Program Profile: Cross-Age Peer Mentoring Program

The program is a school-based peer mentoring program in which high school students provide one-on-one mentoring to late elementary and early middle school students. This program is rated Promising. The mentored children showed statistically significant improvement on measures of spelling achievement and connectedness to school and to parents compared with the control group. However, mentored and control group children did not significantly differ on connectedness to reading, future, or friends.

A Promising rating implies that implementing the program may result in the intended outcome(s).

This program's rating is based on evidence that includes either 1) one study conducted in multiple sites; or 2) two or three studies, each conducted at a different site. Learn about how we make the multisite determination.

Program Goals/ Target Population

The Cross-Age Peer Mentoring Program is a structured cross-age peer mentoring program in which high school students provide one-on-one mentoring to elementary and middle school students, either after school or on weekends, at the mentees’ schools throughout the school year. The program is designed to serve a mix of children who are both identified and not identified as at risk for social problems and academic disengagement. The program also is indicated for children attending schools in districts with high rates of high school dropout, which puts them at greater risk for school failure and dropout. The goals of the program are to promote mentees’ connectedness (to school, parents, and the future) and improve academic achievement by experiencing a supportive relationship with an older peer at school and practicing perspective taking, social skills, and school engagement.

Program Components

The Cross-Age Peer Mentoring Program is a year-round program with a structured curriculum in which high school mentors and elementary to middle school mentees engage with one another within a larger group context. It includes instructions for match (mentor–mentee) and group projects to practice perspective taking and social skills, and to engage in social connectedness activities (such as the two described in the next paragraph). Using a mentor manual, mentors are introduced to one of 36 activities in the curriculum before each meeting with their mentees. Although the pairs work one-on-one during most of the meetings, they also meet with other pairs in a group format during each meeting. Group interactions allow mentor–mentee pairs to participate in small group settings that meet some of their needs for peer engagement.

Two curriculum-based, connectedness promotion components, which are implemented during the first 6 months of the program, are teacher interviews and role-playing stories after reading moral dilemma books. In module 3, the mentee plans a teacher interview, rehearses it with his or her mentor, conducts it with a teacher, discusses the interview with his or her mentor, and then develops a poster and story about the teacher and presents it to peers. The goal of this activity is for the mentees to become more familiar and comfortable with (i.e., connected to) their teachers. In module 5, the connectedness-to-reading activities build on stories in eight brief children’s books. Mentor–mentee pairs read the books together and then role-play alternative outcome scenarios in a peer group format.

Another primary component of the program is family engagement, through quarterly recreational events called SuperSaturdays. For these events, parents of mentees are invited to spend time with their children’s mentors, see the work their children have done, and participate in activities (such as fieldtrips, picnics, and mock carnivals) with their children and their mentors. Parents are also involved through take-home activities and volunteering (such as helping with field trip transportation and food preparation at these Saturday events).

The program can be implemented through two different models. In the faraway model, mentors work with the children monthly during Saturdays and then more intensively during a 2-week summer day program. From September through May, nine Saturday meetings are held with the children and their parents. The monthly daylong sessions (96 hours) and the 2-week summer program (8 hours a day for 6 consecutive days, for 48 total hours) equal 144 planned contact hours between mentees and mentors. In the nearby model of the program, mentors meet with the children after school for 2 hours each day twice a week across 9 months (72 hours) and for 6 hours in quarterly Saturday events (24 hours). Combined with the summer program (48 hours), which is optional but strongly recommended, this approach also provides 144 contact hours.

Key Personnel

School personnel (teachers and counselors) coordinate and oversee implementation of the program, conduct ongoing training and supervision of the mentors, and collect implementation and outcome data.

Program Theory

Both the structure and curricula of the Cross-Age Peer Mentoring Program are designed to foster high school students’ leadership and collaboration skills, interest in serving their communities, and academic connectedness while simultaneously promoting elementary and middle school mentees’ connectedness (to school, family, and the future), self-esteem, and academic achievement. With its foundations in attachment and social-bonding theories of adolescent development (Ainsworth 1989; Jessor and Jessor 1977), ecological connectedness reflects youths’ engagement with and affection for people, places, and activities in their lives (e.g., school, family, friends; see Karcher and Sass 2010). Most curriculum activities foster conventional, future-oriented connections (e.g., to school) to offset the influence of unconventional connections to peers that increase during adolescence (Jessor and Jessor 1977). By seeking to promote connectedness to some of the relevant conventional domains of a child’s life (i.e., self, school, and family), the CAMP program aims to increase student achievement by fostering close relations with mentors and strengthening relationships with significant others in students’ lives, such as teachers, peers, and parents (Rhodes, Grossman, and Resch 2000).

