Peer Mentor Program for the General Chemistry ... - ACS Publications


Peer Mentor Program for the General Chemistry...

1 downloads 92 Views 1MB Size

Article Cite This: J. Chem. Educ. XXXX, XXX, XXX-XXX

pubs.acs.org/jchemeduc

Peer Mentor Program for the General Chemistry Laboratory Designed To Improve Undergraduate STEM Retention Fehmi Damkaci,* Timothy F. Braun, and Kristin Gublo Department of Chemistry Department, State University of New York at Oswego, Oswego, New York 13126, United States ABSTRACT: We describe the design and implementation of an undergraduate peer mentor program that can overlay an existing general chemistry laboratory and is designed to improve STEM student retention. For the first four freshman cohorts going through the program, year-to-year retention improved by a four year average of 20% for students in peer mentored laboratories versus those not in peer mentored laboratories. A majority of students in peer mentored laboratories recommend continuation of the program, 85.4% of 765 students in the first three cohorts. Our program is uncomplicated and requires only modest financial resources to affect significant increases in STEM retention. KEYWORDS: First-Year Undergraduate/General, Laboratory Instruction, Collaborative/Cooperative Learning, Student Career/Counseling, Professional Development, Laboratory Management



chemistry.5 In the PLTL model, students work together in modest sized groups led by a trained peer to work their way through laboratory exercises or chemistry problems with only occasional outside help from the instructor, thus the term “peer led”.6 In PLTL, groups of students spend the majority of the lab in the same group working with the same peer. In our program, the peer mentors move through the entire lab, and their goal is to interact with every student in the class each lab session. PLTL focuses on increasing student learning about chemistry concepts.7 Our program focuses more generally on increasing retention in the STEM fields by seeking to improve student’s understanding of, and attachment to, their major while also facilitating the building of supportive social networks. There is less focus on chemistry specific content in our program, although improving chemistry concepts is a definite side benefit, and the extra assistance with lab experiments and concepts is what helps to win over lab instructors to support the program. The general chemistry laboratories provide a weekly 3 h of interaction, which is part of the student’s schedule and cannot be easily avoided. Indeed our program can be tailored to be delivered in any laboratory setting regardless of discipline. Chemistry laboratories work particularly well as hosts for our program because most chemistry, biology, and earth science majors take at least one semester of general chemistry. At our institution, it is the only lab taken by almost 80% of our STEM majors during their freshman year. The term “peer mentor” should be explicitly defined to better understand our program. Traditional mentors are more experienced members of a profession who are willing to work with new members of the profession; often there is a sizable age gap between mentor and protégé. This is the situation between most faculty and students at a four year university. Peer

INTRODUCTION Retaining science, technology, engineering, and math (STEM) students at the undergraduate level has become a national priority.1 Nationwide, only 52% of students who started a STEM bachelor’s degree program between 2003 and 2009 graduated with a degree in STEM.2 While student’s reasons for leaving the STEM fields are varied, research going back to the 1980s has found that student’s ability to form an attachment to their prospective major and form social networks among their peers in their discipline are two of the greatest determining factors as to whether undergraduate students stay in a STEM field.3,4 With these ideas in mind, we have designed a peermentoring program for the general chemistry laboratory that will serve as a way of increasing attachment to the campus and STEM major and support the formation of social networks among peers. A laboratory component to accompany the lecture for general chemistry has been a staple of undergraduate chemistry programs for many decades and indeed is required for an American Chemical Society accreditation. While much work has been done to modernize the general chemistry laboratory course to make it more relevant and intellectually challenging for students, most general chemistry laboratories give scant notice to the humanistic component, that is, the students and their interaction. Instructor−student interaction in general chemistry laboratories can be more intimate than the lecture, but resources usually limit most programs to putting 20−24 students in each laboratory section with a single instructor, sometimes with teaching assistant (TA) help, sometimes not. The instructor will have limited time to interact with the students and must spend most of their time focused on the mechanics of the experiment being performed. A lab TA can provide more interaction, but most TA’s again focus strictly on the experimental and theoretical work of the lab. The mentoring program outlined in this paper bears some similarity to, but is distinct from, what has been termed “PeerLed Team Learning” (PLTL) laboratories. A number of institutions have experience with PLTL laboratories in general © XXXX American Chemical Society and Division of Chemical Education, Inc.

