Review of the Roots of Youth Violence: Literature Reviews

Volume 5

Mentoring Strategies18

Adult Mentoring

The belief that the development of an adolescent can change for the better with the guidance of a mature and wiser adult is not new. What is new is the belief that these types of relationships can and will change the attitudes and behaviours of violent youth. Supporters of mentorship argue that early intervention can prevent violence in youth, and as a result, many theorize that proper mentorship can alter violent behaviours (Satchwell, 2006). This is evident as mentorship programs increasingly emerge in high-risk communities and target at-risk youth (Brown, 2008). In the United States, nearly half of all youth mentoring programs have been established within the past 10 years. Furthermore, of the established mentoring programs, only 18 per cent have been running for more than 20 years (Rhodes, 2002). Thus, it is only recently that researchers have been able to analyze this trend and evaluate the significance of such programs. In accordance with their growing popularity, many studies that examine the impact of mentoring have identified positive outcomes in regard to reducing delinquent behaviour among youth. While these studies create a sense of promise, very few have been able to provide solid empirical evidence that proves the success of mentorship’s strategies. Due to a lack of in-depth study, the question that lingers amongst researchers is how effective adult mentoring is on youth over time. As this question continues to be unanswered by the academic literature, it is somewhat difficult to conclude that mentorship is the “answer” to youth violence. Despite the lack of research into mentorship’s long term effects, one cannot ignore the initial positive effects such programs have had on troubled youth. The purpose of this literature review is to identify positive behaviour changes in adolescents that have been attributed to effective adult mentoring practices. In turn, this examination will help to expose the areas of mentoring that lack sufficient knowledge. An emphasis will be placed on identifying the conditions under which mentorship programs can work and under which conditions they might have neutral or even negative effects.

The traditional notion of mentorship frequently invokes the positive image of an older, wiser adult providing compassionate guidance to a young individual. Indeed, the modern-day version of mentorship continues to emphasize the importance of a supporting relationship between an adult and adolescent. However, increasingly, mentoring is being seen as an essential component in the enhancement of an adolescent’s educational, social, and personal growth (Brody, 1992). Many different individuals in an adolescent’s life provide guidance, encouragement, and emotional support. “Natural” mentoring occurs when an adult voluntarily offers guidance, encouragement, and emotional support, as part of a young person’s normal life course, and usually occurs between parents or related individuals and children (Beam, Chen, and Greenberger, 2002; Darling, Hamiliton and Hames, 2002) These types of positive relationships, it is argued, lead an adolescent to adopt positive behaviours and acquire specific skills (Darling, Hamiliton and Hames, 2002).

Sadly, these naturally occurring relationships are not available to every developing adolescent. Caring and supportive adults who help develop natural mentoring relationships are few in high-risk areas, as the familial, educational, and community structures are unstable for at-risk youth (Darling, Hamiliton and Hames, 2002). Traditionally, supportive relationships are provided by families and the community; however, the dynamic of these institutions has changed, and as a result, support and encouragement from positive natural relationships has decreased. One in every five Canadian children is born into a single-parent home; most of these families are “fatherless.” These children are at higher risk of growing up in poverty, and are more likely to face emotional and behavioural problems, poor physical health, strained parental and peer relationships, poor academic achievement, and disengagement from school (Canadian Institute of Child Health, 2000; Canadian Council on Social Development, 2002; Lipman et al., 2002; Ross et al., 1998). In addition, declining neighbourhood safety has led to social isolation and restricted opportunities for positive and constructive relationships (Grossman and Tierney, 1998). As a result, there may be a need to develop structured relationships through the use of volunteer mentors who aim to be the supportive, caring individuals who are lacking in the lives of at-risk adolescents. Mentoring has been identified as a “structured and trusting relationship that brings young people together with caring individuals who offer guidance, support, and encouragement aimed at developing the competence and character of the mentee” (MENTOR/NationalMentoringPartnership In this view, mentoring is seen as an essential element of youth development and has emerged as an applicable preventative measure to decrease adolescent delinquency. The question is whether this generalization can apply to all mentorship programs. The following review of studies will focus on the practices, structures, and identified benefits of having adult volunteer mentors. Research will show that adult mentorship programs can have positive impacts on delinquent youth, but only within a focused and highly structured curriculum. Effective mentoring practices include intensive effort to provide suitable matches, proper training and monitoring, structured activities, frequent contact, and long-term relationships between the mentor and mentee. With the majority of mentorship programs created within the past few years, many lack the time and experience needed to build the foundation of a successful program. Despite this, more than 2,000 programs are established in the United States and continue to emphasize the idea that delinquent behaviours can simply change as the result of an adult-youth relationship. Every year, countless additional programs are created and the mentorship paradigm gains more momentum. Baker and Maguire (2005) examine the American public’s increasing confidence in mentorship and trace its popularity from the early 1900s. The researchers identify four major stages in mentorship’s development: Emergence, Establishment, Divergence and Focus.


The Emergence stage of the mentoring paradigm took place during the development of the industrial economy in the United States during the 20th century. A surge of technological advancements created the growth of urban centres that drew people of different ages, genders, and cultures. The influx of people opened doors for exploitation, leading to the unfair treatment of immigrants, women and children (Baker and Maguire, 2005). Overcrowding, stress, and severe poverty led to many social problems that included a rise in petty crimes among adolescents. Typical crimes among adolescents at the time included theft, smoking, loitering and absenteeism from school. These minor youth crimes resulted in harsh adult penalties that included incarceration. Social reformers eventually equated adolescent delinquency to social factors and expressed the need for differential treatment in dealing with youth delinquency. These children were viewed as “victims of circumstance” and needed formal intervention in order to prevent future delinquent acts (Baker and Maguire, 2005). Thus began a child-saving movement that sought to protect children from mistreatment, poverty, and abandonment. In response, programs like Children’s Aid were established, modifications to the educational system were implemented, including mandatory attendance, and youth employment agencies were formed (Baker and Maguire, 2005). Eventually, countless experimental social programs were established that sought to change the vicious cycle of social decline and poverty among the youth of the Industrial Revolution era.

One of the successful experimental projects of the time was the Hull House. Hull House, established in 1889, was a settlement facility started by the female philanthropist Jane Addams. Using her social and economic influence, Addams gained public support to create residences in impoverished areas that would be used by families in need of social, economic, cultural, and intellectual encouragement (Baker and Maguire, 2005). She promoted equal rights for all individuals and demanded the fair treatment of impoverished youth. She felt that poverty-stricken communities could thrive if all members would learn to embrace one another. The principle that drove Addams’s mission was her belief that poverty and lack of opportunity triggered problems in high-risk communities. Addams’ took a particular interest in youth problems because she felt that the increase in youth delinquency and crime were the result of economic disparity and social deprivation as opposed to individual pathology. Addams and her colleagues took it upon themselves to protect children from harsh treatment within the court system and acted as advocates, guardians, and mentors for youth (

Addams and her colleagues were important figures in the creation of the first juvenile court in the United States, established in 1899. They attended court and acted as supporters for children charged with crimes. When they saw a great need for more support, Addams and colleagues raised the support and resources needed to start the Juvenile Protection Association (Baker and Maguire, 2005). With the help of 20 probation officers, Addams sought to change social conditions that she felt amplified delinquent acts among troubled youth. The JPA also helped to stop the selling of alcohol and tobacco to adolescents. Finally, they advocated turning old unused buildings into social recreation centres for youth, led by supportive adults (Levine and Levine, 1992). Thus began the mentoring movement.

The Establishment stage of mentoring rose as the idea of the “helping hand” made its way into the popular culture. The Big Brothers/Big Sisters movement, for example, came at the same time as the establishment of the juvenile court. During this time, the juvenile courts recruited influential men to befriend and support children who were brought before the justice system. Similarly, in New York City, an organization called the Ladies of Charity of New York City, arranged for women to be present and support children appearing in court. It was at this time that the early stages of the Big Brothers/Big Sisters of America organization began. Ernest Coulter, a court clerk for the mayor, was a strong voice in the promotion of adult volunteering to assist and guide troubled youth. His main goal was to provide needy children and adolescents with positive adults to provide a guiding relationship. Jacob Riis, a prominent social reformer during this time, encapsulated Coulter’s adult–adolescent movement by saying,

If each of you were to be a neighbor, brother, to one of these little ones and see him through, forty would be saved from shipwreck. It is not the law the lad needs, but justice, the kind of justice which only a brother can give – the love, the friendship, for which his life has been starving. All the rest will come on the trail of that (Coulter, 1913).

Coulter was later credited with the creation of the Big Brothers/Big Sisters movement. Subsequently, organizations like Chicago Women’s Club, New York Jesuit Big Sisters and Protestant Big Sisters were formed in many cities. Mentorship gained widespread support from social powerhouses, including the Vanderbilt family as well as Jane Addams (Baker and Maguire, 2005). Despite little evidence that proved its effectiveness, the idea of mentorship grew in popularity and approval with the public. As mentorship gained momentum in the public sphere, scientific researchers gained an interest and sought answers on how mentorship might work.

