You are seeing this message because your web browser does not support basic web standards. Find out more about why this message is appearing and what you can do to make your experience on this site better.

www.HFRP.org

The Harvard Family Research Project separated from the Harvard Graduate School of Education to become the Global Family Research Project as of January 1, 2017. It is no longer affiliated with Harvard University.

Terms of Use ▼

Program Description

Overview Sponsor-a-Scholar (SAS) is a college preparatory program that provides students with one-on-one, long-term mentoring, academic support and enrichment activities, college guidance, funds for college-related expenses, and ongoing staff support during high school and through college. SAS is dedicated to motivating students who might not otherwise consider higher education to stay in school and attend college. The program hopes to help Philadelphia high school students “make it” to college. SAS is the centerpiece of Philadelphia Futures, a nonprofit organization dedicated to increasing the percentage of Philadelphia comprehensive (neighborhood) high school students prepared for higher education, while simultaneously reducing the institutional barriers to their academic success.
Start Date fall of 1990
Scope local
Type mentoring, comprehensive services
Location urban
Setting public school
Participants high school students (who participate through attainment of their college degree)
Number of Sites/Grantees not applicable
Number Served over 600 students (since 1990)
Components The SAS program uses a three-pronged approach of (1) long-term mentoring, (2) financial incentives, and (3) academic enrichment and college guidance.

From the day they enter the program in ninth grade, SAS students are exposed to a series of activities that are designed to help them strengthen their academic skills, improve their standardized test scores, identify the college that is right for them, complete the college application process, obtain the financial aid needed to attend the school of their choice, and be prepared for college-level work. On entering the SAS program, each student is matched with a mentor. Mentors are asked to commit to a five-year relationship, beginning when the student enters the ninth grade and continuing through the first year of college. Philadelphia Futures provides mentors with ongoing training, guidance, and information about adolescent development and strategies for building trust. In addition, each student who successfully completes the program is eligible to receive $6,000 for college-related expenses, such as books, fees and transportation home. This money is donated by a “sponsor” and is disbursed on a semester-by-semester basis once the student is enrolled in college. Academic enrichment and college guidance activities include: tutoring, SAT prep courses, workshops on study skills, summer opportunities, college visits, financial aid workshops, and more general college preparatory activities.
Funding Level $365,429 in 1995–1996 and $621,251 in 2002–2003
Funding Sources foundations, corporations, and individuals


Evaluation

Overview The evaluation was designed to test the program's effectiveness in meeting its primary goal-encouraging college attendance.
Evaluators Amy W. Johnson, Mathematica Policy Research, Inc.
Evaluations Profiled Sponsor-a-Scholar: Long-Term Impacts of a Youth Mentoring Program on Student Performance
Evaluations Planned SAS' results (high school graduation rates, college matriculation, first-year persistence, and six-year graduation rates) are tested against the most relevant data available on an annual basis.
Report Availability Johnson, A. W. (1999). Sponsor-a-Scholar: Long-term impacts of a youth mentoring program on student performance. Princeton, NJ: Mathematica Policy Research.


Contacts

Evaluation Amy W. Johnson, Ph.D.
Senior Researcher
Mathematica Policy Research, Inc.
P.O. Box 2393
Princeton, NJ 08543-2393
Tel: 609-799-3535
Fax: 609-799-0005
Email: ajohnson@mathematica-mpr.com
Program Joan C. Mazzotti
Executive Director
Philadelphia Futures
230 South Broad St., 7th Floor
Philadelphia, PA 19102
Tel: 215-790-1666 ext.18
Fax: 215-790-1888
Email: joanmazzotti@philadelphiafutures.org
Profile Updated June 2, 2003

Evaluation: Sponsor-a-Scholar: Long-term Impacts of a Youth Mentoring Program on Student Performance


Evaluation Description

Evaluation Purpose To test the program's effectiveness in meeting its primary goal-encouraging college attendance. More specifically, several dependent variables related to participants were examined: (1) grade point average in tenth, eleventh, and twelfth grade, (2) extent of participation in college preparation activities, (3) sense of self-esteem, (4) college attendance in the first year following high school graduation and second year following high school graduation, and (5) college retention rate between the first and second years of college.
Evaluation Design Quasi-Experimental: The evaluation used longitudinal data on a sample of 434 students from 34 of the Philadelphia public high schools' (26 comprehensive schools, 7 magnet schools, and 1 vocational school) graduating classes of 1994–1997. Of these students, 180 participated in SAS, while the rest made up a matched comparison group. Each SAS participant was matched with two non-SAS students based on race, gender, and school attended. Comparison youth were further matched on academic achievement by selecting the two students with the closest grade point averages, one higher and one lower, to the SAS student. The comparison group sample consisted of 254 youth. (If only one of the two comparison group students showed up at the time the survey was administered, that student became the match. While this prevented the sample from having two matches for each participant, it was assumed that the one match included in the sample was the more “motivated,” and was actually a better match, given that motivation was presumed to be important for SAS students selection.)

