Nov 09, 2011


In addition to the community-based mentoring program, Big Brothers Big Sisters (BBBS) has also developed a separate school-based mentoring program for children and adolescents in fourth through ninth grades. For the school-based mentoring program, matches meet on school grounds either during or after school, rather than meeting at outside locations, as is the case with BBBS community-based mentoring program. In an experimental evaluation of the school-based mentoring program, impacts were found on overall academic performance, quality of class work, number of assignments completed, presence of a special adult, and scholastic efficacy at the end of the first school year. No impacts were found for drug and alcohol use, self-esteem, relationships with peers or parents or behavior outside of school. Moreover, 15 months after the baseline at the second follow-up during the next school year, few impacts were found, the only impact was a greater likelihood of having a special adult in their lives.


Target participant: At-risk school-age children and adolescents.

In addition to the Big Brothers Big Sisters (BBBS) community-based mentoring program, BBBS has also developed a separate school-based mentoring program for children and adolescents in grades fourth through ninth. For the school-based mentoring program, matches meet on school grounds either during or after school. The school-based mentoring program is not simply a tutoring program; nor is it the community-based BBBS program relocated within schools, as more structure is involved such as outlining activities for mentees-mentors. Mentors can be adults, college students, or high school students. In general, in the school-based mentoring program, mentors, either adults or high school youth, meet with their mentee on school grounds one hour per week to provide support, academic assistance, and general friendship. However, the school-based mentoring program does preserve some of the infrastructural components of the community-based model such as screening applicants, training, and supervising the relationships.

A separate, community-based mentoring program has also been developed by Big Brothers Big Sisters. A write-up of the description of the community-based mentoring program, along with evaluation findings, is found on the LINKS database.


Herrera, C., Grossman, J.B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a Difference in Schools: The Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures.

Evaluated population: A total of 1,139 students in 4th-9th grade (9-16 years) were selected from ten Big Brothers Big Sisters agencies across the U.S. Among participants, 85 percent were in grades 4-6 and more than one-half were female (54%). More than one-third of study participants (37%) were white, 23 percent were Hispanic/Latino, 18 percent were black, 13 percent were multi-racial, and 6 percent were Native American. In addition, more than two-thirds of participants (69%) received free or reduced-priced lunch, and about one-half of participants (51%) were rated by their teacher as having difficulties with their overall academic performance at baseline.

Approach: Students were recruited into the BBBS agencies through school referrals. Ten BBBS agencies were selected based on their leadership in management, experience operating a school-based mentoring program for at least four years, experience serving at least 50 students per year and serving both girls and boys, and using two different types of volunteer populations such as schools and employees from nearby businesses. A total of 554 volunteer mentors, 72 percent female and 77 percent white, completed baseline surveys. Two-thirds of mentors were enrolled in high school or college, and approximately two-thirds reported receiving training before or during their meeting with mentees.

Students completed a baseline survey and were then randomly assigned so that, at each school studied, approximately half of the students were assigned to the treatment group and half were assigned to the control group. Each student in the treatment group (n=565) was matched with a volunteer mentor. Students in the control group (n=574) were placed on a waiting list to be matched with a mentor 15 months after the study’s completion.

Meetings between most (approximately 80%) of the students and their mentor took place once a week or more often, with the remaining meetings taking place at least monthly. Approximately two-fifths of meetings (39%) lasted one hour or more. Due to typical delays in starting a program at the beginning of a school year, treatment group students received an average of approximately five months of mentoring.

Students were assessed on a variety of outcomes at baseline, at the end of the first school year (first follow-up period), and just before the winter break of the next school year (second follow-up period). A total of 31 outcomes were measured through both teacher and student report, which are detailed in the table below.

Teacher-rated outcomes Student-rated outcomes
Overall academic performance GPA
Written and oral language School skipping
Reading Teacher-student relationship
Science Scholastic efficacy
Social studies Academic self-esteem
Math Connectedness to school
Quality of class work College expectations
Number of completed
Substance use
School preparedness Misconduct outside school
Classroom effort Perceptions of peer emotional
Task orientation Self-worth
Unexcused absence Relationship with parent
Serious school misconduct
(fighting, sent to principal’s office, suspensions)
Difficult in class
Teacher-student relationship
Positive classroom affect
Prosocial behavior
Social acceptance

Results: At the first follow-up period at the end of the school year, significant impacts were found on overall academic performance (effect size [ES] =.09), quality of class work (ES=.12), number of assignments completed (ES=.14). Impacts were also found on student-reported scholastic efficacy (ES=.11). Treatment group students were also more likely than control group students to report the presence of a significant adult friend that they could look up to, talk with, and who had influence over their choices. There were marginal impacts on written and oral language, science, unexcused absences, and engaging in serious school misconduct. However, there were no impacts on the remaining outcomes assessed.

At the 15-month follow-up, 52 percent of the treatment group students were still receiving mentoring. There were no significant impacts at this follow-up period; however, there were marginal impacts on skipping school and college expectations.

