Developing As a Doctoral Supervisor: Navigating Learning Goals and Professional Growth

Research Paper Recommended citation: Vilhunen, A., Parpala, A., Varis, O., & Taka, M. (2025). Developing As a Doctoral Supervisor: Navigating Learning Goals and Professional Growth. In Kangaslampi, R., Langie, G., Järvinen, H.-M., & Nagy, B. (Eds.), SEFI 53rd Annual Conference. European Society for Engineering Education (SEFI), Tampere, Finland. DOI: 10.5281/zenodo.17631541. This Conference Paper is brought to you for open access by the 53rd Annual Conference of the European Society for Engineering Education (SEFI) at Tampere University in Tampere, Finland. This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License. DEVELOPING AS A DOCTORAL SUPERVISOR: NAVIGATING LEARNING GOALS AND PROFESSIONAL GROWTH A. Vilhunen a,1, A. Parpala b, O. Varis c, M. Taka d, 1 a Aalto University, Espoo, Finland, 0009-0002-6860-2828 b University of Helsinki, Helsinki, Finland, 0000-0001-5822-6983 c Aalto University, Espoo, Finland, 0000-0001-9231-4549 d Aalto University, Espoo, Finland, 0000-0002-6147-9137 Conference Key Areas: Diversity, equity and inclusion in our universities and in our teaching; Building the capacity and strengthening the educational competencies of engineering educators Keywords: Pedagogical training, doctoral supervision, professional development, inclusion, diversity ABSTRACT Engineering education has undergone significant development, and simultaneously, the number of doctorates has rocketed. This has resulted in increased diversity among doctoral students, while the supervisors were already navigating demanding terrain, often without formal training. This study explores doctoral supervisors’ learning goals on supervision, and two research questions guided our study: what kind of learning goals the supervisors have, and do they recognize goals related to diverse supervisees and their support needs? The data were collected from 78 participants in an elective doctoral supervision pedagogical course in Aalto University, Finland, during 2022–2024 through essays and qualitatively content analysed through reflexive thematic analysis with an inductive approach. The key learning goals focused on their personal self-management, balancing between a directive approach and supporting supervisees' independence, and on communication. Additionally, supervisors wanted to learn about pedagogical approaches, and especially lecturers highlighted learning goals on diversity and inclusion. The results indicate that doctoral supervisors need support in their professional development regarding various competencies linked to supervision, and the potential of peer support in their continuous professional development. a, 1 A. Vilhunen, [email protected] d, 1 M. Taka, [email protected] Authors Vilhunen and Taka contributed equally to this research 1 INTRODUCTION Both doctoral and engineering education have undergone significant recent advancements, resulting in a multitude of educational obstacles in terms of both the quality and practices (Akay, 2008; Bastalich, 2017). The number of doctorates has expanded with the global average (as a percentage of the national population) increasing from 0.4 to 0.7 between 2010 and 2021 (The World Bank, 2022). This surge, especially in engineering education, results in an increasing diversity among doctoral students (hereafter also supervisees), thereby challenging supervisors. Diversity has numerous definitions (Gordon et al., 2023; Tamtik and Balasubramaniam, 2024); here we refer to various visible and invisible characteristics that make people unlike each other, i.e., citizenship, gender, (dis)ability, mental health, neurodiversity, language, wealth, career stage, disciplinary position, and sexual orientation (Elsherif et al., 2025). While the increasing time and resource constraints further increase the pressure to produce doctorates, supervisors must acknowledge the diversity attributes and tailor their supervision accordingly (Pearson and Brew, 2002). The key factor controlling doctoral students’ experiences of their studies is their supervisor and their relationship (Sverdlik et al., 2018). The quality and practical approaches of doctoral supervision significantly impact supervisees’ learning outcomes, expertise development, and degree completion (Breitenbach, 2023; Lee, 2008a; Pyhältö et al., 2012). The supervisor will “color the career of the early-stage researcher”, as one respondent concluded in Denis et al. (2018). Further, highquality supervision directly contributes to wellbeing, and a supportive environment is needed to mitigate uncertainty