Individuals, Relationships, and Institutional Responsibility:
How Can Institutional Culture Better Support Mentorship?
Effective mentorship practices contribute to the education and development of the next generation of diverse science, technology, engineering, mathematics, and medicine (STEMM) professionals.1 Because diversity in STEMM workforces has a positive effect on the STEMM ecosystem and on innovation, supporting effective mentorship and mitigating negative mentoring experiences will likely result in STEMM workplaces that are more creative, innovative, and responsive to current and emerging problems.
However, significant institutional change—requiring buy-in from institutional leadership, college deans, department chairs, and individual faculty, as well as new institutional policy—may be needed to ensure broader access to effective mentorship and support systems (Fleming et al., 2012; Packard, 2016). Funding agencies can also play an important role in creating cultures supportive of mentorship (Fleming et al., 2012; Jeste et al., 2009) and can catalyze institutional change in mentorship processes so that outcome measures become routine components of grant applications and reporting requirements.2
This chapter addresses the roles in which multiple participants can serve as well-prepared, informed advocates for effective mentorship,3 one element in inclusive excel-
3 Including university leadership (e.g., presidents, provosts, deans), department chairs, program leaders (e.g., research, training, and graduate program directors), mentors (faculty members, staff, and others who have extensive contact with graduate and undergraduate students), and mentees (undergraduate and graduate students participating in mentoring programs and other mentoring relationships), agencies that fund mentorship programs, and professional or disciplinary associations.
lence in STEMM education and workforces. First, the chapter lays out what a “culture of mentorship” means. Second, it provides an overview of barriers that may be faced in creating change to support effective mentorship and a theory of organization change. Finally, it summarizes possible actions each stakeholder group in the mentorship ecosystem can take to improve mentorship in STEMM. Box 7-1 highlights how theory may inform the concepts that are discussed.
A CULTURE OF MENTORSHIP
The growing science of mentorship indicates that mentorship is a learnable skill, much like teaching and research, capable of improving individual- and institution-level outcomes. In addition, mentorship education can improve mentor competence from the perspective of both the mentor and the mentee. To realize the full potential of mentorship effectiveness, however, changes are needed at all levels of higher education, as well as in external environments such as professional associations.4
Mentorship, as an evidence-based practice, can be systematically integrated into the work of individuals and organizations focused on preparing diverse undergraduate and graduate students to join and be successful in the STEMM workforce. Though mentorship is an activity based on personal relationships—and its successes or failures ultimately hinge on the quality of those relationships—institutions can play a critical role in fostering and supporting mentor-mentee relationships. Institutional culture can promote mentorship by creating settings where faculty members and staff jointly commit themselves to promoting mentoring and facilitate mentors’ abilities to be more effective and culturally responsive in their mentorship of STEMM students. However, faculty members, staff, and others who wish to engage in effective mentorship in the absence of a supportive institutional culture often must work against that culture (DeAngelo, 2016).
CULTURE CHANGE TO SUPPORT EFFECTIVE MENTORSHIP
Academic organizations have strong normative or unspoken rules that are part of the academic culture and that influence expectations for behavior and engagement.5 Changing academic culture involves institutional transformation that is deep, reaching into daily work and value systems, and pervasive in that it is widely adopted across academic units, disciplines, and participants (Choi et al., 2019; Gehrke and Kezar, 2018; Kezar, 2018; Kezar et al., 2018). Institutional transformation in support of effective mentorship involves mentees, mentors, training program directors, departmental chairs, deans, provosts, college presidents, and external partners—all of whom can use mentoring as one intervention to increase retention in STEMM disciplines and help move a more diverse group of students along STEMM career pathways. As was stated in the National Academies report Graduate STEM Education for the 21st Century, “Achieving [cultural change] will require a clear commitment and changes in both policies and practices throughout the [higher education ecological] system, as well as focused actions by every stakeholder” (NASEM, 2018c, p. 127). In short, engaging in organizational change will involve energetic change agents, distributed leadership, adequate support, and commitment to long-term change that will embed quality mentorship practices in daily work (Spillane et al., 2001).
Barriers to Change in Mentorship
Research has documented barriers to change in educational environments regarding advancing students in STEMM. These barriers have been well studied in STEMM teaching reform, which has many parallels to mentorship reform (Brownell and Tanner, 2012; NASEM, 2016, 2018a, 2018c, 2018d). Organizational change can be challenging, particularly with units composed of individuals who value their independence and consider themselves experts in many areas of their work. Here, the committee describes some of the most frequently reported barriers to change, contextualized for mentorship as they encountered them during the listening sessions conducted by the committee and through other venues, and provides some possible approaches for primarily engaging directly with individuals:
Barrier 1: The belief that mentorship is not a problem that needs to be addressed
Many mentors hold views about the effectiveness of their mentorship that are more positive than reality (Kezar, 2018), although some may be more self-critical and reflective. This is where scholarship can be effective in creating a larger conversation on assumptions about mentorship, using research to challenge misconceptions and establishing how to implement new policies or processes to improve STEMM mentee outcomes. Discus-
sions about mentorship education, mentoring tools, and effective mentor behaviors may provide opportunities for improvement that mentors had not yet considered. This report, particularly the material on identity development in Chapter 3 and on effective mentorship behaviors and mentorship education in Chapter 5, can be used to focus such conversations.
