students whose parents model nonviolence were considered more likely to resolve conflict nonviolently. For instance, one boy described how his mother would help him apologize to a friend when he had crossed the line with them: “Most kids get this from their mother and they see their mother apologizing…they’ll apologize also if they do something.” Similarly, another boy explained that his mother’s ability to self-regulate would help him remain calm during conflict and subsequently calm a peer down who wanted to fight him saying, “I guess because my mother has self-control. She taught me that, like, you might have to control yourself sometimes.”

Parental disciplinary practices include parenting practices such as monitoring, supervision, and imposing punishment or other contingencies. Many students stated that parental consequences act as a barrier to fighting. For example, when asked how his family would keep him from fighting when other students are boosting up the conflict, one boy said, “Because if I do fight all I would be doing is embarrassing my momma cause I know she taught me better than that. And I’ll be getting in more trouble than I was going to be if I would have never started fighting.” Students also reported that they are more likely to respond nonviolently when parents are involved, concerned, and provide guidelines for behavior. One girl noted that her family would make it easier for her to help a friend but not let him or her copy her assignment because “if I did let her cheat off my paper my parents, they would probably like, take the things I have that I really like away.” An exception to disciplinary practices as a support to nonviolent responses was mentioned by some students, who reported that concern about parental consequences made them reluctant to make nonviolent responses. For instance, a girl explained that she wouldn’t help a friend with their paper but not let them copy hers because, “if I help them and I’m not supposed to and the teacher sees me and calls my parents, I would get in trouble. I think I would get on punishment.”

Proximal support from an adult authority figure at home was described as having a parent or other adult family member who provides nurturance, reassurance of worth, and guidance in problem solving. Proximal family support was mentioned as a support for nonviolent behavior by nearly all participants. One boy explained how his parents would help him get his friends together to resolve a problem: “They can give me advice, like just help me out or something, to help me get my friends together so they can talk about their feelings so they could stop fighting.” Family support in the form of advice, active intervention, and open communication similarly would enable many youth to enact a nonviolent response with peers. One girl explained how parents’ advice could help students go to peer mediation with someone who started a rumor about them: “They tell their momma why the person picking on them and their momma don’t say, ‘Go fight them,’ but she’ll be like, ‘Why don’t you sit the person down, get a teacher, and ya’ll both talk.’” Similarly, a boy described how his family could help him talk to an adult at school about a peer problem, “Because my family, my dad and my mom, they always teach me to come to them for help. You know, don’t ever think the situation is too bad for them to not help me with it.”

Family support was also viewed as a barrier to fighting. One girl explained how her parents’ direct intervention could prevent her from fighting peers who were picking on her, “They’ll come up to the school and tell the principal what’s going on. Or tell the counselor…so if something happens, more than likely I won’t get in trouble for it.” Many youth also discussed how family advice could serve as a barrier to fighting. One girl described how advice from her family would keep her from fighting another student even if peers were boosting up the conflict:

By helping me to calm down and stuff and telling me what I should not do and telling me that when any person or anybody wants to fight you just tell a teacher or principal and they will do something about it like tell their parents and suspend them.

In contrast, the absence of family support was frequently cited as a support for fighting. One boy characterized parents of students who fight as follows: “Who maybe don’t worry about them or maybe just don’t have a father or something at home…It would mean you don’t have nobody to discipline, no negative worse than your mother.”

A notable minority of students reported an exception to the general rule that family support acts as a support to nonviolent responses. Sixteen youth reported instances where family support could be a barrier to enacting a nonviolent response. These youth reported that their family would advise them to do nothing, or not get involved as opposed to trying to solving the problem. As one boy explained: “There are some (family members) who would try to tell me not to get involved with it [helping two friends resolve a conflict]….cause if I get involved in it we all might start fighting.”

Peer Influences Supporting Fighting

Participants described a variety of factors within the peer domain whose presence encouraged fighting and discouraged the use of nonviolent responses. Themes within this section are arranged from those reflecting the more proximal influence of friends to those reflecting increasingly broad influences within the peer domain (i.e., peer pressure, instigation, direct verbal victimization, bystander pressure, and concerns about image and reputation).

Friends’ support for fighting was most frequently mentioned as a factor that encouraged fighting and to a lesser extent a factor that discouraged nonviolent responses. Support for fighting was typically not described as an aversive or negative peer influence, but rather as a helpful, well-meant offer of emotional support or physical assistance. Students described having friends that would support them in situations where they chose to fight. This often involved encouragement regarding a student’s ability to fight well and win, or other emotional support, as indicated by one girl who noted she might be more likely to fight if her friend said, “Don’t be scared, just go ahead and fight him.” Students frequently indicated that their peers could make it easier to fight if they promised to “back them up,” or help them fight. This was especially true for situations where the student was being teased or when bystanders were encouraging a fight. As one girl noted: “If somebody have your back….Like if somebody want to fight you, that person will always stick by you no matter what and say, ‘I’ll jump in it too.’” More rarely, students indicated that sometimes having each other’s backs means supporting each other in avoiding a fight.