The average annual cost per student was $987. Programs closer to schools tended to have lower costs due to transportation. The most important cost factor was the ratio of students to staff. The more mentor-student relationships that each staff member has to supervise, the lower the cost. However, the authors note it is important to find balance between cost effectiveness and program quality.

Herrera, C., Kauh, T. J., Cooney, S.M., Grossman, J.B., McMaken, J. (2008). High School Students as Mentors: Findings from the Big Brothers Big Sisters School-based Mentoring Impact Study. Philadelphia: Public/Private Ventures.

Evaluated population: A total of 1,139 students in 4th-9th grade (9-16 years) were selected for the study. Among participants, 85 percent were in grades 4-6 and more than one-half were female (54%). More than one-third of participants (37%) were white, 23 percent were Hispanic/Latino, 18 percent were black, 13 percent were multi-racial, and 6 percent were Native American.

The present study specifically examined the impacts on student mentees matched with high school mentors and compared them with student mentees matched with adult mentors. Among high school mentors, 49 percent were juniors, 26 percent were seniors, and 25 percent were either freshmen or sophomores. Additionally, 79 percent of high school mentors were female, 81 percent were white, 74 percent participated in two or more extracurricular activities, and 40 percent had paid employment. High school students often served as mentors as part of a class or community service requirement. High school mentors were generally matched with elementary school mentees, thereby avoiding a similar age pairing of mentors and mentees.

Approach: Please see Herrera et al, 2007 for a description of the 31 outcomes assessed. All mentee-focused outcomes were assessed at the end of the first school year.

Results: At the end of the first school year, there were marginal impacts for Littles (mentees) on only social acceptance and assertiveness among students paired with a high school mentor compared with students paired with an adult mentor. In contrast, there were significant impacts on GPA, classroom effort, and prosocial behavior among students paired with an adult mentor compared with students paired with a high school mentor; there were also marginal impacts on being difficult in class, college expectations, and relationship with parents.

Herrerra, C., Grossman, J.B., Kauh, T.J., & McMaken, J. (2011). Mentoring in schools: An impact study of Big Brothers Big Sisters School-Based Mentoring. Child Development, 82,346-361.

Evaluated Population: A total of 1,139 4th to 9th grade students from 71 schools in 10 cities were evaluated. Participants ranged in age from 8 to 18 years, with an average age of 11. A total of 63 percent of the sample were minorities, with 23 percent Latino, 18 percent African American, and 13 percent multiracial. The sample was 54 percent female, 69 percent were receiving free or reduced price lunch at baseline, and 39 percent came from single parent households.

Approach: Students were stratified by school and randomly assigned to the intervention or control group. Intervention students were then matched with a mentor. Data were collected through self-report and teacher surveys at baseline, 9 months, and 15 months. Data were collected on academic performance, classroom effort, self-perception of academic abilities, unexcused absences, school misbehavior, substance use, misconduct outside school, social acceptance, teacher relationship quality, parent relationship quality, self-worth, presence of a special adult, stressful life events, and extracurricular involvement. (See above for more details.)

Results: There was one significant difference between the intervention and control groups at baseline: the intervention students were less likely to have used substances. A total of 94 percent of the sample completed the 9-month assessment, and 84 percent completed the 15-month assessment. At 9 months, there was a positive impact on academic performance and self-perception of academic abilities, but these were no longer significant at 15 months. There was a positive impact on presence of a special adult, which remained significant at 15 months. All other outcomes were not significant at 9 months or 15 months.

Subgroup analyses were done by gender, race, and age, but there were no significant results for race or gender at either time point. No subgroup differences were found for age at 9 months. At 15 months though, among older children (middle school and high school students), the intervention group had significantly lower class effort compared with those in the control group.


For more information, please visit:

Herrera, C., Grossman, J.B., Kauh, T. J., Feldman, A. F., McMaken, J., & Jucovy, L. Z. (2007). Making a Difference in Schools: The Big Brothers Big Sisters School-Based Mentoring Impact Study. Philadelphia: Public/Private Ventures.

Herrerra, C., Grossman, J.B., Kauh, T.J., & McMaken, J. (2011). Mentoring in schools: An impact study of Big Brothers Big Sisters School-Based Mentoring. Child Development, 82,346-361.

Herrera, C., Kauh, T. J., Cooney, S.M., Baldwin Grossman, J., McMaken, J. (2008). High School Students as Mentors: Findings from the Big Brothers Big Sisters School-based Mentoring Impact Study. Philadelphia: Public/Private Ventures.

For more information about the Big Brothers Big Sisters of America program, please visit:

KEYWORDS: Children (3-11), Adolescents (12-17), Youth (16+), Young Adults (18-24), Mentoring, High-Risk, Reading, Mathematics, Academic Motivations/Self-concept/Expectations, Academic Achievement, Alcohol Use, Marijuana/Illicit/Prescription Drugs, Aggression, Delinquency, Life Skills, Self-esteem, White/Caucasian, Hispanic/Latino, Black/African American, Multi-racial, School-Based, Cost.

Program information last updated 11/9/11.