and challenges (Ives and Rowley, 2005; Levecque et al., 2017; Tikkanen et al., 2024). Doctoral supervisors experience supervision as both highly rewarding and demanding (Halse, 2011). The major caveat is that supervision practices often stem from personal experiences as a supervisee and as a supervisor, and supervisory development is characterized by learning by doing or learning from experience (Amundsen and McAlpine, 2009; Denis et al., 2018; Sefotho, 2018). Further, prior research indicates disciplinary differences in conceptions and fundamental purposes of doctoral education and supervision (Kreber and Wealer, 2023). Individual variation can result in inconsistency in supervision practices (Huet and Casanova, 2022) and as ethical challenges in supervision — such as problems with fairness, power dynamics, and student autonomy (Löfström and Pyhältö, 2017). Doctoral supervision, especially without formal training, is prone to inclusion and equity-related challenges. Unconscious biases, limited understanding of diversity among individuals, and a lack of a welcoming and supporting working culture hinders inclusive doctoral supervision (Gardner, 2008; Lee, 2008). Indeed, supervising doctoral students is a complex task (Bøgelund, 2015) and to adopt more inclusive and effective supervisory practices, supervisors need to recognize their development needs. However, knowledge of supervisors' development needs is limited. This research fills the gap by exploring the following questions: - RQ1: What kind of learning goals supervisors have when they enter an elective Doctoral Supervision pedagogical course? - RQ2: Do the participants recognize development needs related to diverse supervisees and their support needs? Addressing these questions will directly support the development of doctoral supervisor training and provide fitting and timely support for doctoral supervisors. 2 METHODOLOGY We employ a qualitative approach to research doctoral supervisors’ learning goals and needs for professional development. The goal was to map the diversity of the goals, and thus, this explorative approach was ideal for our purposes. 2.1 Data collection The data were collected in a Doctoral Supervision pedagogical course in Aalto university, Finland, years 2022–2024. It is an elective course for anyone with active doctoral supervision duties, however, anyone with a doctoral degree is formally qualified to supervise a doctoral student. We collected written pre-assignments of the participants (N=78, representing 70% of all course participants, and 9% of all the professors and lecturers at Aalto). The main career groups of participants were assistant/associate professors (40%), postdoctoral researchers (42%), and university lecturers and other research-focused personnel, who were studied as one group (18%). Altogether, 82% of the participants were from engineering fields, and 68% were male. The range in previous doctoral supervisees (N=0–5, average 0.3), and current supervisees (N=0–8, average 1.0) was large, indicating various starting points and needs for learning in the course. This research was conducted following the guidelines of Finnish National Board on Research Integrity TENK (2019). Participants submitted their pre-assignments prior to the first course session, and we requested their research permission once the course was completed. Participation in the study was voluntary, and the participants were informed about their rights and the data processing. In this paper, we produced data from the essays from a section of “How would I like to develop as a supervisor?”, which focused on collecting participants’ learning goals identified through their experiences as a doctoral student and a supervisor. All data were anonymized by the responsible researcher to ensure that participants were not directly identifiable. The university’s research ethics committee approved the study. 2.2 Methods We utilized a comprehensive reflexive thematic analysis (Braun and Clarke, 2021) with an inductive approach to systematically explore and map the diversity of learning goals defined by the participants. Two authors independently performed the qualitative content analysis with an open coding approach. We started with a random sample of ten essays and designed a collection of 23 codes based on initial observations and peer debriefing (Lincoln and Guba, 1985). The code group was iteratively expanded during the full data analysis, where the essays were randomly assigned to be reviewed by the two authors individually; however, a 10% random overlap was used to reduce the risk of author bias in the