Abstract
The growing application of social network-based theories and methods (Burt et al., 2013) in scholarship on mentoring illustrates that mentoring goes beyond dyadic relationships comprising a senior mentor and a junior protégé (Higgins & Kram, 2001). However, limited data exist on the state of developmental networks of university faculty. This study examines developmental network characteristics among mentors and mentees participating in an ongoing intervention that aims to enhance career success through improved mentoring. Cross-sectional data come from 81 faculty mentors and mentees at three universities in the Southwestern United States. Using the online Modified Mentoring Network Questionnaire (MNQ), participants provided information on relationships with developers, who are people that have taken concerted action, and provided professional and/or personal guidance to help participants advance in their careers. An individual’s developmental network comprises relationships with developers. We conducted exploratory analyses examining key characteristics of mentors’ and mentees’ developmental networks.
Participants received psychosocial and career support from an average of 4.9 developers (4.8 and 5.1 for mentors and mentees respectively) from 2.3 arenas (2.2 and 2.4 arenas for mentors and mentees, respectively). While the most common arena was the respondents’ current job/position (62%, 64% and 59% for all participants, mentors, and mentees respectively), developers were from graduate school (11%, 6% and 17%); prior jobs/positions (13%, 16% and 9%) and family (8%, 5% and 11%). Our preliminary findings suggest that developers are important for university faculty and that methods and insights from social network analysis can be applied to examine their support networks. As our study is part of an ongoing longitudinal intervention, these findings will inform future analyses that will examine changes in developmental network characteristics and its impact on participants’ careers.
Background
Research and theory on mentoring have looked beyond hierarchal, dyadic relationships to developmental networks, comprising “a set of people a protégé names as taking an active interest in and action to advance the protégé’s career by providing developmental assistance” (Higgins & Kram, 2001, p. 268). Paralleling this broadening scope is an increasing application of social network analysis (Burt et al., 2013) in the mentoring literature (Pifer & Baker 2013; Yip & Kram, 2017). Social network analysis has intellectual roots in diverse academic fields (e.g., sociology, mathematics, and theoretical physics) and encompasses an expanding set of quantitative methods that can be used to examine enduring social processes (Papachristos, 2011). From a theoretical standpoint, social network analysis emphasizes features of relationships over personal characteristics to explain individual and collective outcomes. The widespread integration of social network methods and techniques in research on mentoring has yielded novel insights into how mentoring operates among professionals today.
To date, limited data exist on the state of developmental networks of university faculty. As a result, little is known about whether developmental networks matter in the careers of faculty members. Thus, measuring the structure and content of developmental networks among faculty members is an important first step in assessing how career and psychosocial support shapes faculty career outcomes. Furthermore, such research can potentially inform research and interventions aimed at understanding and enhancing faculty retention and career success. This study examines developmental network characteristics among mentors and mentees participating in an ongoing intervention to enhance career success through improved mentoring among university faculty members. The remainder of the paper proceeds as follows. We first review our approach to measuring key features of developmental networks. We then present findings from the first wave of data collection from an ongoing intervention that aims to enhance mentoring among university faculty. We conclude by discussing the implications of these results and potential avenues for future research on mentoring.
Methods
Cross-sectional data come from 81 faculty mentors (N=44) and mentees (N=37) at three universities in the Southwestern United States: University of New Mexico (HSC and non-HSC campus), Oklahoma University HSC, and Arizona State University (non-HSC) who voluntarily enrolled in the study. Using the online Modified Mentoring Network Questionnaire (MNQ), participants provided information on relationships with developers, who are people that have taken concerted action, and offered professional and personal guidance to help participants advance in their careers over the past year. This questionnaire was a slightly modified version of the Developmental Network Questionnaire (Higgins, 2004). We merged data collected through the MNQ with data collected from a web-based baseline survey that gathered additional data on respondent characteristics (e.g., demographics, career satisfaction, and milestones). We conducted exploratory analyses examining key characteristics of mentors’ and mentees’ developmental networks among our respondents for this study.
