Higher education institutions are increasingly developing professional development (PD) programs for university academics as part of an increased international focus on teaching and learning (Hundey et al., 2020). In this regard, research has called for approaches addressing theories of social learning, experiential learning and collaborative learning for professional growth (Laksov, 2022). In the same manner Mentorship in PD is equally gaining attention (Ince, 2017, Cree-Green et al., 2020). Despite recognition of its benefits, the literature has not conclusively defined the theoretical foundations, duration, or approaches and outcomes of optimal mentorship in PD (Hallman et al., 2020). While several studies have discussed mentorship for early career academics (ECAs), few have been embedded in a context of institutionalized approaches focusing on pedagogical development, and most studies are based on small sample sizes or conducted with no subsequent empirical analysis (Hundey et al., 2020; Pleschova & McAlpine, 2015). Recent literature has thus called for further research to explore mentorship using diverse methodologies, larger samples, and longer-term studies (Pleschova & McAlpine, 2015).

In the higher education literature, several terms such as mentoring, coaching, tutoring, facilitating, and supervising refer to a similar goal of supporting inexperienced educators. Rather than trying to distinguish between these concepts, we find that they all share a common set of imbedded beliefs and practices. Thus, this study adopts the operationalized definition of mentoring used in Pleschova and McAlpine (2015), which emphasises long-term cooperation between teachers and colleagues with more teaching experience and expertise with the aim of educational development and the enhancement of teachers’ pedagogic practice.

The study took place at a Danish University, that provides an institutionalized long-term (12 month) PD program for ECAs. The program is multi-tiered, based on social constructivist theories of adult learning, and follows principles of problem and project-based learning. It comprises participant-centred workshops, collaborative teamwork, individual project work, and mentorship. Uniquely, it employs a collaborative approach to mentorship through a two-mentor setup: a subject mentor from the mentees’ close environment, and a pedagogical mentor from the Centre for Teaching and Learning. While the pedagogical mentor supports the ECA by providing generic pedagogical knowledge, the subject mentor provides support regarding discipline-related teaching and interaction within the immediate work environment (the study program or department). Considering the unique structure of mentorship, this study explores how both mentors perceive their roles by drawing on their experiences, guided by the research question:

What are the views of mentors—both subject-focused and pedagogy-focused—on their role of supporting ECAs in a long-term academic development program?

Conceptual framework

Recent literature calls for a systems-thinking approach to supporting academics’ professional learning (Arnesson & Albinsson, 2017; Hundey et al., 2020; Hallman et al., 2020; Ince, 2017). On this basis, this study conceptualizes the mentor’s role as a situated, contextualized, erratic, enacted, and practiced along three interrelated dimensions: the mentor’s individual characteristics, dynamic relations through collaboration, and contextual interaction.

The individual characteristics dimension describes thecongruence between values and practices (Ince 2017).

The relational dimension focus on how mentors engage with relationship dynamics in groups. Relationship-oriented mentoring is increasingly being reported as an effective PD approach, underlining mutual respect and reciprocal learning (Arnesson & Albinsson, 2017).

The contextual dimension involves mentors’ constant interactions with their environment. Mentorship is recognized as a sustainable resource for ECAs with mentors coordinating organizational activities and supporting community building, facilitating a common language, and building cultures of trust (Laksov 2022).

The proposed framework serves as a conceptual foundation driving the research design and the process of exploring mentors’ subjective views.

With institutional ethical approval, 17 pedagogy and 32 subject mentors (N=49) participated in the study. The participants, 25 females and 22 males, served in various disciplines, including humanities, social sciences, engineering, natural sciences, and health and medicine, and had between one and 35 years of mentorship experience. Q-methodology (henceforth Q) was adopted. This approach aims to capture and contrast individual and collective subjectivity (Brown, 2019). Q enables researchers to explore subjective beliefs, perceptions, and viewpoints before identifying shared perspectives within participant groups. Using abductive inquiry, Q integrates qualitative and quantitative techniques to systematically explore the complexity of subjectivity (Watts and Stenner, 2012), revealing insights less accessible through other methods (Watts and Stenner, 2012). Following Watts and Stenner (2012) and in line with the suggestions of Lundberg et al. (2020), our Q-procedures comprised three steps: Step 1: Developing the Q sample through concourse construction and condensation. - Through several rounds of piloting and discussion, the research team condensed the initial concourse of 79 statements to a final Q sample comprising 33 statements. Step 2: Q sorting and post-sorting activities for participants. - The Q sorting activity was administered face-to-face as an individual reflection activity with the research team members. Participants reflected on their experiences while considering the sorting question: Based on your own experience, what do you believe to be the most useful aspects regarding your role as a supervisor in the University Pedagogical program (UP) for assistant professors in 2019-2022? They were then asked to rank 32 statement cards on a grid ranging from ‘most useful’ to ‘least useful’. Step 3: Q factor analysis and interpretation. - A principal component analysis provided an initial overview of extracted factors with eigenvalues above 1.00 and identified “the point at which the line changes slope” (Watts and Stenner, 2012: 108). Centroid factor analysis was used for formal data analysis to recompute and compare the judgmental rotation screen and varimax rotation, with both the factor loading data and a scatter plot of the two factors displayed as rotated (Brown, 1980). Subsequently, the research team compared different factor solutions through several rounds of calculating and discussing the Q sorts, following principles widely employed as statistical criteria (Brown, 1980; Watts and Stenner, 2012). A three-factor solution was chosen, with no significant factor inter-correlations identified.