Barrier 2: A commitment to and comfort with traditional mentorship practices
Changes in beliefs and behaviors typically involve dissatisfaction with current practices and critique that the status quo may not work anymore for student needs in the 21st century (Gess-Newsome et al., 2003; Weaver et al., 2015; Wieman, 2017). However, many individuals do not desire to or know how to begin to work differently. Regarding higher education reforms, teaching offers an analogous situation in which many faculty care deeply about teaching, but making teaching more public by opening the classroom to systematic evaluation using multiple data sources (Reinholz et al., 2018) would require a marked change in approaches to teaching that recognize the value of evidence-based practice (NRC, 2012). Changing the norms of a department’s mentorship practices from “private practice” to mentorship that is open to review and improvement can be difficult in the face of resistance from individual faculty members and department heads, and institutional leaders may not be aware that there is a problem or who, for one reason or another, is not overtly in favor of inclusive or evidence-based practice.6 Mentorship education, which can be useful in these types of situations, is a solution that the committee explored further in the “Mentorship Education” section of Chapter 5.
Barrier 3: The tendency to place the sole responsibility on the mentee for their mentorship experience
Cases of poor mentorship or negative mentoring experiences are sometimes attributed to the characteristics of the mentee rather than the inadequacies of aspects of the relationship or neglect by a mentor, even though inadequate mentoring and negative mentoring experiences have been repeatedly documented (Eby et al., 2000; Kram, 1985a; Scandura, 1998; Simon and Eby, 2003).7 Although there are no systematic studies in postsecondary STEMM contexts, anecdotal reports indicate the occurrence of negative mentoring experiences may be common. When the quality of mentorship affects the professional development of students, and especially if it involves incidents of discrimination, bias, or harassment, institutions are responsible for addressing the problem on behalf of students and implementing processes at a program level to prevent abuse, neglect, and exclusion (NASEM, 2018d). Having a well-publicized process in place to
6 Faculty culture is also typically collegial and consensus based. As a result, faculty are often unwilling to broadly adopt new practices if specific faculty or a group of faculty oppose them.
address the quality of mentorship signals the institution is proactively averting potential problems. It also represents an important step in expanding notions of institutional responsibility for talent development and student progress.
Barrier 4: The lack of commitment to support implementation of effective mentorship
Research has shown that even effective interventions face barriers to widespread dissemination and implementation, including lack of time, resources, rewards, expertise, and confidence to implement the interventions (AAAS, 2011; D’Avanzo, 2013; Henderson and Dancy, 2007; Henderson et al., 2010; Hutchinson and Huberman, 1994). Despite the research available on how to address these barriers, there is a gap in knowledge about what supports from different levels—individual, programmatic or departmental, and institutional—promote follow-through in implementing innovations in local contexts, and how these supports may be effectively based upon characteristics of the individuals involved. Dissemination and implementation research indicates that contextual factors—including policies, infrastructure, procedures, leadership, interpersonal relationships, and climate—play a key role in supporting others in their implementation process (Brownson et al., 2012). As understood through ecological systems theory, characteristics of individuals—career stage, appointment type, disciplines, gender, race, and ethnicity—may play an important part in interacting with local and distant environments to determine implementation outcomes.8
Approaches to Organizational Change
While there are multiple models of organizational change in higher education and in STEMM learning contexts (Corbo et al., 2016; Gehrke and Kezar, 2018; Henderson et al., 2010; Kezar, 2018; Kezar et al., 2018; Prochaska et al., 2001), thinking of institutions as dynamic learning organizations can help participants foster change using a process that begins with research to assess institutional performance in light of existing practice and results in implementation of evidence-based approaches.9 This process recognizes
9 The committee employed an organization learning lens as a particular change perspective well suited to academic institutions because a learning organization is “an organization skilled at creating, acquiring, and transferring knowledge, and at modifying its behavior to reflect new knowledge and insights” (Garvin, 1993, p. 80). This often involves the use of research or data to change mindsets and behaviors, as well as a focus on changing, rather than preserving, underlying structures or practices. Within this framework, key considerations include both the specific knowledge acquired by various stakeholders as well as how new knowledge becomes embedded in the organizational structures and strategies (Dee and Leišytė, 2016; Garvin, 1993), such as implementing evidence-based practices and regular departmental mentorship education for both faculty mentors and students.
that expanding and diffusing knowledge about mentorship practices can improve institutional performance in graduating more diverse and highly skilled students and successfully placing them in STEMM career pathways.
Most organizational learning initiatives begin with research or data intended to assess institutional performance, help change beliefs, and create the impetus for the adoption of evidence-based practices. Such initiatives are often facilitated by external pressures for improving student success, but changing beliefs does not automatically result in changing behavior. Research has identified key activities that are actively managed by institutions adept at translating new knowledge into innovative ways of behaving (Dill, 1999). These activities can include exploring new knowledge through problem-solving, learning from one’s own and from others’ experiences, experimenting with new processes, and transferring knowledge among actors in units and subunits within the organization.
Institutions often look to similar or peer institutions for solutions to problems and adopt practices from different contexts to improve their performance (DiMaggio and Powell, 1983). With this in mind, organizational learning can translate into efforts to change mentor behavior. Specifically, such efforts may include building knowledge among mentors by providing mentorship education, creating buy-in among faculty and staff, and supporting them in implementing effective mentoring practices by making tools such as templates for mentoring compacts and individual development plans and supports such as coaching and feedback available. Efforts to sustain change might include accountability mechanisms that build mentorship evaluation into annual review, tenure, and promotion decisions. Institutional performance, assessed over time, allows for the development of an understanding whether an implemented change has demonstrable effects in improving outcomes and/or the lived experiences of undergraduate and graduate students.