Peer pressure for fighting included verbalizations and other actions that encouraged fighting. Peer pressure is distinguished from friends’ support by its unwelcome aspect: youth described peer pressure as peer comments and behaviors that encouraged them to behave in ways that they otherwise would not, whereas friends’ support was described as supporting youths’ own decisions regarding fighting. Peer pressure was especially strong in situations where students were the target of name-calling or in which a friend flirted with a girl/boy they knew the student liked. Examples of peer pressure included actual physical manipulation. For instance, one girl related a past experience where a friend actually took her arm to force her to hit another student. Another girl said that although she herself does not believe in fighting, her friends would encourage her to fight: “They might say, ‘Well, they’re flirting with your boyfriend. You should go up and go fight them and everything.’” This was a very common theme that emerged in response to the fill-in-the-blank prompt “Students who fight have friends who ….” Students time and again provided responses such as “fight themselves,” “encourage them to fight,” “tell them to fight,” and “are into fighting,” reflecting the influence of both peer group normative behaviors as well as more direct peer influences.

Peer pressure for fighting was frequently cited as a barrier to nonviolent behavior. For instance, one student noted that her sister would discourage her from taking a nonviolent approach to resolving a conflict with a close friend: “She wouldn’t want for me to stand right there and just try to calm somebody down.” Peer pressure to fight was especially salient as a barrier to seeking help from adults or through peer mediation. Said one student, “I don’t think nobody in that situation be like, you need to go to no peer mediation. They be like, you need to fight and get it over with, let everybody know that you ain’t scared of nobody.” Similarly, one boy explained how pressure from his peers would make him fight a friend:

They’d be saying, “Look at them over there, messing with your girl. You should fight them for it.” You know what I’m saying? And then they be boosting your mind up, making you get even angrier at the person when you wasn’t intending to fight the person.

Peer pressure can take less direct, more implicit forms as well. One girl explained how walking away from a fight would be hard because friends would give “That look where they think you are all scared and stuff and therefore they don’t like you anymore…. If I didn’t fight then my friends would be all mad.”

Peer instigation is an interpersonal process by which friends and peers can exacerbate a potential conflict and increase the likelihood of arguments or fighting by saying or doing negative things, maintaining conflict by continually bringing it up, and generally encouraging conflict. It occurs when peers manipulate circumstances to make a fight more likely. This differs from bystander pressure, which occurs in a crowd. This theme was mentioned most often when responses involved fighting. It was frequently mentioned as a support for fighting in situations in which a friend started flirting with a girl or boy he or she knew the student liked, and situations in which someone started a rumor about the student. One girl described how other students might instigate a fight over a disagreement with a friend: “They’ll try to boost it up. Like go around and say—tell them what I said, but I really ain’t say it….” Another student noted the need to surround yourself with trustworthy peers who won’t instigate conflict, saying:

If they ain’t good friends, then that’s like they going back to saying he saying and she saying, too, cause like if you come up to me and say well, my friend say you say this to her, you said that to her, that’s going to still cause a whole bunch of confusion with you and that person.

Similarly, a girl described how other students instigating a conflict might prevent her from seeking help from peer mediation:

Like if we go to peer mediation and work things out, that’s happened to a whole bunch of people, they come to peer mediation and they try to work it out but then…somebody outside be like, ‘she’s still saying this about you, she’s still saying that about you,’

Direct verbal victimization is a specific form of peer pressure that involves teasing, name-calling, and other forms of ridicule. Direct verbal victimization was more often a barrier to nonviolent behavior than a support to fighting, although it served as both. Students indicated that it would be easier to act nonviolently in the absence of verbal victimization. Several students shared the perception that the nonviolent response of saying “I’m not going to fight” would be met with name-calling and teasing from the peer group. As one boy explained, “If they wouldn’t make fun of me, I couldn’t feel bad about saying, ‘I’m not gonna fight you,’” and another girl observed that kids think that if they don’t fight, “people will tease them, people will call them punks, people will call them names.” In addition, students frequently noted that concern over verbal victimization would discourage them from seeking help from an adult or using peer mediation. One girl explained how the threat of verbal victimization would make it difficult for her go to peer mediation: “The people at my school would probably call me names like she a snitch and stuff like that.”

Bystander pressure to fight is another specific form of peer pressure. In contrast to peer instigation, which may involve subtle and often covert forms of manipulation (e.g., spreading rumors), and other forms of peer pressure that may occur in private, bystander pressure is a very public form of peer pressure that occurs when an individual’s response to a problem situation is influenced by an audience of peers. This theme was mentioned most often in responses that involved fighting, but also to a lesser extent in several close friend scenarios that were paired with nonviolent responses. Many students reported that other students and sometimes children in their communities intensify conflicts by “boosting them up” using chants, taunts, ridicule, threats, and other actions. They noted that many students enjoy fights and view them as a form of entertainment. As one girl explained:

They want to see a fight. They want to see chaos. They want to see people go at each other, so it’s like a big show or something. They like fighting. They’ll try to egg them on, be like, ‘You need to fight her,’ and things like that.

