University Students’ Expectations For Mentoring High-Poverty Youth

Carolyn Hughes and Sara J. Dykstra

Students in service–learning courses represent a source of quality mentors for youth. Pre- and post-mentoring measures confirm high initial expectations.



What are the motivations of college students who mentor youth from high-poverty backgrounds? Our team surveyed university students before and after an elective service-learning course that included voluntary mentoring of high-poverty youth. Mentors were motivated primarily by the opportunity to have a positive impact on youth through (a) being a role model, friend, source of support, and caring adult, and (b) increasing their own understanding of inner-city schools and culture in order to serve youth better. Following the experience, mentors reported having largely achieved these aims. In addition, their responses reflected greater confidence in themselves as mentors, better understanding of the challenges and contexts of high-poverty environments, and a higher level of cross-cultural comfort. Based on these findings, we propose strategies for future mentoring efforts in the context of service learning.



Mentoring programs addressing the healthy development of U.S. youth have burgeoned during the past decade (Rhodes and DuBois, 2006). Their popularity has grown because mentoring is viewed as an inexpensive and effective means of positively influencing at-risk youth (Portwood, Ayers, Kinnison, Warris, and Wise, 2005). At-risk youth — particularly from high-poverty backgrounds — may experience substantial benefits from mentoring (DuBois, Holloway, Valentine, and Cooper, 2002).

Although mentoring at-risk youth has been studied considerably, little is known about why mentors engage in and sustain mentoring relationships. This research gap coincides with the finding that, although an estimated 3 million U.S. youth currently are being mentored, 15 million more are in need of a caring, supportive adult (MENTOR, 2006). Indeed, recruiting and retaining mentors is a prevailing national problem (Wandersman et al., 2006). More information is needed regarding what motivates people to volunteer to mentor and their mentoring expectations and assumptions — particularly with the growing number of at-risk youth (Larson, 2006). The primary goal of mentoring is to address the needs of youth, but if mentors’ expectations of and motivation for mentoring are not addressed, we are unlikely to close the mentoring gap. Because mentoring is a reciprocal, potentially mutually beneficial relationship, it is critical to know mentors’ assumptions and expected benefits.

University students in service-learning courses represent a potential source of quality mentors for youth. Service learning is designed as a reciprocal relationship in which students address a community need and increase their civic engagement and social awareness by reflecting on their service activity (Schmidt, Marks, and Derrico, 2004). Service-learning students are likely to be trained and receive ongoing support in class. Further, there is evidence that the reciprocal benefits experienced through community involvement promote students’ civic engagement and long-term involvement in mentoring after a course terminates (McClam, Diambra, Burton, Fuss, and Fudge, 2008).

Few studies exist of university-based service-learning mentoring programs for at-risk youth (see DuBois et al., 2002; Rhodes and DuBois, 2006). DuBois and Neville (1997) examined the relation between characteristics of mentor-mentee relationships and mentors’ ratings of perceived benefits for youth. However, this study did not give details of the related service-learning course or where or what mentoring activities occurred. Further, despite growing U.S. school dropout rates, particularly in high-poverty high schools (Orfield, 2004), and the potential for university students as mentors, we found no published study of a university-based service-learning program in which mentoring of high-poverty youth occurred in a high school setting. If we intend to increase enrollment is such programs, it is critical to determine why university students may choose to engage in mentoring.

Noting the growing need for mentors of at-risk youth and the lack of research on service-learning mentoring programs for high school students, we conducted an exploratory examination of the assumptions and expectations of university students voluntarily enrolled in a service-learning course designed to mentor youth attending high-poverty high schools. We specifically wanted students who were mentoring voluntarily, not as a course requirement, in order to determine why people may choose to engage in mentoring.

We used a survey to answer the following questions:

1. What are the expected benefits of the mentoring process for mentors and mentees?

2. Do rankings of mentors’ priorities for mentoring change over time?

3. What assumptions do mentors have regarding the mentoring process itself, particularly with respect to interacting with youth from high-poverty backgrounds?

4. What is the mentors’ comfort level in high-poverty environments?

5. To what extent do mentors’ assumptions relate to their racial, ethnic, or socioeconomic background?

Responses to a questionnaire were compared before and after the mentoring experience to determine if expectations were met. In addition, we collected and examined qualitative data.