Received: May 16, 2017 Revised: October 29, 2017

A

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

Figure 1. Presemester peer mentoring topic survey for fall and spring semesters, combined results from fall14 to spring16.

students each week; selections of topics are listed in Figure 1. The total meeting time is generally about an hour each Friday. During this meeting time, peer-mentoring topics are discussed and peer mentors are encouraged to suggest tips and points to be included in the weekly discussions with students. Handouts of discussion points on each topic are provided for the peer mentors at these weekly meetings, and occasionally handouts are also distributed to the students. During the actual laboratory period, the peer mentors can help students with the day’s experiment, and such help is a useful lead-in to talking about the peer mentoring topic, but the peer mentors are expected to move around the laboratory and try to talk to all the students in that lab section each week about the peer mentoring topic for that week. Their role is not to be disruptive and to become a natural and distinct part of the lab to the students and instructors within a few lab sections. Peer mentors are not traditional teaching assistants since they are not primarily responsible for communicating theory and practice of the experiment. Rather, peer mentors are there to spend time with each student or group of students discussing the peer-mentoring topic of that week. Of course at least some time is also spent talking about other subjects and this reflects the social networking aspect of the program. Peer mentors are trained to be aware of being “captured” by individual students

mentors, on the other hand, are much closer in age and experience to the mentee. Junior and senior undergraduate peer mentors can be very effective when working with freshman because they have recently been freshman and are more in tune with the generational experiences and struggles of both groups. Research has shown that peer mentors are more easily able to productively interact with new undergraduate students.8 Furthermore, students who experience peer mentoring are more likely to show greater satisfaction with their university and greater affective commitment to their discipline than are students who do not experience peer mentoring.9 As Reid suggests, “real magic in a mentoring program comes in attention to details, in attentiveness and planning, in learning and practicing and reflecting”.10



PEER MENTOR PROGRAM FOR GENERAL CHEMISTRY LABORATORIES The structure of the peer mentor program for our general chemistry laboratories is rather straightforward. One peer mentor is assigned to each laboratory section. The peer mentors attend the preparatory meeting for laboratory instructors each week and then stay for a separate session on the peer-mentorship topic on which they are to work with the B

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

or groups and of “escape” techniques to avoid being monopolized by a small segment of the lab population. From a student selection perspective, this system is quite well blinded. Students choosing a lab section during course selection do not know which lab sections participate in the peer-mentor program at the time of registration. Lab sections with peermentors are selected randomly based on the availability and schedule of the peer-mentors. The general chemistry laboratories at SUNY Oswego are taught primarily by graduate students in the department’s Master’s program. As such, they are the only instructors in the lab with the students. There are usually 12 or 13 experiments each semester, as our peer-mentoring program uses the first lab to gather student input, we cover about 11 or 12 nonexperimental topics, as listed in Figure 1, each semester. Laboratory sections are capped at 20 students. The peermentoring program was designed to not require any adjustment in the general chemistry laboratory schedule.

Table 1. Rubric for Assigning Letter Grades for Peer Mentors Grade A

A− B+ B B−

Criteria 1. Their lab section participated in presemester survey. 2. Attended weekly lab meetings on a regular basis. 3. No absences during the semester when their lab section met. 4. Submitted post semester surveys on time. 5. Turned in reflection paper on time. 6. Gave 100% while in the lab helping students with their experiments. 7. Gave 100% while in the lab conversing with students on the assigned weekly topics. Did not perform well in 1 out of the 7 Did not perform well in 2 out of the 7 Did not perform well in 3 out of the 7 Did not perform well in 4 out of the 7

experienced over the course of the semester and how it affects their identity as scientists and educators.