Another girl noted that it is difficult to refrain from fighting when presented with strong inducement from a crowd because it makes it hard to regulate emotions: “It would be hard ‘cause I wouldn’t want all these kids surrounding me. I would get hot and stare and stuff. That would be real hard.” Students also mentioned the confusion caused by bystander pressure as a contributing factor to fighting. Said one girl, “It’ll be so confusing that all you can do is fight, because you want everybody to leave you alone.” This student explained how friends might help:

It’s kind of hard to [calm down] when somebody is just in your face constantly screaming at you. So it would be kinda hard for me to try to calm down when my friend is trying to tell me not [fight] and everybody else is telling me to.

Students explained that fighting is more likely with bystander pressure because it affects the other person in the conflict as well, who might “listen to the crowd and hit you.”

Bystander pressure to fight was also mentioned as a barrier to nonviolent behavior. One boy explained how a peer audience would make it hard to act as peacemaker between two friends: “People at my school, they are in a circle, one that nobody breaks it up. They just still want to fight. They will think it is stupid trying to break it up.” In contrast, several students indicated that it would make it easier to say “I’m not going to fight” if there were not students around shouting or otherwise intensifying the altercation.

Concern over tough image and reputation was raised in response to virtually all of the scenarios, but was most often mentioned in situations involving pressure to fight. One boy described how concern over image can support an aggressive response to teasing: “If you don’t fight, people gonna say stuff about you. If you do fight, they still will, but you know that you won…. you don’t want everybody to think you a punk or nothing.” Some students stated that their school is known for fighting: “Our school has a bad reputation for fighting. Every time somebody say [name of school]: ‘Everybody at that school be fighting!’ So they think that you have to fight.” Students also justified fighting if someone started a rumor about them because otherwise the rumor could harm their reputation. Some students reported it is necessary to be proactively verbally or physically aggressive to maintain a “cool” image. One girl suggested that some people picked on each other at her school, “because they think it probably make them popular.”

Students discussed how concern about their image and reputation could inhibit certain nonviolent responses. One girl explained how concern about peer reactions would make it hard to talk to a parent about a problem with a friend: “They all say…if you tell your parents then you’re scared. That will keep me from telling my parents’ cause I would want to keep a good reputation and this is the kind of school where people just take on people to be picking fights and make fun of them.” Concerns about image and reputation also made it difficult for students to respond to a public conflict by saying “I’m not going to fight.” One girl offers this possible explanation: “Other people think you’re scared or something, so you just go ahead and fight him, try to show off, try to prove that you’re better or … that you’re not scared.”

Peer Influences Supporting Nonviolence

In contrast to peer influences supporting fighting, participants described fewer factors in the peer domain whose presence encouraged nonviolence and discouraged fighting. Themes within this section included friends’ support, peer pressure, and concern about image and reputation. Factors that supported fighting including specific forms of peer pressure such as instigation, bystander pressure, and direct verbal victimization did not have parallels when it came to factors positively related to nonviolence.

Friends’ support for nonviolent behavior was discussed by nearly all the students interviewed. Students mentioned various forms of friends’ support that would make it easier to make a nonviolent response, including having friends with values that supported nonviolent behavior, friendship, and achievement. For example, one student explained how her friends would support a nonviolent solution to a conflict with a friend who told a secret: “My friends probably would help me out. They’ll say that you just need to ask her why she broke the promise and if she, like, tries to tell you, just listen, don’t get all angry. Just listen to what she has to say before you follow it with actions.” Students indicated that peers could also provide instrumental support for assertive responses in conflict situations, for example, by accompanying them to talk it out peacefully with a peer. This emerged as a particularly important factor in situations in which youth feared that a nonviolent response would be met with verbal victimization from peers. One girl described how it would be difficult to tell someone to stop spreading a rumor about them in the absence of such instrumental support: “‘Cause when you go over there they might be laughing and picking with you and all that. And then you won’t be wanting to talk, say nothing to them, cause you might be too embarrassed.” Another girl explained how peers can support talking to an adult at school if other kids pick on them: “Try to help you talk to a teacher. Just try to be there for them. Don’t try to tell them, ‘Well you should go fight her ‘cause she shouldn’t’ve said this and she shouldn’t’ve said that.’” Choosing a nonviolent response was also easier if peers refuted the perception that doing so was uncool or weak. In contrast, students felt it would be more difficult to respond nonviolently if peers were critical or dismissive of such responses. This was particularly true for responses that involved seeking support from adults or using a peer mediation program. As one boy explained: “They might say something bad about peer mediation or something…. just ragging and stuff like that.” The lack of friends’ support was also seen as influencing other types of nonviolent responses. For example, one girl indicated that her friends would advise her to “just let them argue and fight” if her friends were in conflict, rather than attempting to ameliorate the situation.

