A College Outreach Mentorship Program for High School Students : Considering College Student Mentors ’ Roles and Perspectives

This case study examined a uniquely designed university-based college outreach work-study mentoring program in practice that is rooted in university-school partnerships. The study focuses on the roles and perspectives of the college student mentors in the program, as such perspectives are limited in the literature. The mentors assist high school students, many from underserved backgrounds, as they navigate the pathways of college admission. Interviews with mentors, as well as mentor training documents and report data, revealed the overall positive role mentors believed they played in improving college access for their mentees. Findings also shed light on improvements for the university-based program and its school partners.

Mentoring has become a valuable tool to increase college access and success, providing prospective and current college students the necessary support to achieve their goal of attending and graduating from an institution of higher education (Coles, 2011). Providing such support is particularly key for students from low-income backgrounds and students of color who have been historically underrepresented in higher education, as "the challenges faced during the transition to college often affect [these students] to a greater extent than their counterparts [who are better] equipped with the cultural capital necessary to succeed in predominantly white institutions" (Corwin, Colyar, & Tierney, 2005, p. 5). Yet while the benefits of mentoring for the mentee are well documented (Coles, 2011), few studies examine how mentors experience, learn, and benefit from mentoring students to increase college access.
Family members, school personnel, and peers can assist students with accessing college, yet it is often individuals who work in pre-college outreach programs who help students navigate this process through various instructional approaches including mentoring (Loza, 2003). In the state of Texas, efforts to increase college participation and success among the state's constituents have specifically included the establishment of a distinct work-study mentorship program through the Texas Higher Education Coordinating Board (THECB) called the Collegiate G-Force workstudy mentorship program. The program is rooted in and dependent on the power of university and school/community partnerships.
The mentorship program allocates funds to "pay wages to college students [who] are employed on a part-time basis…across the state serving as mentors in Go Centers [college-focused centers], community centers, high schools, and institutions of higher education" (THECB, 2009, p. 1). Institutions of higher education in the state primarily seek funding for the program and are responsible for the organization, implementation, and assessment of the program, which includes the training of mentors and collaborating with schools and community entities to determine mentoring sites and the specific structure of services provided. College students who have financial need, as determined by the host institution, can apply to be mentors and reap multiple benefits from their role. At present, little is known about the experiences or perspectives of the mentors in this specific mentor program aside from the work of Amaro-Jiménez and Hungerford-Kresser (2013) that examined the G-force mentor program at the University of Texas at Arlington. Therefore, this case study uniquely contributes to the dearth of research on mentors in this distinct, work-study mentor program and sheds light on the types of experiences mentors in similar university-school mentor programs can have. This study also recognizes the value of mentors' perspectives, as they can inform how the university-based mentoring program can improve in serving and working with its school partners.

University Community Context
This study specifically examined the G-Force mentor program at a large, four-year, doctoral granting public university in central Texas, otherwise known here as Hillside University (Hillside). Hillside is situated within the larger community referred to here as Rolling Hills which is home to approximately 54 thousand residents, about 41% which identify as Hispanic, 50% White (non-Hispanic), 5% Black/African American, about 2% Asian, and 2% of residents are biracial or multiracial (U.S. Census Bureau, 2015). Among Rolling Hills residents who are 25 years or older, about 25% have a high school diploma, 23% have some college credit but no degree, and only 22% have a bachelor's degree (U.S Census Bureau, 2015). Additionally, about 37% of residents live below the poverty line, and the median income for residents is $28,923 (U.S Census Bureau, 2015).
Within the last five to 10 years, Hillside University and leaders in the Rolling Hills school district and community collaborated to identify pressing needs of residents, and among those was the need to increase collegegoing rates for Rolling Hills high school students. This was in part based on the fact that from 2008-2011 college-going rates in the district ranged from 36% to 43%, which were well below the national average of 66% as of 2013 (Texas Higher Education Coordinating Board, n.d.). Rolling Hills High School, the only comprehensive public high school in the city, serves 2,239 students, of which 71.4% are Hispanic, 20.8% are White, 6.0% are African American, and approximately 1% are Asian (Texas Education Agency, 2015). About 61% of students are considered "economically disadvantaged" and 46% "at-risk" (Texas Education Agency, 2015). When considering the most recent data from the 2014-15 school year regarding advanced course and dual credit course completion, White students and Asian students have the highest completion rates at 61.2% and 100% respectively. In comparison, these rates were 48.3% for Hispanic students and 47.8% for African American students (Texas Education Agency, 2015). College-ready graduate rates from this same school year showed a similar disparity among students of different races with 49.0% of Whites, 21.0% of Hispanics, and 27.0% of African Americans considered collegeready in both English/Language Arts and Mathematics; college-ready graduate rates were not available for Asian students because of the limited size of this population (Texas Education Agency, 2015).