Determining Nonexperimental Topics

Recruiting Peer Mentors

A questionnaire is distributed during the first laboratory of the semester, and it consists of a list of possible nonexperimental peer-mentoring topics for that semester and students are asked to rank them on a bubble sheet, based on interest, on a scale of 1 to 5 (5 being most interested). We then tailor the list of topics for each week that semester based on the student feedback, including some topics we feel need to be covered even if the students do not select them (the results are shown in Figure 1). The peer mentors are prepped on and discuss each topic at the weekly meetings as they come up through the semester. Topics such as study tips, preparing for finals, summer research, and course selection are typically among the most selected topics.

Finding sufficient, quality undergraduate STEM students to be peer mentors has been one of the more challenging aspects of our program, especially after we expanded the program to try to cover all general chemistry laboratories after the second year. We typically run about 16−20 laboratory sections in the fall and 14−15 in the spring. Some peer mentors are able to cover two lab sections, but we usually will need at least 12 peer mentors for the fall semester. Most peer mentors are chemistry or biochemistry majors, although we usually also have one or two biology or zoology majors as well. As with many public university chemistry departments, more than half of the students going through our general chemistry series are not chemistry or biochemistry majors, so some diversity of peer mentors helps. We are now headed into our fifth year of the peer-mentoring program and most of our peer mentors have now come up through the peer-mentored laboratories, which has greatly facilitated peer-mentor recruiting. To introduce the peer mentors to the students, the program prepared a peermentor poster, which included each peer-mentors photo, a description of themselves, and their future goals. The peermentor poster was placed on the hallway outside of the general chemistry laboratories where students could read about their peer mentors while waiting for their lab periods.

Assessment of Peer Mentoring

Students who are in peer-mentored laboratories are asked to fill out a satisfaction survey at the end of the semester. We use the results of the survey to modify the program for following semesters (the results are shown in Figure 1). The ultimate goal of our program is to increase STEM student graduation rates. As there is at least a four-year lag between beginning and finishing an undergraduate program of study, we have chosen year-to-year retention (freshman to sophomore; sophomore to junior; etc.) as an intermediate measure of program success. In obtaining information from SUNY Oswego’s office of Institutional Research, we are also able to monitor GPA, disqualification, students switching to non-STEM degrees, and voluntary nonreturning students, giving us the ability to also look for correlates among these other issues pertaining to STEM student retention.



RESULTS Here we report STEM student retention data over the first four full years of the general chemistry peer mentoring program at SUNY Oswego, which began in the fall of 2012 (data for freshman cohorts from 2012, 2013, 2014, and 2015). Year-toyear retention of STEM students from their freshman to sophomore years, with corresponding loss pathways, with and without experiencing a peer mentored general chemistry laboratory, is shown in Table 2. The primary focus of the project, improving first year retention (freshman to sophomore), is summarized in Figure 2. The first year of the program (2012) was very successful, showing a 48% retention difference (two tailed t test p = 2.73 × 10−13) between students who did have a peer mentored general chemistry lab and those that did not (nM = 96, nC = 118). However, only 44.8% of all eligible STEM students were going through a peer-mentored lab in the 2012 academic year. The peer-mentoring program was expanded in the 2013 academic year and again showed a substantial first year retention

Assessment of Peer Mentors

Junior and senior undergraduate STEM students who took general chemistry and performed well (grade with B+ or higher) may become a peer mentor by enrolling in a teaching of chemistry course for one or two credits. Peer mentors are assessed (graded) on the rubric shown in Table 1. Items 1−5 are easily quantified. Items 6 and 7 are scored based on the lab section’s satisfaction survey, which includes questions about the performance of the peer mentor, lab instructor feedback, and the observations of the peer mentor director. Peer mentors are also required to write reflection papers about their experience over the semester. The reflection paper is meant to get the peer mentors to think more about what they C