FACILITATING CHANGE FOR MENTORSHIP
Organizational changes are facilitated by institutional change agents—primarily university leaders at various levels and the faculty and staff working directly with students. One strategy for achieving change in academia has been for institutions to create groups or teams to develop solutions or to foster connections among faculty in disciplinary or interdisciplinary professional learning communities on mentorship on and off campus (Bauman, 2005; Gehrke and Kezar, 2018; Kezar, 2018; Kezar et al., 2018).10 This approach has supported individuals willing to experiment and share successes at the department level and has been well documented as a successful strategy in encouraging the use of evidence-based practice in science, technology, engineering, and mathematics (STEM) (Borrego and Henderson, 2014; Cox, 2004). It is also employed by the National Research Mentoring Network to increase implementation of evidence-based mentorship education (Spencer et al., 2018).
10 Professional learning communities are also sometimes referred to as “communities of practice.”
Culture change in general is rarely easy (Haizlip et al., 2012), in large part because it entails a vision-driven process that advances through successive stages11 and accounts for a variety of different levels of institutional perspectives.12 Change also requires continued support from organizational leaders as the new culture becomes institutionalized. Moreover, culture change in academia presupposes that a common vision is shared among institutional leadership, deans, department chairs, faculty, staff, and students (Henderson et al., 2010). One sign of the difficulty in achieving lasting culture change in academia is found in the numerous National Academies reports that repeatedly call for culture change in academia to better support undergraduate and graduate student success in STEMM (NASEM, 2018a, 2018c, 2018d, 2019; NAS-NAE-IOM, 2007).
Each stakeholder group in the mentorship ecosystem can take actions to create the changes needed to improve mentorship in STEMM and its outcomes, and each has opportunities to leverage their position to affect institutional change. The remainder of this chapter outlines possible actions and opportunities for five participant groups: university leaders; department chairs; research, training, and graduate program directors; faculty mentors;13 and undergraduate and graduate students. Each participant group is provided with a set of potential actions (see Boxes 7-2, 7-3, 7-4, 7-5, and 7-6), many of which were offered during the committee’s listening session activities. In addition, the committee discusses potential mechanisms for change that can be facilitated by funding agencies and disciplinary associations and organizations. However, evidence of the outcome for each actor is often lacking, particularly for interventions at the leadership level in STEMM. Where it is possible, the committee builds on what is known from other domains of scholarship.
As Chapter 4 noted, some academic institutions have created cultures that support and value mentorship in alignment with the findings and recommendations made in
11 For example, organizational research points to eight stages of transformation to achieve organizational culture change: (1) establish a sense of urgency, (2) form a powerful guiding coalition, (3) create a vision, (4) communicate the vision, (5) empower others to act on the vision, (6) plan for and create short-term wins, (7) consolidate improvements and produce more change, and (8) institutionalize new approaches (Kotter, 1995).
12 For example, scholarship from the field of physics education research developed a six-fold change perspective: (1) scientific management, (2) evolutionary, (3) social cognition, (4) cultural, (5) political, and (6) institutional (Corbo et al., 2016).
13 Faculty are not the only members of a campus community who can serve as mentors. However, nearly two-thirds of individuals who identified as having a mentor as undergraduate students categorized them as a “professor.” This varied by population: it held true for more White students (72 percent) than underrepresented students (47 percent) and more continuing-generation students (67 percent) than first-generation students (61 percent) (Gallup, 2018).
this report.14 However, colleges and universities that institute policies to support effective mentorship in STEMM remain the exception rather than the norm. Institutional policies and practices are among some of the stronger determinants for implementing effective mentorship programs because they signal to internal and external constituents that quality mentorship and its outcomes are valued in the academic workplace.
Evidence regarding institutional processes that effectively support mentorship come primarily from the literature on mentorship in business settings. This literature contains extensive research on institutional and administrative factors that increase the likelihood that organizations can implement and sustain effective mentorship programs (Hegstad and Wentling, 2005). While it is true that mentorship outcomes can vary by setting (Eby et al., 2008), overall findings across settings support the supposition that universities are organizations with employees (including faculty) and that there are important lessons to learn from similar organizational settings that systematically employ mentorship in professional development.
At the institutional level, a commitment from leadership can have a profound effect on the quality of mentorship and ultimately the development of undergraduate and graduate students (Scandura et al., 1996). Research from the organizational perspective shows the critical role institutional leaders play in creating and sustaining cultural change (Gelfand et al., 2007; Jayne and Dipboye, 2004; Kozlowski and Doherty, 1989; Ostroff et al., 2013; Stamarski and Son Hing, 2015; Taylor et al., 2011). For example, university leadership could emphasize that a culture of teamwork, trust, and successful mentorship are among the cornerstones of successful institutions (Allen and Poteet, 1999; Kirchmeyer, 2005).