analysis. Further, the learning goals outside the diversity category were inductively analyzed to recognize indirect diversity-related goals. All the qualitative analyses were done in ATLAS.ti (Web version, 2025) and no artificial intelligence (AI) was used in the research. The learning goal categories were clustered into main groups using a hierarchical clustering approach (HCA). It produces clusters by first combining the learning goals with the highest correlations, i.e., the closely joined goals and clusters have stronger correlation with each other. We performed this clustering using a multiscale bootstrap approach (999 runs) and with the average method (the distance between clusters is the average of the distances between the points in one cluster and the points in another cluster) with pvclust package (Suzuki and Shimodaira, 2019) in RStudio 4.4.3 (R Core Team, 2025). The bootstrapping approach is useful as it does not require much assumptions about the distribution of the data (Efron, 1979). The results are reported with their approximately unbiased (au) p values, and for the clusters with au p values >95, the hypothesis “the cluster does not exist” can be rejected with a 0.05 significance level (Figure 1). Compared to conventional methods, such as k means, HCA is free from pre-defined assumptions of e.g. the number of clusters or their shape, and it is applicable to smaller data sets, too. Thus, our dendrogram and the number of clusters are data-driven. 3 RESULTS 3.1 Emphasis on the pedagogical approaches The participants described a diversity of their learning goals, each reporting 4.8 goals on average. We inductively categorized these 362 observations into 26 groups, which were statistically grouped into three main clusters: I. Key competencies and research process support, II. Supervisee’s wellbeing and learning support, and III. Supervisor’s professional development and collaboration (Figure 1). The results emphasized the various starting points and motivations to participate in an elective doctoral supervision course, and interestingly, all studied goals were well represented in the data. Of all career groups, post-doctoral researchers mentioned the most learning goals (N=162), followed by the professors (N=156) and others (N=69). The key goals of the cluster on competencies and research process support focused on control, namely balancing between a directive approach to supervision and supporting supervisees' independence and freedom to manage their own tasks (N=34). Further, participants reported numerous needs to develop their communication skills (N=25) and often backed this up with examples from their own experiences. The main learning goals for supervisees’ well-being and learning support called for knowledge of pedagogical approaches (N=26). Additionally, their learning goals emphasized not just the ability to support supervisees’ learning (N=21), but also in the context of diversity (N=19). The third cluster focused on supervisors’ professional development and collaboration, emphasizing the need to develop self-management skills (N=34). The participants reported their needs to manage their uncertainty and issues with confidence as a supervisor (N=15), together with an understanding of supervision relationships (N=11). Fig 1. A hierarchical dendrogram showing the identified learning goal categories. Each horizontal distance indicates the degree of correlation between the learning goals and clusters, i.e., the sooner one cluster joins another learning goal, the stronger their correlation. The clusters that join closely have a stronger correlation with each other, and they are numbered in the order of their appearance in the analysis (1–24). The values in parentheses are the approximately unbiased (au) p values. The coloured rectangles enclose a group strongly supported by the data (approximately unbiased p ≥ 95), and we have named them according to their content. When comparing different career groups, both professors and postdocs emphasized the need to develop their self-management competencies (N=14 and 18 participants, respectively), and for example, designing a supervision strategy that would support them with an increasing number of supervisees. The second most prominent learning goal dealt with finding the balance between control and supporting independence (N=15 and 17, respectively). 