University, school district, and community leaders¹ concerns spurred the development of a P-16 Center (Center) at Hillside University in 2010, which aims to develop and deliver programming for Rolling Hills and surrounding school districts to increase college access and success. The Gforce mentor program is one of several initiatives coordinated and carried out by the Center at Hillside that aims to increase college access for high school students in the surrounding community, with a particular focus on historically underserved students, including students of color and first generation college students. Given the mission, this study was guided by the following questions: (1) How do G-Force mentors perceive their role in assisting students and families with college access within the program?
(2) How can the university-based mentoring program improve based on G-Force mentors' perspectives and experiences?

University-Based College Outreach Mentoring Programs
A majority of the studies examining college student mentors' experiences when working with school aged students (from elementary to high school) do so within the context of service-learning courses (Banks, 2010;Hughes, Boyd, & Dykstra, 2010;Hughes & Dykstra, 2008;). Few studies, like those of Bergerson andPeterson (2009) andAmaro-Jiménez andHungerford-Kresser (2013), examined university-based mentoring programs for school aged students that have a specific college-outreach focus and were not implemented within the context of a service-learning course. Thus, college outreach mentor programs are also distinct when compared to other larger more comprehensive college-outreach programs that include mentoring as one component of what they do (Swail & Perna, 2002;Swail, Quinn, Landis, & Fung, 2012). College outreach programs such as Upward Bound or Gear Up provide additional programming including academic enrichment and support, college visits, and preparation for college-entrance exams in order to increase college knowledge and access for the students they serve (Swail & Perna, 2002;Swail et al., 2012).
The literature on mentoring programs indicates that programs vary widely, with some utilizing single-sex mentoring, or mentor matching based on shared interests, while others let such relationships naturally emerge. The type, frequency of contact, and duration of mentoring provided (i.e., individual and/or group) also varies, with most mentoring programs lasting for one semester with weekly contact. Existing research also suggests college student mentors working with school aged students reap multiple benefits from such experiences including increased cultural awareness, enhanced character and professional development, as well as an opportunity to reflect on one's priorities and appreciate one's own lived experiences (Banks, 2010;Hughes & Dykstra, 2008;Hughes et al., 2010). Mentors in Bergerson and Peterson's (2009) study all "discussed being rejuvenated, encouraged, and motivated" by their roles as mentors, in addition to experiencing personal and skill-based growth (p. 55).
As previously noted, Amaro-Jiménez and Hungerford-Kresser (2013) examined one of the G-force mentorship work-study programs in Texas, like the program examined in this case study. They drew on the written reflections and activity logs of 34 mentors and other artifacts to determine mentors' perceptions of the success of the program and the challenges mentors faced in their roles while working with their mentees. Mentors worked at their designated high school sites about 15 hours a week and, given the structure of the work-study program, met with any students at their school sites that sought their assistance to discuss or complete college-related activities.
Mentors described the program's success in terms of it "reaching its goal of continuing and oftentimes beginning the conversation about the possibility of pursuing a postsecondary education among all students" (Amaro-Jiménez & Hungerford-Kresser, 2013, p. 7). Mentors also believed the program had achieved success through its ability to promote a college-going culture at their school sites. A college-going culture can be described as one that is widespread, palpable, and accessible to all students that "cultivates aspirations and behaviors conducive to preparing for, applying to and enrolling in college" (Corwin & Tierney, 2007, p. 3). Challenges identified by the mentors related to the implementation of the program at their school sites and included an inability to access necessary resources on the school campuses, not being seen or treated as professionals by school staff, and facing resistance on the part of some teachers and even some students. Mentors addressed some of the initial challenges by working with school administration. The latter findings clearly indicate the need to develop clear and on-going communication and collaboration between university and school stakeholders that participate in such programs and partnerships to ensure that not only the needs of mentees are being met, but that those of mentors are as well.