The idea of mentorship gained support and appeared to be a valid tool to combat youth delinquency. However, recidivism rates among youth continued to grow. As a result, the Divergence stage of mentoring emerged, as mentoring veered from the interests of social reformers towards the science of crime and delinquency prevention (Baker and Maguire, 2005). Psychology, “hailed as the science of the mind,” gained momentum during the early 20th century. There was a particular interest in children. Psychologist Lightner Witner opened the first psychological clinic at the University of Pennsylvania in 1896. In regard to adolescent delinquency, he saw delinquency as a result of environmental factors and therefore as open to rehabilitation through scientific evaluation and treatment (Witner, 1911). Chicago Juvenile Court Judge Merrit W. Pinckney was the first to act on the phenomenon and formed a committee to establish a unique institute that would assist in the physical and psychological examination of children brought before the court. In 1909, the Juvenile Psychopathic Institute was founded by Ethel Dummer and William Healy. The institute set out to collect data on the characteristics of delinquency, its cause, and its possible treatments. The institute gained extensive support from within the medical, psychological and social service fields. Adolescent delinquency was no longer viewed as a social construct, but as an individual flaw that could be best understood from scientific evaluation by medical personnel (Levine and Levine, 1992; Jones, 1999).

In response to increased interest in scientific evaluations, physician Clarke Cabot conducted the first methodical study on the effects of mentorship. The Cambridge– Somerville Youth Study (CSYS), named after the two participant communities, began in 1936 and sought to discover how delinquency developed and how to improve youth development and prevent delinquent behaviour. A sample of 500 children was drawn, and each child was paired with respect to numerous factors that included age, intelligence, and family histories. Each pair was randomly selected to either appear in the treatment group, which received a mentor, or in the control group, which received no mentor. For the next six years, the treatment group participated in various social, educational and health services with the guidance of an adult mentor. Cabot hypothesized that delinquent children were a product of their environment and that the treatment group would benefit from the guidance of a mentor (Powers and Witner, 1951). The study, however, only found marginal differences between the treatment and control group, leading many to question the benefits of mentorship. In 1975, McCord interviewed the original participants of the CSYS study, who were now middle-aged, to examine long-term effects. Data was collected from 1975 to 1981 and included testimonials, court records, mental hospital records, alcohol treatment facilities records, and death records. Testimonials from the treated group presented positive results, but compared with the control group, McCord (1992) found that the treatment group was actually more likely to have engaged in serious street crime, died five years younger, and was more likely to be treated for alcoholism, schizophrenia, and manic depression. McCord even found that the longer an adolescent was involved in mentorship, the more negative the outcomes. It should be noted, however, that mentorship was only a part of the treatment, one of many services offered to the adolescent, making it hard to pinpoint which service was detrimental and which was not. This study was damaging to the mentoring phenomenon, and it continues to be discussed in contemporary research. However, the efforts of programs like Big Brothers/Big Sisters continued to express the importance of mentorship relationships. As a result, further studies were conducted to evaluate mentorship’s effectiveness. Many of these studies produced much more positive results.

This led to mentorship’s final historical stage: Focus. The CSYS opened doors to many interpretations and analyses. While it was an impressive study for its time, it was chaotic in structure. The mentor assigned to each child acted as a social worker, and as a mentor, and also completed various data-reporting duties, making it difficult to separate positive and negative interventions. As a result, it was difficult to reach definitive conclusions in regard to mentoring. The 1960s, however, saw a stronger focus on mentorship. In 1963, President John F. Kennedy signed the Community Mental Health Centers Act, which would supply the necessary funds to develop community health centres. President Lyndon Johnson followed by supporting programs designed to combat poverty through improved health care and education (Meritt, Greene, Jopp and Kelly, 1999). Similar to the child-saving movement in the late 1800s, community psychology advocates equated pathology to the social environment rather than to psychological flaws. As a result, non-professional volunteers were being sought to aid in treatment. Mentors were being recruited, and studies later emphasized evaluating the characteristics of good mentors. Goodman (1972) conducted a notable study where he recruited college students to act as mentors to troubled school-aged boys as they went through programs that offered developmental skills. One hundred and sixty school-aged boys participated, with 88 boys receiving treatment with a mentor and 74 receiving treatment with no mentor. The study found marginal differences between the two groups. However it did raise questions surrounding the preparation involved in mentoring programs, as well as the characteristics and needs of selected mentors. Contemporary research continued to focus on these aspects of mentorship, but more recently, studies have delved deeper and have looked into the characteristics of the mentees themselves.

Risk Factors

In this section, risk factors that may contribute to violent behaviour among adolescents are examined. Many at-risk youth tend to have similar characteristics and life experiences. While the characteristics give little insight into the individual adolescent, identifying certain factors may help to identify youth who are at risk of engaging in violence and help gauge areas of concern with a mentee. The Public Health Agency of Canada (2008) posits that violence among youth is something that is learned from a young age. There is no one factor that results in a violent adolescent; however, the presence of multiple risk factors increases the chances that an adolescent will engage in violent behaviour. Research has shown that there are a number of factors that can lead to youth violence (Satchwell, 2006). These factors are characteristics that consistently arise in youth who are brought before the justice system. The factors are not meant to be interpreted as the sole predictors of violent behaviour, though they often coincide. Nonetheless, the factors give a better insight into the lives of troubled youth. It has been identified that the most consistent predictors of violent behaviour for adolescents include:

In regard to familial and environmental factors, consistent predictors of violent behaviour for adolescents include:

The pioneers of mentorship’s establishment held the belief that delinquent behaviour was learned, and therefore could be prevented. What they failed answer was how the process of mentorship worked, and how to prevent future delinquent and violent behaviour. The following section introduces popular theories that help to understand mentorship and why it has been viewed as an influential preventative tool with regard to adolescent delinquency.

How Mentoring Works

Resilience Theory

Resilience theory holds the belief that certain individuals possess specific characteristics that allow them to recover from and survive situations like family violence, poverty and emotional abuse. Researchers have identified certain traits, circumstances, and situations that help vulnerable children and youth recover. Protective factors that are found to advance resilience include 1) characteristics of the individual, such as intellect and character; 2) characteristics of family, such as its stability, level of closeness, and socioeconomic advantages; and 3) characteristics of the community, such as relationships with non-related adults who act as positive role models (Masten and Costsworth, 1998). The act of mentorship corresponds well with the idea of resiliency. Resilient children often have at least one important non-related adult in their lives (Rhodes, 2002). Reviewing literature on European and Israeli children in war, Rhodes found that children adapted to the stress because, often, a non-related adult represented “efficacy and control” in a time of disorder and confusion. The same philosophy can be applied to children of disorganized environments. In the midst of negativity, a mature and caring figure may represent structure and encouragement.

The idea behind mentoring is that it can positively change youth who experience negative risk factors. The Compensatory Model of resiliency holds that positive influences in a youth’s life can offset negative influences within the family or wider community. For example, a youth who is surrounded by peers who engage in violent conduct is at risk of engaging in similar behaviour. However, that youth may resist, with the presence of positive influences such as an encouraging mentor (Garmezy, Masten and Tellegen, 1984). The Protective Model of resiliency holds that supportive individuals may help to decrease the “cause and effect” of negative influences. (Zimmerman, Bingenheimer and Behrendt, 2005). In a study of adolescent youth, Zimmerman et al. (1998) found that the association between having violent friends (cause) and violent adolescent male behaviour (effect) was lower among youth who had a strong bond with and support from their immediate families (protective factor). The theory strongly supports mentoring, as mentoring can act as a crucial component for positive development in adolescents. In essence, a mentor can offer youth who face negative external pressures the guidance, support and care that is needed to counteract the negative consequences of risk factors.

Psychologist Michael Rutter (1987) found that at-risk children with at least one good relationship were less likely to develop behavioural problems. In an earlier study, Rutter (1983) found that minority children from single-parent, low socio-economic backgrounds were less likely to drop out of school if they had the presence of a positive influential extended family member or other supportive adult. Based on self-report studies, it is safe to say that the influence of a positive, supportive and caring (non-parental) adult can have many beneficial effects on youth experiencing difficult situations. However, one should not assume that naturally occurring relationships can be reproduced by programs that artificially place an adolescent with an adult mentor (Hamilton and Hamilton, 1992) Mentoring should not be used as a substitute for proper parenting and community support (Rhodes et al., 2002). It is only recently that studies have been able to identify benefits of mentoring, which will be discussed later in this review.