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

Table 2. Year-to-Year Student Retention: Students in Mentored Chemistry Labs versus Students in Nonmentored Labs as Controla Cohort in STEM Retention (%) Cohort F′12 (nM = 96, nC = 118) 1st Y 2nd Y 3rd Y 4th Y Graduated F′13 (nM = 160, nC = 103) 1st Y 2nd Y 3rd Y F′14 (nM = 207, nC = 79) 1st Y F′15 (nM = 216, nC = 62) 1st Y

Mentored

Cohort Non-STEM Retention (%)

Control

Mentored

89

41

6

78 60 57 47.3 76

32 23 23 17 59

61 51 71 76

Control

Cohort Academic Disqualified (%) Mentored

Control

12

1

25

8 21 22.5 17.2 8

17 22 20 12.6 14

3 3 4

27 28 31.5

9.6

39 36 56

18 19 11

27 29 20

10 10 7.5

55

10

13

6

Cohort Nonreturning (%) Mentored

Control

Cohort on Campus Retention (%) Mentored

Control

All STEM Campus Retention (%)

3

22

95

53

71.4

11 16 16

25 27 26

12

7

16

86 81 77.5 64.5 84

49 45 43 29.6 73

60.3 56.1 55.7 42.4 78.2

12 12 16

16.4 20 10.6

23 23.5 8

79 70 82

66 65 76

70.4 65.4 79.6

8.3

17.7

86

68

81

14.5

“Cohort” is the year of student enrolled in general chemistry lab in SUNY Oswego, “n =” is the number of students who took general chemistry lab (nM = mentored lab students, nC = control lab, nonmentored, students), “first Y” refers to retention from freshman to sophomore year, “second Y” refers to retention from sophomore to junior year, and “third Y” refers to retention from junior to senior year; “fourth Y” retention is a calculated pass through rate combining graduating students and continuing students. “Graduation rate” is the percentage of starting students who graduate in 4 years. “STEM Retained” is the percentage of returning students within cohort who stayed in a STEM major (even if they switched from one STEM major to another). “Non-STEM Retained” is the percentage of returning students within cohort who switched majors to a non-STEM field but remained at SUNY Oswego. “Academic Disqualification” applies to students within cohort whose GPA falls below thresholds that vary according to number of credit hours completed and results in those students being dismissed from SUNY Oswego. “Nonreturning” are students within cohort who are eligible to return to SUNY Oswego the following year but chose not to return. “Cohort on Campus Retention” is the total percentage of students within cohort returning to SUNY Oswego, sum of “STEM Retained” and “Non-STEM Retained”. “All STEM Campus Retention” includes all STEM Majors (whether or not they took general chemistry, but arrived in SUNY Oswego in that year) and thus these numbers are a different from” Cohort Campus Retention”.

a

Figure 2. First year retention rates comparison of peer mentored lab students versus nonpeer mentored lab students. Values in parentheses are two tailed t test significance (p) values for the difference in retention between peer mentored students and nonpeer mentored (control) students.

difference of 17% (p = 2.0 × 10−3) between students having a peer-mentored lab their first year and those that did not, with 60.8% of all students in a peer-mentored lab (nM = 160, nC =

103). In the third year of the program, 2014, we expanded the peer mentoring to almost all general chemistry laboratories, 72.4% of students that year (nM = 207, nC = 79), and saw a D