A Framework for Understanding the Role of College Mentors
This case study drew on Karcher, Kuperminc, Portwood, Sipe, and Taylor's (2006) framework that was developed to further understand and examine the many forms of youth mentoring programs that exist, particularly to consider the effectiveness of such programs. In the context of this study, the framework is not used as a tool to measure the effectiveness of the G-Force mentor program. Instead, the framework serves as a guide to further understand the nature of the mentor-mentee relationships formed through the G-Force program, and more specifically to examine the research questions that focused on mentors' perceived roles and how the mentors' perspectives can inform improvements in the mentoring program. Karcher et al.'s (2006) framework suggests that when examining a mentor relationship, it is essential to consider the context in which the mentoring takes place (i.e., field-based or site-based mentoring), the structure of the mentor-mentee relationship (i.e., one-on-one, adult-with-youth mentoring, cross-age peer mentoring), and the goals of the mentoring. The context for mentoring relationships often falls into the category of either fieldbased or site-based programs (Karcher et al., 2006). In field-based mentoring, an agency or organization sponsors the mentoring program and coordinates the mentor-mentee match, but the mentor and mentee are able to set up their own meeting times and locations (Karcher et al., 2006). Mentors and mentees interact within the confines of a particular space, such as a school or community center, during set times in sitebased mentoring. Site-based mentoring programs also often tend to have greater structure and focused goals (Karcher et al., 2006). Some of the most common structures for mentoring relationships with youth include cross-age peer mentoring, group mentoring, e-mentoring, and intergenerational mentoring; each has its advantages and limitations (Karcher et al., 2006). In cross-age peer mentoring, the mentor and mentee are often both considered youth but are not of the same age. Group mentoring can vary in terms of structure, but often consists of mentoring provided to six to 10 youth who meet over a certain time period as a group with a mentor or team of mentors (Herrera, Sipe, & McClannahan, 2000). In e-mentoring the mentor-mentee relationship develops online, although the structure can vary greatly form one program to another. Intergenerational mentoring involves matching older adults who are typically 55 years or older with youth.
It is the "goals of a program [that] shape the activities that occur in the match [between mentor and mentee], and these activities fall along a continuum from task-or skill-focused to relational and developmental in nature" (Karcher et al., 2006, p. 710). Thus, the goals of mentoring usually fall within two primary categories that are not necessarily mutually exclusive: developmental and instrumental. Developmental mentoring focuses "on facilitating the relationship between mentor and mentee as a way of promoting the youth's development" (Karcher et al., 2006, p. 714). Developmental mentoring can consist of mentors and mentees engaging in social or recreational activities that offer an opportunity for mentors and mentees to get to know each other on a personal level to uncover shared interests. The goal of such mentoring is focused on facilitating the mentees' social, emotional, and cognitive development.
The main goal of instrumental mentoring is on "the learning of skills or the achievement of specific goals" (Karcher et al., 2006, p. 714). While instrumental mentoring has been described as prescriptive, it can still be flexible so that "the mentor helps the mentee to accomplish tasks or goals of the mentee's choosing by providing advice, guidance, explanations, or suggestions" (p. 714). It is believed that the focus on helping mentees achieve the more immediate goals or skills attained through instrumental mentoring can then facilitate "long-term or more distal developments in social, emotional, and academic skills" (p. 714).
It is important to reiterate how a mentoring program can draw on both developmental and instrumental mentoring approaches, although Karcher et al. (2006) suggest that programs tend to focus on one goal more than the other. In choosing one of these two different orientations "the outcomes desired by the mentor or program" are usually revealed (p. 714). As outcomes are tied to programmatic goals, Karcher et al. argue that: Program developers should be clear about which theoretical approach their program will reflect. Researchers and evaluators must examine whether or not activities reflecting these two approaches actually predict both immediate and long-term goals; current research has not sufficiently explored their impact on the outcomes of youth mentoring. (p. 714) In addition, Karcher et al. (2006) identify other essential characteristics of mentorship programs that should be considered when examining the structure and outcomes of mentoring programs, which include infrastructure (i.e., screening, matching, and ongoing training and support provided to mentors) and dosage (i.e., frequency, intensity, and duration of mentoring).