Attachment Theory

Attachment theory holds that an early bond with an influential family member, like a parent, sets the psychological foundation for one’s perception of self-worth (Bowlby, 1969). Self-esteem is important during a youth’s development because it gauges how an individual will act in the face of insecurity. In a situation where one might feel powerless, an individual with a strong sense of self can feel a sense of power. Those with a positive outlook on life will be able to control negative reactions to challenges and develop adaptive coping skills (Satchwell, 2006). This theory corresponds well with the idea of “natural” mentoring (as defined in the introduction) and exemplifies the importance of having a positive individual as an influence. Research conducted by Yates, Egeland and Sroufe (2003) found that positive attachment developed early in children leads to an increased ability to adjust after periods of turmoil. The finding is reminiscent of the philosophy behind resilience. Bowlby (1988) expands on the theory and states that the beginning of a child’s ability to build on a relationship is based on early experiences with his or her parents, which demonstrates the importance of strong familial bonds. However, when this is lacking in the life of disadvantaged children, non-familial adults can act as an ego ideal and a child can learn positive adult behaviours and develop valuable skills for the future (Rhodes et al., 2002). It is the foundation of attached relationships that gives an adolescent a sense of worth. Those with positive perceptions of self-worth engage less in risky behaviour (Satchwell, 2006).

Search Method

The purpose of this literature review is to identify effective mentoring practices that lead to violence prevention among youth. A search for articles relating to mentoring programs was conducted in the Scholars Portal under the Psychology and Sociology databases. A search for the term “youth mentoring” was conducted, and a combined total of 276 scholarly articles were retrieved. A specific time period was not selected, in order to track mentorship’s history and progress. The search was refined to include at-risk youth, limiting the search to 77 articles. Of these articles, only three specifically discussed mentoring as a tool for violence prevention. As a result, a section in this literature review will include studies that discuss the overall benefits of mentorship for at-risk youth. A deeper analysis of the three articles that specifically address violent behaviour among youth will be conducted later in this review. With a limited amount of articles with which to make a fair assessment of the impact of mentoring on violence prevention, this review will include a discussion on the more general outcomes of mentoring for at-risk adolescents.

Despite the lack of studies that specifically address violence prevention as an outcome of mentoring, a review of all studies that involve at-risk youth is important. Research shows that certain behaviours and environmental circumstances are valid predictors of violent conduct (Brown, 2008). Poor academic achievement and anti-social behaviours, for example, have been identified as risk factors that lead to violent behaviour (McCord, 1992). Review of the positive and negative outcomes that have been identified in relation to mentorship will now be reviewed and discussed.

General Mentorship Studies

Jackson (2002) studied the outcome of a mentoring program designed for at-risk junior high school children prone to delinquent and disruptive behaviours in school. Using a non-control design, 13 at-risk students (six boys and seven girls) were randomly selected, all of whom were children in danger of school suspension or expulsion for delinquent behaviour that included bullying, physical fighting, smoking, and drinking on school property. The program recruited university students, who successfully completed several child development, psychopathology, and intervention courses to become mentors. The mentors were required to spend a minimum of 15 hours a week with their mentees for the duration of the academic year. Activities included going to mall, having dinner, going out for lunch, watching movies, working on homework, or simply spending some time together. For the duration of the study, both mentor and mentee were supervised by a licensed child psychologist. The mentor was also required to meet with the psychologist two hours per week to review information on child psychology, causes, characteristics of delinquent behaviour, intervention techniques, and also problem-solving strategies for managing difficulties in the mentor–mentee relationship.

By the end of the study, Jackson (2002) found changes in internal behaviour (anxiety, depression, somatization), and external behaviour (aggression, hyperactivity, conduct problems), but found minimal change in adaptive skills (adaptability, leadership, social skills, study skills). Based on the observations of both the adolescent’s parents and teachers, Jackson measured the behavioural changes at four intervals throughout the year. Parents found that most negative internal and external behaviour traits decreased over time, most notably aggression and attention problems. Teachers found similar results, and found that the number of school infractions decreased significantly with all except one mentee. By the end of the program, all of the children had few or no new school infractions recorded against them.

Summary of the Jackson Study

Keating et al. (2002) conducted an experimental control design study that evaluated an extensive mentoring program that targets at-risk youth. Participants of the mentoring program were referred children and adolescents who were deemed to be at risk of engaging in delinquent behaviour. Reasons for referral included fighting, behaviour problems in school or in the community, emotional problems, poor grades or school attendance, or engagement in minor crimes (theft). The program concentrated on providing children in need of positive adult influences with well-trained mentors. Mentors were screened for their time commitment and appropriateness for involvement in the program. They were required to attend training sessions designed to educate them about child development, warning signs of child abuse, how to interact with troubled youth, and finally, conflict solutions for dealing with troubled youth. Mentors had to check in every week for seminars and follow-up. Every effort was made to match an adolescent with someone according to gender, ethnicity, age, geographical location, and common interests, as well as personal preferences of both the mentor and mentee. The program required a minimum of three hours each week with a mentee, and consisted of activities that included going to sporting events, movies, the park, and group activities sponsored by the program to promote social and life skills.

The experiment matched 34 adolescents with a mentor and compared them with 34 adolescents who were placed on a waiting list for a mentor. Participants were between the ages of 10 and 17, 65 per cent male and 35 per cent female. Thirty-two per cent identified themselves as Caucasian, 24 per cent as African-American, 37 per cent as Latino, three per cent as Asian, and three per cent as “other.” The study conducted a pre-test/post-test design, which had the adolescent as well as the adolescent’s parent and teachers participate in pre-test/post-test interviews with a six-month gap. Youth completed a Hopelessness Scale for Children questionnaire (Kazdin et al., 1986), Piers-Harris Self-Concept Scale (Piers, 1984), and the Self-Report Delinquency Scale (Elliot et al., 1985), and parents and teachers completed a Child Behavior Checklist (Achenbach and Edelbrock, 1991).

After six months, parents and teachers reported a decrease in both internalizing and externalizing behaviours among the intervention group. The same degree of change was not reported with regard to the non-intervention group. Over all, mentoring seemed to affect various levels of problematic behaviours among at-risk youth. Mentoring appeared to keep behavioural and emotional problems from getting worse. It should be noted that the children were receiving other forms of support that included school and family counselling. Mentorship was not the only form of intervention, and the control group was also involved with the additional program. Therefore, it is evident that the presence of a mentor increased benefits. Mothers of African-American youth reported less improvement in internalizing and externalizing behaviours then their non-African-American counterparts did. However, the mothers reported low incidence of those behaviours in the pre-intervention stage. This should not be focused on, as African-Americans represented a small portion of the study. However, these results do raise questions about whether the effects of mentorship differ with ethnicity.

Summary of the Keating Study

The next study that we examined took place in Australia. Lemmon (2005) studied a mentorship program that provided at-risk adolescents with job opportunities that ranged from the fast food industry to office services. The youth involved in the program included individuals with noted substance abuse problems or criminal histories, as well as victims of serious assault, neglect, and abuse. The program aimed to reduce the cycle of social disadvantage and provide employment opportunities that may not have been easily accessible. Youth were paired with a mentor who acted as a role model and an employment advisor. Lemmon initially interviewed 10 women in the program and followed up three years later to track their progress. Lemmon found that all 10 women stopped or reduced their substance use. Lemmon concluded that mentorship was not the sole agent of change, but that it did enhance positive effects, because it promoted connectedness, guidance during difficult times, and the building of close personal relationships. In essence, mentorship may increase its benefits when it operates in conjunction with other treatment initiatives.

Summary of the Lemmon Study

The next study involves the goal of drug-use prevention among at-risk youth. LoScuito et al. (1996) evaluated the outcome of a mentoring program that aimed to prevent drug use among at-risk grade six students attending three public middle schools in Philadelphia. Most of these students are at-risk for substance abuse issues, as they face poverty, high-crime neighbourhoods, and low academic achievement. Through the support of volunteer mentors, the program is designed to increase resiliency among youth by helping with the development of awareness, self-confidence, and skills needed to resist drugs and overcome social disadvantages. Mentors were carefully recruited, screened and trained. The mentors were continuously supervised and supported by project staff. Each mentor was responsible for one or two students and was required to meet with them at least four hours per week during the school year. Activities included attendance at sporting events, assistance with school work, or attending community service activities together. The program also promoted a life skill curriculum that focused on stress management, self-esteem, problem-solving, substance knowledge, health information, and social networking. Through class discussion, homework assignments, and role-play activities, the program allowed students to apply these skills in real-life situations. Community service activities were another component, and included biweekly, hour-long visits to residents in nursing homes in order to spend time with the elders in their community. Lastly, the program included a workshop for parents of the troubled youths to develop effective parenting styles. The mentor and mentee participated together in all aspects of the program.

