Strategies for Preventing Peer Aggression, Bullying, and Victimization

  • Evidence Rating for this Practice:

    Promising (2 Tests of the Practice in 2 Independent Studies)

    In one of the tests of the practice, the practice of preventing peer aggression was associated with better outcomes. In this test of the practice, the outcome evidence and the methodology used for assessing effects of the practice satisfied criteria for a designation of Promising. In the other test of the practice, the practice met criteria for conceptual framework, but had a small sample size which reduced the strength of the rating for design quality: the test of this practice was thus classified as Insufficient Information.

    Description of Practice:

    Preventing peer aggression, bullying, and victimization is based on intentional program efforts to develop behaviors, skills, and attitudes that reduce or prevent engagement in (1) aggressive or bullying behavior and (2) experiencing victimization by peers. While specific definitions vary, bullying behavior is generally understood to include unwanted, aggressive behavior that involves an imbalance of power (either real or perceived) and that is repeated over time. Bullying behaviors include threats, spreading rumors, verbal or physical attacks, and social exclusion ( Drawing on prior research and theory, Elledge and colleagues defined peer victimization as repeated exposure to interactions with peers that involve an intent to harm, result in harmful effects on the recipient of these interactions, and are endorsed (explicitly or implicitly) by the peer group (Elledge, Cavell, Ogle, & Nugent, 2010).

    While participation in a mentoring program or relationship may be recommended as beneficial for youth who are engaging in bullying or aggressive behavior directed at peers, being victimized by peers, or both (“bully-victims”), the focus of this review is on explicit and intentional practices that are designed to increase young people’s ability to interact positively with their peers or to disrupt or change the overall peer ecology that forms the context for experiences of bullying, aggression, or victimization.


    The primary goal of the practice is to promote positive outcomes for the mentee by having their mentors, with program support, engage in activities that are designed to either (1) prevent or reduce aggressive or bullying behavior toward peers or (2) assist the youth in coping with the experience of peer victimization and cultivate social skills and relationships that enable the child to have more positive relations with their peers.

    Targeted Forms of Mentoring and Youth Populations:

    This practice is potentially applicable to all forms of mentoring with youth who are at risk for or are engaging in peer aggression or bullying behavior, or who are experiencing victimization by peers.

    Theory and Background Research:

    Prior research on outcomes of mentoring programs has shown effects on reducing aggressive behavior among youth. In the Big Brothers Big Sisters Impact Study (Tierney, Grossman & Resch, 1995), youth who were mentored were one-third less likely than youth in the non-mentored control group to have hit someone during the 12 months prior to follow-up assessment. In a recent meta-analysis conducted by Tolan and colleagues focusing on youth at risk for delinquency, mentoring was found to have a modest, but significant, effect on aggression, although the authors noted that this size of effect was comparable to other interventions targeting high-risk youth (Tolan, Henry, Schoeny, Lovegrove, & Nichols, 2014). Follow-up moderator analyses indicated that stronger effects on outcomes were observed when emotional support and advocacy were present as mentoring components. At present, however, there is a lack of research identifying and evaluating the specific mechanisms by which mentors may help their mentees reduce aggressive behavior.

    One interesting example is the Lunch Buddy program, which was originally developed as a control condition in a study of a mentoring intervention for aggressive children (Cavell & Hughes, 2000). In the original control condition, mentors met with children in the cafeteria over a short (30-minute) lunch period where other children were present, and children were paired with a different mentor for each of the three semesters of the study. Unexpectedly, results showed positive effects (decreases in both teacher-and parent-rated aggression) for both the treatment condition as well as this control condition. Cavell and colleagues (e.g., Cavell & Henrie, 2010) have suggested that the positive outcomes for Lunch Buddy mentoring resulted from the mentor’s interactions in the lunchroom creating positive changes in the peer ecology of the mentored children. The current mentor manual for Lunch Buddy mentoring (Cavell, n.d.) notes that the goals of the program are to increase mentees’ positive social interactions and to enhance their social reputation. As part of the manual, mentors are provided with several specific examples of strategies they can use to facilitate these goals, such as engaging other peers at the lunch table in conversations with the mentee.

    With regards to peer bullying and aggression, mentors may be able to leverage their relationship with youth to challenge beliefs that support bullying (e.g., victim blaming, hostile attributional biases, etc.) and to serve as a role model for non-aggressive social interactions. For youth who are experiencing victimization, mentors may be able to provide a supportive relationship in which they can assist youth in building positive relationships with other peers and instruct youth in the use of adaptive coping skills, which have been shown to be negatively impacted by peer victimization (Troop-Gordon, Sugimura, & Rudolph, 2017). Mentoring programs in which mentor-youth interactions take place in settings where youth interact with peers (e.g., schools) may enable mentors to witness peer interactions that involve either bullying or victimization and then be able to intervene directly in the youth’s peer ecology by responding within the social context to both their mentee and other youth. This kind of intervention is consistent with an ecological view of youth bullying (e.g., Espelage, 2014), which notes the importance of addressing ecological context, rather than focusing exclusively on individual behaviors, when addressing bullying. If mentors are working with children in a peer context, they may also have the capacity to have follow-up one-on-one interactions with their mentees using some of the strategies described above (e.g., developing adaptive coping strategies). The presence of the mentor as an adult monitor in social contexts may have the capacity to reduce bullying behavior simply by having an adult present in the context for peer interaction; furthermore, if the mentor is held in high regard by the peer group (either as an older peer or as a valued adult), the presence of the mentor may help build social capital for victimized youth, thus reducing their risk of being bullied (Elledge et al., 2010).