Method
The researchers used a descriptive case study design (Yin, 2013) to examine the overarching research questions: (1) How do G-Force mentors perceive their role in assisting students and families with college access within the program? (2) How can the university-based mentoring program improve based on G-Force mentors' perspectives and experiences? A case study design is useful when examining "how" and "why" questions in order to understand the phenomena, or case under study. This design also draws on multiple forms of data collected through various means so that evidence is triangulated, contributing to issues of trustworthiness (Yin, 2013). The case under study in this inquiry was the G-force mentor work-study program at Hillside University. However, this study was not a program evaluation, as the aim was to focus on unearthing the nuanced experiences and perspectives of mentors, which could then inform the program and the university-school partnership. The study was conducted by a research team consisting of two female faculty members and one female doctoral student at Hillside who do not work with or have any affiliation with the mentor program, aside from conducting the study.
As previously noted, the mentor program is one of several initiatives delivered by the P-16 Center (Center) at Hillside, a Center that is entirely grant funded and overseen by a faculty member who is the Center's director. The Center also has a full-time program coordinator who oversees the mentoring program, among other Center initiatives. As the mentor program is designed as a grant-funded, work-study program for college students, mentors are recruited and asked to apply for the program and then screened and hired by the program coordinator through an interview process. The mentor application asks that mentors provide: general contact information and specific information needed for the workstudy program, including whether a student is a first generation college student and if and what financial aid they receive. Other questions on the application refer to a student's academic standing, educational goals, current and previous employment, and specific strengths and abilities that would qualify the student for the position. Other documents requested as a part of the application include a resume, contact information for three references and the student's academic advisor, official transcript, and class schedule.
According to the Center's website, the mentor program "enables and encourages high school students to realize their full potential and enhances their ability to pursue higher education." The role and duties of college student mentors are to "assist one-on-one with 9th-12th grade students to create a college-going culture. They teach high school students how to enroll in a college or university as well as how to apply for scholarships and financial aid." Additionally, mentors assist with dispelling myths about college life and help students access university websites, or in some cases select a career path. Mentors also organize and/or help facilitate parent/student workshops on high school campuses or in the community.
At the time of the study, the mentor program served five high school campuses; mentors are assigned to a campus, although mentors are not assigned specific mentees or a finite number of students at their respective schools. Instead, mentors work in collaboration with high school personnel, particularly counselors, to help identify and call upon particular students to offer their mentorship with college planning. The mentors are also available to any student that seeks their assistance when the mentors are on their designated campuses. Thus, mentors are often physically situated in the vicinity of a high school's counseling office or a college center if the school has one.
Monthly data collected from October to December of 2014 for the G-Force mentoring program provides a clear indication of the most common mentoring interactions occurring and information about the mentees being served. Overall, mentors most often assist high school seniors who are of Hispanic descent with college applications and "other mentoring activities" such as general mentoring, tutoring, goal setting, and career exploration. From October to December, 135 to 196 mentoring interactions focused on college applications, while the number of interactions focusing on other mentoring activities ranged from 158 to 361. Generally, there was an equal gender distribution among mentees seeking services during this time.
It is also important to note that many of the college student mentors themselves are first generation college students, many are Latina/o, and all also receive some form of financial aid. The schools that the mentoring program serves have large populations of students from similar backgrounds; all schools have a student population that is at least 54% Hispanic, and all but one school have at least 57% of students from economically disadvantaged backgrounds based on being eligible for free and/or reduced priced lunch.

Data Sources and Analysis
During the fall of 2013, there were approximately 21 G-force mentors, and all mentors who had at least one full semester of experience working in the program were solicited to participate in the study during a mentor program meeting. An additional 12 former mentors who were still students at Hillside University were also asked to participate via email or phone. In all, 14 current or former college student mentors (10 females, four males) and one female graduate research assistant, who worked directly with the G-force mentor program to help coordinate mentor schedules and activities and provide additional support to mentors, participated in the study. Nine mentors and the graduate research assistant identified as first generation college students, and seven mentors and the graduate research assistant identified as students of color, specifically Latina/o. The Center's director and peer mentoring program coordinator were interviewed for additional information and clarification regarding the mentoring program and its coordination.
Interviews with mentors and the graduate research assistant were digitally audio-recorded, face-to-face, and semi-structured in nature (Rubin & Rubin, 2005). An interview protocol was derived by the research team was used to guide these interviews and consisted of 20 questions that focused on gauging mentors' experiences in the program: what they liked, disliked, and learned from being a mentor, and what the program could stand to improve. Example questions included: Tell me about your role as a G-Force mentor, and what are your main responsibilities? Describe the relationship you have with G-force mentees. What do you like most about being a G-Force mentor? The same interview protocol was tailored to gauge the graduate research assistant's perspective regarding the mentoring program. Interviews ranged in duration from approximately 20 to 45 minutes.