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

positive retention difference of 15% (p = 1.0 × 10−2) between students who did have a peer-mentored lab and those that did not. The 2015 freshman STEM cohort saw a 21% gap (p = 1.1 × 10−3) between peer mentored lab students and students lacking such an experience with 77.5% of the cohort in a peermentored lab (nM = 216, nC = 62). As can be seen in Table 2, we also collected data on students who switched majors out of STEM programs (non-STEM retained) as well as academically disqualified (DQ) students and students who chose not to return to SUNY Oswego for the next year. Students who did not have a peer mentored general chemistry lab were about twice as likely to be in one of these groups than those in a peer-mentored lab, with considerable variance from year to year both between categories and within categories. We also show retention data for the sophomore to junior year (2012 and 2013) and junior to senior (2012 only) years. While retention falls steadily in subsequent years, note that students who have a peer-mentored general chemistry laboratory remain consistently above the campus STEM student retention (which includes some STEM majors who do not normally take the general chemistry series) as well as the nonpeer lab group. When cohort campus retention is compared with overall STEM campus retention (Table 2, last two columns), students who had peer mentors showed greater on campus retention rates than overall STEM majors, while students who did not have peer mentors showed lower retention rates. Lastly, we show a 4 year graduation rate for the 2012 cohort, which follows the same trends, peer-mentored cohort graduation rate is substantially higher (∼30%) than the nonpeer laboratories group as well as (∼22%) than the general student population. The presemester topic surveys for fall and spring semesters are combined separately as shown in Figure 1. Students’ interest in topics to be discussed by peers does not show significant change from year to year. The top five topics for fall and spring semesters are almost the same. Even though most students lack of math preparedness (based on our in-coming STEM freshman math placement assessment), students do not feel a need to discuss or seek help in math preparation or increasing their ability. Qualitative data to support the assertion that the students themselves also feel that the peer-mentoring program for the general chemistry lab is useful. Most students over four semesters (85.4%, n = 765, average = 4.3/5) felt that the peermentored laboratories were a benefit and should be continued. Similarly, 84% of the students (n = 765, average = 4.2/5) find the peer mentoring help them in their learning chemistry.

The 48% retention in STEM major difference between students who did have a peer-mentored lab versus those that did not (control) in the first year of the program was good, but too high to be realistically reproducible. We suspect that we were the beneficiaries of a positive sorting effect that year, which resulted in the peer-mentored laboratories having many more successful students than other, nonpeer mentored laboratories. This supposition appears to be borne out with the results from the next three years with more peer mentored laboratories. Second, third, and fourth year program retention numbers average a 17% positive retention difference. We suspect that the longer-term average positive benefit is more likely to be around a 15% increase in freshman to sophomore year retention in STEM major. When the first four years of data for first year retention in STEM major are combined, the retention rate was significantly higher among those mentored (p = 3.67 × 10−13 (two tailed t test) p < 1.0 × 10−4; 95% confidence interval for difference (0.185, 0.310)).12 We must also consider whether factors other than the peer mentoring are at work in this system. One benefit to the way this program was implemented is it overlays on an existing general chemistry lab series, students enrolling in the laboratories do not know whether they will be in a peer mentored lab. This blinding removes the effect of student choice as a plausible reason for why students in peer-mentored laboratories do better. Likewise, instructors and peer mentors are blind as to which lab sections they will be in, largely removing choice as an option there as well. Sorting effects are certainly possible, as noted for the first year of our program, but should decrease as the program nears saturation where nearly all laboratories are covered. Other effects are possible such as time of day or day of week variations. The peer mentored laboratories are randomly assigned based on peer availability and a brief review does not show any day of week or time of day patterns as to when nonpeer mentored laboratories occur. Part of the college process is self-discovery and it is inevitable that some young people will join a major without fully understanding what it takes to complete that major. Those students should switch to another major as they come to better understand their major and their own strengths, weaknesses, and goals. Thus, achieving 100% retention in any major is not the end goal, rather institutions should work to prevent the loss of students due to more avoidable factors such as academic failure, failure to get into major classes, failure to understand what the major entails, or inability to form a connection with their major and or peers within their major. Our peer mentoring program seeks to increase student attachment to and understanding of their major and what it takes to succeed in science by learning from those who have already done so. Given the link between peer mentoring and positive engagement with a student’s major, it should not be surprising that one of the most noticeable reductions in STEM student loss pathways would be the reduction in the number of students switching to a non-STEM major. It is somewhat surprising that there is not a stronger correlation between peer mentoring and nonreturning students, but the decision on whether to return to college is a bigger choice than switching majors and is impacted by more variables. Lack of correlation with academic disqualification is also reasonable as less prepared students are simply more likely to fail courses regardless of whether they experience peer mentoring. The more interesting question with the DQ students is whether peer mentoring is increasing rates of these students seeking