Service-Learning Class

This study was conducted at a private urban university in the southeastern United States serving approximately 12,000 students of whom 69% were white, 8% black, 6% Asian, 5% Hispanic, and 12% unreported. Most (99%) undergraduates lived on campus.

Participants in the study were enrolled in an elective service-learning class entitled “High-Poverty Youth: Improving Outcomes.” The purpose of the course was to improve outcomes for youth attending high-poverty high schools through mentoring and to increase participating students’ awareness of (a) the effects of poverty on youth, and (b) economic disparities across neighborhoods, schools, races, and ethnicities. Students met in class twice weekly in the 16-week course. The instructor facilitated discussions on mentor training, racial segregation, white privilege, unemployment, the working poor, high school dropout rates, and related topics, augmented by readings, videos, and guest speakers.

Each student mentored a high school student once or twice per week at the mentee’s school. The course required participants to complete ongoing reflective journals and to share their mentoring experiences in classroom discussions and focus groups. The instructor graded the journals based on activities, feelings, and experiences in relation to class content and awareness of the effects of poverty on youth and their families. Although the course required 22 hours of mentoring, over half completed 5 to 10 hours more, as reflected in their journals.

Each university student was matched one-on-one with a mentee based on similar interests and class schedules. Mentoring was conducted during class study time when mentees were allowed to work individually on class assignments. Mentor and mentee typically interacted in a quiet corner of the mentee’s classroom, the school library, or computer center, as directed by the teacher. As identified in mentor journals, activities included befriending, tutoring, supporting mentee class performance, and assisting in the college application process. In addition, pairs spent time attending sporting events, going to the mall, seeing movies, or eating out. Mentors and mentees also communicated with each other via e-mailing, phoning, and text-messaging. Approximately one-third of mentors reported to the class instructor or mentees’ teachers that they maintained their mentoring relationships one year or more beyond the semester either by spending time with their mentees (if continuing to live nearby) or electronically (if having left the area).



All 29 students in the class consented to participate in the study. The class consisted of 4 sophomores, 11 juniors, 13 seniors, and 1 graduate student; 26 were female (18 white, 7 black, and 1 Asian), and 3 were white males; 14 reported having attended a public high school, 13 a private high school, and 2 did not specify. High schools attended were 16 suburban, 7 inner city, 5 urban not inner city, and 1 unspecified. Family income during high school was reported as upper income (15), middle income (11), lower income (2), unspecified (1). Students were not asked to quantify income levels. All but four students reported prior experience with high-poverty youth, typically tutoring. None reported previously serving as a mentor to youth.



Mentees attended one of two comprehensive high schools in the local metropolitan school district of 73,000 students. Both schools offered courses in academic and career preparation and served students from high-poverty neighborhoods. High school A had a graduation rate of 42% and enrolled 1,267 students, of which 78% were black, 19% white, and 3% other ethnicities (Hispanic, Asian, Native American). High school B had a graduation rate of 50% and an enrollment of 1,407 — 70% black, 23% white, 5% Hispanic, and 2% Asian. Both schools were identified as “needing improvement” based on graduation rates and test scores prescribed by No Child Left Behind. Eighty percent of residents in the students’ neighborhoods were black, 40% unemployed, and 44% of families lived below the poverty level. Typically, the head of household was a single female receiving or previously receiving public assistance (U.S. Census Bureau, 2006).


Questionnaire development. We developed a questionnaire to assess expectations and assumptions about mentoring youth from high-poverty backgrounds. First, we chose items from questionnaires used in previous investigations of mentoring relationships (e.g., MENTOR, 2006; Rhodes, 2002). Second, we drew additional items from end-of-semester reflection papers written by former students in the same service-learning mentoring class to ensure relevance of content. Third, we field-tested items by asking undergraduates not in the class to complete the questionnaire and provide feedback, which we incorporated into the final version of the survey. Fourth, we constructed a second survey changing wording to the past tense for the post-mentoring survey.

There were two parts to the questionnaire. In the first part, we asked mentors to provide gender, ethnicity, year in school, major, type of high school attended, type of community lived in during high school, and family income level. The second part contained four sections relating to the mentoring experience. Section 1 addressed mentor assumptions regarding expected benefits to themselves, such as building friendships, experiencing personal growth, and applying knowledge from class to real life experiences. In the pre-survey (11 items on a 1-5 scale where 1 = “not at all important” and 5 = “very important”), we asked mentors to rate, “How important are the following potential benefits to YOU, the mentor?” In the post-survey, we asked, “How important have the following benefits been to YOU, the mentor, during your mentoring experience?” (Table 1).