3.2 Diversity in the learning goals Our analysis of supervisees’ wellbeing and learning support category revealed both direct and indirect diversity-related development needs based on how they were worded in the essays. Direct goals were labelled as diversity (N=19), and they consisted of goals that explicitly mention word diversity or reflect goals relating to supervision of diverse individuals. These goals were mainly about learning to cope with supervisees from different backgrounds and with individual traits (N=8) and being able to cater to diverse learners’ needs (N=6), e.g.: "I would like to understand about cultural differences when supervising doctoral work" "How to include a shy student in some joint collaboration with other researchers such that it does not make unwanted pressure on the student?" "Tailor supervision to individual student needs" Diversity-related learning goals (n=19 by 15 participants) were mainly pointed out by “other” faculty (mainly lecturers and research fellows; n=8), and professors (5 professors). Despite only two postdocs (representing 15% of all postdocs) mentioned diversity-related learning goals, these two represent 37% of the identified diversity goals. Goals that co-occurred the most often with diversity focused on students’ learning support, inclusion, and supervisors’ knowledge about challenges, pedagogical approaches, and self-management (Figure 2). Fig 2. Co-occurrence of diversity-related and other learning goals in the participants’ essays. The graph shows how a diversity-related learning goal contained other learning goals, too. The learning goals are grouped based on Figure 1. Indirect diversity goals were scattered under various codes, i.e., pedagogical support, learning support, sensitivity, knowledge about challenges, inclusion, and wellbeing. However, not all observations under these codes were linked to diversity. Some of these indirect goals addressed needs to learn, e.g., tailoring support for individual needs, to provide emotional support through mentoring for supervisees in diverse difficult situations, help supervisees navigate various mental health issues, and foster a non-discriminative environment. 4 DISCUSSION AND CONCLUSIONS In this study, we explored doctoral supervisors’ learning goals regarding supervision. The essay analysis produced a vast spectrum of learning goals, most of which dealt with supervisors’ personal work balance in managing different tasks in supervision. This aligns well with prior research on the demanding nature of doctoral supervision (Bøgelund, 2015; Halse, 2011). Even if the complexity of doctoral supervision and the general need for supporting supervisors’ professional development have been acknowledged earlier, our results highlight the need for supervisory support on a range of topics. The participants successfully understood the broad spectrum of responsibilities in doctoral supervision, as e.g. Lee (2008) and Wisker and Robinson (2014) have discerned. For example, learning about effective pedagogical approaches and efficient communication in various supervision situations are often identified by various actors. Furthermore, participants recognized diversity-related learning goals regarding supervisees' abilities, ways of working, needs, goals, backgrounds, and other individual traits. However, considering the broad spectrum of diversity, some aspects, e.g., disciplinary and epistemological stances, disabilities, status, gender, and sexuality remained unrecognized. This suggests that while the awareness and interest in diversity, inclusion, and equ(al)ity may have increased in PhD education, supervisors still struggle with fully understanding the breadth of diversity, making it challenging to cater diverse supervisees’ needs. Our results support prior findings on how diversity can be perceived differently (Tamtik and Balasubramaniam, 2024) among higher education teachers, as some define it narrowly and some more complexly (Gordon et al., 2023). This might stem from the complexity and lack of a universally understood definition (Haug, 2017; Jahnukainen et al., 2023; Reindal, 2016), and confusion on how to implement inclusivity through pedagogical action (Haug, 2017; Reindal, 2016). Prior diversity-related research on supervision has mainly focused on race and culture (Showunmi et al., 2024), lacking a broader perspective. Supporting supervisors’ ability to better understand and navigate diversity could be one potential way to renew supervision practices and PhD education into more inclusive for diverse supervisees. The differences among the studied career groups are potentially controlled by the differences in participants’ supervision experiences. Post-doctoral researchers are only at the beginning of their academic career, having less experience on supervising doctoral students, compared to, e.g., lecturers and research