DISCUSSION In light of the fact that retention of STEM students has become a national as well as an institutional priority at most college campuses, the results from the first four years of our peermentoring program for general chemistry laboratories are quite encouraging. The peer-mentor program provides networking opportunities for freshman with their upper class peers in a structured program. According to Thomas et al. students perform better academically and are more likely to persist in school when “those students who possess broader, wellconnected networks...are able to more easily make connections with others due to the multitude of paths reaching to many parts of the overall network”.11 E

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

Figure 3. Comparison of campus retention rates for STEM students, the students in peer mentored laboratories and the general campus population.

stated “(peer mentoring) has led me to gain a broader insight into what students find most frustrating when learning chemistry and has challenged me to think of ways to help.” Several peer mentors mentioned that they wish they had a similar program when they were a freshman so they would not have struggled in finding a solution to their problems in their first semester on campus: “It made me think back to my freshman year, 4 years ago, where I used to struggle with basic calculations that I “thought” were hard...I feel that this program not only benefitted the students, but also benefitted me.” They have also suggested “students are less intimidated to approach a student mentor as opposed to approaching the professor or TA to ask a question”. Our study and program reinforces Colvin’s outcome of “both peer mentors and students saw benefits, ranging from individual gains to helping students become connected to the campus as a whole”.15 Recruiting peer mentors during the first two years of the peer mentored laboratories program was a challenge since the program was new on campus. However, in its third year, as the students who have had peer-mentored laboratories became juniors, it was easier to recruit enough peer-mentors. Student schedules always pose a challenge for finding a lab section and being able to attend the weekly peer-mentor meeting. The program was expanded to other freshman level laboratories, such physics and earth science, however successes of these are highly dependent on the motivation behind the program and recruitment of mentors within their majors. The recruitment of peer mentors usually is done before and during course registration period by letting juniors and seniors know about the program, its benefits, and the course number to register for to be part of the peer-mentoring program. The program we have laid out for peer mentoring in the general chemistry lab to increase STEM student retention is distinct from the peer-led teams in PLTL. Our program focuses less on specifically improving chemistry knowledge, as our peer mentors are not charged with traditional teaching duties, and more on increasing student understanding and ownership of their major. PLTL can be quite effective at increasing students’ understanding of chemistry concepts if properly implemented.5 However, PLTL can also require fairly extensive reworking of a department’s general chemistry program, taking considerable faculty effort, and requiring significant financial support from administration. These barriers likely explain why PLTL has not been widely adopted. Our peer-mentoring program, by

academic help such as tutoring, this would require a more targeted analysis with more data. The peer-mentoring program has contributed to closing the gap between first year STEM students’ campus retention and general first year campus retention (80% in 2013−2015), see Figure 3, regardless of whether students are retained with a STEM major. This aspect of the peer lab mentoring program benefit was unanticipated, but understandable. The peer lab program may be increasing the student’s attachment to the university and helping them to establish peer networks that they wish to maintain. The increase in on campus retention can be brought up as an additional benefit of the program when discussing support for the program with administrators. Students with peer-mentor experience also showed greater first year retention when compared to overall general campus retention (+2−3% difference), STEM majors’ campus retention (+4−6% difference), and control group (took general chemistry without a mentor, + 7−15% difference). When all four years of data are grouped together, the retention rate of the cohort on campus is significantly higher among those who were mentored (p = 3.68 × 10−13 (two tailed t test; p < 1.0 × 10−4); 95% confidence interval for difference: (0.135, 0.259)).12 The program’s secondary impact is creating relationships between freshman and upper-class peers as early as the first weeks the freshman’s college experience and providing a continued structure for interaction over two semesters. At the exit interviews, it was evident that students created strong social relations with peer mentors, which continued after the lab and lead to reaching out to them for additional help or socializing with them or attending student club activities more often than before.13 One of the peer-mentor stated, “Students tend to approach and ask personal questions to peer mentors pertaining their college life and future. It is also easier for them to contact peer mentors outside of the classroom than the teaching assistants”. The program’s benefits to peer mentors are substantial as well, providing teaching skills, which are needed for those who want to become graduate students in a PhD program. The peer mentors also reinforce their early chemistry knowledge, helping them better understand their current courses. There is evidence to suggest that this “reinforcement and exercising” of knowledge improves the later performance of peer mentors in other discipline related courses.14 Peer mentors also likely benefit by creating a better connection between themselves and their majors much as the freshman students. One peer mentor F