The study explored mentors’ views on their roles in a PD program. Q methodology identified three significantly distinguishing collective interpretations on what constitutes good mentorship: 1. Providing advice based on one’s own teaching experience 2. Promoting mentee self-reflection and self-regulation 3. Co-creating and mediating. While the study identified a few statistically supported consensuses mentors also reported opposing perspectives. In particular, subject mentors promoted a pedagogical strategy of providing direct advice based on their own experience and academic status (Kamyounias et al., 2008; Mathias, 2005). We argue that there is a risk that PD activities based on this pedagogical strategy becomes stagnant and reproductive since ECAs are encouraged to “do the same as their mentors”. A mechanism which could be one of the reasons behind the fact that universities are quite resistant to pedagogical change (Goffe & Kauper, 2014). The findings open for a debate on the types of mentorship knowledge needed to become an effective mentor within PD (Ince, 2017). This may be particularly salient for those without prior experience in PD, such as subject mentors, who struggle in their roles and may risk taking contradictory strategies and approaches (Mathias, 2005). The results also reflect concerns about identifying and choosing mentors, a topic debated in the literature (Bean et al., 2014). Based on the study the following conclusions are drawn: - The combination of subject and pedagogical mentors represents a holistic approach to PD, as mentors provide distinct different kinds of feedback. - Subject mentors should not defer to providing advice solely based on own experiences but should act as organizational connectors treating mentorship as a peer-learning opportunity - creating room for personal growth and organizational change. - Mentors need to be educated to secure effective and holistic PD in mentor driven programs.

Arnesson, K., & Albinsson, G. (2017). Mentorship: A pedagogical method for integration of theory and practice in higher education. Nordic Journal of Studies in Educational Policy, 3(3), 202–217. Bean, N. M., Lucas, L., & Hyers, L. L. (2014). Mentoring in higher education should be the norm to assure success: Lessons learned from the faculty mentoring program, West Chester University, 2008-2011. Mentoring & Tutoring: Partnership in Learning, 22(1), 56–73. Bickerstaff, S., & Cormier, M.S. (2015). Examining faculty questions to facilitate instructional improvement in higher education. Studies in Educational Evaluation, 46, 74-80 Brown SR (2019) Subjectivity in the human sciences. Psychological Record 69: 565–579. Cree-Green, M., Carreau, A. M., Davis, S. M., Frohnert, B. I., Kaar, J. L., Ma, N. S., ... & Nadeau, K. J. (2020). Peer mentoring for professional and personal growth in academic medicine. Journal of Investigative Medicine, 68(6), 1128-1134. Goffe, W. L., and Kauper, D., (2014). A survey of principles instructors: Why lecture prevails. Journal of Economic Education, 45 (4), 360-375. Hallman, S., Massoud, L., & Tomiuk, D. (2020). An integrating model for excellence: Mentorship to enrich the three pillars of education. Journal of e-Learning and Higher Education, 1–8. Hundey, B., Anstey, L., Cruickshank, H., & Watson, G. P. (2020). Mentoring faculty online: a literature review and recommendations for web-based programs. International Journal for Academic Development, 25(3), 232-246. https://doi.org/10.1080/1360144X.2020.1731815 Ince, A. 2017. Managing Risk in Complex Adult Professional Learning: The Facilitator’s Role. Professional Development in Education 43 (2): 194–211. Kamyounias, P., McGrath‐Champ, S., & Yip, J. (2008). ‘Gifts’ in mentoring: Mentees’ reflections on an academic development program. International Journal for Academic Development, 13(1), 17–25. Laksov, K.B,, Elmberger, A.., Liljedahl, M. & Björck, E. (2022). Shifting to team-based faculty development: a programme designed to facilitate change in medical education. Higher Education Research & Development, 41(2), 269-283. Lundberg, A., de Leeuw , R., & Aliani, R. (2020). Using Q methodology: Sorting out subjectivity in educational research. Educational Research Review, 31, Article 100361. Mathias, H. (2005). Mentoring on a programme for new university teachers: A partnership in revitalizing and empowering collegiality. International Journal for Academic Development, 10(2), 95-106. Pleschová, G., & McAlpine, L. (2015). Enhancing university teaching and learning through mentoring: A systematic review of the literature. International Journal of Mentoring and Coaching in Education, 4(2), 107-125. Watts S and Stenner P (2012) Doing Q Methodology: Theory, Method and Interpretation. London: Sage.