However, merely communicating the value of mentorship will produce limited organizational change unless institutional accountability mechanisms align with statements about the value of mentorship in an institution’s overall efforts. For example, the University of Maryland Baltimore County has been recognized as a national leader of university-wide inclusiveness initiatives that have mentorship elements, sometimes described collectively as “university as mentor” (Bass et al., 2007). Evaluation of the university’s programs, such as the Meyerhoff Scholars Program, has shown positive effects on the retention and success of STEM students of color (Maton et al., 2000, 2012, 2016; Mervis, 2019; Santo Domingo et al., 2019).15
Rewards and Accountability
Research on work motivation and its relationship to desired employee behaviors suggests that employees must understand what factors matter in performance appraisal and rewards systems to be sufficiently motivated to change their work behavior (Kerr, 1995; Raymond and Kannan, 2014; Van Eerde and Thierry, 1996). Although some institutions have implemented awards for quality mentorship as a means of recognizing and placing value on effective mentorship,16 systems that highlight and reward exceptional mentorship often do little to communicate criteria for effective mentorship or to support effective mentorship by faculty who are not awardees.
Indeed, studies of accredited colleges of business have shown that when institutions closely align their performance appraisal and promotion and tenure guidelines with their emphasis on mentorship,17 faculty and staff are more likely to view mentorship as a serious commitment requiring that they allocate time to mentorship activities and continuously develop their mentorship skills (Raymond and Kannan, 2014). Some institutions require faculty to report on the progress of their students and their eventual employment in the workforce, but as the committee’s listening sessions with faculty showed, it is unclear how many institutions use such information in performance reviews.18 Furthermore, processes that require faculty to provide the number of undergraduate, graduate, and postdoctoral researchers they are currently supervising do not effectively incentivize or even measure the quality of mentorship processes and outcomes in these relationships (Raymond and Kannan, 2014; Thomas et al., 2007). In other words, most institutions could revise their promotion and tenure and performance appraisal guidelines to not only track the number of students a faculty member mentors, but also track key indicators of effective mentorship. Such indicators could include whether the mentored scientists are coauthors on manuscripts and grants and their placement into positions, as well as process measures that assess mentoring relationship quality from the perspective of the mentee and the mentor (Scandura et al., 1996).
Institutional commitment to mentorship can only translate into meaningful results if the ratio of mentors to undergraduate and graduate students is reasonable and if mentors can allocate meaningful increments of time to mentorship activities. Thus, institutions may also want to consider the mentor-to-mentee ratio at the college and
16 For example, the University of Houston Undergraduate Research Mentor Award (see https://www.uh.edu/provost/faculty/current/awards/mentoring/), the University of Georgia Graduate School Outstanding Mentoring Award (see http://grad.uga.edu/index.php/current-students/financial-information/graduate-school-recognition-awards/outstanding-mentoring-award/), the North Carolina State University Graduate School’s Outstanding Graduate Faculty Mentor Award (see https://grad.ncsu.edu/research/mentor-award/), or the Virginia Tech Outstanding Mentor Award (see https://graduateschool.vt.edu/about/awards/outstanding-mentor-award.html); accessed August 17, 2019.
17 See the “Department-Level Change” section in this chapter for more specific suggestions.
university level, while acknowledging that some mentors can handle a different mentorship load than others (Pulsford et al., 2002).19 In addition, it is important to recognize that mentees engage in mentorship with staff and other members of the campus community. Therefore, the institutional commitments to mentorship will have to include staff and other members of the campus community. Institutions might reflect on their overall activities to support mentorship by recognizing and measuring all forms of mentorship, including informal and formal relationships that occur beyond the research advisor and student.20
Recruitment, Hiring, and Onboarding
One way for improving mentorship in academia would be hiring individuals with a commitment to mentorship or evidence of prior success as a formal or informal mentor. Similar to current practices for diversity and inclusion, postings for faculty positions could highlight how institutions view mentorship as a key component of faculty job performance, and applicants might indicate their previous or intended contributions to the mentorship and development of diverse undergraduate and graduate students.
Onboarding processes for new faculty and staff at many institutions include multiday orientation sessions that entail training on critical processes, procedures, and organization goals. However, as noted in the committee’s listening sessions and at the public workshops, many institutions do not stress that effective mentorship for undergraduates and graduate students is a high-value priority that aligns with key institutional goals during the onboarding process or during orientation.21 Systematic mentorship education is rarely a component of onboarding processes, despite evidence suggesting that well-trained mentors can affect undergraduates and graduate students’ perceptions very positively (Raymond and Kannan, 2014).22 Additionally, research evidence lends support to the notion that, for mentorship education to be effective, it does not have to be long and time-consuming (Allen et al., 2006). Therefore, this type of institutional commitment to mentorship does not place a large additional burden on incoming faculty and staff. At the same time, undergraduate and graduate student orientation programs do not discuss frequently enough how successful mentoring relationships can be created, cultivated, and nurtured so that they benefit students and their mentors (Packard, 2003a).
19 The various mentoring structures that can help address a high mentee-to-mentor ratio are discussed in the “Non-Dyadic or Multiple-Mentor Mentorship” and “Online or E-Mentorship” sections in Chapter 4.