The program is considered developmental because it aims to promote the psychosocial development of both mentors and mentees in two ways. First, it is expected that through program participation both mentors and mentees will form developmentally specific social and perspective-taking abilities, and will experience interpersonal support that serves to promote academic and future-oriented self-esteem and connectedness. Second, the tiered program allows elementary school student mentees to continue in the program for several years (through high school), developing into protégés (mentors in training) in middle school, and into mentors themselves in high school. Initially titled the Developmental Mentoring program, the name was changed to the Cross-Age Peer Mentoring Program given that “developmental mentoring” already had a specific connotation in the research literature.

Developmental Mentoring as conceptualized within this program has five core elements: 1) structured mentoring activities, 2) ongoing supervision for the mentors, 3) monitoring program development, 4) controlling the frequency of contacts, and 5) facilitating parent involvement. These elements are supported by research on youth mentoring programs (DuBois et al. 2002). Participating schools must provide a suitable context for mentors and mentees to meet regularly, consent to including children who are both at low and high social and academic risk, and assign at least one school staff member to structure or supervise the mentoring program.

Program Note

The Cross-Age Peer Mentoring Program has been adapted to reduce obesity (Smith 2011) and to prevent delinquency (Sar and Sterrett 2014). In addition, mentoring programs based on the Cross-Age Peer Mentoring Program framework and core elements program have been developed and implemented in Chicago (Douglas 2014; see www.yolobe.com), Australia (Willis et al. 2012), and in both San Antonio and Atlanta by Boy with a Ball (see www.boywithaball.com).

Study 1

Karcher, Davis, and Powell (2002) evaluated the Cross-Age Peer Mentoring Program (titled “Developmental Mentoring Program” at the time of the research) with public elementary school students as mentees and high school students as mentors. Teachers identified 30 fifth graders functioning at or above grade level, but who were attending schools in districts with high rates of high school dropout. The students were randomly stratified by grade and gender to the mentoring or the control group. However, only 13 students per group participated in follow-up assessments after 1 year because four students were dropped from the study when their families relocated out of the school district. The students in the mentoring and control groups did not differ statistically in age, gender, or ethnicity. There were nine girls and four boys in the mentoring group, and six boys and seven girls in the control group. The participating students were African American (42 percent), Mexican American (39 percent), and white (19 percent).

Mentors were 18 high school and 4 eighth grade students from a private Christian high school, who made a 2-year commitment to participate in the program. Twelve of the mentors were female. In terms of ethnicity, almost half were white (49 percent), followed by African American (29 percent), and Mexican American (22 percent). Mentors participated in the selection of their mentees, and all mentors worked with the same mentees for the 1-year duration of the study.

This study evaluated the faraway model of the program, as described in the program components section above, meeting monthly on Saturdays and during a 10-day summer program.

Outcomes included measures of math and spelling achievement, as well as student self-report measures of connectedness to parents, friends, school, and the future. The math and spelling achievement scales from the Wide Range Achievement Test (WRAT–3) were used to assess pretest and posttest math and spelling performance. The reading scale from the WRAT–3 was not used, as it could not be administered in a small group format. The Hemingway: Measure of Preadolescent Connectedness was used to measure the strength of connections to parents, friends, school, and the future.

Pretests were conducted in the spring before group assignment. Posttests were conducted the following spring, 1 year later. A preliminary omnibus test of differences revealed significant differences between the mentoring and the control groups on connectedness to friends and math achievement at baseline. Therefore, these two outcomes were excluded from analyses of program effects. Neither these measures nor others were included as covariates in analyses of program effects. Multivariate analyses of variance with change scores from pretest to posttest were conducted to determine the overall effect of participation in the Cross-Age Peer Mentoring Program on each outcome.

Study 2

Karcher (2005) evaluated the Cross-Age Peer Mentoring Program (titled “Developmental Mentoring Program” at the time of the research) with both high- and low-risk fourth and fifth grade students as mentees, and with middle and high school students as mentors. To identify high- and low-risk children, teachers provided assessments of risk in three domains (family, academic, school/peer behavior) using a 15-indicator checklist. Students for whom teachers identified three or more risks were considered high risk. Seventy-three children from fourth and fifth grades were recruited for the study from a pool of 198 children. Of these, 33 were randomly assigned to the mentoring group (based on the number of available mentors); the rest formed the control group. Twenty-eight of the 33 children in the mentoring group remained in the program at posttest. Complete pretest and posttest data was available from 24 children in the mentored group.