Using a pre-test/post-test control group design, willing grade six students were randomly selected and placed into one of three groups. Group 1 was a control group that received no intervention (189 students). Group 2 participated in the life skills program, community service, and the parent workshop (193 students). Group 3 participated in the life skills program, engaged in community service, and attended parent workshops. However, this particular group also received an adult mentor (180 students). The pre-test found no significant differences in the three groups. A post-test was administered at the end of every academic year for three years. The final sample (students completing both the pre-test and the post-test) included 562 students, of whom 52.2 per cent were African-American, 32.8 per cent were Caucasian, and 15 per cent identified themselves as “other.” Results show that in Group 2 (students who engaged in life skills programs, community service and parenting workshops), attitudes differed from the control group that received no intervention. Group 2 had better attitudes towards elders, their own well-being, reactions to situations involving drug use and community service. However, these students did not improve on attitudes towards school or the future and showed slightly more drug use. Group 3 students (students who engaged in the program with a mentor) showed significant improvement in most categories, the highest being in attitude towards school, future, elders, and themselves, as well as in attitudes towards drug use and in frequency of substance use. Mentored students also showed fewer days absent from school, and teachers found that students who were more involved with their mentor had the fewest absent days. Over all, the study supports the hypothesis that mentorship positively affects the attitudes and behaviours of at-risk youth.

Summary of the LoSciuto et al. Study

The following study examines the effectiveness of one of the oldest mentorship programs. In a two-year study, Grossman and Tierney (1998) examined the impact of Big Brothers/Big Sisters youth participants. Using a random control pre-test/post-test design, they found that youth with mentors were less likely to start using illegal drugs or alcohol, and were less likely to hit someone or skip school. They also found that students were more confident in their school performance and developed better relationships with their families. Mentors were required to meet with their mentee two to four times per month for at least a year, with a meeting usually lasting three to four hours. The program did not aim to target specific problems with a youth, but rather to simply have an adolescent develop a friendship with an adult friend. Many times, this would help to filter any problems or concerns that were specific to that child. Volunteer mentors were carefully screened to ensure that they would form positive relationships, have the ability to meet necessary time commitments, and would be safe for youth participants. Volunteers completed intensive training and continued to be supervised and supported. Youth participants were also screened to ensure that they wanted mentors and that their parents wanted mentors for them.

Eight BBBS agencies were selected to participate in the study, using a random design. Half of the applicants to the agencies were randomly selected for the control group and placed on a waiting list for a mentor for 18 months. The other half were randomly selected to be matched with a mentor and were studied for 18 months. The adolescents ranged from 10 to 16 years of age. The final sample included 959 youth participants (487 treatment and 472 control). A little more than half of the participants were boys (62.4 per cent). Fifty-six per cent of the participants were visible minorities (71 per cent Black, 18 per cent Latino; rest were from a variety of other racial groups). Forty per cent were in households that received public assistance. Twenty-five per cent had been in homes where there had been divorce or a family history of alcohol or substance abuse, or had been victims of physical, emotional, or sexual abuse.

The findings showed the following:

Anti-Social Behaviours
Academic Attitudes, Behaviours and Performance
Family and Peer Relationships
Social and Cultural Enrichment

Over all, the study provides some evidence that mentorship can be beneficial to a youth’s development.

Summary of the Grossman and Tierney Study

David J. De Wit et al. – Review of Big Brothers/Big Sisters – a Canadian Perspective

The following study conducted by De Wit et al. holds great significance, because it is one of few studies that examine Big Brothers/Big Sisters programs from a Canadian perspective. The American equivalent of this study (Grossman and Tierney, 1998), examined earlier in this paper, found that the program was beneficial, as mentored youth were less likely to start using illegal drugs or alcohol, and less likely to hit someone or skip school. Mentees were more confident in their school performance and developed better relationships with their families. De Wit et al. (2007) conducted a randomized control study of two Canadian Big Brothers/Big Sisters agencies, which included 71 families and 30 adult mentors. Thirty-nine families were randomly placed in a BBBS program, while 32 families were placed on a waiting list for a BBBS mentor. Questionnaires were administered to children and their parents before the initial study as well as 12 months after analysis. In a 12-month follow-up period, another questionnaire was given to parents and mentors and face-to-face interviews were conducted with the children. The study found that there were non-significant outcomes as between the intervention and control groups in most areas. However, results did reveal that the intervention group slightly benefited in regard to symptoms of emotional problems, symptoms of social anxiety, teacher social support and social skills.

Child behavioural problems were measured by the response of the child and parent to a Strengths and Difficulties Questionnaire (Goodman, 2001; Goodman, Meltzer and Bailey, 1998). Depression was measured based on the children’s response using the Centre of Epidemiology Studies Depression Scale (Radloff, 1977). Academic performance was measured by child and parent reports. Peer self–esteem was measured using the peer dimension of the HARE self-esteem scale (Shoemaker, 1980). Children’s social support was measured using the Survey of Children’s Social Support (Dubow and Ullman, 1989). Children’s attachment to school was measured using a scale designed to assess attitudes towards school (De Wit et al., 2007). Children’s social skills were measured using the Elementary Level Student and Parent Forms of Social Rating Skills Rating System (Flanagan et al., 1996; Gresham and Elliot, 1990). Notable characteristics of the sample include the following: 51 per cent of the children were boys, 77 per cent came from single-parent homes, 35 per cent were visible minorities, and finally, the majority came from low socio-economic status backgrounds. The majority of the mentors were from white European ethnic backgrounds and were educated (88 per cent were in university or had completed university. Most were involved in full-time employment and lived in households with incomes of more than $40,000.

Summary of the De Wit et al. study

Overview of “Non–Violent” Mentorship Studies

In accordance with the Attachment theory, mentoring is believed to enhance an adolescent’s social and emotional development. By providing care and support, mentors can combat negative views that adolescents may hold about the self, thus increasing an adolescent’s ability to connect with positive and influential peers (Rhodes, 2002). If the mentor exemplifies knowledge, skills and behaviours that equate with future success, the adolescent will benefit. This is especially true for lower-income adolescents who may have insufficient role models, which may affect their belief in opportunity and success (Rhodes, 2002). From the studies reviewed, there is an overall consensus that mentorship does benefit at-risk youth. In many instances, we see pro–social changes that lead to a decrease in many destructive behaviours. For instance, decreased use of drugs and alcohol was noted in many of the studies. Positive views about the self and the future contributed to better academic achievement and better relationships with family and peers. However, it is important to note that mentorship was simply a component in additional curricula in all of the reviewed studies. The benefits of mentoring seem to increase when it operates in combination with a curriculum that stresses the development of positive skills and strategies for the future. How do these findings relate to violent behaviour? In many of the previous studies, various forms of defiant behaviours were used as a measurement to define at-risk youth. With the belief that aggression is developed from childhood, there is a strong indication that the strongest risk factors for youth violence include school withdrawal, poor academic achievement, absenteeism, tardiness, and poor peer relations (Borum, 2000; Denenberg, Denenberg and Braverman, 1998; Herrenkohl et al., 2000). Borum (2000) specifically states that school violence is the result of academic failure and poor school attachments. In accordance with Resilience Theory, the presence of a positive figure decreases the chances of an adolescent’s engaging in delinquent behaviour. If a positive individual intervenes during the various stages of youth development, there is a potential for negative behaviours to be reversed (Dubois, Holloway, Valentine and Cooper, 2002). The following section examines the use of mentorship as preventative intervention with respect to future violent behaviours.

Mentoring Programs and Violence Prevention

The following studies specifically examine mentorship and its effects on youth violence. The previous section of this review highlighted many positive results that help to increase confidence in mentorship’s use. We see that mentorship can have a major impact on at-risk youth; however, many questions still surround its effectiveness. There are varying levels of social disadvantage for children placed with a mentor. How can you address the needs and risks appropriately for every child? Violent behaviour may be a result of an array of factors that includes poverty, crime, abuse and behavioural issues (Rhodes et al., 2002). Not one study in the previous section was able to pinpoint the trajectory of change. By failing to address this topic, there continues to be a level of superficiality with respect to the mentorship paradigm (Rhodes, 2002). It is important to study all aspects of mentorship because that will shed light on what works and what does not. Without proper analysis, studies will continue to produce assumptions on how mentorship works. The following section will take a different approach and will review the quality and structure of the research design, and in turn, will help extract issues that continue to be unresolved. The following studies are presented by quality of evaluation. All three studies are similar in their area of concern and examine mentorship’s role in violence prevention. However, the three are significantly different in presentation and methodology. A low-quality study will be examined first. The study includes a pre– test/post-test design; however, the findings rely solely on self-reported testimonials, which limits scientific evaluation. A medium-quality study will follow, which examines a mentoring program whose distinct objective is to prevent violence among young African-American males. Again, the design lacks scientific evaluation. Finally, a high-quality study is examined, which compares mentored youth with non-mentored at-risk youth using a highly structured pre-test/post-test design.