The second section queried mentors regarding the expectations they held for benefits their mentees might experience, including having a consistent, caring adult, improving academic performance, and completing the college application process. Section 2 asked, “How much do you expect your MENTEE(S) to gain in the following areas?” on the pre-survey, and, “How much did your MENTEE(S) gain in the following areas?” on the post-survey on nine items using a 1-5 scale where 1 = “No gain at all” and 5 = “Gain a lot” (Table 2). Section 3 questioned mentors’ assumptions with respect to the mentoring process, the role of a mentee’s cultural background in the mentoring relationship, and challenges facing inner-city city youth. Mentors were asked, “Do you agree/disagree with the following statements?” for both the pre- and post-survey using a 5-point scale marked “Strongly disagree” (1) and “Strongly agree” (5) in response to 11 items (Table 3). The final section asked mentors to report their level of comfort serving in a mentoring role, interacting with people with different backgrounds, and discussing race-related issues. Mentors were asked, “What is your level of comfort?” on five items on both the pre- and post-survey using a 5-point scale where 1 = “Very uncomfortable” and 5 = “Very comfortable” (Table 4).

Questionnaire administration. We administered the questionnaires to mentors at the beginning of a regularly scheduled class on the university campus. Students filled out the pre-survey during the second week and the post-survey during the last week of class. Surveys were coded to allow a comparison of all students’ pre- and post-surveys, while maintaining mentors’ anonymity. During administration, each student was given a copy of the questionnaire by the course instructor and asked to complete it independently. Questionnaire administration took about 20 minutes.


Data Analysis

We calculated frequencies, means, and standard deviations for all items for both pre- and post-mentoring and for demographics. We conducted independent samples t-tests to determine significant changes across time. We then rank-ordered items on Sections 1 and 2 (expected mentor and mentee benefits) according to mean scores and compared pre- and post-survey results to identify changes in mentor priorities. Frequency of responses and mean scores were calculated for items in Sections 3 and 4 (assumptions about mentoring process and level of comfort) pre- and post-mentoring. In addition, we tabulated these responses per level of response (i.e., 1-5) for each item rather than by rank order to determine agreement among mentors. Finally, ratings of demographic subgroups (e.g., ethnicity, type of high school attended, family income) were analyzed for response patterns.


T-tests revealed no changes from pre- to post-survey across items, indicating that expected outcomes closely matched achieved outcomes. Findings related to (a) rankings of benefits to mentors and mentees, and (b) assumptions about the mentoring process and level of comfort are discussed below. Finally, we bring in qualitative data from journals and interviews.


Benefits to Mentors

Table 1 shows mean pre- and post-mentoring ratings of potential benefits to mentors rank-ordered according to pre-mentoring scores. Prior to mentoring, mentors rated all potential benefits listed as important to them. Mean ratings of all 11 benefits ranged from 3.10 to 4.66 (5 = “very important”). Benefits rated as most important to mentors were those that addressed positive gains to their mentees, either directly or indirectly: “learn the value of positive supports and resources for youth from high-poverty backgrounds” (mean = 4.66), “learn to be a positive role model” (mean = 4.66), “gain an understanding of inner-city schools or community centers” (mean = 4.34), and “gain an understanding of the experience of a different cultural group” (mean = 4.21). Benefits rated as least important to mentors were those that focused directly on benefits to the mentors themselves: “reevaluate my own priorities” (mean = 3.83), “increase self-confidence” (mean = 3.52), and “explore a potential career choice” (mean = 3.10). Despite receiving lower rankings, however, these benefits were still rated as “somewhat important” or higher.

Few changes in ratings on items were found from pre- to post-mentoring. The three highest-rated benefits in the pre-survey, all mentee-focused, retained their ranking in the post-survey (i.e., learning the value of positive supports and resources for youth, being a positive role model, understanding inner-city city schools/community centers).

The benefits valued least post-mentoring (i.e., increase self-confidence and explore a potential career choice), which focused directly on the mentors themselves, remained at the bottom of the rankings post-mentoring. Ratings for all items were examined for differences based on demographic subgroups (e.g., black vs. white, inner city vs. suburban school background, prior experience with high-poverty youth, family income level). No response patterns were found across groups.