Three significant differences were found between the SAS group and the comparison group in their background characteristics: comparison groups were significantly more likely to have a mother who was currently working, SAS youth were significantly more likely to indicate that they come from a family with strong support, and SAS students repeated significantly fewer grades in their educational careers. These differences were included as controls in the regression analyses testing for program effects and were not found to have altered any of the results.
Data Collection Methods Document Review: Coordinators' notebooks were collected and analyzed for the evaluation. Information collected included students' fulfillment of program responsibilities, effort invested by both parties in the mentoring relationship, mentor's relationship with the student's family, and student's need for and receipt of tutoring services to address academic problems.

Interviews/Focus Groups: Qualitative interviews were conducted with a subset of students from among a “continuum of success” in relationships with mentors (from weak and unsuccessful to strong and successful). This continuum of success was determined by analyzing coordinators' notebooks. These interviews focused on the nature of the mentoring relationships, the frequency of visits, and aspects of program supports. One guidance counselor at each of five high schools was also interviewed in 1996 to look at how the program was received by school staff and to learn about college preparation services, support, and encouragement provided by Philadelphia public schools.

Secondary Source/Data Review: Academic transcripts were collected directly from the school district for all students. These transcripts include demographic data, attendance data, and grades (by course) for each school quarter.

Students' school characteristics were analyzed through an examination of school district data, including racial composition of the school, school size, number of beginning teachers, average SAT scores, percentage of students dropping out each year, and status as a magnet, vocational, or comprehensive school.

Surveys/Questionnaires: Students were surveyed during each of the four years of the evaluation, through a self-administered questionnaire while they were still in school and a telephone survey after they had left school. The questionnaire asked about background characteristics, parental involvement in school-related issues, participation in activities and programs, sense of self-esteem, perceptions of peers, and plans and preparation for college. The post-high school surveys asked about college experiences, where applicable, and current activities and plans. Students also answered questions about their mentors and their mentoring relationship. Survey response rates were 98% in year one, 99% in year two, 92% in year three, and 95% in year four.

Mentors were surveyed once during their assigned students' senior year in high school. These surveys collected mentors' background information, their training for SAS, perceptions of the mentoring relationship, the extent of their involvement, and changes they would recommend. These surveys were mailed, and the response rates were 84% for the class of 1994, 89% for the class of 1995, and 85% for the class of 1996. Mentors for the class of 1997 were not surveyed since only the previous years' cohorts could have entered college during the period of data collection.
Data Collection Timeframe Data were collected from 1993 to 1997.


Findings:
Formative/Process Findings

Activity Implementation In the student surveys, 54% of SAS students reported the mentoring relationship had a big influence on whether or not they attended college, while another 31% indicated it made some difference.

In these same surveys, 70% of SAS students reported that the $6,000 in financial assistance made a big difference on their college attendance, while another 20% said it made some difference.

When asked to choose which component made a bigger difference in their attending college, students were very reluctant to choose. Despite this reluctance, 50% chose the mentoring component and 50% chose the financial assistance component.

Twenty-six percent of mentors said they contacted their student once a week, 54% said they contacted their student less than once a week, but more than once a month, and 20% said less than once a month.

Five percent of mentors indicated they saw their student once a week, 52% said they met less than once a week, but more than once a month, and 44% said they met less than once a month.

Eighty-four percent of students indicated that their mentor maintained some sort of contact with them during the first year of college.
Costs/Revenues In 1996, the per student cost to operate the program, excluding the $6,000 scholarship contribution, was approximately $1,485.

Since the program's inception, Philadelphia Futures has raised over $3 million in funds for its SAS students. Of this amount, approximately $1.5 million has already been disbursed to collegians, and over $1.5 million is committed to existing high school students and collegians.
Recruitment/Participation The average length of participation for the students studied in the evaluation was just under four years, with 52% participating for four to five years, 36% participating for three to four years, and 13% participating for less than three years. Of the 180 students in the evaluation, only 12 students completely left the program.

Students are nominated by their teachers and counselors. The program targets middle-achieving students (with grades primarily in the B–C range) who exhibit evidence of motivation through participation in extracurricular activities, good attendance, completion of program forms clearly and on time, ability to share information about themselves, an interest in relating to an adult, and an expressed interest in participation in the program and working toward the goal of college attendance. Students must be financially eligible, based on their qualification for the free and reduced-price lunch program, and their parents or guardians must express general support for the program's goals. Finally, the program selects students who staff believe will benefit from a long-term relationship with an adult and would probably not make it to college without additional support and encouragement.
Staffing/Training About a quarter of the students experienced a change in mentors: 23% had one change in mentor and 3% had two or more. These changes are due to mentors moving away, mentors' personal circumstances, and occasionally mentors' waning commitment, resulting in a request to leave the program.