The Center's director and the program coordinator were also interviewed together in a semi-structured format; the interview was audio recorded and lasted 32 minutes. The director and program coordinator were asked to share their perspectives regarding the development and impact of the mentor program in the community, the program's infrastructure and strengths, training and support provided for mentors, benefits to mentors, and challenges with delivering the program on the high school campuses.
Programmatic documents were also examined and included the G-Force mentor handbook that is used in the training of new mentors and data gathered by the program for university and state reporting. The mentor handbook specifically included information regarding the following content: mentor responsibilities, requirements and expectations, disciplinary procedures, reporting harassment and discrimination, personal boundaries with clients (mentees), ethics, boundaries and issues, dress code, history of the G-force mentor program in the state of Texas, terms of employment, general policies and procedures for mentors, and instructions and examples of time sheets and contact logs. The manual also contained information and resources that mentors could refer to in their work with mentees such as: a senior college prep timeline, a list of Texas colleges and universities with their application deadlines, SAT and ACT testing information, the Apply Texas and Common applications for college admissions, information related to resume writing and college essays, and FAFSA and scholarship information.
All interviews were transcribed and checked for accuracy prior to analysis. Each of the researchers was previously trained in qualitative data analysis using thematic coding (Boyatzis, 1998); all researchers used Microsoft Word to assist them in the coding process. Therefore, an inductive analysis of data began with a first cycle of open coding (Saldaña, 2009), where a code was assigned to "a summative, salient, essence-capturing, and/or evocative attribute" for a portion of textual data (Saldaña, 2009, p. 3). A second cycle of coding followed where codes were revisited to identify overarching themes that reflected a deeper understanding of the data. To address issues of trustworthiness, identified themes were continuously compared back to the data sources. Additionally, in addressing inter-coder agreement, the three researchers examined the data independently and then collectively during the data analysis process to compare and finalize themes and to confirm the assignment of themes from the data.

Limitations
It is important to note that this study has a number of limitations. First of all, findings are not meant to be generalizable given the case study approach used, whose purpose is to examine a context-specific case or phenomena (Yin, 2013). However, it is possible that the perspectives of the mentors in this study might be similar to those of mentors in other college outreach programs that are comparable in structure and mission to the G-Force mentor program. The study is also limited in that it focuses deliberately on mentors' perspectives, given the dearth of research in this area. Thus, findings from this study are restricted in that they do not include the perspectives of mentees, school partners, or other community stakeholders. As previously noted, this study also does not purport to be a program evaluation, although mentors' perspectives are considered valuable in offering greater insights to the benefit of the G-Force mentor program.

Findings
With Karcher et al.'s (2006) framework in mind, we determined that the goals and many structural aspects of the G-Force mentoring program at Hillside University were based on the THECB's guidelines for the workstudy mentorship program. Thus, as the THECB stipulates, the program is site-based, with mentors working at their designated high school sites about 15 hours a week. There was also a team leader at each high school, often a mentor with more extensive experience in the program, who helps guide newer mentors at each site. Mentoring was primarily instrumental in nature, with mentors focused on assisting any and all students at their school sites with college-related activities. This meant that the frequency and duration of mentoring varied based on mentee needs. Therefore, some mentees were seen once, while others were seen repeatedly over the course of weeks, months, or even years. Karcher et al.'s (2006) framework suggests that this flexible structure in the mentoring program might seem inconsistent and make it difficult to measure effectiveness of the program. Although, if the programmatic goals are primarily instrumental then the duration and frequency of the mentoring relationship would need to vary based on each mentee's college-related needs. As data were analyzed to answer the two guiding research questions related to mentors' perceived roles, and their perspectives on how the program can improve, the following themes emerged: developing the mentoring relationship, assisting with college applications and financial aid, administrative duties and other college awareness activities, and strengths and areas for improvement for program and partnerships.

Developing the Mentoring Relationship
In discussing their perceived roles, mentors first and foremost described how they worked to develop positive, trusting, and meaningful relationships with their mentees. They also focused on keeping the relationship with their mentees friendly during the sometimes stressful process of discussing postsecondary plans and completing financial and college applications by "cracking jokes, and making sure that you know, it's a positive experience even though it's not." While mentors recognized that there was little age difference between them and their mentees, they wanted to attain a level of respect and trust in their roles while connecting with students on a more personal level so the mentees would not feel intimidated. As one mentor admitted: I try and keep it light with the kids just because I've noticed a lot of the time when I'm meeting with them first like, "have you taken your SAT, ACT? Do you know where you want to go?" And a lot of the time it's, "no, no, no" …so I just try and say, "It's okay. We still have time to do this. We can help you out." So definitely try and keep it light and fun for them so they want to come back in.