Friends’ support for nonviolent behavior was also seen as a factor that would discourage fighting. Such support often took the form of advice about negative consequences of fighting. For example, one girl noted her friends would support the response of saying “I’m not going to fight,” because fighting is “on your record. You can’t kind of mess up in school because it follows you through your whole life.” Students indicated that their friends could help them calm down and resist negative influences, as well as provide comfort and perspective, by telling them not to worry about the social consequences of not fighting. Other students mentioned group norms supporting nonviolent behavior that would impede them from fighting, even under provocation from a larger crowd. One boy explained that peers can provide instrumental support for nonviolent behavior by telling a teacher about a conflict so the target student can save face themselves: “[they can help by] telling the teacher. Cause I probably don’t really want to fight this person. And if they tell the teacher I wouldn’t be able to fight them.”

Lack of friends’ support for fighting was discussed as a detriment to fighting. This took the form of specific prohibitions against fighting, as reflected by statements such as “Don’t fight. Do you see what you’re getting ready to get yourself set up for, get yourself suspended?” Some students also described how peer norms could discourage fighting. One boy noted, “me and my friends, we don’t be fighting or nothing. We don’t be trying to start no trouble. We just keep trying to do our work. Trying to get good grades on our report cards and stuff.” Lack of support for fighting was frequently mentioned in situations that involved conflict among close friends. For example, one girl explained how her friends would discourage fighting: “They could…tell me how stupid it is to fight when y’all were just friends.” Students noted that although some friends might encourage fighting, “good friends” would tell them to talk it out instead of fight.

Peer pressure for nonviolence was mentioned by only four students who described how peers could support nonviolent behavior by encouraging a response adolescents would be reluctant to make on their own. One girl explained how her friends could pressure her into talking to a counselor if she’s mad: “…if I don’t want to go to the guidance, they’ll make me go, like they’ll tell the guidance what’s going on, so the guidance can come get me.” Other examples included encouraging friends to talk to an adult or go to peer mediation.

Concern over prosocial image and reputation was mentioned by only three students who noted that concerns about image and reputation could serve a positive function, either as a barrier to fighting or support to nonviolent behavior. These students were motivated to protect a nonviolent image by, for example, going to peer mediation to clear up a false rumor, or by avoiding a fight. One girl explained how such concerns might prevent her from fighting:

Some of these teachers and some of the people that work in the office say it’s just project children that act like that. And I don’t want them to think that just because I’m from the projects, that’s all I want to do is fight.

School Influences

Unlike the themes described in the peer and family domains, no unique themes emerged in the school domain as primarily supporting fighting. Rather, the presence of both themes that emerged were primarily described as supports for nonviolent behaviors and barriers to fighting.

Proximal support or supervision at school. Nearly all of the students indicated that close relationships with staff, supervision and active intervention by staff, and a school environment that promotes peaceful norms supported nonviolent behavior. Many students explained how support from school staff in the form of advice or active intervention would encourage them to enact a nonviolent response to conflict with close friends. For instance, one girl explained how advice from her teacher could help her apologize to a friend who was angry with her for saying something that crossed the line: “She will be like, go ahead and say you’re sorry to her. If you apologize then that will make you like the bigger and better person ‘cause you’re not going to fight over something so stupid.” Similarly, a boy explained how active intervention by school staff would help him deal with a friend who broke their promise and shared something private with others: “She [the school counselor] probably would go to him herself and talk to him and ask the person why did he tell somebody when the person told him not to tell.”

Although the presence of support helped students resolve problematic situations with close friends, two youth reported that the absence of this support made it difficult for youth to resolve these kinds of situations nonviolently. For instance, when asked how teachers could help kids talk to a friend who broke a promise, one boy noted, “If they cared…Well, they [teachers here] don’t care about fights.” Several other students indicated that a lack of support from school staff would make it difficult to respond nonviolently to peer conflict. For example, one girl explained how a lack of support from teachers would prevent her from seeking adult support if other students were teasing her: “Some of these teachers here don’t even like listening. They just want to send you to the office and then the office gonna be like, ‘Well why didn’t you talk to the teacher?’”

Proximal support from teachers and others was also seen as an influence on fighting. One girl explained how such support could prevent her from fighting a peer:

If you have one good teacher that is crazy about you, she just thinks that you do everything right…She gonna be like, “Well why you want to fight and I thought you was a different kind of person?” So your teacher’ll be riding you constantly, because she’ll want to know why you chose to do that.

In contrast, another girl described how the absence of support would make students more likely to fight another student who started a rumor about them: “Some teachers, they just let you do whatever you want. If you choose to get good grades, you choose to get good grades, if you choose to fight, you choose to fight…They really don’t care.”