The Modified Mentoring Network Questionnaire
The MNQ uses an egocentric method to collect data on focal respondents’ professional networks. Egocentric networks focus on individual respondents (i.e., the egos) and their set of relationships with others (i.e., the alters). The egocentric Developmental Networks that we examine comprise faculty members and their relationships with developers. The first section of the MNQ centers on name-eliciting and asks respondents to identify developers. Importantly, respondents were instructed to think broadly about the people who constitute their developmental networks, as developers can include people from current or prior employment settings, friends, family members, members of the broader community, or other social arenas. Developers potentially include individuals who occupy superior roles as well those who are peers and subordinates, as long as they help in the respondents’ careers by advancing their professional and/or personal development (e.g., helping negotiate work-life balance issues). Respondents entered the initials (or another brief identifier) for at least two and up to six developers. Respondents then answered a series of questions about individual developers and their relationships with developers. We used responses to these questions to measure key dimensions of our respondents’ developmental network described below. We first discuss two measures of support, which center on network content or the types of resources transferred through ties. We then discuss key measures of network structure that capture the strength of ties with developers, the overall level of interconnectedness among developers, and the larger social arenas from which developmental relationships originated.
Developmental Network Content: Career and psychosocial support. One main way in which developers enhance protégés’ careers is through the provision of support. Career support involves providing sponsorship, professional exposure, and protection against situations that challenge or jeopardize career objectives (Dobrow Riza & Higgins, 2019). Beyond career support, developers also provide varying degrees of psychosocial support, which involves friendship and care beyond the workplace and is more emotional and social than career support. Research demonstrates that developers vary in their provision of support (Dobrow Riza & Higgins, 2019; Kram, 1985). Furthermore, developers often specialize in the types of support they offer, with some developers offering more psychosocial support than career support, and vice versa (Higgins & Thomas, 2001).
The MNQ asks respondents about dimensions of career support provided by each developer. Respondents indicated the extent to which each developer: 1) “Provides you with opportunities that stretch you professionally,” 2) “Creates opportunities for visibility for you in your career,” 3) “Opens doors for you professionally,” 4) “Acts as a sponsor for you,” and 5) “Acts as a buffer for you from situations that could threaten your career achievement.” Responses ranged from 1 (never, not at all) to 7 (to the maximum extent possible). We measure career help provided by each developer by taking the sum of the five items (alpha = 0.851).
Respondents also specified the extent to which each developer provides psychosocial support in the following ways: 1) “Cares and shares in ways that extend beyond the requirements of work,” 2) “Counsels you on non-work-related issues,” 3) “Offers you support, respect, and encouragement,” 4) “Is a friend of yours,” and 5) “Confirms and affirms your identity and sense of self.” Responses ranged from 1 “never, not at all” to 7 “to the maximum extent possible.” We measure psychosocial help for each developer by taking the sum of the five items (alpha = 0.882). Notably, values of 20 or below are low career or psychosocial support and 21 and greater as high support (Higgins, 2004).
Developmental relationships can be differentiated according to high versus low levels of psychosocial and career help that developers provide. Developers who provide high amounts of both psychosocial and career support are categorized as “mentors,” while “allies” provide low levels of both career and psychosocial support. Developers who are “sponsors” provide high career assistance and low psychosocial support, and “friends” provide high psychosocial support and low career support.
Developmental Network Structure: Tie Strength, Size, and Density. We measured multiple dimensions of developmental network structure. The first two measures capture dimensions of tie strength with each developer, namely communication frequency and emotional closeness (Granovetter, 1973; Marsden & Campbell, 1984). Communication frequency was measured using responses to the question: “How often do you communicate with each person (including talking face-to-face, via email, over the phone, teleconferencing, or text messaging)?” Ordinal responses included: 1 = “Less than once a year,” 2 = “Once a year,” 3 = “a couple of times a year,” 4 = “once a month,” 5 = “once every 2 weeks,” 6 = “once a week,” 7 = “several times a week,” and 8 = “everyday.” Emotional closeness was measured by taking responses from the question: “How emotionally close are you to each person?” Responses were ordinal and included 1 = “distant,” 2 = “less than close,” 3 = “close,” and 4 = “very close.”
Two key network-level measures are size and density. Network size (or degree) is the count of developers in a respondent’s developmental network. Size can potentially range from 2 to 6. Density captures the overall level of cohesion among developers. Density is at its maximum when each developer knows every other developer and at its minimum when all developers are unfamiliar with one another. To measure network density, we first counted the number of dyads in the developmental network that know one another and divided that value by the number of unique dyads in the network. Network density can range from 0, indicating that no developers know one another, to 1, indicating that all the developers know one another.
Social Arena and Range. For each developer, respondents indicated the social arena or the primary way the respondent knows each developer. Social arenas could include the following categories: family; childhood friends; college friends; graduate school friends; graduate school professors; previous job or organization; current job or organization; volunteering; religious organization; and others (please specify). If a developer was a member of more than one social arena, respondents were instructed to pick the arena that best captures their relationship. We calculate the range of social arenas, which is a summary measure indicating the count of unique social arenas from which developers are known. The measure potentially ranges from 1 (all developers come from a single social arena) to 6 (a respondent has six developers, and they all come from different social arenas).