Ethical issues may arise in mentoring relationships, and institutions are responsible for ensuring student educational progress (Anderson and Shore, 2008; Johnson, 2017; McDonald and Hite, 2005; Schlosser and Foley, 2008). Compliance with federal laws on discrimination and sexual harassment requires institutions to establish processes for reporting and handling cases of conflict or behaviors of ill intent with neutral parties or ombudspersons (NASEM, 2018d). In most cases, though, only informal processes exist for those involving negative mentorship experiences or mentor-mentee conflict. Because poor outcomes for mentees are associated with negative mentoring experiences (Eby and Allen, 2002; Eby et al., 2010),23 processes for confidential intervention and resolution would place the mentee in a more optimal context for learning and development. Possible actions for university leadership are listed in Box 7-2.
Many faculty and students identify at least as strongly with their department as with their institution. As was stated in the National Academies report Graduate STEM Education for the 21st Century, “The department is the primary organizational unit on a campus. It serves as the primary affiliation for most faculty and students, serving as a key connection to a [student’s] identity within his or her field of research [or discipline]” (NASEM, 2018c, p. 134). Thus, departments and department chairs can serve as key drivers of change at institutions of higher education. It is through departments that most institutional policies are operationalized, including promotion, tenure, and reward systems, as well as academic oversight. Thus, departments and department chairs will be important catalysts in developing a culture of inclusive excellence through mentorship.
Department chairs, whose function at the university is equivalent to first-line supervisors in many organizational and business settings, can serve as critical levers in the mentorship process. They can receive information about how mentorship practices can be taught and improved and about the roles they can play in developing their faculty and staff not only as STEMM professionals but also as mentors of the next generation of STEMM professionals. For example, if mentorship in performance appraisal and promotion and tenure processes and guidelines are to affect outcomes, department chairs may want to provide junior and senior faculty, as well as staff, with the knowledge that their involvement in development and learning activities such as mentorship will result in positive outcomes for their mentees (Aryee et al., 1996). Department chairs will likely also want to provide mentors with feedback and tools to monitor and upgrade their mentorship skills.24
Research suggests that whole department adoption is a highly effective way of embedding reforms central to the teaching and learning mission (Wieman, 2017). A recent report from the American Astronomical Society Task Force on Diversity and Inclusion in Astronomy recommended that efforts to catalyze organizational change to improve student retention should focus on departmental practices, including mentorship (Rudolf et al., 2018). The report makes recommendations for departments striving to provide effective mentorship and expand networking opportunities. Specifically, the report makes several recommendations at the department level, including providing or requiring mentor education for faculty and other parties involved in mentoring, and providing mentee training to help mentees be more proactive in their mentoring relationships.
Department chairs are also well positioned to implement changes in the reward structure for faculty and staff. Effective mentorship can be documented along with evidence of influence of faculty work in teaching, research, and service.25 On many campuses, mentorship is included in teaching effectiveness, documented in terms of advisee and student progress and even student placement in postdegree pathways. However, attendance at mentor training workshops and evaluative work can also provide further evidence of efforts to improve mentorship effectiveness. For example, many campuses solicit letters from students at key promotion points for faculty, but any evaluative metrics can be useful in the review process. Additionally, prestigious mentorship awards such as the Presidential Awards for Excellence in Science, Mathematics and Engineering Mentoring, American Association for the Advancement of Science Mentor Award,26 and others can make a convincing case for national mentorship recognition. In research, the common ways that mentorship is recorded are indicated by coauthorship with undergraduate and graduate students on a mentor’s curriculum vitae in promotion and merit considerations. Research statements can also include information about how the mentor extends research opportunities and outcomes. Finally, many faculty are engaged in mentor training of colleagues, which becomes a vital service for improving the quality of mentorship in departments and programs.
Department chairs can also collaborate with faculty and departmental staff to create an “error management” culture around mentorship so that when mistakes occur, they are shared openly as opportunities to improve policies, processes, and outcomes for everyone involved in mentorship activities (Keith and Frese, 2008). A department that develops an organizational error-management culture will value mistakes and construe them as opportunities to gain and improve departmental and individual outcomes (van Dyck et al., 2005). To this end, chairs must create opportunities to learn from current and new practices and reflect on results (Bauman, 2005; Dill, 1999).27 Effective error-management cultures stand in contrast to systems where mentorship quality, process, and outcomes are assessed, but the resulting data are utilized punitively rather than with the goal of developing better mentors and better mentorship processes and outcomes. Possible actions for department chairs are listed in Box 7-3.
25 For example, as part of a promotion or tenure package.
26 More information about the Presidential Awards for Excellence in Science, Mathematics and Engineering Mentoring is available at https://www.nsf.gov/awards/PAESMEM/. More information about the AAAS Mentor Awards is available at https://www.aaas.org/archives/mentor. Accessed on August 3, 2019. A summary of the 2016 and 2017 STEM Mentors Alumni Meetings provides a reflection of the insights of several recipients on effective mentoring practices (AAAS, 2018).
Research, Training, and Graduate Program Directors
Faculty and staff can have leadership roles as the directors of student programming focused on research, training, or graduate education. Many of the research and training programs, often funded partially or entirely by external funding agencies, have mentorship as a core element. Programs that involve research with faculty or comprehensive student support either implement some mentorship activity or assume that mentorship will occur. The directors of these programs often enable or persuade colleagues to take part in the program and, when placements or interactions between faculty and students do not work out, they are faced with developing a solution that will help students while maintaining their own relationships with colleagues.