School consequences were seen as supporting nonviolent behavior and discouraging fighting, and included knowledge of school rules and rule enforcement. Many students indicated that knowledge of consequences prevented them from fighting. For example, one girl explained how school consequences would prevent her from fighting despite a crowd of students encouraging her to fight: “If I fight the person, I’d get suspended. And if I get suspended I won’t be able to do my work and stuff.” The enforcement of school consequences can also support youth in responding nonviolently. A boy explained how school consequences affect his decision to help a friend with an assignment without letting him or her copy his: “That’s one reason not to cheat: teachers give you a zero if you let them cheat.” Similarly, a girl reasoned about a person who thinks she started a rumor about her:

Because if I did something bad … I could get probably locked up or suspended for three days at home which I’m not going to do, I’m going to go to that person and talk to them cause I don’t want to be in trouble for something that I didn’t do.

It was suggested that youth’s knowledge of potential school consequences can also prevent a nonviolent response. For instance, one girl explained why some kids would not talk it over with an adult that they trusted about a peer who was being fake with them noting that, “they might get punished.” Youths’ concerns in this regard pertained to the possibility that they would be misunderstood or unfairly blamed by an authority figure at school if they tried to ask for help. However, responses like these were less common than those suggesting that knowledge of school consequences supports nonviolent behavior and serves as a barrier to fighting.

Neighborhood Influences

The neighborhood domain was represented by two themes. Exposure to violence/trauma, delinquency, and drug or alcohol abuse was generally viewed as a support for fighting and barrier to nonviolence. In contrast, proximal support, supervision, and monitoring emerged as a support for nonviolent behavior and a barrier to fighting.

Exposure to violence/trauma, delinquency, and drug or alcohol abuse. Students discussed how being around people engaging in violent and delinquent acts and using drugs or alcohol can influence youth. Many adolescents reported that such exposure supported them in behaving violently themselves. One boy explained how living in a violent neighborhood supports him in fighting: “Because I see them doing it and I pick up the habit that they do.” In contrast, many fewer adolescents considered negative neighborhood influences a barrier to responding aggressively to peer conflict. One girl explained how witnessing violence in her neighborhood would prevent her from fighting another student who was spreading rumors about her: “In my community…if people get to fighting, then they bring out weapons and they try to shoot you or whatever. So they should just stop the violence and talk.”

Exposure to violence/trauma, delinquency and drugs or alcohol abuse was also frequently reported as a barrier to nonviolent behavior by some youth. One girl explained why she would be unlikely to tell an adult if other kids were teasing her:

It’s not really that easy ‘cause there’s so much rough stuff that go around in my neighborhood, like people get killed, shot almost every day… If somebody say something to somebody, they be ready to shoot and pick up stuff and hit, so it’s kind of hard for you to try to tell somebody if they don’t, ‘cause you see what they doing and you just a child.

Conversely only four students indicated that exposure to violence, drugs and delinquency in their neighborhood would support them in responding nonviolently to peer conflict. One girl explained how seeing the bad behavior in her neighborhood would support her in trying to calm her friend down after she said something that crossed the line: “I mean with them [neighbors] fighting and stuff like that, you see where they probably end up, or you see what’ll happen to them.”

Proximal support, supervision, or monitoring within the neighborhood occurs where resources such as programs are available, positive alternative activities are provided, and norms reflect a sense of community and caring. The impact of neighborhood support was seen across different types of scenarios with neighborhoods acting as an extended family for advice, supervision and support. Both the availability of community support as well as the quality of that support were raised as important factors in determining behavior. For example, one girl explained how lack of support doesn’t necessarily involve negative messages about behavior, but rather can occur when “there’s nobody to talk it over with” in your neighborhood. When resources exist, they can provide support for nonviolent behavior and prevent aggressive behavior. Support can take the form of advice, and community members can act as ad hoc parents. For instance, one girl explained how neighborhood support assisted her in talking things out with a peer who believed that she had spread a rumor about them: “If I go and knock on my neighbor’s door or something, they’ll tell me about the problem, they’ll tell me how to work it out…like if my mom ain’t home and I need someone to talk to.” Community members can also act as positive role models. One girl noted that it would be easier to get friends together because “in my neighborhood, a lot of people are friends. I see the old man friends, and I’ll be like, ‘If they can keep their friendship together, then I can, too.’” Support can also take the form of structured programs. For example, one student suggested that she would be less likely to fight because her Girl Scout leader discourages it. Although community support can act as a barrier to fighting, one student noted that the environment can vary even block by block: “My neighborhood, that’s all the fighting, but my block is clean, nice, and good people. So they really don’t want me to fight because they don’t like it when I fight.” The presence of prosocial adults in the community can discourage fighting from a monitoring perspective as well. One student said it would be harder for her to get in a fight in the community because “adults make it hard ‘cause they don’t like seeing fights. If they know you were getting in a fight, and they know you, they’ll tell you to go home or they’ll call your mama.”

In contrast, the absence of neighborhood support was viewed as a barrier to nonviolent behavior and a support for aggressive behavior. A boy explained the role his neighborhood plays in his decision to fight as follows: “Oh, I know I got some friends in the neighborhood and they’d probably be like, ‘Hey, man, don’t be talking down. You got to fight ‘em.’”