Expected Benefits to Mentees

Mentors’ pre- and post-mentoring ratings of expected gains to mentees are displayed in Table 2 as ranked by pre-mentoring ratings. When questioned regarding their assumptions about mentees’ gains from mentoring, mentors had high expectations for the anticipated benefits. Prior to mentoring, mean ratings of potential gains ranged from 3.07 to 4.34 where 5 = “gain a lot.” The top-ranked area in which gain was expected was “improve goal setting and attainment” (mean = 4.34). The next highest-ranked areas of expected gain all focused on benefits to mentees via the mentoring relationship itself: “build a friendship with a mentor” (mean = 4.31), “have a positive role model” (mean = 4.28), and “have a consistent, caring adult to interact with” (mean = 4.07). The areas in which mentors expected mentees to make the least gain were related less directly to the mentoring relationship itself: “increase self-confidence” (mean = 3.62), “improve academics” (mean = 3.34), and “improve social relationships with their peers” (mean = 3.07).

Little change in ratings of mentors’ perceptions of mentee gains occurred from the pre- to post-survey. The three lowest-ranked areas of expected gain, which focused less on mentoring itself (i.e., increasing self-confidence, improving academics, improving peer relationships), maintained their rankings post-mentoring. Mentee benefits ratings were analyzed for demographic subgroup trends, but none emerged.


Assumptions About Mentoring High-Poverty Youth

Table 3 displays responses to items investigating mentors’ assumptions about mentoring high-poverty youth, grouped according to (1) the mentoring process itself, (2) the role of cultural background, and (3) the inner-city environment. Prior to mentoring, responses indicated that mentors assumed that the mentoring process would be reciprocal between mentor and mentee. Specifically, mentors strongly agreed that they could learn as much from their mentees as mentees could learn from them (item 7, mean = 4.76) and that mentoring benefits mentors and mentees equally (item 3, mean = 4.24). Responses also suggested that mentors believed they had the skills to mentor, as indicated by item 9 (“I feel confident that I can be an effective mentor,” mean = 4.62) and item 2 (“the most important part of mentoring is just to be yourself,” mean = 4.48). Mentors were in less agreement that the success of mentoring primarily depends on the willingness of the mentee (item 5, mean = 3.86), that mentoring is a difficult process (item 6, mean = 3.69), and that there are certain skills and training needed to be an effective mentor (item 4, mean = 3.41).

Responses indicated strong agreement that peoples’ cultural backgrounds are an important part of who they are (item 8, mean = 4.72). At the same time, only two mentors “somewhat agreed” and 25 “strongly” or “somewhat” disagreed that to be an effective mentor it is important to be of the same cultural background as a mentee (item 10, mean = 1.59). Mentors also indicated that they felt safe working in an inner-city school (item 11, mean = 4.28) and that they were familiar with the challenges facing inner-city youth (item 1, mean = 4.03).

Few changes in ratings occurred from pre- to post-survey (see Table 3). Some patterns did emerge, however, relative to demographic subgroup responses. Black mentors and white mentors who had attended inner-city schools (n = 9) indicated greater familiarity with the challenges facing the inner city (item 1, mean = 4.78) compared to white (and one Asian) mentors who had not attended school in the inner city (n = 20, mean = 3.70). In addition, black mentors (n = 7) disagreed less strongly with the importance of being the same cultural background as a mentee (item 10, mean = 2.29) than did other mentors (n = 22, mean = 1.36).


Level of Comfort

Mentors’ reported levels of comfort in serving as a mentor and in relation to race and social class issues were notably high prior to mentoring, as shown in Table 4. No mentor reported feeling uncomfortable working with people of a different culture or race (item 4, mean = 4.69) or acting in a mentoring role (item 2, mean = 4.59). Only one mentor indicated feeling “somewhat uncomfortable” working with people from a different socioeconomic class (item 3, mean = 4.62) or understanding the needs of youth from high-poverty backgrounds (item 1, mean = 4.17). Only three mentors indicated some degree of discomfort in discussing issues related to race and race relations (item 5, mean = 4.14).