fellows (included in “Other”) who have a strong emphasis on pedagogical duties in their work, or to professors who often experienced acting as the main supervisor. Thus, professors may consider themselves as generally more competent as supervisors. The main categories of the identified learning goals hold high practical implications in developing doctoral supervisor support and pedagogical training. Self-management strategies could be developed with the help of peers and their experiences; in general, universities and doctoral programs should facilitate communities of peer support and arenas for seeking and sharing support and knowledge. Second, the doctoral supervision training should focus on pedagogical knowledge instead of emphasizing academic leadership. Third, the increasing need for diversity competencies must be considered when updating our pedagogical training for supervisors. Addressing diversity in doctoral supervision context holds immense value in developing tertiary education into more inclusive, and hence, supporting not only wellbeing and sense of belongingness but also possibly better integration of diverse supervisees in academia. As this is part of a broader project, we will next perform an in-depth analysis of the identified indirect learning goals, focus on differences among career groups, and examine the contribution of previous supervision experience on the reported learning goals. Special emphasis will be put on supervisors' understanding of inclusion and diversity in the context of doctoral supervision. 5 ACKNOWLEDGEMENTS We thank all the participants for the data. This research has been funded by The Land and Water Technology Foundation (Maaja Vesitekniikan tuki ry foundation, MVTT) and the Ministry of Education and Culture of Finland through the Digital Waters (DIWA) Doctoral Education Pilot related to the DIWA Flagship (decision no. 359248) funded by the Research Council of Finland's Flagship Programme. REFERENCES Akay, A., 2008. A renaissance in engineering PhD education. Eur. J. Eng. Educ. 33, 403–413. https://doi.org/10.1080/03043790802253475 Amundsen, C., McAlpine, L., 2009. “Learning supervision”: trial by fire. Innov. Educ. Teach. Int. 46, 331–342. Bastalich, W., 2017. Content and context in knowledge production: a critical review of doctoral supervision literature. Stud. High. Educ. 42, 1145–1157. https://doi.org/10.1080/03075079.2015.1079702 Bøgelund, P., 2015. How Supervisors Perceive PhD Supervision – And How They Practice It. Int. J. Dr. Stud. 10, 039–055. https://doi.org/10.28945/2096 Braun, V., Clarke, V., 2021. One size fits all? What counts as quality practice in (reflexive) thematic analysis? Qual. Res. Psychol. 18, 328–352. https://doi.org/10.1080/14780887.2020.1769238 Breitenbach, E., 2023. Factors Influencing Doctoral Program Completion. https://doi.org/10.5772/intechopen.113824 Denis, C., Colet, N., Lison, C., 2018. Doctoral Supervision in North America: Perception and Challenges of Supervisor and Supervisee. High. Educ. Stud. 9, p30. https://doi.org/10.5539/hes.v9n1p30 Efron, B., 1979. Bootstrap Methods: Another Look at the Jackknife. Ann. Stat. 7, 1– 26. Elsherif, M., Middleton, S., Phan, J.M., Azevedo, F., Iley, B., Grose-Hodge, M., Tyler, S., Kapp, S., Gourdon-Kanhukamwe, A., Grafton-Clarke, D., Yeung, S.K., Shaw, J., Hartmann, H., Cairney, M., 2025. Bridging Neurodiversity and Open Scholarship: How Shared Values Can Guide Best Practices for Research Integrity, Social Justice, and Principled Education. Finnish National Board on Research Integrity, 2019. Guidelines for ethical review in human sciences. the Finnish National Board on Research Integrity 3/2019. Gardner, S.K., 2008. “What’s too much and what’s too little?”: The Process of Becoming an Independent Researcher in Doctoral Education. J. High. Educ. 79, 326–350. https://doi.org/10.1080/00221546.2008.11772101 Gordon, S.R., Yough, M., Finney-Miller, E.A., Mathew, S., Haken-Hughes, A., Ariati, J., 2023. Faculty perceptions of teaching diversity: Definitions, benefits, drawbacks, and barriers. Curr. Psychol. 42, 243–252. https://doi.org/10.1007/s12144-021-01406-2 Halse, C., 2011. ‘Becoming a supervisor’: the impact of doctoral supervision on supervisors’ learning. Stud. High. Educ. 36, 557–570. https://doi.org/10.1080/03075079.2011.594593 Haug, P., 2017. Understanding inclusive education: ideals and reality. Scand. J. Disabil. Res. 19. Huet, I., Casanova, D., 2022. Exploring the professional development of doctoral supervisors through workplace learning: a literature review. High. Educ. Res. Dev. 41, 774–788. https://doi.org/10.1080/07294360.2021.1877629