Discussion

The goal of this study was to identify social-environmental factors that influence urban African American adolescents’ responses to problem situations involving peers. Systematic and iterative analysis produced a theme structure that was generally well represented in the data in terms of the number of interviews in which themes were identified. The findings underline the significant role that the social context plays in shaping aggressive behavior. Youth described peers, friends, parents, teachers and neighbors as important players in aggressive exchanges, serving as supports and barriers to both violent and nonviolent behavior. Many of the factors that promoted fighting (e.g., friends’ support for fighting, bystander pressure to fight, instigation, concern over reputation) were also seen as barriers to nonviolent responses. Similarly, factors that supported nonviolent responses (e.g., friends’ support for nonviolent responses, proximal support from adult family members, parental values against fighting, parents’ disciplinary practices, support at school) also tended to inhibit violent behavior. Although many of the general themes described in this study have been identified in previous quantitative research, the current study provides a deep and detailed description of the social contexts relevant to challenging social situations that are necessary to the development of ecologically valid interventions.

Themes identified in the peer domain underscore the important role peers play in influencing both violent and nonviolent responses to problem situations. Participants generally described a peer climate that strongly supported fighting and discouraged nonviolent responses. Conflicts were often intensified by bystanders “boosting them up” through taunts, ridicule, threats, and other actions. This is consistent with Kochenderfer-Ladd (2004), who found that although elementary school youth responded to victimization with a variety of emotions, the most common response was anger, and the most frequent response was retaliating with aggressive behavior. This study also highlighted the extent to which adolescents may be particularly susceptible to peer influences due to concerns about their status or reputation. Friends’ support, pressure, and concerns about image and reputation are key developmental processes during early adolescence (Prinstein et al. 2001). Prinstein and Cillessen (2003) noted that although aggression has been linked to peer rejection, there is also a subgroup of youth who achieve popularity by using aggression to achieve instrumental goals. Some teens spend considerable effort creating and projecting a particular image and will engage in behaviors in response to peer pressure to maintain or protect this image (Fagan and Wilkenson 1998). This was evident in many of the interviews in which adolescents noted their reluctance to engage in nonviolent responses because of concerns about peer reactions. Although there were parallel structures for friends’ support for nonviolence, there were few strong nonviolent influences to parallel the majority of peer influences supporting fighting. For example, students did not identify bystander pressure to make a nonviolent response or peer instigation of nonviolent behavior (e.g., youth do not usually stand in groups chanting, ‘Don’t fight, Don’t fight’). Proximal influences from friends produced more mixed effects. Some participants noted how close friends might encourage aggression, in some cases backing them up and encouraging them to fight.

Family influences were generally discussed as having the potential to promote nonviolent responses and discourage aggression. The themes that emerged in the family domain are consistent with a substantial body of research demonstrating that family support, supervision, and consistency of discipline are associated with lower levels of engagement in aggressive behavior (e.g., Reese et al. 2000). Previous research has found parental messages supporting aggression to be a strong predictor of adolescents’ aggression (Orpinas et al. 1999). However, less attention has been paid to parents’ messages about nonviolent alternatives to fighting. These messages were identified in interviews with students in the current study as important predictors of their behavior. The salience of parental modeling of both violent and nonviolent behavior has also been found in numerous investigations (e.g., Eisenberg and Fabes 1998). Although families have the potential to promote positive outcomes, this potential is often not realized. Many participants noted the absence of positive family influences and poor parental disciplinary practices as factors that would make it difficult to make nonviolent responses. This is consistent with the findings of a qualitative study by Farrell et al. (2007) that described the challenging family environments within which many poor, urban adolescents are raised. Poverty and lack of access to resources was seen as a particular strain on many families. Indeed, previous work has suggested that impaired parenting may be the mechanism through which poverty and exposure to violence produce negative outcomes for many urban youth (e.g., Grant et al. 2005).

Several barriers and supports were embedded in the school environment. Teachers and other adults at school were viewed by some students as an important potential support for nonviolent behavior. Warm and supportive relationships with teachers promote connectedness to school that can enhance motivation for academic achievement and promote adjustment (Hamre and Pianta 2001), and serve as a protective factor for emotional distress and risk-taking behaviors including substance use, violence, and sexual activity (Resnick et al. 1997). Unfortunately, not all interactions with teachers are positive. Some participants noted the lack of support from teachers and, in some cases, concerns about negative responses from teachers that would make them less likely to use nonviolent responses. Similar findings were reported in Farrell et al.’s (2007) qualitative study that identified problem situations related to negative interactions with teachers.