Few changes were reported in ratings from pre- to post-mentoring. Notably, however, although eight mentors indicated being either “neutral” or “somewhat uncomfortable” in understanding the needs of high-poverty youth (item 1) on the pre-survey, all 29 mentors reported feeling comfortable in this area post-mentoring. Level of comfort in discussing race-related issues decreased slightly to a mean of 4.0 (“somewhat comfortable”) from pre- to post-survey. Demographic subgroup analysis revealed that all seven black mentors and two white mentors who had attended inner-city schools reported being either “somewhat” or “very comfortable” on all five items. Only 1 Asian mentor and 9 of the remaining 20 white mentors responded with a similarly high level of comfort. All mentors who reported “neutral” or lower levels of comfort discussing race and race relations pre- and post-survey were white.


Qualitative Findings

Comments made by mentors during focus groups and in their journals supported the positive mentoring experiences indicated in questionnaire responses. For example, one mentor shared, “The main way I benefited from my mentoring experience is the sense of fulfillment that has resulted from it, because I actually feel as though I have had a positive impact on D.’s life. Reading articles and hearing people say that mentors and positive role models can really make a difference in the lives of high-poverty youth is not nearly as compelling as actually experiencing the difference firsthand.”

And a mentee shared, “Mentoring is a new part of my life now. I encourage all students to do this if someone hasn’t been there for them or if they never had much attention.”

Another offered, “It is not what I expected. When I first started, I thought my mentor would be mean — a person standing over me. But when we met, we just clicked. We joked and laughed and we were serious too. The best thing was when I said I felt stupid and she said, ‘No, you’re awesome. You’re not stupid — you’re my mentee. Just study, and if you need me to come make you, I will!’ Now my chemistry grade went from failing to around a C because of her.”

A teacher said: “V. was just doing what he had to to get by and now that’s all changed since the mentoring. This is very important for so many students — they need pointed in the right direction and shown that they can do more.”

Another teacher added: “People don’t realize how much time the mentors put in. They go beyond the call of duty, helping the students with financial aid, taking them on trips and out to eat, and things like that. What a huge thing it is for the mentees just to get taken out to eat!”

Informal interviews with mentees, mentors, and teachers supported these positive comments.



In this study, students in a university-based service-learning class had high expectations for a semester-long mentoring experience prior to the actual mentoring in high-poverty high schools. Post-mentoring measurements of these expectations confirmed these expectations, as reflected in the nonsignificant differences in mean scores from pre- to post-survey. Qualitative comments in journals and focus groups supported these measurements. These findings are especially important because we found no published study that measured perceptions of service-learning college students mentoring youth in high-poverty high schools before and after the mentoring experience. There is also very little research that examines the assumptions volunteers have about what benefits they might gain from the process for themselves and their mentees (DuBois and Rhodes, 2006).

This research, among other results, helps us understand what motivates people to volunteer to mentor youth and how to retain mentors (e.g., Grossman and Rhodes, 2002; Parra, DuBois, Neville, Pugh-Lilly, and Povinelli, 2002; Wandersman et al., 2006).

University students who voluntarily enrolled in an elective service-learning mentoring course indicated being motivated primarily by having a positive impact on youth by (1) being a role model, friend, source of support, and caring adult, and (2) increasing their understanding of inner-city schools and culture in order to serve youth better. Importantly, mentors also felt after the experience that they had achieved these aims.

Other lessons. In recruiting mentors it may be important to emphasize the expected outcomes our mentors valued highly, such as friendship, rather than lower-ranked ones, such as improving mentees’ academic performance or exploring their career options (see Tables 1 and 2). Further, considering that an estimated one-half of mentoring relationships end prematurely (Rhodes, 2002), it is critical to seek mentors’ views regarding whether they achieved their identified aims. Our findings also support previous research that shows that matching mentor/mentee race and ethnicity are not critical factors in successful mentoring (DuBois et al., 2002). No differences were found with respect to perceived mentoring outcomes based on race or ethnicity of mentors. Blacks were slightly more likely to report greater familiarity with inner-city environments and higher levels of comfort across cultures than whites; however, white mentors from inner-city high schools had ratings similar to blacks on these items. Interestingly, mentors across all demographic groups (white, black, low- to upper-income families) had similar ratings across benefits achieved for mentors and mentees despite mismatches in class and race across the majority of mentor pairs. Most mentees were low-income blacks, and most mentors were higher income whites.

These findings held true despite white mentors’ slightly lower familiarity and comfort levels in inner-city environments. Therefore, from the mentors’ perspective, cross- and same-race pairing was equally effective. In fact, the only item on which the overwhelming majority answered “strongly disagree” was, “It is important to be of the same cultural background as your mentee in order to be an effective mentor” (Table 3, item 10).