    Corresponding Elements of Effective Practice:

    This practice is most relevant to the areas of Training and Monitoring and Support within the Elements of Effective Practice.

    Key Personnel:

    The successful implementation of this practice is likely to require staff to have a strong base of understanding of the dynamics of peer aggression/bullying and peer victimization and to be able to effectively communicate the rationale for the use of the related practices to improve the experience of their mentees in their relations with peers.

    Additional Information:


  • Study 1

    Evidence Classification: Insufficient Evidence

    Evaluation Methodology:

    Elledge and colleagues (2010) tested the effects of a school-based intervention, Lunch Buddy mentoring, using a quasi-experimental design, with a small sample (n = 12) of fourth and fifth grade children identified as experiencing peer victimization based on child and teacher ratings. From a larger study, a pool of 42 children were identified as being in the top 20% of peer victimization ratings. Principals and school counselors ranked the eligible children in terms of perceived risk and need for a mentor. Recruitment into the study group was based on a first-come-first-serve strategy as parents were contacted and consented. Two matched comparison groups of 12 students each were identified: one group from the same school and one from a different school. The comparison groups. The groups were established based on similarity on the child-teacher victimization index, gender, grade, and ethnicity.

    The intervention condition used the Lunch Buddy mentoring model, in which children are paired with a college student mentor who visits their mentee twice weekly during scheduled lunch times. In this study, mentoring took place during the spring semester of the school year. Girls were matched with female mentors and boys with either male or female mentors. Lunch Buddy mentors received two hours of training prior to the start of the program, and continued to meet weekly in small group sessions with a graduate student in order to discuss their mentoring experiences and problem-solve about challenging situations. For the mentoring sessions, mentors sat with their assigned child and his/her peers at the child’s regular assigned lunchroom table.

    Study outcomes were measured in the fall (prior to the start of the program) and at the end of the intervention period in the spring. Peer victimization was assessed by youth self-report, by peer report, and by teacher report. Analyses of covariance were used to examine group differences in peer victimization at post-test controlling for pretest scores.

    Evaluation Outcomes:

    Child Report of Victimization
    Elledge and colleagues (2010) reported that there were no significant differences between the study groups on child-reported peer victimization.

    Peer Report of Victimization
    Elledge and colleagues (2010) reported a significant group effect for peer reports of victimization. Follow-up analyses indicated that mentored children were viewed by their peers as significantly less bullied than children in the different school control condition.

    Teacher Report of Victimization
    Elledge and colleagues (2010) reported that the teacher ratings of victimization did not differ among the study groups.

    Study 2: Promising

    Evaluation Methodology:

    Sheehan and colleagues (1999) tested the effects of a peer mentoring intervention for violence prevention within the context of a larger community-based program in an urban neighborhood with a high level of community violence (The Cabrini Green Youth Program, CGYP). All of the adolescents (Aged 14-21) who participated in the CGYP’s Children Teaching Children (CTC) program served as peer mentors to younger children (aged 7-13) in the program. A comparison group (youth not currently participating in the CGYP) was recruited from the same community and matched on age, gender, and census tract.

    During the program, the adolescent peer mentors designed lessons to teach the younger mentees about violence prevention. Over the course of the 18-month program, 12 activities were produced in which information was provided through activities such as skits, games, and rap music.

    Outcome measures of interest were assessed at baseline, mid-study (9 months) and at the end of the intervention (18 months). Assessment of youth measures included child reports of exposure/attitudes to violence (Determining Our Viewpoint of Violent Events; DOVVE) and acceptance of aggression (Normative Belief About Aggression Scale; NOBAGS). Children were also rated by their teachers on the Revised Problem Behavior Checklist. Youth in the intervention program (n = 50) were compared with youth in the control condition (n = 75), although the sample size was reduced for both groups due to attrition.

    Evaluation Outcomes:

    Child Report of Exposure/Attitudes to Violence
    Sheehan and colleagues (1999) reported that at post-test, youth in the intervention group had significantly lower scores relative to the control group on the measure assessing exposure and attitudes to violence.

    Child Report of Acceptance of Aggression
    Sheehan and colleagues (1999) reported that at post-test, youth in the intervention group had significantly lower scores relative to the control group on the measure assessing acceptance of aggression.

    Teacher Report of Problem Behavior
    At post-test, Sheehan and colleagues (1999) reported teacher-reported scores of children’s problem behavior were significantly lower for the intervention group relative to the control group. The control group problem behavior scores increased over time, indicating higher ratings of problem behavior at post-test.