Students noted broad influences on their behavior at the neighborhood and community level. Our findings regarding negative neighborhood influences are consistent with previous studies that have identified the numerous challenges faced by youth in many urban communities (e.g., Allison et al. 1999; Attar et al. 1994; Farrell et al. 2007). Themes that supported aggression included exposure to violence, delinquent behavior, and drug or alcohol use. Witnessing violence is a stressful and traumatic experience that may result in high levels of negative emotional arousal placing youth at elevated risk for engaging in aggressive behavior (Lemerise and Arsenio 2000). Youth who witness high levels of violence may also come to believe that violence is a normative and even desirable response (e.g., Schwartz and Proctor 2000). It is important to note that not all neighborhood influences were seen as negative. Descriptions of positive behavioral supports at the neighborhood level are infrequently mentioned in the research literature and are counter to many popular and potentially stereotypic notions of urban communities. These findings raise the question as to whether a focus on the correlation between negative neighborhood factors and youth aggression has diverted attention from community factors that support nonaggression and opportunities at the community level to support nonviolence.

A major strength of this study was its use of qualitative methodology that built upon a series of previous, ecologically-grounded studies designed to examine the experiences of urban African American adolescents. These findings identified the range of messages about what is normative or appropriate that adolescents receive from different niches in their environments. The detailed responses of students in these interviews provide a useful basis for filling in some of the blanks left by current quantitative studies. In particular, adolescents’ descriptions of critical peer, family, school, and neighborhood influences could guide the development of more ecologically valid measures of risk and protective factors for use with this population. In particular, the detailed descriptions provided by the participants in this study could do much to enhance the content validity of existing measures, many of which have been informed by researchers’ perspectives of the environmental challenges faced by these youth (Farrell et al. 2007).

Although the qualitative approach employed in this study provides a valuable perspective, it also has some inherent limitations. Most notably, the use of interviews to identify factors that influence adolescents’ behavior assumes that the participants have insight into the factors that influence their behavior and are willing to share them. Although further study is needed to verify the themes that emerged, we believe the present findings provide a good starting point for further exploration. Given the specific context within which this study was conducted, further work, both qualitative and quantitative, is needed to explore the extent to which the findings of the present study represent the experiences of sub-groups of youth that vary in their risk and protective factor profiles and represent different developmental periods.

The findings of this study have important implications for the developers of prevention programs for adolescents. They indicate that youth face a broad range of factors from multiple contexts that influence how they will respond to difficult peer situations. The strong influence of these factors suggests that violence prevention programs focused primarily at the individual level are unlikely to be successful if not implemented as part of a more comprehensive effort that simultaneously produces changes in peer, school, family, and community domains (Boxer and Dubow 2001; Farrell and Camou 2006; USDHHS 2001). Empirical work provides support for targeting the social context or considering the social context as a moderator in preventive interventions (e.g., Metropolitan Area Child Study Research Group [MACS] 2007). At a conceptual level, this work underscores the need to consider the social embeddedness of behavioral responses and the importance of community and cultural concepts such as communalism or relational orientation. For example Jagers et al. (2007) recently found that communal values among fifth grade boys were directly associated with lower levels of violence. Providing youth with permission and support to use nonviolent responses and associated prosocial resources offers promise for reducing youth violence.

Many school-based violence prevention programs emphasize individual-level skills thought to promote nonviolent responses to problem situations (Boxer and Dubow 2001). Although such programs have shown some encouraging results with younger children (e.g., Conduct Disorders Research Group 1999), this approach has had less success when implemented with adolescents (e.g., Multisite Violence Prevention Project 2009), particularly those in inner-city school systems (e.g., MACS 2007). The current study and the related report by Farrell et al. (2008) highlight some of the challenges faced by developers of programs directed at this population. A basic assumption of programs that promote nonviolent alternatives to fighting is that such responses will be effective. Farrell et al. (2008) found that many urban African American adolescents were unlikely to use nonviolent responses to problem situations because they did not feel that the responses would successfully resolve the problem, would have negative repercussions, and in some cases would actually make matters worse. They indicated that they were unlikely to seek help from an adult such as a teacher because they would be viewed as a “snitch,” would not go to peer mediation because they would be seen as scared, and would not use strategies such as talking it out because it could result in a fight. Many of the nonviolent responses were seen as potentially leading to a fight or damage to their status. In contrast, many participants believed that, all things considered, fighting was the most effective way to handle many situations.

The current study identified a variety of environmental factors that make such conclusions appear plausible. Participants described how peers instigate and maintain conflicts, and the perceived cost in terms of peer status associated with nonviolent responses. Moreover, more positive influences from peers, family members, and school and community environments that might support nonviolent responses may be absent for some adolescents. Developers of prevention programs that attempt to train individual youth in the use of nonviolent behavioral responses to problem situations require a more empirical basis for determining whether the responses they promote will produce the desired outcomes, and whether the risk for more undesirable and uncontrollable outcomes should be considered in light of the challenges of particular environments, including those faced by many urban adolescents. This study also identified many contextual factors that make it difficult to enact many of the skills taught in violence prevention programs. For example, situations in which bystanders are actively inciting an argument would pose a challenge for efforts to employ emotion regulation. Although this suggests the need for prevention programs targeting individuals to provide sufficient training and practice to enable youth to master the use of these skills in challenging situations, such as when they are not supported by their peers, we must ask ourselves whether this is plausible and the most effective point of intervention, or whether the target should be social norms and school environments.