Program directors can take steps to prevent or mitigate these problems by (1) ensuring there are guidelines that clarify expectations of mentors and mentees, (2) informing participants about regular assessment activity as part of program requirements, (3) establishing activities that incentivize good working alliances, and (4) including mentorship education as an expectation for participants. Program directors can also regularly provide midlevel administrators, such as deans and department chairs, with program information, including information about mentoring metrics, to establish the program as vital to training at the institution and an exemplar for mentorship beyond the program.
It is essential that departments continuously provide faculty with information on how they can best recruit, mentor, and contribute to the success of diverse undergraduate and graduate students in their respective research groups (Johnson‐Bailey and Cervero,
2004; Thomas et al., 2007). This is particularly critical for mentees from underrepresented (UR) groups28 and first-generation mentees, populations that are less likely to have a professor as a mentor during their undergraduate experiences (Gallup, 2018). While evidence of the challenges of identifying, recruiting, developing, and supporting diverse undergraduates and graduate students is often discussed in social science communities (Bauman, 2005), such evidence is less often part of the dialog in STEMM disciplines, making progress more difficult.
Another action worthy of testing is for graduate program directors to provide incentives for groups of faculty to function as mentorship teams.29 Such a step could hasten the transition from a system where one principal investigator mentors and supervises undergraduate and graduate students to a group approach, which could increase the likelihood that several mentors’ skill sets would meet the mentorship needs of a mentee. Such practices not only limit the likelihood of abuse (Johnson and Nelson, 1999), but help undergraduates and graduate students grow through exposure to the mental models, methodological approaches, and other attributes of multiple mentors from several disciplines. Program directors can also pay attention to the stages of mentorship and ensure that the evolving needs of undergraduate and graduate students are met as they move toward increasing independence.30 Possible actions for research, training, and graduate program directors are listed in Box 7-4.
Faculty can have tremendous influence on the culture of mentoring through their own practice—by what they implement, role model, and value in their research teams and in what they support and promote within their programs and departments. Institutional change can begin with a faculty innovator or group of faculty change agents who lead the organization either from a position of authority or at the grassroots level, with a longer-term intention of influencing others that have direct contact with students (Kezar and Lester, 2009).
In addition, any institutional change toward a culture of mentorship that fails to recognize the needs of the faculty and focuses solely on the needs of the students will not be successful. Faculty can be integral in advocating for specific policies or programs that maintain the relational nature of mentorship. Many institutional levers for change entail faculty approving and implementing new policies and adopting new practices that become a part of their daily work in developing STEMM talent. Moreover, faculty who
28 This report refers to UR groups as including women of all racial/ethnic groups and individuals specifically identifying as Black, Latinx, and American Indian/Alaska Native. Where possible, the report specifies if the UR groups to which the text refers are Black, Latinx, or of American Indian/Alaska Native heritage.
have developed successful practices often find these practices begin to be more broadly adopted across peer institutions, as institutions have a tendency to become more alike rather than dissimilar over time (DiMaggio and Powell, 1983).
While institutional or department-level rewards and incentives will likely be crucial motivators for faculty in dedicating time toward improving their mentorship, there are also several potential intrinsic or implicit motivators for individual faculty members to consider. These include improvements in the overall efficiency of their mentees and productivity of the research teams, socioemotional growth, improved teaching and communication skills, improved clarity around personal-professional boundaries, and the development of the next generation of STEMM professionals.
As they reflect on their own experiences, UR faculty may put additional time into mentoring UR students, offer the mentoring they may have desired in their own professional development, and seek to supplement student needs and fill gaps that are not being fully addressed by the institution. Additionally, UR faculty may be asked disproportionately to mentor by UR mentees who perceive them as more effective than non-race- or non-gender-matched mentors or by colleagues who think that UR faculty are better able to mentor UR mentees.31 While mentorship is beneficial for students—especially
undergraduate students—UR faculty who mentor may not reap benefits for the service they render within their academic departments.
One study, for example, found that UR faculty pay a high emotional, financial, and professional price that outweighs much of the altruistic satisfaction received from helping students (Schwartz, 2012). Faculty members said that helping UR students navigate the new academic culture, helping them through personal or family problems, or finding resources to enable them to stay in college took a toll on them emotionally. Another study found that mentoring took a toll on Black faculty members’ personal and family life because it meant dedicating more hours to tasks that faculty members could complete faster without student help (Hunter et al., 2007). Investigators have described the taxing nature of the lack of time and financial resources needed to fund undergraduate research (Lei and Chuang, 2009). The financial costs of mentoring incurred by UR faculty include devoting extra time beyond the workweek to mentoring and using their own income to help fund undergraduate research.
Professional costs arose from spending 10–16 hours (or more) per week doing undergraduate research mentoring that was not valued by the university and that took time away from teaching and publications, which could have severe ramifications for the faculty member’s career (Schwartz, 2012). In fact, studies show that academic reward systems do not value campus service activities such as mentoring (Acker and Armenti, 2004; Clark and Corcoran, 1986), and some faculty have reported that their institutions implicitly or explicitly discouraged faculty from devoting much time to service and mentoring by not acknowledging such efforts in promotion and tenure decisions and not allowing relief from clinical, administration, or teaching activities to allow time for mentoring (Gandhi and Johnson, 2016). Research indicates that women faculty feel particularly pressured by the demands of service, mentoring, and teaching, while men faculty were more protective of their research time (Misra et al., 2011), though the way they protect that time was unclear. One study found that women viewed service, including mentoring, primarily as a burden, and even though they recognized that it would not benefit their tenure packages, women still volunteered for service because they saw it as vital to sustaining diversity (Misra et al., 2011).