Respondent Characteristics
Apart from the data collected through the MNQ, respondents completed a baseline survey that collected information on respondent demographics and their current faculty position. We present data on respondent age, gender, race/ethnicity, and faculty rank and track.
Analysis
This study is part of an ongoing intervention and data collection effort. Given the absence of information on faculty members’ developmental networks, our study presents baseline characteristics of our respondents’ developmental networks, broken down by mentor and mentee status. The Human Research Protection Office at the University of New Mexico, which served as the Institutional Review Board of record, approved this research.
Results
Descriptive statistics of our sample broken down by mentor and mentee status are displayed in Table 1. Among respondents who are mentors, nearly two-thirds (65.91%) are female, and an equal proportion (65.91%) are non-Hispanic Whites. The mean age of mentors is 50.00 years. Nearly four-fifth are senior faculty, i.e., professors (32.56%) and associate professors (46.51%). More than half (52.27%) of mentors are tenure track faculty.
Table 1.
Descriptive Statistics
| Mentors (n=44) | Mentees (n=37) | |||
|---|---|---|---|---|
|
| ||||
| Mean/% | (SD) | Mean/% | (SD) | |
|
| ||||
| Age | 50.00 | (9.36) | 40.42 | (8.54) |
| Male | 34.09% | 37.84% | ||
| Female | 65.91% | 62.16% | ||
| Race/Ethnicity | ||||
| White (non-Hispanic) | 65.91% | 64.86% | ||
| Hispanic (non-Black) | 13.64% | 13.51% | ||
| Black (non-Hispanic) | 6.82% | 2.70% | ||
| Asian | 6.82% | 8.11% | ||
| Other | 10.81% | |||
| Faculty Rank | ||||
| Instructor | 4.65% | 0.00% | ||
| Assistant Professor | 16.28% | 77.78% | ||
| Associate Professor | 46.51% | 16.67% | ||
| Professor | 32.56% | 0.00% | ||
| Other | 0.00% | 2.78% | ||
| Lecturer | 0.00% | 2.78% | ||
| Faculty Track | ||||
| Tenure | 52.27% | 38.89% | ||
| Clinician Educator | 36.36% | 47.22% | ||
| Lecturer/Instructor | 2.27% | 11.11% | ||
| Research | 2.27% | 2.78% | ||
| Other | 6.82% | 0.00% | ||
Among mentees in the sample, most respondents are female (62.16%) and non-Hispanic Whites (64.86%). The mean age is 40.42 years. Nearly four-fifth of the mentees are junior faculty, i.e., assistant professors (77.78%) and lecturers (2.78%). Nearly half (47.22%) of the mentees are on the clinician-educator track.
Developmental Network Characteristics
The means and percentages of network characteristics are displayed in Table 2. Overall, there are relatively minor differences in most network characteristics between mentors and mentees in our sample. Mentors, on average, report 4.75 developers, and mentees have 5.06 developers in their networks. The mean levels of density are 0.62 for mentors and 0.65 for mentees.
Table 2.
Developmental Network Characteristics
| Mentors (N=44) | Mentees (N=37) | |||
|---|---|---|---|---|
|
| ||||
| Mean/% | (SD) | Mean/% | (SD) | |
| Network Level Measures | ||||
| Size | 4.75 | (1.40) | 5.06 | (1.17) |
| Range | 2.23 | (1.08) | 2.39 | (1.18) |
| Density | .62 | (.32) | .65 | (.28) |
| Developer/Tie Level Measures | ||||
| Social Arena | ||||
| Childhood Friends | .90% | 0.00% | ||
| Colleague/collaborator | 2.25% | 1.00% | ||
| College Friends | .90% | 1.50% | ||
| Current Job or Organization | 64.41% | 58.50% | ||
| Family | 5.41% | 11.00% | ||
| Graduate School Friends | .90% | 6.00% | ||
| Graduate School Professors | 5.41% | 10.50% | ||
| Other | .45% | .50% | ||
| Previous Job or Organization | 15.77% | 9.00% | ||
| Profession: Generic | 1.80% | 1.00% | ||
| Religious Organization | .45% | .50% | ||
| Volunteering | 1.35% | .50% | ||
| Support | ||||
| Psychosocial Support | 24.62 | (7.45) | 24.82 | (7.50) |
| Career Support | 20.09 | (7.52) | 21.43 | (7.85) |
| Relationship Type | ||||
| Mentor | 35.96% | 37.77% | ||
| Ally | 20.20% | 19.15% | ||
| Friend | 31.53% | 28.72% | ||
| Sponsor | 12.32% | 14.36% | ||
| Communication Frequency | 5.38 | (1.81) | 5.47 | (1.65) |
| Emotional Closeness | 2.83 | (.90) | 2.88 | (.88) |
Examining the mean levels of network range, developers on average come from 2.23 social arenas for mentors and 2.39 social arenas for mentees. Table 2 also displays the percentages of the social arenas from which developers are known. The most common social arena from which mentors’ developers are known is the current job or organization (64.41%), followed by previous job or organization (15.77%), and family (5.41%), graduate school professors (5.41%), and collaborators (2.25%). Compared to mentors in the sample, a larger percentage of mentees’ developers are family members (11.00%), graduate school professors (10.50%), or graduate school friends (6.00%), whereas a smaller percentage of mentees’ developers (58.50%) are from their current job or organization or their previous job or organization (9.00%).