Study 1: A Qualitative Evaluation of a Mentor Program for At-Risk Youth: The Participants’ Perspective by Diane de Anda (2001)

De Anda (2001) examines the first year of a mentor program called Project R.E.S.C.U.E (Reaching Each Student’s Capacity Utilizing Education). Firefighters served as mentors to at-risk high school youth situated in a low-income urban environment with elevated rates of youth and violent crime. Characteristics of the youth involved in the program included risk of dropping out of school, poor academic achievement, criminal delinquent behaviour, gang affiliations and substance abuse. De Anda (2001) theorized that the mentors would provide at-risk youth with a role model who could provide experiences that would not otherwise be available to them. This included psychological and emotional support to encourage behavioural and attitudinal improvements. The purpose of recruiting firefighters was to present a figure who would encourage an adolescent’s social and emotional development and educational and career motivation. The mentors and mentees were encouraged to meet on a weekly basis to spend time together to talk, complete homework, talk about vocational aspirations and engage in activities that involved common interests, but the time involved was left up to individual schedules. The study involved nine African-Americans, eight Latinos, and one bi-ethnic mentee. De Anda collected qualitative data to evaluate the findings. Using audio-taped interviews, de Anda interviewed the mentees before the study began and one year later, after observation.

Initially, de Anda found that many of the mentees joined the mentorship program in order to develop a communicative friendship with a mature and stable individual. Some mentees believed that better communication with a “mature” and “non-judgmental” adult would lead to self-improvement, allowing them to become better people. In the post-test interview, de Anda found that many mentees had positive things to say about their mentors and the program. Three mentees credited the program with changing their behaviour. One mentee acknowledged that the program taught him “how to stay out of trouble and respect others and stop the violence.” Another stated, “I feel I got a different meaning of life – meaning not fighting or stuff that would hurt anyone.” While the comments varied, de Anda found that all mentees had a positive experience. Many mentees felt that they developed long-term friendships with emotionally supportive individuals. De Anda states that the mentees gained positive values, goals and perspectives through their mentorship experience.

The mentors in the program also benefited from participation in the program. Two mentors were interviewed for the purpose of the study. The two joined the program because it offered them an opportunity to help their community by learning about the lives and experiences of the troubled youth in their neighbourhood. By gaining a better knowledge of the youth’s perspective, the mentor would be able to offer the guidance needed. One mentor wanted his mentee to have increased opportunity to grow and become a more productive citizen and student. As a result, he stressed confidence and assertiveness, and aimed to introduce his mentee to other people outside of his disadvantaged environment. The other female mentor wanted her female mentee to finish school, build self-esteem and improve on her way of dressing. By the end of program, eight of the mentees graduated from high school, four of whom were accepted in state college and two in part-time college. The remaining 10 continued in high school and vowed to decrease their violent behaviours. Over all, mentorship changed the lives of violent youth. By building strong bonds with mentors, mentees were able to make positive development changes emotionally and socially. Mentors provided opportunities that encouraged educational and career goals, resulting in a significant change in the life of each youth. De Anda concludes that the caring and supportive relationships changed the violent attitudes and behaviour into pro-social behaviours. De Anda (2001) attributes this change to the relationship between the mentor and mentee, since it led to healthy developmental progress. Having “volunteer” mentors showed the mentees that others cared about their welfare. To ensure success, de Anda (2001) suggests the following:

Summary of the de Anda Study

The study allows one to see how important a solid, open, communicational relationship is to at-risk youth. They require non-judgmental individuals who are available to listen and offer experiences the youths may lack in their own environments. A stable and caring relationship was an essential component, as it led to progressive development from negative and violent behaviours to more pro-social attitudes (de Anda, 2001).

Unfortunately, it is studies like the one conducted by de Anda that increase the belief that mentorship is the answer to youth violence. De Anda may have offered significant results, but still did not answer the question of why or even how mentoring changed the attitudes and behaviours of the delinquent youths. This study still makes it difficult to isolate the point at which the violent adolescent changed. The process of change has yet to be discovered. This type of study relies heavily on testimonials and does not facilitate a proper analysis of mentorship’s effectiveness. A major question that arises is how did the mentoring relationship progress and lead to change. From the perspective of industrial/organizational psychology, Kram (1988) suggests that mentorship relationships grow through various stages of development. There is an initiation phase, a phase of mutual admiration, where both the mentor and mentee work to make an impression on the other, and then, several phases where the nature of the relationship changes as the roles and responsibilities of the mentor and mentee become more clear. This may shed light on adult mentorship relationships, but solid empirical study in regard to the phases of adult mentorship is severely lacking. More information is required to pinpoint the stages of mentorship and the process of change.

The study fails to discuss any problems that may have risen between mentor and mentee, which is very probable with violent and delinquent youth. The actions of adults will be perceived and responded to in various ways; it simply depends on the adolescent’s ability to be open to new relationships (Rhodes, 2002). Every adolescent is different, thus making it difficult to believe de Anda when she states that the program resulted in “all” youth participants changing their violent behaviours. De Anda’s study sensationalizes mentorship without providing any empirical evidence. The following study improves on de Anda’s evaluation strategy.

Study 2: The building resiliency and vocational excellence (BRAVE) program: A violence-prevention and role model program for young, African American males by J.P. Griffin (2005)

Griffin (2005) examines the BRAVE (Building Resiliency and Vocational Excellence) pilot program, which acts as a substance abuse, violence prevention and role model program for young (16 to 20 years old) African-American men. Through a one-on-one relationship with an adult mentor, the program aims to establish positive roles in the male African-American community by offering coaching and career planning to instill a sense of purpose and success for the future. The program hopes to build resilience in these young men and combat negative social experiences. Proven resilient characteristics in individuals include positive social skills, solid analytical skills, and an overall ability to manage difficult life experiences. The idea is that if certain characteristics help youth to overcome adversities, disadvantaged young African-American males from underprivileged environments can adopt the same traits. The program looks to build on social skills, analytical skills and ways to cope with peer pressure. The foundation of the BRAVE program was that Black adolescent males will become less likely to engage in violent behaviour if they:

Mentors would be used to help teach, nurture and guide the mentees. In addition, the program included a school-based career development program that emphasized goal setting and life skills training. Mentors were 21 years of age or older African-American males, recruited through public announcements at an African-American university or by word of mouth. Before the start of the program, mentors were required to undergo background checks to ensure no past criminal activity and to attend a two-week training session to become acquainted with the programs goals. Influential adults were sought, and included an architect, a lawyer, a computer operator, and an information system director. The purpose of this recruitment style was to show young Black males that they could achieve similar success. Mentors were given clear instructions on how to communicate with youth and received continuous support from the program staff.

Initially, the program was conducted during two weekday evenings at an alternative school. It included 60 adolescent participants who were all noted for engaging in violent behaviour. Drug use was also prevalent among the participants. Almost all had experienced academic failure or were academic underachievers. Mentors and mentees attended group training sessions where they discussed substance abuse and violence prevention, life skills (social skills, communication, how to handle anxiety, how to be assertive), self-image, self-improvement, and decision-making skills. The classroom component included role-playing, discussion, and homework assignments. BRAVE program staff found that many of the participants continued to be involved in violence during the duration of the program. Several students reported being shot during the program and others discussed the probability of it happening in the near future. Female program administers, who would frequent classroom sessions, were regularly referred to as “bitches” and “hos.” As a result, the program implemented a “manhood development” curriculum that stressed the value of respect for women. Problems arose in the study, as Griffin felt that the sample size (60 students) was too small, making it difficult to statistically evaluate the effectiveness of the program. In regard to the study itself, it was difficult to find another alternative school that was willing to participate in the program to act as a control group. As a result of the difficulties of the pilot program, coordinators sought younger participants in a middle school in a disadvantaged neighbourhood. This increased the number participants. Despite the age change, the study found that the younger participants still encountered the same stresses as the previous age group. The younger mentees also suffered from academic failure, disengagement with school, and living in single-parent, low-income homes, and they were at risk of substance abuse and violent behaviour.

The revised program included structured social events in order to enhance the relationship between the mentor and mentee. Activities included participation in community events, attending the movies and attending sporting events. School sessions were now two to three days a week and included long-term and short-term goal discussions, which were meant to encourage the skills present in resilient youth. The sessions were also used to develop a sense of purpose and future through career planning. The program stressed responsibility, respect, maturity, and independence, and simply focused on redefining the idea of the Black male. Topics included male responsibility, the treatment of women, maturity and dignity. Skills developed in the classroom were practiced in the working world, as mentors were encouraged to expose their mentees to their place of employment.

Using student self-reports, the program assessed the frequency of engagement in violence and risky behaviour after the program. Results indicate that students reported less involvement in violent behaviour.

Summary of the Griffin Study

A clear objective was stressed, as was the process, in implementing the curricula for the mentoring program. As a result, it was possible to examine the various procedural steps to achieving a beneficial program. The program did not solely rely on mentorship. While it was a big part of the program, it was simply in association with other training perspectives. This identifies a need to include training programs that emphasize the development of career goals and self-improvement. We see from the pilot program that problems can arise with delinquent youth. Some were unruly toward program administrators and others continued to be involved in violence. In response, modifications were made to improve the curricula and address the needs of the youth participants. Acknowledging the difficulties encountered in the pilot program helped to create a new model for intervention. This supports the idea that mentorship programs must be open to change as the needs and concerns of the mentees may call for program modifications.