In contrast, another study documented UR faculty mentors’ narratives on the benefits of working with high-performing graduate students in terms of sharing their work with the world, collaborating, learning, and “loving it” (Lechuga, 2011). Further, benefits accruing to mentor and mentee in effective relationships between faculty include higher rates of presentation, publication, and support in promotion and tenure (Tillman, 2001). Some UR faculty have reported that their departments and universities encourage mentoring by providing protected time for mentoring, offering mentoring awards, and establishing mentoring as valuable in promotion decisions. Many UR faculty also report that they participate in mentorship activities through professional
organizations that value social identity and that welcome and include UR mentors at multiple career stages.32
Even UR faculty who choose to spend time mentoring UR students as part of their own mission as change agents for diversity may find that the time they take to mentor UR students goes unrewarded by their institution. When UR faculty provide service such as mentorship, and when they are sought out for that service because of their identity, even by students outside of their own departments, there is “identity taxation” or a “cultural tax”—the extra time and effort spent on the needed service but not spent on other activities such as research that may lead to promotion. In this case, UR faculty may feel a conflict between needing to attend to the usual tenure-track duties related to research, teaching, and services, and wanting to assist in the mission of increasing diversity. The Strategies and Tactics for Recruiting to Improve Diversity and Excellence project at the University of Michigan, part of the National Science Foundation’s (NSF) Organizational Change for Gender Equity in STEM Academic Professions program for advancing women faculty, addresses the cultural tax, among other conflicts, by appointing tenured faculty from well-represented groups to bring attention to issues affecting institutional diversity and inclusion (Baez, 2000; Diggs et al., 2009; Hirshfield and Joseph, 2012; Reed et al., 2018; Sturm, 2006).
Possible actions for faculty are listed in Box 7-5.
Undergraduate and Graduate Students
Graduate and undergraduate students can significantly affect the culture of mentorship, both in their individual mentoring relationships and in their departments and educational programs (Lunsford and Baker, 2016). Evidence suggests that students can actively engage in “mentoring up,” an approach that helps mentees gain the knowledge and confidence to take equal responsibility with mentors for developing effective mentoring relationships (Lee et al., 2015, chap. 7).33 Students can also learn to become more effective in their relationships by participating in mentorship education to advance their skills and confidence in being effective mentees while translating those skills into relationships with more junior colleagues they begin to mentor. As members of departments and programs, they can influence change by discussing their mentoring experiences and the criteria they use when choosing mentors and by providing honest feedback about their mentorship experiences in the department or program.
Students have the opportunity to benefit substantially from the developing research on mentorship, both as mentees and as mentors to others. When the number of faculty
32 For example, the National Society of Black Engineers, the Society for Advancement of Chicanos/Hispanics and Native Americans in Science, the Society of Hispanic Professional Engineers, and the American Indian Science and Engineering Society.
or diversity of faculty is not sufficient, graduate students and postdoctoral researchers play a critical role in interacting with undergraduate students. For example, they can help undergraduate students build skills and learn about graduate education (Aikens et al., 2016; Dolan and Johnson, 2010) and they benefit themselves by mentoring these students (Dolan and Johnson, 2009; Limeri et al., 2019).34 Many intervention programs also involve undergraduate students as peer or near-peer mentors. Research on biomedical students who reported receiving advice from juniors or seniors found that those students adjusted better to academic life and had a heightened sense of belonging in their freshman year (Hurtado et al., 2007).35 Possible actions for students are listed in Box 7-6.
Funding agencies can play a powerful role in advancing cultural change by proactively encouraging or even requiring institutions to systematically develop undergraduate and graduate students, and especially students from historically UR backgrounds to diversify
35 While this study did not focus on mentorship per se, it illustrates the benefits of interacting with more advanced peers.
the U.S. science workforce (Hrabowski and Henderson, 2019). There are several examples of funding agencies already engaging in the realm of mentorship, particularly in the use of mentoring tools.36 For example, the NSF requires a mentorship plan, including academic and professional development activities, to be provided to all postdoctoral researchers supported by an NSF-funded project (NSF, 2019). In 2014, the National Institutes of Health (NIH) began to require individual development plans for all NIH-funded graduate students and postdoctoral researchers as a means of providing a structure for identifying and achieving their career goals (NIH, 2014). More recently, the National Institute of General Medical Sciences (NIGMS) began requiring mentor preparation for all mentors of trainees on NIH T32 grants.37
37 NIH T32 grants include the Ruth L. Kirschstein National Research Service Award (NRSA) Predoctoral Institutional Research Training Grants, Medical Scientist Training Program, Initiative for Maximizing Student Development (IMSD), Graduate Research Training Initiative for Student Enhancement (G-RISE), Institutional Translational Research Training Program, and Training Program for Institutions That Promote
Some funding agencies require mentorship plans that include explicit mentoring by multiple researchers to ensure they have access to a broad set of technical skills and experiences, along with exposure to essential role models from diverse backgrounds (Campbell and Campbell, 2007).38 Despite the increased emphasis on multimentor approaches favored by funding agencies,39 institutional obstacles to executing these plans or to providing and encouraging access to more than one mentor are still deeply ingrained in the culture of many academic departments, colleges, and institutions, which emphasizes the primacy of the apprenticeship model of graduate education (de Janasz and Sullivan, 2004).40
The Howard Hughes Medical Institute Gilliam Fellowships for Advanced Study for graduate students represents one example of how a funding organization is emphasizing and supporting mentorship through program requirements. The Gilliam Fellowships program now requires its mentors to engage in a year-long mentor education program based upon Entering Mentoring and culturally responsive mentor education (HHMI, 2019).41 The Alfred P. Sloan Foundation’s approach involves supporting nine campus-based University Centers of Exemplary Mentoring. These centers provide scholarships, faculty and peer mentoring, professional development activities, seminars, and other resources that promote completing graduate study (APSF, 2019).