Mentors also spent time getting to know their mentees and their interests before beginning to discuss and/or complete college applications. Most mentors reported this was an important first step in the process; without it, the mentor/mentee relationship likely would not progress. Once they found out the mentees' interests, mentors generally asked mentees to discuss their future goals with them. The concept of goal setting was significant to the mentoring relationship because the mentors wanted to ensure "that they [mentees] know that they're there for a purpose and, you know, keeping their goals in mind is really important." Assisting mentees with developing visions for their futures and providing them with the tools and resources to reach their goals comprised a large part of the mentoring relationship. As one mentor stated, "Some students have no path that they want to follow so we have to help those students out a little more, and just guide them to what they like and what they will pursue in the future."

Assisting with College Applications and Financial Aid
Another large aspect of mentors' roles related to the direct assistance they provided to mentees with completing college applications and financial aid. As one mentor noted, "Once we know where they [mentees] want to go, we help them out with their application, their essays all that stuff. And then come spring time it's FAFSA and scholarships." Ensuring that mentees understood all of the types of financial aid that was available to them along with other application considerations was also a key component of the mentor role, particularly since the program was serving many Latina/o and first generation college students. Therefore, as one mentor explained, mentors worked to "make sure that they [mentees] know what financial aid is available to them, what schools they want to go to, what major they want to have, just how to be successful in college in general." The mentors acknowledged that mentees arrived with varying knowledge about the financial aid and college application process. Some mentees also came to them having completed various portions of these applications. Many mentors began by "working on [college] applicationson essays with them [mentees]. Some of them just needed-help with their resume. Some of them were at the bottom [sic] where they hadn't even taken the SAT or ACT." Regardless of the mentees' stage in the application process, mentors worked with them throughout the year to help students reach their goals for life after high school and navigate the financial aid process to do so.

Administrative Duties and Other College Awareness Activities
The mentors acknowledged their work entailed administrative as well as more traditional mentoring duties. Regarding administration, the mentors stated that their primary responsibilities were to keep track of the students with whom they were meeting and documenting the work that they did in each session. One mentor explained that, "The main responsibility is just making sure that when I call out my students I'm documenting who I'm calling out, what I did with that student, and making notation of what I did that time." Part of the documentation included maintaining logs of their meetings with mentees. As the program coordinator stated, "They have to turn in a contact log for every kid that they work with. Its tells us specifically who they met with and what services they provided with each student." Mentors also discussed the need to keep track of deadlines for applications, scholarships, and financial aid to ensure their mentees were submitting their information in a timely manner. Establishing a balance between accomplishing the tasks necessary to apply for college and secure funding while maintaining a positive relationship with the mentees was important to the mentors.
The mentors also worked with their mentees' families, though in a more limited capacity, during scheduled conversational meetings in the community where English and Spanish-speaking parents and students were invited to learn more about college and the application process. These sessions gave the mentors an opportunity to connect with families at the various high school campuses. Though one mentor found the meetings to be "challenging because it just required a lot of time," most mentors felt the meetings with parents were beneficial and aided the work that they were doing with students. "It was really rewarding to see that the parents wanted to get involved."

Strengths and Areas for Improvement for Program and Partnerships
By far, the most significant strength of the program identified by the mentors was the opportunity they had to help students. As one mentor shared, "We have all these different campuses and there's, you know, hundreds of students in each of them and in a lot of ways we're the reason they're going to college, you know for a lot of them." One mentor shared some of the comments the mentees had made, such as "If you wouldn't have called me out [of my class to speak with me], I wouldn't have known that I needed to do this. Or if it wasn't for you I wouldn't be going to school." As mentioned earlier in the section on data sources, during initial training mentors received a handbook with supporting documents and resources to assist them in their work with student mentees. During the year, mentors occasionally received additional training, such as how to handle different scenarios that might arise on campus, and they received financial aid training in the spring. Trainings were often provided by the program coordinator, or when possible, other university officials with specific expertise. Though some mentors felt they "received pretty good training," others felt "trainings are spotty." Some mentors ended up feeling like "they were kind of just thrown in there" and lacked adequate training to feel comfortable when first starting.
