First-year undergraduate STEM students changing majors to nonSTEM fields have become a concern heard around the country (Dagley, Georgiopoulos, Reece, & Young, 2016). First identified over 20 years ago (Seymour & Hewitt, 1997), today’s statistics reveal that upwards of 60% of STEM majors change their area of study or drop out of school altogether (Hayes, 2007; Schneider, Bickel, & Morrison-Shetlar, 2015). Institutions around the country have studied the problem and have tried to address this issue in multiple ways (Blackburn, 2017; Bruffee, 1999). Peer mentoring continues to be one of the most popular methods institutions use to retain STEM students in their chosen programs and increase overall student persistence in STEM (Bahr & Norton, 2006). For example, Dagley et. al., found higher retention and graduation rates for students who received peer mentoring, particularly those from underrepresented groups. Despite this “popular” method, little has been done in the way of large-scale research efforts to better understand peer mentoring, particularly from the perspective of the peer mentors themselves.

According to Packard, Marciano, Payne, Bledzki, and Woodard (2014), peer mentoring is a process where an older, more experienced person in the field or practice provides “emotional, academic or career growth support for a new person (e.g., a student) through their shared activities” (p. 434). Mentoring provides a different experience for students to learn and grow for both mentees and their peer mentors (Stigmar, 2016). Despite the focus on peer-based mentoring and persistence in STEM education, the majority of research has focused on the outcomes of first-year students who participate in peer mentoring and not the peer mentors who actually deliver these services (Page & Hanna, 2008).

According to Amaral and Vala (2009), most of the research conducted on the outcomes for mentors have been somewhat relaxed. In some instances, mentor findings have been referred to as “add ons” or “additional” findings to the core investigation. The few studies conducted have found that peer-based mentors in STEM have changed epistemological beliefs of the mentors themselves about teaching and learning, as well as improved communication, organization, and leadership skills; however, it has also been noted that these studies have typically involved small samples of mentors, with approximately 5 to 40 mentors total (Amaral & Vala, 2009).

This article provides a program overview for a large university-wide peer mentoring program, whereby 372 mentors served over 3,000 first-year students across three years.

A research-intensive (R1) university located on the upper eastern seaboard with a total enrollment of approximately 6,400 undergraduate students recognized the need to better support first-year students, particularly in the crucial first semester. In 2014, the institution secured a five-year grant for $1.2 million from the Howard Hughes Medical Institutes (HHMI) to design and implement an Integrative Program for Education, Research and Support Involving Science and Technology (I-PERSIST). The purpose of this learning community program was to provide all first-year students enrolled in one or more of the introductory, faculty-taught, STEM courses (i.e., Calculus I, Chemistry I, or Physics I) with weekly small group, tutoring-support sessions run by a peer mentor. The main objectives of I-PERSIST were to: improve student experiences in these three “gatekeeper” courses, help incoming STEM students develop key study and social skills shown in the literature to help students persist, improve student academic achievement, and increase the percentage of first-year STEM students continuing at the institution. The long-term goal of I-PERSIST was to increase student persistence in STEM fields. Retention of first-year students has remained stable even though the number of incoming students has increased.

In order to provide enough peer mentors for the incoming 1,000 first-year students, an average of 124 mentors were recruited annually. Peer mentors were mostly second or third year undergraduates who applied in January to serve as a mentor starting in the fall (see Table 1). Students interested in serving as mentors applied by completing a comprehensive application form. The responses to application questions, along with the GPA (seeking at least a B grade in the course they applied to mentor and overall GPA > 3.0) were used to select applicants for a two-phase interview process with staff from the Office of Student Life and the respective faculty from the various disciplines. The comprehensive nature of the application-selection process allowed the selection of student mentors with the commitment, personal qualities, and academic background to serve as effective mentors. In year one, the program faced some challenges recruiting mentors and only 99 were available; however, in years two and three, 136 and 137 mentors were recruited, respectively. A third of the mentors in years two and three were returning mentors from the previous year. This strategy resulted in a core of experienced mentors who provided guidance to first-time mentors. It was believed that students saw mentoring as not only a way to give back to their college community, but also an activity they could use to build their resume for graduate school or for potential employers seeking the skills the mentoring program helped develop (e.g., increased communication skills, teamwork, collaboration).