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

campus retention, most likely by creating supportive social networks between freshman and upper-class peers, thereby increasing their connectedness to the campus or their major.

contrast, needs little in the way of structural changes to the host program or outside resources to implement. Over the four years of our program, STEM student retention on campus has increased by 10 percentage points. Several other new programs, not described in this paper, have also contributed to this change, these programs are (a) increased STEM subject tutoring (number of students who received tutoring is substantial but much smaller than the number of students who received peer-mentored and also the degree of tutoring of each students varied considerably), (b) early career (limited to ten freshman and sophomore per year) summer research support, (c) a summer bridge camp for freshman focusing on math skills (students impacted by this program is very small portion of all STEM students for the period of this analysis), and (d) greater infusion of math concepts into early science courses (impacts all STEM students, including our nonpeer mentored students). Deconvoluting the individual contributions of these programs is essentially impossible, but as noted earlier, about 80% of STEM majors at Oswego take the general chemistry lab sequence so a substantial part of the 10 percentage point retention increase likely comes from the peer mentoring program. Implementation of our program required cooperation across several science departments and considerable help from administration. Over the first four years of this program, we have identified several points, which have led to successful implementation. The first is administrative support, without someone to push for resources and attention at the administrative level these types of programs cannot flourish. Second, securing dedicated communication between science departments and with administration. The National Science Foundation grant, which funded this program, required us to establish an internal advisory board, which we took to mean all STEM departments and administration. While there were certainly existing lines of communication between departments and with administration, one of the most important side benefits of setting up the peer mentored laboratories program has been the improved communication between STEM stakeholders. Indeed, SUNY Oswego is moving to institutionalize the meeting process that began with the STEM student retention grant. Lastly, and most importantly, the actual coordinator of the program must have the time, inclination, and skills to successfully manage the peer-mentoring program. We were lucky enough to have such a person in house in our existing chemistry lab coordinator; however, not all lab coordinators or indeed all faculty would have the required skills or time to make the peer-mentoring program work. Our lab coordinator receives additional compensation for the time it takes to manage the peer program, but all-in-all, the peer mentored laboratories program is amazingly inexpensive to run.



AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Fehmi Damkaci: 0000-0001-6944-7829 Notes

The authors declare no competing financial interest.



ACKNOWLEDGMENTS This project was supported by NSF STEP Grant No. 1161127. We thank Mehran Nojan, Director of Institutional Research and Assessment at SUNY Oswego, and Alla Gul for help with data gathering, design, and assessment of STEM retention projects, Dr. Eric Olson for help with the qualitative assessment (entrance interviews, exit interviews, and surveys), and Dr. Scott Preston for statistical analysis of the retention data.