Turning to support measures, mentors report higher mean levels of psychosocial support (24.62) than career support (20.09). Similarly, on average, mentees receive more psychosocial support (24.82) than career support (21.43) from their developers.
Examining the relationship types among mentors in our sample, a little more than one-third (35.95%) of developers are “mentors” (i.e., provide high career and psychosocial support), and 20.20% of developers are allies (i.e., provide low career and psychosocial support). Slightly less than one-third (31.53%) of mentors’ developers are friends (i.e., provide low career support but high psychosocial support), and 12.32% of mentors’ developers are sponsors (i.e., provide high career support but low psychosocial support). There was a similar distribution of relationship types among mentees in the sample, with 37.77% of mentees developers being classified as mentors, 19.15% being classified as allies, 28.72% being classified as friends, and 14.36% being classified as sponsors.
Mentors and mentees report similar average levels of communication frequency and emotional closeness with developers. The mean communication frequency among mentors was 5.38, and the mean emotional closeness score was 2.83. Among mentees, the respective scores are 5.47 and 2.88.
To help contextualize these findings, we display two example developmental networks among a mentor and mentee pair in our sample in Figure 1 in the Appendix. In this figure, developers are differentiated according to relationship type, with square icons depicting mentors, circles depicting friends, diamonds depicting sponsors, and triangles depicting allies. The social arenas from which developers are known are also provided, as are ties among developers, which are used to measure range and density. The focal respondents and their ties to developers are omitted for display purposes. Panel A displays the mentor’s developmental network, which includes two friends and two sponsors, and no allies or mentors. Three of the developers are colleagues/collaborators, and one is a graduate school friend, which results in a range of 2. The mentor’s developmental network has a density of .334, given that only 2 of the 6 possible ties among developers are present. Panel B displays the mentee’s developmental network, which includes three graduate school professors and 3 developers from the current job, resulting in a range of 2. Half of the developers are sponsors, one is a mentor, another is a friend, and one developer is an ally. Finally, the mentee’s network has a density score of .267, given that 4 of the possible 15 ties among developers are present.
Discussion
Our preliminary findings suggest that developers play an important role in the careers of university faculty mentors and mentees and that methods and insights from social network analysis can be applied to examine their support networks. Notably, mentors and mentees report similar mean levels of communication frequency, emotional closeness, and career and psychosocial support from their respective developers. Both mentors and mentees report higher psychosocial levels than career support. Additionally, the network size, range, and density are also similar among mentors and mentees. Notable differences emerge when examining the social arenas among mentors and mentees, with mentors reporting more of their developers coming from their current or previous jobs and mentees reporting more of their developers coming from graduate school and family contexts.
Our data indicate that university faculty are currently using developmental networks rather than dyadic relationships. Research that integrates insights from social network analysis and mentoring literature emphasizes that a single, dyadic hierarchical mentoring relationship rarely satisfies workers’ complex needs (Kram, 1985; Murphy & Kram, 2010; Sorcinelli & Yun, 2007). Instead, professionals often seek and obtain support and assistance from multiple sources (mentors, sponsors, and peers) and social arenas beyond the immediate workplace, such as professional organizations, friends and family members, and larger communities. Recognizing that individuals look beyond formal, hierarchal, and dyadic mentoring relationships throughout their careers, researchers are increasingly focusing on the importance of developmental networks (Higgins & Kram, 2001) to understand how mentoring factors into successful careers.