A part-time coordinator works closely with each group of approximately 30 students. This staff position entails maintaining monthly contact with both the mentor and the student in order to monitor the progress of the relationship. They also review nominations, select students for participation, refer students for outside social services, and maintain notes recording the relationship between mentors and students and the students' progress toward college enrollment and attendance.

Mentors were evenly split in their responses regarding whether they felt they needed training to become a mentor.


Summative/Outcome Findings

Academic Compared to nonparticipants, SAS participants had significantly higher GPAs in both tenth and eleventh grades (79 versus 77, p<.05 and 78 versus 76, p<.10, respectively), although these gains were not sustained through twelfth grade. The evaluators conclude that SAS' policy of encouraging students to enroll in high-level courses and become involved in extracurricular activities may have resulted in this loss of impact.

SAS participants were significantly more likely than comparison group students to participate in college preparations activities (e.g., college visits, SAT preparation courses).

SAS students were significantly more likely, after controlling for other differences, than comparison group students to attend college during the first year after high school (85% versus 64%, p<.001). SAS students were also significantly more likely than the comparison group to attend college in the second year after high school graduation (73% versus 56%, p<.10).

SAS students averaged four applications to colleges, compared with three by students in the comparison group (p<.05).

SAS students were more likely to seek academic assistance during both the initial year and second year of college than the comparison group (74% versus 59% in year one, p<.05 and 70% versus 48% in year two, p<.05).

Although SAS students attended college in greater numbers than comparison groups, retention rates between the two groups did not differ significantly, although the retention rate was higher for SAS students than comparison group students (77% versus 69%).

SAS participation was found to significantly improve tenth grade GPA (p<.05), college preparation activities (p<.001), first and second year college attendance (p<.05 in both cases), and college retention (p<.10) for students coming from families that were minimally supportive. For students coming from families with moderate levels of support, SAS participation was related to increased college preparation activities (p<.001) and first-year college attendance (p<.001). None of the outcome measures was significantly improved for students from families with high levels of support.

Among the least motivated students, as measured by high levels of absenteeism, SAS participation was related to significantly better outcomes than among comparison group students on tenth grade GPA (p<.05), eleventh grade GPA (p<.001), and first-year college attendance (p<.001). For those in the mid-range of absenteeism rates, participation was related to increased first-year college attendance (p<.05). For those with low absenteeism, participation was not significantly related to outcome measures.

Among students with low ninth grade GPAs, participation was related to increased tenth grade GPA (p<.001), eleventh grade GPA (p<.001), first-year college attendance (p<.01), and second-year college attendance (p<.05). For those with the highest ninth grade GPA, participation was not related to program outcomes.

Among students in schools with the highest dropout rates, program participants did significantly better than their peers on tenth grade GPA (p<.10), eleventh grade GPA (p<.05), and on first-year college attendance (p<.05). For those in schools with mid-level dropout rates, SAS students did significantly better on first-year college attendance (p<.001), second-year college attendance (p<.05), and college retention (p<.05). For those in schools with low dropout rates, SAS students performed better than nonparticipants on involvement in college preparation activities (p<.05).

For SAS students in a comprehensive high school, participation was related to higher tenth grade GPA (p<.05), eleventh grade GPA (p<.1), first-year college attendance (p<.001), second-year college attendance (p<.001), and college retention (p<.05). For SAS students in magnet or vocational high schools, participation was not related to any of these outcome measures.

SAS students whose mentors contacted them most often did significantly better on tenth grade GPA (p<.001), eleventh grade GPA (p<.001, first-year college attendance (p<.001), second-year college attendance (p<.05), and college retention (p<.05). SAS students with moderate mentor contact had significantly higher rates of first-year college attendance (p<.001). SAS students with low mentor contact had significantly higher tenth grade GPAs (p<.05).

SAS students who saw their mentors most often had higher GPAs in the tenth, eleventh, and twelfth grades (p<.001, p<.001, and p<.05, respectively). SAS students who saw mentors a moderate amount had significantly higher eleventh grade GPAs (p<.10). Students who saw their mentors the least had no significant gains, and actually had significantly lower self-esteem (p<.10), suggesting to the evaluator that relationships lacking commitment may be more harmful than no relationship at all.

Neither mentor's prior involvement in a mentoring program, age, having children at home, or urban residence was related to more positive outcomes for youth. Students of the same race as their mentors did not consistently do better or worse than students with mentors of a different race.
Youth Development SAS students and comparison group students did not differ in measures of self-esteem after participation in the program.

SAS students were more likely than those in the comparison group to live away from home after high school, either in college housing or their own apartment (p<.05).

© 2016 Presidents and Fellows of Harvard College
Published by Harvard Family Research Project