REFERENCES

(1) National Science Board. Preparing the Next Generation of STEM Innovators: Identifying and Developing Our Nation’S Human Capital; National Science Foundation: Washington, DC, 2010. https://nsf. gov/nsb/publications/2010/nsb1033.pdf (accessed October 2017). (2) Chen, X.; Soldner, M. STEM Attrition: College Students’ Paths into and out of STEM Fields; U.S. Department of Education: Washington, DC, 2013. https://nces.ed.gov/pubs2014/2014001rev.pdf (accessed October 2017). (3) Tinto, V. Leaving College: Rethinking the Causes and Cures of Student Attrition; University of Chicago Press: Chicago, IL, 1987. (4) Stage, F. K. University attrition: LISREL with logistic regression for the persistence criterion. Res. High. Educ. 1988, 29, 343−357. (5) (a) Wilson, S. B.; Varma-Nelson, P. Small Groups, Significant Impact: A Review of Peer-Led Team Learning Research with Impacts for STEM Education Researchers and Faculty. J. Chem. Educ. 2016, 93, 1686−1702. (b) Robert, J.; Lewis, S. E.; Oueini, R.; Mapugay, A. Coordinated Implementation and Evaluation of Flipped Classes and Peer-Led Team Learning in General Chemistry. J. Chem. Educ. 2016, 93, 1993−1998. (c) Lewis, S. E. Retention and reform: An evaluation of peer-led team learning. J. Chem. Educ. 2011, 88, 703−707. (d) Gafney, L.; Varma-Nelson, P. Evaluating Peer-Led Team Learning: A Study of Long-Term Effects on Former Workshop Peer Leaders. J. Chem. Educ. 2007, 84, 535−539. (6) Gosser, D.; Roth, V.; Gafney, L.; Kampmeier, J.; Strozak, V.; Varma-Nelson, P.; Radel, S.; Weiner, M. Workshop chemistry: Overcoming the barriers to student success. Chem. Educ. 1996, 1, 1−1710. (7) Gosser, D. K., Jr.; Kampmeier, J. A.; Varma-Nelson, P. Peer-Led Team Learning: 2008 James Flack Norris Award Address. J. Chem. Educ. 2010, 87, 374−38010. (8) Parker, P.; Hall, D.; Kram, K. Peer Coaching: A Relational Process for Accelerating Career Learning. Acad. Manag. Learn. Edu. 2008, 7, 487−503. (9) Sanchez, R.; Bauer, T.; Paronto, M. Peer-Mentoring Freshmen: Implications for Satisfaction, Commitment, and Retention to Graduation. Acad. Manag. Learn. Edu. 2006, 5, 25−37. (10) Reid, E. S. Mentoring peer mentors: Mentor education and support in the composition program. Composition Studies 2008, 36, 51−79. (11) Thomas, S. L. Ties that bind. J. Higher Educ. 2000, 71, 591−61. (12) A 95% upper confidence interval for difference suggests that for every (no more than) five students who are not peer-mentored,



CONCLUSION We have created and institutionalized a peer-mentorship program, which is structured around general chemistry laboratories. It provides a weekly 3 h meeting time, which is part of the student’s regular schedule without need to schedule an additional time slot. Peer mentors are also trained at a weekly lab meeting regarding their role and nonexperimental topic to be discussed with students. Both students and peer mentors showed great interest in the program and gained benefit from the program based on exit surveys. The program showed a significant impact on both STEM retention and on G

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX

Journal of Chemical Education

Article

replacing them with five who are peer-mentored will result in an additional one STEM major retention. (13) Varma-Nelson, P. Peer-Led Team Learning. Metrop. Univ. 2006, 17, 19−29. (14) Amaral, K. E.; Vala, M. What Teaching Teaches; Mentoring and the performance gains of mentors. J. Chem. Educ. 2009, 86 (5), 630− 633. (15) Colvin, J. W.; Ashman, M. Roles, Risks, and Benefits of Peer Mentoring Relationships in Higher Education. Mentoring & Tutoring: Partnership in Learning. 2010, 18, 121−134.

H

DOI: 10.1021/acs.jchemed.7b00340 J. Chem. Educ. XXXX, XXX, XXX−XXX