Program

Jul 08, 2009

OVERVIEW

The Study of Mentoring in the Learning Environment (SMILE)
is a large-scale, randomized study of school-based adult mentoring for students
10-18 years of age. By the end of the school year, students in the treatment
group had received an average of 8 meetings and an average match length of about
3 months. Small, but positive, main effects of mentoring were found on
self-reported connectedness to peers, self-esteem, and social support from
friends. There were no impacts on other areas, including grades and social
skills. Analyses of subgroups found those who received the added mentoring
component had significant positive impacts among elementary school boys
(connectedness to school, connectedness to culturally different peers, empathy,
cooperation, and hopefulness) and high school girls (connectedness to culturally
different peers, global self-esteem, and self-in-the-present, and support from
friends).

DESCRIPTION OF PROGRAM

Target population: Elementary, middle, and high
school-age children (10-18 years).

The Study of Mentoring in the Learning Environment (SMILE)
is a large-scale, randomized study of school-based adult mentoring for students
10-18 years of age. Students received a mentor and support services through
Communities in Schools of San Antonio (CIS-SA). Support services for children
include educational enhancement activities, supportive guidance, enrichment
activities, and/or tutoring. Mentor-mentee meetings took place one hour per week
either during or after school; meetings outside school were prohibited. Mentors
received a one-hour orientation prior to mentee assignment. Mentoring was
unstructured and occurred wherever space permitted such as the library,
cafeteria, or CIS-SA school office. Mentors completed activity logs after each
meeting to document frequency and activity type.

EVALUATION(S) OF PROGRAM

Karcher MJ (2008). The Study
of Mentoring in the Learning Environment (SMILE): A Randomized Evaluation of the
Effectiveness of School-based Mentoring. Prevention Science,9(2),
99-113.

Evaluated population: 516 elementary, middle, and
high school students from 19 schools in the San Antonio area participated in the
study. The study population was predominately Latino, with twice as many girls
as boys, and the majority of students came from families with an annual family
income of less than $20,000.

Approach: Students were parent-, teacher-, or
self-referred to participate in a program of social and academic enrichment
support services provided by Communities in Schools of San Antonio (CIS-SA).
CIS-SA housed a case manager in each of the 19 participating schools and was
responsible for providing a range of support services to CIS-SA-enrolled
students. After completing pre-test assessments, youth were randomly assigned to
receive either a school-based mentor in addition to CIS-SA support services
(n=252) or CIS-SA support services only (n=264). In September, students
completed a battery of self-report questionnaires measuring 21 outcomes,
including: Connectedness (to school, teachers, peers, culturally different
peers); Self-esteem (global self-esteem, self-in-the-present,
self-in-the-future, peer, school, family, physical); Social Skills (empathy,
assertiveness, cooperation, self-control); Social Support (friends, family); and
Hope and Mattering (hope, mattering). Students were then administered a
post-test in late April. Additionally, students’ math and reading grades were
also collected.

Among mentors, more than one-half were Latino (54%), more
than one-third were white (35%), 5% were black, and 6% were of another race.
Nearly three-quarters of the mentors (73%) were female, and 43% spoke Spanish.
Additionally, 70% were college students, 13% were military personnel, 15% were
full-time employed adults, and 2% identified as other. Mentors were recruited by
CIS agency staff at military bases, local business, colleges, and within local
organizations, with a large majority being college students. Mentors were not
provided incentives for participation.

Results: By eight months after the start of the
study, youth receiving the additional mentoring services had met with their
mentor, on average, eight times–spanning, approximately, three months. At
post-test, mentored youth had higher end-of-year scores in all outcomes compared
with youth receiving support services only. However, only four statistically
significant main effects of mentoring were found: connectedness to peers
(p<.01), global self-esteem (p<.05), self-in-the-present (p<.01), and perceived
support from friends (p<.05). Main effects were not significant for the
remaining 17 outcomes. Computed effect sizes were small for connectedness to
peers (d=.25), global self-esteem (d=.16), self-in-the-present (d=.25), and
perceived support from friends (d=.18). Among all 21 outcomes, averaged effect
size was further reduced (d=.10). Despite these significant impacts on
self-esteem and peer relations, the author recommends caution, as some research
suggests peer-referenced self-esteem and peer relations can be predictive of
increased behavior problems, risk-taking, and school disengagement.

Findings from a three-way cross-level interaction analysis
of gender-by-school level on treatment condition, indicate greater benefits on
several outcomes among elementary school boys and high school girls. Elementary
school boys receiving the added mentoring component reported higher
connectedness to school (d=.86), connectedness to culturally different peers
(d=.58), social skills (empathy [d=.77] and cooperation [d=.71]), and
hopefulness (d=.73) compared with those receiving standard support services
only. Among high school girls, those receiving the added mentoring component
reported higher connectedness to culturally different peers (d=.34), self-esteem
(global [d=.27], self-in-the-present [d=.34]), and support from friends (d=.39)
than those receiving standard support services only.

Additionally, a few adverse effects were found of mentoring
for older boys and younger girls. More than one-half of the treatment
coefficients for high school boys receiving mentoring were negative (for
example, connectedness to school, self-in-the-future, cooperation), however,
only connectedness to teachers was significantly lower among high school boys
receiving mentoring services compared with their peers receiving support
services alone. One significant adverse effect, lower self-control, was found
among middle school girls receiving mentoring services compared with middle
school girls receiving support services only.

SOURCES FOR MORE INFORMATION

For more information about the Study of Mentoring in the
Learning Environment (SMILE), please visit:

http://www.utsasmile.org/
.

For more information about Communities in Schools of San
Antonio, please visit:

http://www.cissa.org/
.

For more information about the Hemingway Measure of
Adolescent Connectedness survey and scoring, please visit:

http://www.schoolbasedmentoring.com/
.

References

Karcher MJ (2008). The Study of
Mentoring in the Learning Environment (SMILE): A Randomized Evaluation of the
Effectiveness of School-based Mentoring. Prevention Science,9(2),
99-113.

KEYWORDS: Children (3-11),
Adolescents (12-17), Youth (16+), Elementary, Middle School, High School, Co-ed,
Hispanic/Latino, School-based, Mentoring, Tutoring, After School Program, Case
Management, Social Skills, Self-Esteem, Self-Efficacy, Reading, Mathematics,
Academic Achievement.

Program information last
updated on July 8, 2009.

Subscribe to Child Trends

Short weekly updates of recent research on children and youth.