The Center director and coordinator both spoke to how formal professional development trainings for mentors were offered less frequently now than when the program was first established in 2010, in part due to a decrease in funding for the grant-funded mentoring program. The Center Director explained: The great part was they [THECB grant funders] gave a lot of money for professional development of the students...They got all kinds of training, and all kinds of funding for it. They really have not been-in the last three years, I haven't seen any [similar] programs.
The program coordinator added, "I think what it was, is that additional funding paid for the overhead and training, and all that. When that money ended, they just stopped [providing the formal professional development]." Admittedly, the coordinator shared: I wish we had more. I try to provide as much as we can…You know, we don't have a budget for that. So, I do try to find leadership opportunities for them. Training purposes. I'm very much about community service, so part of their position requires them to agree to do community service every month, so I do try to-no I would say [I] push that, but it is part of the, of having that position. You know, they need to be just as, we're trying to push and encourage students to pursue higher ed. They also have to be the model for that.
The coordinator expanded on how she tried to develop trainings on "different topics that they [mentors] need." She also would ask mentors "for recommendations, like is there something that you all have been needing additional training on." Some suggested trainings that were provided focused on the new policies related to Deferred Action for Childhood Arrivals (DACA), and helping students in foster care.
Despite the need for additional trainings that some mentors expressed, all mentor participants appreciated the support received from the graduate assistants in the program and the program coordinator. The program coordinator and Center director worked to make sure the mentors were balancing their work as mentors and as college students, "always emphasizing that, you know, school comes first." Though mentors saw flexible scheduling as a positive aspect of working for the program, many acknowledged it could cause problems at the school campuses because mentors would show up at different times. This meant some mentors might be at the campus individually at times, without additional support. Additional school site concerns were mentioned by mentors including: difficulty accessing necessary resources on the school campuses, occasional conflicts with the school counselors, and dealing with resistance from disinterested students and teachers who did not want to yield instructional time to accommodate mentor-mentee visits. To resolve such issues, several mentors suggested that in the future it would be beneficial if they were formally introduced to students and staff in the schools at the beginning of the year by the school administration in order to "make a better connection with the teachers from the get go." The Center director and coordinator expanded on their relationships with school partners and how "at the school district level we have pretty good relationships with administration." However, there were often tensions in working with the school counselors at one of the high schools the mentor program served. The Center director shared how at this particular school site the counselors focused primarily on serving "the kids who are in the top 10%," which she believed was in part due to logistical as well as philosophical reasons. The director noted the high counselor to student ratios (1:500) at this school that logistically kept counselors from having the time to meet the college access and preparation needs of all students. However, the director also shared, "I don't know that they [the counselors] think it's as important to serve the kids that we're really focusing our efforts on." In this respect, the director felt that the counselors at this particular school did not necessarily agree that the Gforce program should focus so heavily on serving historically underserved students, including students of color and first generation college students. The program coordinator also spoke to the specific logistical issues that they faced with the head counselor at this particular school, making it "one of our hardest schools to work with." The coordinator cited how the school counselor was ultimately "very controlling" because she "insists on hiring her own mentors. We can't just place somebody there. She needs to interview them, and accept them." However, working with other school partners was more of a collaborative process. With one school, a novice counselor "didn't know anything about college or career counseling" and was admittedly, "very open about that. She was like I'm going to need your help." The coordinator expanded, "Together we developed a system, and we had really good mentors at the time, that they all just kind of came together and they played well. They did everything very well." However, issues can still arise with new or existing staff on a campus. In such cases, as the coordinator explained: [I ensure] mentors are trained to go to their team leaders first [mentor leaders], and their team leaders are training to be able to handle those types of situations. If they can't they bring it to the GA [graduate assistant] and the GA brings it to me. Or, they can just bring it to me. I have an open-door policy-they don't have to bring it to the GAs but a lot of the times they do.
The program coordinator tried to shield the mentors from as much of the friction with school partners as possible.
However, the largest frustration expressed by mentors about the mentor program was its limited funding. Mentors felt "held back a lot because of our budget." That frustration was shared by the program coordinator: "I think if we had unlimited funds, or more resources, I think that we could provide a stronger leadership training component for students." The mentors expressed an interest in seeing the program expand to additional high schools and adding additional mentors so that they could reach more students. As one student noted, "I wish that there was more money put into it because it's really impactful."