  • External Validity Evidence:

    Variations in the Practice
    The studies reviewed examined practices targeting children who were experiencing bullying or victimization by their peers (Elledge et al., 2010), and children who were identified as being at risk of engaging in aggressive behavior in their interactions with peers (Sheehan et al., 1999). Efforts to explicitly reduce the incidence of bullying behavior among youth have limited or no representation in the research to date. Conclusions about the effectiveness of this practice should therefore be limited to the specific behaviors that were the focus of the studies reviewed.

    Youth served by the programs that were the focus of the studies reviewed were from a variety of backgrounds and had varying levels of environmental and individual risks, including family low-income status or living in a single parent home. All programs in the studies reviewed targeted youth who were elementary or middle school-aged; the ages of mentees ranged from 7 to 13 years. The studies reviewed, however, did not test for differences in effect of strategies for preventing peer aggression, bullying, and victimization in relation to these types of youth characteristics, thus making the applicability of findings to different subgroups of youth unknown.

    All of the mentors represented in the studies reviewed were volunteers. One study used young adult mentors (college students; Elledge et al., 2010) and the other used adolescents between 14-21 years of age (Sheehan et al., 1999). The studies included in the review did not test for differences in effect of supports for youth thriving in relation to mentor characteristics, thus making the applicability of findings to different subgroups of mentors unknown.

    Program Settings/Structures
    The mentoring programs evaluating this practice have been U.S. based and either school-based or community-based. Positive effects of the practice have been evident in the context of both school- and community-based programs although these differed in their intervention targets (i.e., the school-based mentoring program focused on bullied children and the community-based program on acceptance of and engagement in aggressive behavior). While the school-based program was based on a one-to-one mentoring format, the community-based program used a small group of adolescents to serve as peer mentors for a larger group of youth, typically interacting in a group setting. Understanding of the effectiveness of the full range of practices for preventing peer aggression, bullying, and victimization in relation across the broader spectrum of potential program structures and settings (one-to-one versus group mentoring and school-based, community site-based, or community-based programs), however, is limited.

    One of the studies reviewed (Elledge et al., 2010) assessed youth outcomes specific to experiencing peer victimization (based on reports from the youth themselves, their teacher and their peers) and the second study assessed outcomes relevant to engagement in aggressive behavior (Sheehan et al., 1999). Findings have been varied in relation to both types of outcomes and outcomes in other domains (e.g., mental health, engaging in bullying behavior) have to date received limited or no consideration. There is thus little understanding of the extent to which effects of this practice may vary (or be similar) across different types of outcomes.

  • Resources Available to Support Implementation:

    While NMRC does not currently have resources specifically focused on strategies to reduce peer aggression, bullying, and victimization, resources addressing related issues can be found under the Resources section of this website. These resources include: the Peer Mentoring Handbook, Supporting Young People in the Wake of Violence and Trauma, Tools for Mentoring Adolescents, Ready to Go: Mentor Training Toolkit, Training New Mentors: Effective Strategies for Providing Quality Youth Mentoring in Schools and Communities, and the Search Institute’s REACH Resources Overview.

  • Evidence Base:

    Elledge, L. C., Cavell, T. A., Ogle, N. T., & Newgent, R. A. (2010). School-based mentoring as selective prevention for bullied children: A preliminary test. Journal of Primary Prevention, 31, 171-187. doi:10.1007/s10935-010-0215-7

    Sheehan, K., DiCara, J. A., LeBailly, S., & Christoffel, K. L. (1999). Adapting the gang model: Peer mentoring for violence prevention. Pediatrics, 104, 50-54. doi:10.1542/peds.104.1.50

    Additional References:

    Cavell, T. A. (n.d.). Peer Safety Project Mentor Manual. Retrieved from

    Cavell, T. A., & Henrie, J. L. (2010). Deconstructing serendipity: Focus, purpose, and authorship in Lunch Buddy Mentoring. In G. G. Noam, M. J. Karcher, & M. J. Nakkula (Eds.), New directions for youth development: Theory, practice, and research (pp. 107-121). Hoboken, NJ: Wiley.

    Espelage, D. L. (2014). Ecological theory: Preventing youth bullying, aggression, and victimization. Theory Into Practice, 53, 257-264. doi:10.1080/00405841.2014.947216 (n.d.). What is Bullying. Retrieved from

    Tierney, J. P., Grossman, J. B., & Resch, N. L. (1995). Making a Difference: An Impact Study of Big Brothers Big Sisters. Philadelphia: Public/Private Ventures.

    Tolan, P. H., Henry, D. B., Schoeny, M. W., Lovegrove, P., & Nichols, E. (2014). Mentoring programs to affect delinquency and associated outcomes of youth at-risk: A comprehensive meta-analytic review. Journal of Experimental Criminology, 10, 179–206. doi:10.1007/s11292-013-9181-4.

    Troop-Gordon, W., Sugimura, N., & Rudolph, K. D. (2017). Responses to interpersonal stress: Normative changes across childhood and the impact of peer victimization. Child Development, 88, 640-657. doi:10.1111/cdev.12617

Insights for Practitioners

Click here for additional insights and tips for those working in, developing, or funding programs that may use this practice.

Request no-cost help for your program

Advanced Search