Mentor training was a key component of the I-PERSIST program, and all mentors were required to participate in extensive training before they were able to lead mentoring sessions. Training took place in two parts (see Table 1). In early May, the selected mentors were provided an introduction and overview of the I-PERSIST program, its goals and objectives, how the program functioned, and their roles and obligations. Following this, mentors participated in a three-day training late in the summer just before the beginning of the semester (see Table 1). Mentors received intense training on how mentor sessions should function and how to help mentees develop study skills, time management skills, and problem-solving abilities. Mentors also participated in mock simulations with staff to better understand how to address first-year students’ concerns and perspectives, particularly around skill-building exercises. Overall topics covered in trainings included: working with first-year students, understanding issues surrounding diversity and inclusion, engaging in various approaches to learning, creating an engaging classroom, and working with international students. Mentors were paid a stipend for their participation in the three-day summer training. The stipend would not be available once the grant had expired; however, it was anticipated that this would pose minimal adverse effects because mentoring had become a popular activity amongst the student body. During the fall semester, the mentors did not receive financial compensation for their participation, but instead got course credits by registering in a mentoring course specifically designed for the program. Thus, the mentoring program was very cost-effective to the institution. Whenever possible, underrepresented minorty (URM) students were assigned to a URM mentor or to a mentoring group with another URM mentee. Also, mentees who were identified as potentially at-risk were not placed alone in mentoring groups with students who were overachievers.

Mentors were each assigned two groups of eight to ten students and held sessions with each group for one hour once a week. These small group sessions were held in classrooms or meeting rooms throughout campus. Mentees could be in all three of the introductory courses, and therefore, receive all three mentoring experiences; however, in most cases mentees were in one or two of the introductory courses. During the weekly sessions, mentors worked with their mentees to reinforce material and concepts taught during weekly course lectures. Mentors also taught first-year transitional content (study skills, time management, test-taking strategies, etc.), similar to subjects found in traditional first-year seminar courses. In addition to weekly mentoring sessions, mentors were required to hold two hour-long office hour sessions a week. These office hours were designed to provide students with individual time to ask content-related questions.

During the academic year mentors received “just-in-time” professional development. Mentors met once a week with faculty from their subject area, as well as staff from the Office of Student Life. During these supervisory meetings, mentors received information about upcoming lesson plans and reported back on the progress and outcomes of their sessions. Mentors provided faculty and staff with feedback about students who were absent, unengaged, or seemed to be struggling. This allowed staff the opportunity to follow up with these targeted students and get them back on track. An electronic form was also created so mentors could report at-risk students in a timely manner.

A coordinator was hired for the Student Affairs Division to run the day-to-day operations of the I-PERSIST program, working with staff, faculty, mentors, and mentees on a regular basis. An advisory committee was also formed. This committee included faculty, including some who taught these introductory courses, members from the Offices of First-Year Experience, Student Support Services, and Institutional Research. This committee met monthly throughout the academic year to address programmatic concerns, review formative and summative student outcome data, plan activities, and modify programming where necessary.

Program evaluation was another key activity. While Institutional Research served as the program’s internal evaluator, an external evaluator was also hired to collect formative and summative data and to author an annual evaluation report. The mentor survey was administered in the fall of 2015–2017 and consisted of a series of close-ended items and open-ended items. The purpose of the survey was to gather mentor perceptions of the overall mentoring process, skills development, relationships established, and connections to the institution formed as a result of participating as a mentor. The survey items used a Likert-type scale. Content validity was established through a thorough review of the literature and professional review. Cronbach alphas for each of the four sections of the survey were .64 and greater. A copy of the mentor survey can be found online (available at ).

Survey data for mentors were collected internally by institutional research, along with demographics data (i.e., gender, ethnicity). Surveys were administered at the end of the fall semesters to mentors and overall 83% (309) responded. Demographics for the mentors who responded to the survey compared to demographics of incoming classes is shown in Table 2.

The first analysis examined the demographics of the mentors who responded to the survey. As shown in Table 2, females made up a significantly larger proportion of the mentors than males, and white students served as mentors more than any other ethnicity group; however, 12.9% of mentors identified themselves as Asian. While the majority (68%) of incoming first-year students were male, the majority (58%) of mentors were female.

Likert-type items from the mentor survey were examined using descriptive statistics and were used annually by the interdisciplinary faculty team that monitored the program. An Analysis of Variance (ANOVA) was performed to determine if there was a significant difference for item means across the three years of data (i.e., 2015–16, 2016–17, 2017–18).

Overall, mentors reported positive outcomes associated with skill development. Most agreed that as a result of being a mentor they had increased their own leadership, study, presentation, and time management skills. From year to year, mentors had an increase in this perception. An ANOVA revealed that the difference in 2017–18 means for leadership and presentation skills were significantly greater than for 2015–16 or 2016–17 (see Table 3).

In addition to skills, mentors also reported agreement with topics related to content knowledge and self-regulation. An ANOVA revealed a significant difference in mean for 2017–18 compared to the other years for increased knowledge of the subject matter, greater motivation to be successful, and better able to cope with stress (see Table 4).

Mentors also reported a stronger connection to others in the college community. More specifically, a posthoc analysis revealed that all four items supporting improved academic relationships were found to be significant for 2017–18 compared to the other years (see Table 5).