There are growing examples of empirically guided institutional initiatives to support culturally responsive mentorship, including the National Research Mentoring Network and the Building Infrastructure Leading to Diversity programs, both sponsored by the NIH Diversity Program Consortium. Emerging evidence from one program that serves UR STEMM students has documented the positive effect of campus partners supporting faculty engagement in providing research training environments that affirm UR students’ cultural and science identities and their sense of belonging (Estrada et al., 2017).
Diversity. More information is available at https://grants.nih.gov/grants/guide/pa-files/PAR-19-228.html; accessed May 9, 2019.
38 For example, NIH Career Development (K) awards.
40 Although the reasons for limiting access to multiple mentors are multifaceted, key factors include individual research mentors’ attitudes and belief systems regarding mentorship (Johnson and Huwe, 2002). Many faculty continue to believe that the individuals they mentor should contribute exclusively to their own research productivity and that the student should not spend time working on projects associated with a secondary mentor. In other cases, faculty believe that working with multiple mentors on research will result in breadth rather than depth of training. At the same time, pressure to publish based on grant funding is perceived as high by most investigators, and this pressure often drives research mentors to expect scientists-in-training in their groups to dedicate themselves to the groups’ objectives entirely (Johnson and Nelson, 1999). Some have reported working on the lab’s research for some 70 hours a week, leaving students with limited time to pursue other interests (Mason et al., 2009) or develop mentorship or collaborative relationships outside their research advisor’s research group. Additional discussion of multiple mentorship structures is presented in Chapter 4.
Although funding agencies have successfully created programs to support undergraduate and graduate students from UR backgrounds with the explicit goal of transitioning them to faculty roles (Fleming et al., 2012), accountability mechanisms that require institutions to emphasize effective mentorship and expect principal investigators to mentor successfully are still not as prevalent as they could be. For instance, even though some grant programs require descriptions of mentorship plans, funding agencies have issued limited recommendations for process and outcome measures that can be used to evaluate mentorship progress within a grant-funded project. Furthermore, there are no apparent processes in place for determining whether and how well principal investigators have implemented supportive mentorship activities, particularly for undergraduates and graduate students from underrepresented backgrounds. With exceptions such as NIH T32 pre- and postdoctoral training grants, institutions and principal investigators applying for funding are rarely required to include documentation on the diversity of those involved in mentoring relationships or present evidence about the effectiveness of their mentorship activities. More generally, many funding mechanisms do not routinely require applicants or their institutions to describe their mentorship systems, including the systems that incentivize and reward effective mentorship or the processes in place to support and evaluate culturally responsive mentorship. As noted above, NIGMS has recently implemented a requirement for mentorship education and evaluation in NIGMS training grants as one way to engender a noticeable shift toward more effective practices. How this new requirement affects mentoring programs remains to be seen.
Disciplinary Association Support
STEMM disciplinary associations and organizations have been catalysts for supporting and empowering faculty in education reform, often offering opportunities for faculty to showcase innovations and learn from peers and providing venues for discussion of mentorship research and interventions. They also provide mentoring experience both through standalone programs and through affiliation with conferences and other gatherings, including the following:
- STEMM-focused professional societies, for example, the American Physical Society,42 the American Astronomical Society,43 and the American Chemical Society44
42 For example, case studies in mentorship. More information is available at https://www.aps.org/programs/education/ethics/mentoring/; accessed May 9, 2019.
43 For example, a task force and report focused on diversity and inclusion in graduate education. More information is available at https://aas.org/education/aas-task-force-diversity-and-inclusion-graduate-astronomy-education; accessed May 9, 2019.
44 For example, New Faculty workshops. More information is available at https://www.acs.org/content/acs/en/education/educators/coursesworkshops/csc-new-faculty-workshop.html; accessed May 9, 2019.
- Education-focused societies and organizations, for example, the American Society for Engineering Education,45 the Society for the Advancement of Biology Education Research,46 and the Center for the Integration of Research, Teaching, and Learning47
- National initiatives dedicated to helping faculty and their institutions implement change, for example, Project Kaleidoscope of the American Association of Colleges and Universities,48 and the Association of Public Land-Grant Universities49
These organizations support professional learning communities that value mentorship and can extend the knowledge base about and implementation of effective practices.
49 For example, degree completion initiatives (see http://www.aplu.org/projects-and-initiatives/center-forpublic-university-transformation/), access and diversity initiatives (see http://www.aplu.org/projects-andinitiatives/access-and-diversity/), and STEM education initiatives (see http://www.aplu.org/projects-andinitiatives/stem-education/); accessed May 9, 2019.
This page intentionally left blank.