Thirty-three high school students were recruited as mentors through announcements in the school. A majority of the mentors were female (73 percent). The mentors were fairly evenly distributed in terms of grades. Seven were 9th graders, eight were 10th graders, nine were 11th graders, and six were 12th graders. In addition, six 8th graders, who were nominated by their middle school counselors, also served as mentors. Mentors were predominately white (34 out of 36); one mentor identified as Hispanic, and one identified as biracial. Two mentors dropped out of the program because of conflicts with extracurricular activities.

The control group was given the opportunity to participate in a tutoring program, but their attendance in that program was not tracked (and the tutoring program was not fully implemented until after this study was completed). Complete pretest and posttest data was collected for 30 of the 40 participants in the control group.

This study evaluated the nearby model of the program, as described in the program components section, with posttests completed before any children participated in the summer component.

Mentors and mentees participated in an initial 6-hour Saturday orientation session, during which they also self-selected one another, with 90 percent of mentees receiving their first- or second-choice mentor.

Outcomes included measures of connectedness to reading, school, parents, and friends, which were assessed using children’s self-reports on The Hemingway: Measure of Preadolescent Connectedness (Version 3). Total connectedness was also measured by using the average of the four domain subscales. Pretest and posttest assessments were conducted in groups of 15 to 20 and included youths in both mentoring and control groups. When possible, these were monitored by teachers and school counselors. A one-way multivariate analysis of covariance was conducted to assess the overall effect of participation in the Developmental Mentoring program on all four outcomes. Follow-up univariate analyses of covariance were conducted to test for between-group differences on individual outcomes. Students’ academic risk status and total connectedness scores at pretest were included as covariates.

For the faraway model evaluated by Karcher, Davis, and Powell (2002), all mentors participated in a 2-day training at the beginning of the program and also received monthly 1-hour group supervision. For the nearby model evaluated by Karcher (2005), all mentors received 8 hours of initial training, and most also participated in monthly 2-hour supervision.

Attendance information was collected for mentors and mentees and was calculated as the percentage of total number of days each attended for the nearby model of the program evaluated in Karcher (2005). Of the 48 meetings that occurred during the 6½ months of the study, mentors on average attended 72 percent of the meetings (range: 24 percent to 94 percent) and mentees on average attended 75 percent of the meetings (range: 10 percent to 95 percent)

The following implementation materials are available from the developer: the program manual, the mentor training guide, the connectedness curriculum, and mentor handbooks. Information about these materials can be secured at the program Web site (http://www.crossagepeermentoring.com) and can be purchased at http://developmentalpress.com

Technical assistance and training (both mentor and staff training) are available from the international youth development organization, Boy with a Ball (www.boywithaball.com), based in Buford, Georgia.

Additional Findings

Karcher (2005) examined whether there was a relationship between the attendance of mentors and change in self-esteem, self-management, and social skills of the mentees. Mentor attendance was a significant predictor of changes in mentees’ social skills and self-esteem, particularly in self-perceptions of attractiveness. Reported levels of social skills and self-perceptions of attractiveness were higher in the case of students whose mentors had higher attendance in comparison with those whose mentors had lower attendance.

Karcher (2009) examined academic connectedness and self-esteem among high school students who served as cross-age peer mentors in the implementation of the Cross-Age Peer Mentoring Program evaluated by Karcher (2005). High school student mentors from grades 9–12 who provided mentoring to students in grades 4 and 5 for a school year were compared with a group of students in grades 10 and 11 who did not provide mentoring as part of the CAMP program. Results showed that mentor students reported greater connectedness to friends, culturally different peers, their self-in-the-future, and school at posttest than the comparison group students. Additionally, students who had been mentors reported greater extracurricular self-esteem, sports/athletics self-esteem, and school self-esteem than comparison group students at posttest. The above results were seen in analyses controlling for mentor gender and age as well as pretest scores.

These sources were used in the development of the program profile:

These sources were used in the development of the program profile:

Following are CrimeSolutions-rated programs that are related to this practice:

Mentoring for Youth Development

This practice provides youth with a positive and consistent adult or older youth relationship to promote healthy youth development and social functioning and to reduce risk factors. The practice is rated Effective in reducing delinquency and improving educational outcomes; Promising in improving psychological outcomes and cognitive functioning; and No Effects in reducing substance use.

Evidence Ratings for Outcomes

	Crime & Delinquency - Multiple crime/offense types
	Education - Multiple education outcomes
	Mental Health & Behavioral Health - Psychological functioning
	Mental Health & Behavioral Health - Cognitive functioning
	Mental Health & Behavioral Health - Social functioning
	Drugs & Substance Abuse - Multiple substances

Why might a practice's outcome ratings differ from the ratings of specific programs encompassed by that practice?