Once again, the weakness in this study is the lack of statistical evidence to prove the program’s effectiveness. The limited findings produced by the study were based solely on self-report surveys. The program sought to change characteristics of violent Black males, but the study did not use any measurements to show a process of change. Ultimately, if we cannot measure resilient characteristics, it is hard to identify resiliency as the product of change. The research falls short because there may be other factors in an individual that lead to resilient behaviour. The ability to adopt resilient behaviour may be the result of multiple factors. Werner and Smith (1989) question resiliency because they found that youth who flourish in the face of adversity tend to have hobbies, other interests and “a unique capacity” to engage with others. This could be the result of an array of factors. It could be the result of intelligence, character, physical appearance, demeanor, etc. However, the success with the program’s younger participants may help bolster resiliency theory. Once violence is instilled in one’s life, it may be difficult to change. Children are at an impressionable stage, and if positive life skills are instilled at a younger age, violent tendencies may be reversed (Satchwell, 2006).

We saw that in the pilot project, older youth continued to engage in violence and many refused a mentor. The younger participants were very receptive to the program, resulting in behavioural changes for the majority. This may be very important to acknowledge because it may exemplify the fact that younger participants are easier to build positive relationships with. Many of the older participants in the pilot program had already adapted to their environment and behaviours. Many continued to engage in violence during the duration of the program. Changing one’s perception of the self is not impossible, but often difficult. Intervention may be best when an adolescent is younger because they go through rapid changes in development and are more susceptible to influences, whether they are negative or positive (Rhodes, 2002). Having a positive, accomplished and successful individual as a mentor can have a major impact on a developing youth. George Herbert Mead (1934) surmised that adolescents developed an image of themselves through the perspectives of a significant individual in their lives. More positive influences can lead to a youth’s developing the desire to create and sustain that constructive image. Mentoring may be best as a strategy to prevent the development of future violent behaviours among youth.

Study 3: A school-based violence prevention model for at-risk eighth grade youth by S. A. Rollin et al. (2003)

Rollin et al.’s (2003) study centred on the effectiveness of violence prevention programs for at-risk grade eight students. The effectiveness was evaluated through the comparison of students, at three separate public schools, who were considered at-risk youth. At-risk status was based on one or more risk factors, which included the following: involvement in the juvenile justice system, one or more instances of fighting or unruly conduct in school, high absenteeism, or over-age in grade (Rollin et al., 2003). Students who participated in the program were randomly selected by school officials. The intervention group received community-based mentors who served as one-on-one career and emotional advisers. The intervention group was compared with a control group of students who did not receive a mentor. The study measured 1) unexcused absences; 2) number of in-school suspensions; 3) number of days of in-school suspensions; 4) number of out-of-school suspensions; 5) number of days of out-of-school suspensions; and 6) total number of infractions committed on school property. These were used to measure violence based on past research that identifies disobedience and defiance as early signs of childhood aggression. Numerous studies show that total number and days of out-of-school suspension is the result of fighting and physical violence (Costenbader and Markson, 1994; Dupper and Bosch, 1996; Imich, 1994; Skiba, Person and Williams, 1997). Similar findings were related to total number and days of in-school suspension (McFadden, Marsh, Price and Hwang, 1992). Unexcused absences were also used as a measurement, as researchers found that lack of supervision increases the opportunity to engage in delinquent behaviour (Skiba and Peterson, 1999). Infractions on school property were included as a measurement, as numerous researchers have identified antisocial and delinquent behaviours as strong indictors of violent conduct. School infractions included the following: arson, battery, breaking and entering, disorderly conduct, fighting, larceny and theft, sexual offences, battery/harassment, threat/intimidation, vandalism, and weapons offences. Other identified risks for school violence included academic failure, school expulsion, withdrawal, absenteeism, tardiness, poor peer relations and participation in delinquent activity (Borum, 2000; Denenberg, Denenberg and Braverman, 1998; Herrenkohl et al., 2000).

The intervention groups were placed in voluntary year-long internships that included a mentor. Students were place on work sites for the academic year, and the mentors were individuals from the work site who each agreed to supervise and guide an adolescent throughout the year. Students in the program received school credit as well as a small stipend every two weeks. Earnings were based on their performance and behaviour at their internship site. Students selected for the program had to go through a three-week orientation program. After passing the orientation, students were matched with an internship site and mentor based on career interests. Mentors and mentees met approx four days each week for two hours at each session. The program also hired a part-time teacher for each school to coordinate a curriculum that included skills training, behavioural lessons, and opportunities to explore career values, interests, and skills. A comparison was made between the intervention group (78 males/females) and control group (78 males/females) at the beginning of the school year and again at the end. The post-test showed a difference in the intervention group compared with the control group in all measurements. The most noted changes included the following:

School 1 – substantial decrease in total number of in-school suspensions, out-of-school suspensions, and number of school infractions for the intervention group.

School 2 – substantial decrease in total number of out-of-school suspensions and number of infractions for the intervention group.

School 3 – substantial decrease in total number and total number of days of out-of-school suspensions and number of infractions on school property for the intervention group.

Summary of the Rollin et al. Study

This intensive study was conducted using a controlled pre-test/post-test design, allowing for a proper comparison. What differed in this study is that it included three separate sets to compare and analyze, thus increasing its statistical power. In addition, the measurements in the study were justified and effective in addressing violent conduct. The study was well executed; however, it failed to address numerous issues. Most of the participants were African-American, limiting the scope of the analysis. Results may differ in the general population. The previous study reviewed in this literature review identified a possible need to modify programs based on “race,” gender, etc. The sample size was also very small, making it difficult to relate to the general population. The previous studies in this section relied heavily on self-reported surveys. This particular study did not include a self-report survey, which may have been beneficial in association with the other measurements. It would have been valuable to gain a better understanding of the personal perspectives of the youth participants as well as the teachers and mentors.

What lacks in all these studies, as well as most studies within the mentorship paradigm, are long-term effects. Very few studies look into the lives of the participants after the evaluation period. Research allows one to see preliminary benefits, but the failure to review after evaluation makes it difficult to see if behavioural changes continue into early adulthood. Researchers are exploring the benefits of mentorship, but may be missing the negative influences of mentorship. Researchers may want to consider the effects on other family members as an adolescent grows through a program. A sibling may experience feelings of hurt and jealousy as one family member is exposed to increased opportunities (Rhodes, 2002). Parents may feel inadequate in their parental abilities, as their child is being guided and supported by a non-familial individual. The mentee might experience problems from other peers and face ridicule and alienation from others in his or her community (Fordham and Ogbu, 1986). An array of issues may rise in the course of mentoring. It is vital for researchers to examine all aspects during and after evaluation in order to identify unforeseen positive and negative changes. The following section evaluates an intensive meta–analysis study conducted by DuBois et al., which gives a broader perspective on mentorship.

Dubois et al. -A Meta–Analytic Review

Dubois et al. (2002) conducted a significant meta–analysis that proved to be essential to the scope of mentorship. This intensive study reviewed 55 different empirical studies and provided an unbiased view into the effects of mentoring programs for youth. Over all, the study found that mentoring offered marginal benefits to youth. However, DuBois’s study did not discredit mentorship’s influence; what it did was help identify certain conditions and practices that increased mentorship’s success. All mentorship programs were different in scope, making it difficult to conclude that mentorship was the “answer” to combatting youth delinquent behaviours. The study examined the differences among the established mentorship programs and sought to extract the practices that produced better results.

The structure of the Big Brothers/Big Sisters program has continuously been hailed as the ideal because of its model of “best practices” for youth mentoring (Tierney, Grossman and Resch, 1995). As a result, Dubois aimed to compare mentorship programs to the practices of the BBBS. He found that mentorship programs differed in their basic goals and philosophy, whether they were generalized to promote positive youth development or more focused on specific goals such as employment and educational enhancement, all mentorship programs gained a positive response (Saito and Blyth, 1992). Dubois examined the procedures for recruiting and supervision, and found that mentorship programs differed in the level of training and supervision. Additional suggestions have examined the matching of mentors and mentees based on gender, race/ethnicity or mutual interests. The study looked into frequency of contact, monitoring the relationship, and the support and involvement of parents and guardians (Freedman, 1992; Hamilton and Hamilton, 1992, Saito and Blyth, 1992). In regard to the characteristics of mentored youth, DuBois found that most programs targeted the participation of at-risk youth based on individual and environmental circumstances. Other targets have included youth from single-parent homes and youth of specific racial or ethnic groups. Finally, Dubois examined the outcome of adult mentorship. Many programs have identified a wide variety of positive outcomes that include emotional and behavioural modifications, increased academic achievement, employment, and career development. The analysis examined whether the benefits of mentoring were evident across the various anticipated outcomes.