Discussion and Conclusion
This case study reiterates the positive impact that university-based college outreach mentoring programs can have not only on the students served, but also on the college students serving as mentors. The context of this current study is unique in a number of ways in that the mentoring program: 1) is a work-study program that purposefully recruits college student mentors who are first generation college students and/or students that qualify for needed financial aid, 2) utilizes a programmatic design in which mentors are not assigned to particular mentees, and thus can serve a wide array of high school students on their assigned campuses, which generally serve large populations of underserved students, and 3) facilitates mentor-mentee relationships that can vary in duration depending on the needs of the mentee.
Findings revealed the significant role that the college student mentors felt they played in the lives of their mentees through the development of positive, trusting, and collaborative relationships whereby mentees could explore their postsecondary goals and interests and move forward with obtaining, understanding, and completing college and financial aid applications. The relationships mentors formed with mentees and the impact they were able to have on mentees were personally motivating and served as impactful experiences for the mentors.
Mentors also provided insight into what they found most noteworthy about the program, which centered on the impact they were able to have on the high school students, and first generation college students in particular. The mentors' discussion of the impact they were able to have on the mentees reflect the concepts of both developmental and instrumental mentoring outlined in Karcher et al.'s (2006) mentoring framework, specifically the goals dimension. While the primary goal of the G-force mentoring program could be described as instrumental, mentors felt they were able to help mentees grow and develop through the mentoring relationship (i.e., developmental mentoring) as well as help mentees learn skills related to the college application process (i.e., instrumental mentoring).
Findings also suggest that the process of helping students grow and impart knowledge about college to mentees takes time, and the efforts made to personally connect with mentees helped ensure mentees would continue to seek guidance from mentors. This emphasis on process and time reflects the dosage characteristic of Karcher et al.'s (2006) mentoring framework, suggesting the importance of the frequency and total duration of such mentoring relationships. At the same time, the G-Force mentoring program provided great flexibility for the development of mentor-mentee relationships that focused on meeting the needs of individual mentees as opposed to requiring a set dosage for each mentor-mentee.
Mentors also noted challenges and constraints they faced in their roles as mentors, which served to inform program delivery and the nature of the university-school based partnerships. A number of the on-site challenges noted reflected those identified in Amaro-Jiménez and Hungerford-Kresser's (2013) study, specifically related to access to resources on the school campuses, facing resistance from disinterested students and teachers, and conflicts with the school counselors.
As such, the mentorship program at Hillside University should consider the suggestion made by several mentors to be formally introduced to school staff and students at the beginning of each school year to help develop rapport and support for the mentors and the program. However, input from the Center director and program coordinator provided additional details regarding the nuances of fostering and sustaining collaborative relationships with school partners, particularly when there is turnover among school staff and mentors. As such, it seems that the Center director and program coordinator could benefit from meeting with the school administrators and school counselors from each site once a year, or once a semester to communicate individual and collaborative goals, needs, strengths, and identify trends in their efforts to serve the needs of mentees and mentors.
A number of mentors also indicated that additional training would have helped them feel more confident and prepared in their roles. Karcher et al.'s (2006) mentoring framework identifies ongoing training and support as a critical aspect of a mentoring program's infrastructure, and it was clear from this study that infrastructure is an important aspect to consider in university-based mentoring programs. It would also be beneficial for the program to gather more consistent data from mentors to gauge the specific type of additional training they feel most necessary to improve their work, not just upon taking on the role of mentor, but also throughout the mentoring process. Another suggestion for improvement that was noted but cannot be as easily addressed was related to the financial constraints of the mentor program given its sole reliance on grant funding. As this study helps shed light on the significant impact of the program on mentors from Hillside University and mentees in the Rolling Hills community, it is plausible that Hillside University and the Texas Higher Education Coordinating Board might invest additional funds towards this effort.
Finally, as there is a gap in the research on peer mentorship programs for college access, findings from this study are pertinent to other scholars in the field as well as university-based practitioners and administrators developing or implementing similar programs. However, as previously noted, findings from this study are limited since data was only obtained from mentors and university staff and thus only reflects their perspectives. Therefore, future research could extend this work by examining different university-based mentoring programs at a variety of institutions to see how they compare in development and practice while garnering the perspectives of all stakeholders, including mentees and school stakeholders, to expand impact of such programs and on the role of the university-school partnerships. Efforts to track mentees and mentors as they progress through their postsecondary pursuits would also provide insight into the long-term impact of university-based mentoring programs and the P-16 partnerships that support this work.