Lastly, mentors’ satisfaction with the institution were examined. Posthoc analysis revealed that across all three items related to connecting to the institution, the 2017–18 mean was statistically significant compared to the other two years. Mentors in year 2017–18 may have been in more agreement about the outcomes and benefits of their mentoring experience. Many of them had served as mentors previously compared to the two other years. Approximately, one third of the mentors were returning mentors, which program officials perceived as a positive sign. This high rate of returning mentors increased the quality of the program, thereby enhancing the mentoring experience for both mentees and mentors (Table 6).

In addition to close-ended Likert item, several open-ended questions were asked. One item gathered in-depth information from mentors about the challenges they faced with their respective mentees. Mentors reported challenges with mentees. For example, mentors noted that many times mentees were overconfident, not doing their homework, not having time management skills, not handling the transition from high school to college well, not possessing good stress management skills, and not using university resources. For example, one mentor wrote:

“Many of my students studied ‘hard’ for the first and second exams, then didn’t do as well as they wanted to. The root of the problem was their studying was rooted in reward, not challenge. They would mainly just read the material and mostly only the concepts they were already somewhat comfortable with. This created an inflated sense of preparedness that crumbled in the exam room.”

Another item gathered was about the benefits mentors believed they gained from the experience. These benefits included, but were not limited to giving back to the institution; connecting with faculty and first-year experience personnel; making friends; and developing leadership, presentation, and social skills. When asked if they would consider mentoring again, the majority of mentors indicated that they would. More specifically, one mentor wrote:

“Absolutely. I can say without hesitation that this program has changed my life for the better and I thank everyone involved for providing me with this opportunity. In learning about all of the things I should be passing on to my students, I learned so much about what I could be doing better for myself. Not only that, but being in front of a room of people looking up to me for advice and guidance placed a level of responsibility on me that I’ve never had before. That responsibility gave me something to live up to, and once I set my sights on excellence it’s been almost impossible for me to aim at anything else.”

STEM businesses and industry continue to report that undergraduates entering the profession severely lack basic communication skills (both oral and written), as well as the ability to collaborate, problem-solve using a team approach, and develop positive working relationships with others (Lingard & Barkataki, 2011). Results from this project demonstrated that mentors received valuable skills that will serve them well as they enter their respected professional fields. Developing these skills may also be another reason the institution has seen the steady increase in interest by students to serve as mentors.

In addition, this project has also helped to expand the outcomes generally associated with mentors. Research on peer mentoring has focused on increasing mentors’ skills and content knowledge; however, this project also documented the ability of mentors to foster positive relationships with faculty and staff, and form an overall better bond with the institution as a whole.

This program could be developed with minimal additional funds if the faculty and student life staff are committed. Office of Student Life/Success would be best suited to take the lead in organizing and administering the program because of the laborious nature of selection and training students, and their direct alignment with the goals of the program. If students are willing to serve for the purpose of the experience and professional development, this could be reflected in their transcript as a two-credit mentoring course as in this program. Department chairs/deans must also have a commitment to provide a modest teaching release (i.e., one course credit per year), as well as recognize the pedagogical and service contributions of the faculty participating. Some minimal funding would be required to purchase t-shirts for the program, lunch during the few training days, community-building activities, and for office supplies (binders, handouts, etc.). Interested institutions may want to consider starting their mentor program as a pilot, with one subject area.

One of the core components to implementing the I-PERSIST program was that students be willing to serve as mentors without financial compensation. This was an essential aspect for long-term sustainability and for feasible implementation at other universities. While this program was implemented in a four-year school, there is no reason it could not be replicated in a community college. Before scaling up a current program or implementing a new program of this size, it is recommended that the institution determine whether there may be enough interest from students to serve as mentors. A shortage of appropriate volunteer students to serve as mentors would pose a great challenge to successfully implementing this model.

This large-scale effort required a high level of collaboration and coordination within the organization from both STEM faculty and Student Life staff. Program team members, faculty, and members from the Offices of First-Year Experience, Student Support Services, and Institutional Research met monthly throughout the academic year to address programmatic concerns, review formative and summative student outcome data, and modify the programming accordingly. This level of commitment is essential for running a successful campus-wide mentoring program.

Acknowledgments

The authors would like to acknowledge the efforts of Rensselaer staff in the Office of Student Life and Institutional Research and Assessment, as well as faculty from the Departments of Chemistry and Chemical Biology, Mathematical Sciences, Physics, Applied Physics, and Astronomy for making the I-PERSIST program possible.

Dean T. Spaulding (ds6494@yahoo.com) is the vice president of external evaluation and new projects for Z Score Inc. in Albany, New York. Jelane A. Kennedy is an associate professor in the Department of Counselor Education and Family Therapy at Central Connecticut State University in New Britain,Connecticut. Amanda Rozsavolgyi is an evaluation assistant at Z Score Inc. and a senior research support specialist at the Center for Human Services Research at the University at Albany in Albany, New York. Wilfredo Colón is professor and head of chemistry and chemical biology at Rensselaer Polytechnic Institute in Troy, New York.