Dubois et al. (2002) located articles, through the Internet, dating back to 1970, which was the year that research on the outcomes of mentorship began to appear. His search of articles ended with 1998. A final examination of 55 articles was conducted. Each article was coded into six major categories that included a) report information (year of publication); b) evaluation methodology (type of research design, sample size, etc.); c) program features; d) characteristics of participating youth; e) mentor–mentee relationships; and f) assessment of outcomes.

Results show that there are only modest benefits for an average youth participating in a mentorship program. Dubois et al. (2002) found that the variations of mentorship programs produce varying effects. Evidence shows that only certain mentorship programs have a high level of success. They include programs that employ both theory-based and empirically based “best practices,” and that are also characterized by the establishment of a strong relationship between mentor and mentee. Programs that are poorly implemented can in fact be detrimental to youth. Youth from backgrounds of environmental risk and disadvantage appear to benefit most from mentorship programs. The following section goes more into detail.

Program Features
Characteristics of Youth
Summary of Results

Mentorship’s degree of impact falls short of what was expected in regard to psychological, educational and behavioural adjustments. However, DuBois et al. (2002) did find that mentoring has its benefits, but more so under an array of conditions. Mentoring can help an adolescent in regard to emotional, behavioural, educational, and career development if there continues to be ongoing training, structured activities for mentors and mentees, and clear expectations for frequent contact. Strong support by program administrators for the mentor, as well as the mentee, is crucial. Parental involvement can also increase the benefits of mentoring. Training needs to continue, even when relationships have been established. In regard to the mentor-mentee relationship itself, the structure of the relationship must involve frequent contact, emotional closeness and longevity. Mentees who have very strong relationships with their mentors, prove to benefit the most. Every aspect of the relationship can make an important contribution, thus leading to positive outcomes. Many times, the focus has been on screening, training and matching. While these continue to be vital, it is also very important to continue to train and provide support, even when relationships have been established. This has been lacking in many mentorship programs. Over all, adult mentorship programs can be an important component in a youth’s development when more consistent and natural positive mentoring is lacking in an adolescent’s life. At-risk children appear to benefit the most from these types of relationships. Youth who only face individual dysfunctions do not appear to benefit from mentorship, because such adolescents required more structured and professional help. However, youth who face difficult life circumstances benefit most from positive influences. As a result, mentorship programs may be used as a preventative measure to combat youth delinquencies.

What Works in Mentoring

Based on the results of the various studies reviewed in this review, and more specifically Dubois’s meta-analysis, it is evident that mentorship’s success relies on a well-planned and structured program. The following are identified to be the most effective practices for mentorship programs in reducing delinquent behaviour:

What Is Less Effective

The following have been identified as producing limited or marginal success in mentorship programs:

It is evident that a highly structured program is vital for a mentorship program’s success. The relationship between the mentor and mentee is of utmost importance, and a program should do all in its power to facilitate the bond. As a result, programs must properly screen and train mentors. Mentors must be made aware of the intense commitment and be willing to do what is needed to develop a strong bond and provide consistent support. Programs need to continuously monitor relationships in order to ensure stability and provide support for the mentor and mentee. Programs should also be responsible for providing structured activities and skill-building in order to promote development. It is important to note that benefits are gradual and require patience and extensive time commitments. As a result, long-term relationships should be encouraged.

Failure to produce a properly structured program will lead to detrimental effects for youth. Programs are less effective if they do not help to assist the development of a strong mentor-mentee relationship. If the necessary steps are not taken to ensure a solid match, youth can suffer additional emotional and social disadvantages.


Based on research into the mentoring paradigm, it is easy to see why mentorship is viewed as a viable “answer” to adolescent violence. However, the conclusion should be viewed with caution. At-risk adolescents can benefit from the presence of adult mentors; however, mentoring is not the answer per se, for it should not be used as a substitute for proper parenting and community support. Based on the examination of studies in this review, it is evident that mentoring does produce positive effects, but they are contingent on an array of factors that require extensive time, planning and commitment. If the necessary efforts are not made to create a solid program, mentoring can prove to be detrimental to at-risk children. Over all, mentoring seems to be a valid tool in preventing violence among at-risk youth; however, more research must go into its long-term effects in order to solidify its significance. Finally, a great deal of what we know about mentoring is based on findings from the United States. While some of these findings may be generalized to the Canadian context, it is clear that more evaluation research on mentoring programs is required in this country.


Achenbach, T. M. and C. S. Edelbrock. (1991). Behavior problems and competencies reported by parents of normal and disturbed children aged four through sixteen. Monographs of the Society for Research in Child Development, 46, 467–469.

de Anda, D. (2001). Qualitative Evaluation of a Mentor Program for At- Risk Youth: The Participants Perspective. Child and Adolescent Social Work Journal, 18(2), 97–117.

Arriaga, X. B. and S. Oskamp (Eds.). (1998). Addressing Community Problems: Psychological Research and Interventions. Thousand Oaks, CA: Sage Publications.

Baker, D. B. and C. P. Maguire. (2005). Mentoring In Historical Perspective. In D. L. Dubois and M. J. Karcher, Handbook of Youth Mentoring (pp. 14–29). Thousand Oaks, CA: Sage Publications.

Beam, M. R., C. Chen and E. Greenberger. (2002). The Nature of Adolescents' Relationships with their "very important" Non-parental Adults. American Journal of Community Psychology, 30(2), 305–325.

Borum, R. (2000). Assessing Violence Risk among Youth. Journal of Clinical Psychology, 56, 1263–1288.

Bowlby, J. (1969). Attachment and Loss: Vol 1. Attachment. New York: Basic Books.

Bowlby, J. (1988). A Secure Base: A Parent – Child Attachment and Healthy Human Development. New York: Basic Books.

Brody, J. (1991). New York State Mentoring Training Manual. New York: New York State Education Department.

Brown, J. D. (2008). Mentoring and Violence Prevention/Intervention: Promising Practices for Program Development and Children and Youth. London: The University of Western Ontario.

Canadian Council on Social Development. (2002). The Progress of Canada's Children 2001 – Highlights. Ottawa: Canadian Council on Social Development.

Canadian Institute of Child Health. (2000). The Health of Canada's Children, (3rd ed.). Ottawa: Canadian Institute of Child Health.

Costenbader, V. K. and S. Markson. (1994). School suspensions: A survey of current policies and practices. NASSP Bulletin, 78, 103–107.

Coulter, E. (1913). The Children in the Shadow. New York: McBride, Nast.

Darling, N., S. F. Hamilton, T. Toyokawa and S. Matsuda. (2002). Naturally Occurring Mentoring in Japan and the United States: Social Roles and Correlates. American Journal of Community Psychology, 30(2), 245–270.

Denenberg, T. S., R. V. Denenberg and M. Braverman. (1998). Reducing Violence in U.S. Schools. Dispute Resolution Journal, 53, 28-35.

De Wit, D. J., E. L. Lipman, M. Manzano, J. Bisanz, K. Graham, E. O’Neil, et al. (2007). Feasibility of a randomized controlled trial for evaluating the effectiveness of the Big Brothers Big Sisters Community Match Program at a national level. Children and Youth Services Review, 29, 383–404.

Dubois, D. L., B. E. Holloway, J. C. Valentine and H. Cooper. (2002). Effectiveness of mentoring programs for youth: A meta-analytic review. American Journal of Community Psychology, 30(2), 157–197.

Dubow, E. F. and D. G. Ullman. (1989). Assessing social support in elementary school children: The survey of children's social support. Journal of Clinical Child Psychology, 18(1), 52–64.

Dupper, D. R., and L. A. Bosch. (1996). Reasons for school suspensions: An examination of data from one school district and recommendations for reducing suspensions. Journal for a Just and Caring Education, 2, 140–150.

Elliot, D. S., D. Huizinga and S. S. Ageton. (1985). Explaining Delinquency and Drug Use. Beverly Hills: Sage Publications.

Flanagan, D. P., V. C. Alfonso, L. H. Primavera, L. Povall and D. Higgins. (1996). Convergent validity of the BASC and SSRS: Implications for social skills assessment. Psychology in the Schools, 33, 13–23.

Frecknall, P. and A. Luks. (1992). An evaluation of parental assessment of the Big Brothers Big Sisters program in New York City. Adolescence, 27(107), 715–718.

Freedman, M. (1992). The Kindness of Strangers: Reflections on the Mentoring Movement. Philadelphia: Public/Private Ventures.

Freedman, M. (1993). The Kindness of Strangers: Adult Mentors, Urban Youth and the New Volunteerism. San Francisco: Jossey-Bass.

Fordham, S. and J. U. Ogbu. (1986). Black students' school success: Coping with burden of acting white. Urban Review, 18(3), 176–206.