Key features of developmental networks (e.g., the provision of career and psychosocial support) are associated with positive career outcomes among working professionals. Our data indicate that developers provide career and psychosocial support to university faculty, more of the latter than the former. The developmental networks of university faculty are of high range (number of social systems from which developmental network relationships originate), density (or cohesion among developers), emotional closeness, and frequency of communication. Research has linked range, density, emotional closeness, and frequency of communication to career outcomes such as professional identity (Dobrow & Higgins, 2005), mutuality (Dobrow et al., 2012), optimism (Higgins et al., 2010), and leadership development (Ghosh et al., 2013) in multiple settings (Murphy & Kram, 2010; Shen & Kram, 2011), professions (Higgins & Thomas, 2001; Higgins et al., 2008), and career stages (Chandler & Kram, 2005).
The strengths of this study relate to the use of developmental network characteristics to study university faculty and a high proportion of women and racial/ethnic underrepresented minorities from multiple institutions in the Southwestern United States. However, the study limitations include small sample size and a lack of longitudinal data on the impact of network characteristics on career outcomes.
Conclusion
As our study is part of an ongoing longitudinal intervention, these findings will inform future analyses that will examine change in developmental network characteristics and their impact on participants’ career outcomes. For instance, longitudinal analyses will examine how levels of developmental network support and structural characteristics correlate with career satisfaction and the achievement of career milestones. This research will help identify how and for whom dimensions of developmental network structure matter for faculty members’ career success and satisfaction. Additionally, a subset of respondents will provide qualitative data that can help clarify the mechanisms underlying changes that respondents make to their developmental networks and how these changes impacted their future careers. Quantitative findings from this analysis and subsequent waves of data collection will enhance our qualitative data collection efforts by enabling interviewers to identify recent changes that respondents made to their developmental networks and elicit information motivations and strategies for making changes and how network changes affected their careers. This research promises to advance the understanding of how mentoring operates among university faculty today and enhance mentoring programs by providing insight into how faculty cultivate and utilize developmental networks across career stages. Attending to faculty members’ developmental networks can advance institutional efforts at improving faculty retention and reducing burnout by offering faculty opportunities and strategies for forming and effectively maintaining relationships with developers within and beyond their academic institutions.
Supplementary Material
Figure 1.
Examples of Developmental Networks Among a Mentor and Mentee Pair Panel A. Mentor’s Developmental Network Panel B. Mentor’s Developmental Network
Acknowledgement
This study was supported by NIH/NIGMS U01GM132175 (Sood, PI); and 2U54GM104944 (Sy, PI).
References
- Burt RS, Kilduff M, & Tasselli S (2013). Social network analysis: Foundations and frontiers on advantage. Annual Review of Psychology, 64, 527–547. [DOI] [PubMed] [Google Scholar]
- Dobrow Riza S, & Higgins MC (2019). The dynamics of developmental networks. Academy of Management Discoveries, 5(3), 221–250. [Google Scholar]
- Granovetter MS (1973). The strength of weak ties. American Journal of Sociology, 78(6), 1360–1380. [Google Scholar]
- Higgins MC (2004). Developmental network questionnaire. Harvard University Press. [Google Scholar]
- Higgins MC, & Kram KE (2001). Reconceptualizing mentoring at work: A developmental network perspective. The Academy of Management Review, 26(2), 264–288. [Google Scholar]
- Higgins MC, & Thomas DA (2001). Constellations and careers: Toward understanding the effects of multiple developmental relationships. Journal of Organizational Behavior, 22(3), 223–247. [Google Scholar]
- Kram KE (1985). Mentoring at work: Developmental relationships in organizational life. Scott, Foresman. [Google Scholar]
- Marsden PV, & Campbell KE (1984). Measuring Tie Strength. Social Forces, 63(2), 482–501. [Google Scholar]
- Murphy WM, & Kram KE (2010). Understanding non-work relationships in developmental networks. Career Development International, 15(7), 637–663. [Google Scholar]
- Papachristos AV (2011). The coming of a networked criminology? In MacDonald J (Ed.), Measuring Crime and Criminality (pp. 101–140). Transaction. [Google Scholar]
- Pife MJ., & Bake VL. (2013). Managing the process: The intradepartmental networks of early-career academics. Innovative Higher Education, 38(4), 323–337. [Google Scholar]
- Sorcinelli MD, & Yun J (2007). From mentor to mentoring networks: Mentoring in the new academy. Change: The Magazine of Higher Learning, 39(6), 58–61. [Google Scholar]
- Yip J, & Kram KE (2017). Developmental networks: Enhancing the science and practice of mentoring. In Clutterbuck D, Kochan F, Lunsford L, Domínguez N, & Haddock-Miller J (Eds.), The SAGE Handbook of Mentoring (pp. 88–104). SAGE Publications. [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.