Our finding that mentoring programs should recruit from all races and socioeconomic classes is especially encouraging because whites tend to volunteer as mentors more often than other races and because youth most in need of mentoring tend to be minorities (Rhodes, 2002). Still, a perception persists that pairs should be matched by race, class, and other factors (e.g., Diversi and Mecham, 2005). Our findings do suggest, however, that some mentors unfamiliar with predominantly black inner-city environments initially may need additional training.

The most highly ranked pre- and post-survey ratings of benefits focused on benefits for mentees versus benefits to the mentors (see Table 1). Previous research suggests that mentoring relationships are of higher quality, more effective, and longer lasting if mentors do not have their own self-interest as a primary motivator in entering into the relationship but rather the interests and needs of the mentee (Karcher, Nakkula, and Harris, 2005).

However, Rhodes and DuBois (2006) suggested college mentors may be more motivated to fulfill their service-learning requirements than to serve their mentee’s needs. In contrast, mentors in our study place mentee well-being above their own, as suggested by higher ratings on “being a caring adult” and “being a role model” than on reevaluating personal priorities. Qualitative comments also supported this distinction: “Mentoring was a way for me to respond and do something to change the outcome of a student, rather than just study the problems.” Mentors also believed that the benefits of mentoring were reciprocal (Table 3, items 7 and 3). As indicated in post-survey ratings, service-learning effects in our findings are seen in enhanced civic engagement, promotion of positive mentoring relationships, and satisfaction with benefits experienced.

Although black mentors’ ratings of level of comfort and sense of familiarity with inner-city environments were slightly higher than those of whites, mentors’ pre- and post-survey ratings of their own ability to mentor and their level of comfort in cross-cultural environments in general were quite high. For example, no mentor disagreed with the statement, “I feel confident that I can be an effective mentor” (Table 3, item 9) in either the pre- or post-survey, and only one mentor prior to and only one mentor after mentoring “somewhat” disagreed that the most important part of mentoring was to be oneself (item 2). Studies have associated high self-efficacy with effectiveness of mentoring (Parra et al., 2002), suggesting that students who chose to be in the elective service-learning class were those who felt confident about mentoring. Also, the majority of mentors had prior experience with youth from high-poverty backgrounds, a factor that relates to effectiveness of mentoring (DuBois et al., 2002). It may be advisable in future service-learning mentoring programs to assess mentors’ level of confidence in engaging in high-poverty environments and to provide extra support where needed.

A university-based service-learning class may be an ideal setting for promoting effective mentoring relationships with high-poverty youth. Specifically, the class incorporated recommended practices found in the mentoring literature, such as ongoing monitoring, structure, clear expectations, and support (DuBois et al., 2002; Rhodes and DuBois, 2006). In this class, observation of mentoring activities was provided on site by the class instructor’s biweekly supervision, as well as examination of mentors’ journals. Mentoring contact and duration expectations were established and tied to class grades. Class discussions and focus groups addressed mentor support needs, and the instructor provided written feedback to journal entries. Organized mentor/mentee activities like campus tours and trips to sports events were scheduled periodically.

Academic content covered the effects of poverty on youth and their families, as well as race, class, and gender. That may account for mentors’ reported increases in knowledge of the needs of high-poverty youth and the challenges they face. (Tables 3 and 4, item 1). Post-mentoring reflections submitted for class also indicated that mentors better understood the context in which negative outcomes for low-poverty youth may occur. The academic content of the class in combination with daily reflections on the mentoring experience appears to have addressed the issue raised by Rhodes and DeBois (2006) of the “fundamental attribution error,” or locating the problem within the individual rather than considering the context in which a behavior occurs. Mentoring activities, reflective journals, and academic activities may have combined to educate mentors about racial and class disparities beyond book learning alone. In supporting mentors, moving beyond a “blame the victim” perspective may be critical to the success of mentoring relationships, particularly because most potential mentors are white middle-class persons likely to be paired with low-income black or Hispanic youth. A combined academic and service-learning experience may also increase social justice awareness (Eyler, 2002) among students, an explicit goal of service-learning pedagogy, as well as their mentoring beyond the duration of a college class.


Study Limitations and Recommendations for Future Research

A tip-off that research is exploratory is the number of suggestions at the end for followup study. Ours is no exception. Here are our thoughts about further study and research design considerations:

• To validate self-reports more thoroughly, include direct observation of mentoring as it occurs.