The findings of this work provide significant support for targeting the social environment as a critical resource in intervention. In addition, the findings suggest that we broaden our unit of analysis in examining conflict situations from the level of the dyad to include the broader social context. Current disciplinary procedures and intervention programming focus more on dyads than on the significant role played by others in the social context. For example, Talbott et al. (2002) noted that social conflict and relational aggression often preceded physical aggression among urban adolescent females, and that using suspensions to address the physical altercation did not resolve the ongoing socially embedded conflict. Interventions can increasingly focus on the social context of peers, and ways in which friends, teachers, family and community support youth in effectively handling difficult situations. Much of the work in our field focuses on the factors associated with negative behavioral outcomes (e.g., that negative peer influences support aggression). The current work, although finding somewhat less support within the broader peer domain for nonviolence, shifts our attention to the prosocial opportunities. Youth discussed how peers, teachers and parents are each important and significant sources of support in choosing nonviolent responses. Although more than 1 in 3 youth has been in a fight within the last 12 months, we must remember that the majority of youth have not resorted to fighting (Eaton et al. 2006). This perspective highlights the need for further efforts to improve our understanding of how to support youth in their efforts to use effective nonviolent responses in challenging and ongoing social situations.

Appendix

Problem and Response Pairs for Interviews

Effective Non-Violent Responses

Situation: Someone started a rumor about you and other students are keeping it going and making the rumor worse. Now, it seems like all the kids are talking about you.

Response: I would tell my friends to stop spreading the rumor.

Situation: You told a good friend you were interested in going with a girl/boy. You see your friend flirting and trying to talk with this person like they want to go out with them.

Response: I would confront my friend in a positive way. For example, I’d say I want to be friends and tell my friend that I don’t want them to flirt with someone I want to go with.

Situation: You told a friend something private and they told it to other people. This friend had promised they would not tell anyone but went behind your back and told other people.

Response: I’d talk to my friend and ask why they broke their promise not to tell.

Situation: Somebody is spreading a rumor about a student and you got blamed for it. Now you have a big problem with this person who thinks you were talking about them behind their back.

Response: I’d talk it out with the person the rumor was started about and explain I didn’t start it.

Situation: Someone is fake with you, sometimes acting like a friend and sometimes saying mean things about you. You can’t trust them because they change how they act all the time.

Response: I would talk it over with an adult I trust like one of my parents, a teacher, or my counselor and tell the what’s going on.

Situation: You and a friend are joking and cracking on each other. You accidentally say something that you didn’t think would cross the line but your friend gets really mad at you. You didn’t mean it, you were just joking around but you crossed the line and now your friend wants to fight you for real.

Response: I would say I’m sorry for what I said, I didn’t mean it.

Situation: Two of your friends are fighting and they try to put you into the middle of it. You feel pressure from both sides because they can’t get along and they each want you to take their side.

Response: I would get them together so they could talk but not fight and try to help them solve the problem.

Situation: Someone started a rumor about you and other students are keeping it going and making the rumor worse. Now, it seems like all the kids are talking about you.

Response: I would go to peer mediation with the person.

Situation: A friend asks to cheat off a paper you work really hard on. Their friendship is really important to you but this was your work and it took you a lot of time to do.

Response: I’d help them with the paper but not let them copy mine.

Situation: You and a friend are joking and cracking on each other. You accidentally say something that you didn’t think would cross the line but your friend gets really mad at you. You didn’t mean it, you were just joking around but you crossed the line and now your friend wants to fight you for real.

Response: I’d try to calm my friend down.

Situation: You and another kid get into an argument at school. Other students are there boosting it up saying, “Fight, fight, fight.”

Response: I would say, “I’m not going to fight.”

Situation: Other students are disrupting class and making it hard for you to concentrate and get your work done.

Response: I’d ask the teacher to move me.

Situation: Other kids at school tease and pick on you. They call you names and make fun of you.

Response: I would tell an adult at school, like the teacher or a principal.

Situation: Another kid at school says something that is disrespectful about your family.

Response: I wouldn’t let it bother me, because I’d know they ere wrong.

Violent Responses

Situation: You and another kid get into an argument at school. Other students are there boosting it up saying, “Fight, fight, fight.”

Response: I would fight the person.

Situation: Someone started a rumor about you and other students are keeping it going and making the rumor worse. Now, it seems like all the kids are talking about you.

Response: I would fight the person who started the rumor.

Situation: You told a good friend you were interested in going with a girl/boy. You see your friend flirting and trying to talk with this person like they want to go out with them.

Response: I would fight my friend.

Situation: Other kids at school tease and pick on you. They call you names and make fun of you.

Response: I would fight the other kids doing it or saying it to me.