Furrano, K., P. A. Roaf, M. B. Styles and A.Y. Branch. (1993). Big Brothers Big Sisters: A Study of Program Practices. Philadelphia: Public/Private Ventures.

Garmezy, N., A. S. Masten and A. Tellegen. (1984). The study of stress and competence in children: A building block of developmental psychopathology. Child Development, 55, 97–111.

Goodman, G. (1972). Companionship Therapy: Studies in Structured Intimacy. San Francisco: Jossey-Bass.

Goodman, R. (2001). Psychometric properties of the strengths and difficulties questionnaire. Journal of the American Academy of Child and Adolescent Psychiatry 40(11), 1137-1145.

Goodman, R., H. Meltzer and V. Bailey. (1998). The strengths and difficulties questionnaire: A pilot study of the validity of the self report version. European Child and Adolescent Psychiatry, 7(3), 125–130.

Gresham F. M. and S. V. Elliot. (1990). Social Skills Rating System. Circle Pines, MN: American Guidance Service.

Griffin, J. P. (2005). The building resiliency and vocational excellence (BRAVE) program: A violence-prevention and role model program for young, African American males. Journal of Health Care for the Poor and Underserved, 16(4), 78–88.

Grossman J. B. and A. Johnson. (1998). Assessing the effectiveness of mentoring programs. In J. B. Grossman (Ed.), Contemporary Issues in Mentoring (pp. 25–47). Philadelphia: Public/Private Ventures.

Grossman, J. B. and J. P. Tierney. (1998). Does mentoring work? An impact study of the Big Brothers Big Sisters program. Evaluation Review 22(3), 403–426.

Hamilton, S. F. and M. A. Hamilton. (1992). Mentoring programs: Promise and paradox. Phi Delta Kappan, 73(7), 546–550.

Herrenkohl, T. I., E. Maguin, K. G. Hill, J. D. Hawkins, R. D. Abbott and R. F. Catalano. (2000). Developmental risk factors for youth violence. Journal of Adolescent Health, 26, 176–186.

Imich, A. J. (1994). Exclusions from school: Current trends and issues. Educational Research, 36, 3–11.

Jackson,Y. (2002). Mentoring for delinquent children: An outcome study with young adolescent children. Journal of Youth and Adolescence 31(2), 113–122.

Jane Addams Hull House Association – History. Retrieved February 24, 2008 (

Jones, K. (1999). Taming the Troublesome Child: American Families, Child Guidance, and the Limits of Psychiatric Authority. Cambridge: Harvard University Press.

Kaddin, A. E., A. Rodgers and D. Colbus. (1986). The hopelessness in children: Psychometric characteristics and concurrent validity. Journal of Consulting and Clinical Psychology, 54, 241–245.

Keating, L. M., M. A. Tomishima, S. Foster and M. Alessandri. (2002). The effects of a mentoring program on at-risk youth. Adolescence, 37(148), 717–734.

Kram, K. E. (1988). Mentoring at Work: Developmental Relationships in Organizational Life. Lanham, MD: University Press of America.

Lemmon, A. (2005). Mentoring 1: Whitelion Individualised Mentoring and Employment Program. Youth Studies Australia 24(4), 40–44.

Levine, M. and A. Levine. (1992). Helping Children: A Social History. New York: Oxford University Press.

Lipman, E. L., M. H. Boyle, M. D. Dooley and D. R. Offord. (2002). Child well-being in single-mother families. Journal of the American Academy of Child and Adolescent Psychiatry, 41(1), 75–82.

LoSciuto L., A. K. Rajala, T. N. Townsend and A. S. Taylor. (1996). An outcome evaluation of Across Ages: An intergenerational mentoring approach to drug prevention. Journal of Adolescent Research, 11(1), 116–129.

Masten, S. A. and D. J. Coatsworth. (1998). The development of competence in favorable and unfavorable environments: Lessons from research on successful children. American Psychologist, 53(2), 205–220.

McCord, J. (1991). Family relationships, juvenile delinquency, and adult criminality. Criminology, 29(3), 397–417.

McCord, J. (1992). The Cambridge – Somerville Study: A pioneering longitudinal-experimental study of delinquency prevention. In J. McCord and R. E. Tremblay (Eds.), Preventing Antisocial Behaviour: Interventions From Birth Through Adolescence (pp. 196–208). New York: Guilford Press.

McFadden, A. C., G. E. Marsh II, B. J. Price and Y. Hwang. (1992). A study of race and gender bias in the punishment of handicapped school children. Urban Review, 24, 239–251.

McPartland, J. M. and S. M. Nettles. (1991). Using community adults as advocates or mentors for at-risk middle school students: A two year evaluation of Project RAISE. American Journal of Education, 99, 568–586.

Mead, G. H. (1934). Mind, Self, and Society from the Standpoint of a Social Behaviorist. Chicago: University of Chicago Press.

MENTOR/National Mentoring Partnership. Retrieved February 26, 2008.

Merrit, D. M., G. J. Green, D. A. Jopp and J. G. Kelly. (1999). A history of Division 27 (Society for Community Research and Action). In D. A. Dewsbury (Ed.), Unification through Division: Histories of the Divisions of the American Psychological Association (pp. 73–98). Washington: American Psychological Association.

Piers, E. V. (1984). The Piers-Harris Children's Self Concept Scale, Revised Manual. Los Angeles: Western Psychological Services.

Powers, E. and H. Witner. (1951). An Experiment in the Prevention of Delinquency: The Cambridge – Somerville Youth Study. New York: Columbia University Press.

Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1, 385–401.

Rhodes, J. E. (2002). Stand by Me. London: Harvard University Press.

Rhodes, J. E., G. A. Bogat, J. Roffman, P. Edelman and L. Galasso. (2002). Youth mentoring in perspective: Introduction to the Special Issue. American Journal of Community Psychology, 30(2), 149–155.

Rollin, S. A., C. Kaiser-Ulrey, I. Potts and A.H. Creason. (2003). A school-based violence prevention model for at-risk eighth grade youth. Psychology in the Schools, 40(4), 403–416.

Ross, D. P., P. A. Roberts and K. Scott. (1998). Variations in Child Development Outcomes Among Children Living in Lone-Parent Families. Hull: Human Resources Development Canada, Applied Research Branch, Strategic Policy, W-98-7E.

Rutter, M. (1987). Psychological resilience and protective mechanisms. American Journal of Orthopsychiatry, 57, 57–72.

Rutter, M. and H. Giller. (1983). Juvenile Delinquency: Trends and Perspectives. New York: Guilford Press.

Saito, R. N. and D. A. Blyth. (1992). Understanding Mentoring Relationships. Minneapolis: Search Institute.

Satchwell, K. (2006). Mentoring Literature Review. Alberta Children's Services. Online:

Shoemaker, A. L. (1980). Construct validity of area-specific self-esteem: The HARE self-esteem scale. Educational and Psychological Measurement, 40, 495–501.

Sipe, C. L. (2002). Mentoring Programs for Adolescents: A Research Summary. Journal of Adolescent Health 31(Supp16), 251–260.

Skiba, R. J. and R. L. Peterson. (1999). Zero-tolerance: Can Punishment lead to safe Schools? Phi Delta Kappan 80, 372–376.

Skiba, R. J., R. L. Peterson and T. Williams. (1997). Office referrals and suspensions: Disciplinary intervention in middle schools. Education and Treatment of Children, 20, 295–315.

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

Werner, E. E, and R. S. Smith. (1989). Vulnerable but Invincible: A Study of Resilient Children. New York: McGraw-Hill.

Witner, L. (1911). Criminals in the making. The Psychological Clinic, 4, 221–239.

What is Youth Violence? Retrieved March 10, 2008

Yates, T. M., L. B. Egelannd and A. Sroufe. (2003). Rethinking resilience: A developmental process perspective. In S. S. Luthar (Ed.), Resilience and Vulnerability: Adaptation in the Context of Childhood Adversities (pp. 243–266). Cambridge: Cambridge University Press.

Zimmerman, M. A, B. J. Bingenheimer and E. D. Behrendt. (2005). Natural mentoring relationships. In D.L. DuBois and M. J. Karcher (Eds.), Handbook of Youth Mentoring (pp. 143–158). Thousand Oaks, CA: Sage Publications.

Zimmerman, M. A., K. J. Steinman and K. J. Rowe. (1998). Violence among urban African American adolescents: The protective effects of parental support. In X. B. Arriaga and S. Oskamp (Ed.), Addressing Community Problems: Psychological Research and Interventions (pp.78–103). Thousand Oaks, California: Sage Publications.

18 This chapter was written with the assistance of Kanika Samuels, BA, undergraduate program in criminology, University of Toronto.


Volume 1. Findings, Analysis and Conclusions

Volume 2. Executive Summary

Volume 3. Community Perspectives Report

Volume 4. Research Papers

Volume 5. Literature Reviews