• Conduct informal interviews several times in the study instead of just at the end.

• Make the service-learning course two semesters in length.

• Recruit more representative volunteer mentors; ours were primarily white females attending a private university.

• Compare attitudes on race and poverty of students enrolled in general university classes with service-learning students.

• Our study suggests self-efficacy is an important concept in mentoring dynamics. Future studies might make this a central concept.

• Leave some journal entries ungraded, as grading them may have affected content.



According to the literature, mentoring has a positive impact on America’s youth. Moreover, mentoring is gaining in popularity. But there is a gap between the number of available mentors and students in need of mentoring. Although a primary goal of mentoring is to address the needs of youth, mentors’ expectations of and motivation for mentoring are also important. Failing to assess them will further contribute to the mentoring gap. Mentoring programs joined with research-based service learning can stimulate lifetime social issue awareness and community participation by students, mentees, their teachers, and university faculty, adding to our understanding of both mentoring dynamics and cross-cultural issues.



Diversi, M., & Mecham, C. (2005). Latino(a) students and Caucasian mentors in a rural after-school program: Towards empowering adult-youth relationships. Journal of Community Psychology, 33, 31-40.

DuBois, D.L., & Neville, H.A. (1997). Youth mentoring: Investigation of relationship characteristics and perceived benefits. Journal of Community Psychology, 25, 227-234.

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

DuBois, D.L., & Rhodes, J.E. (2006). Introduction to the special issue: Youth mentoring: Bridging science with practice. Journal of Community Psychology, 34, 647-655.

Eyler, J. (2002). Reflection: Linking service and learning — linking students and communities. Journal of Social Issues, 58, 517-534.

Grossman, J.B., & Rhodes, J.E. (2002). The test of time: Predictors and effects of duration in youth mentoring relationships. American Journal of Community Psychology, 30, 199-219.

Karcher, M.J., Nakkula, M.J., & Harris, J.T. (2005). Developmental mentoring match characteristics: Correspondence between mentors’ and mentees’ assessments of relationship quality. Journal of Primary Prevention, 26, 93-110.

Larson, R. (2006). Positive youth development, willful adolescents and mentoring. Journal of Community Psychology, 34, 677-689.

McClam, T., Diambra, J.F., Burton, B., Fuss, A., & Fudge, D.L. (2008). An analysis of a service-learning project: Students’ expectations, concerns, and reflections. Journal of Experiential Education, 30, 236-249.

MENTOR. (2006). The National Agenda for Action: How to close America’s mentoring gap. Retrieved September 12, 2007, from

Orfield, G. (2004). Losing our future: Minority youth left out. In G. Orfield, (Ed.), Dropouts in America: Confronting the graduation crisis (pp. 1-11). Cambridge, MA: Harvard Education Press.

Parra, G.R., DuBois, D.L., Neville, H.A., Pugh-Lilly, A.O., & Povinelli, N. (2002). Mentoring relationships for youth: Investigation of a process-oriented model. Journal of Community Psychology, 30, 367-388.

Portwood, S.G., Ayers, P.M., Kinnison, K.E., Waris, R.G., & Wise, D.L. (2005). YouthFriends: Outcomes from a school-based mentoring program. The Journal of Primary Prevention, 26, 129-145.

Rhodes, J.E. (2002). Stand by me: The risks and rewards of mentoring today’s youth. Cambridge, MA: Harvard University Press.

Rhodes, J.E., & DuBois, D.L. (2006). Understanding and facilitating the youth mentoring movement. Social Policy Report, 20, 3-19.

Schmidt, M.E., Marks, J.L., & Derrico, L. (2004). What a difference mentoring makes: Service learning and engagement for college students. Mentoring and Tutoring, 12, 205-217.

Wandersman, A., Clary, E.G., Forbush, J., Weinberger, S.G., Coyne, S.M., & Duffy, J.L. (2006). Community organizing and advocacy: Increasing the quality and quantity of mentoring programs. Journal of Community Psychology, 34, 781-799.

U.S. Census Bureau. (2006). 2004 American Community Survey. Washington, DC.


About the Authors

Carolyn Hughes is professor of special education at Vanderbilt University. Sara J. Dykstra is coordinator of special education at Achievement Preparatory Academy in Washington, D.C. Hughes can be reached at

Leave a Reply

Your email address will not be published.