Relationship between doctoral supervisors’ competencies, engagement in supervisory development and experienced support from research community

ABSTRACT

Supervision is one of the most important determinants of a successful doctoral process. However, there have been few large-scale survey studies on the development of PhD supervisors. In this study, doctoral supervisors’ perceptions of their supervisory competence, professional development activities and professional support from the research community are explored. A total of 561 doctoral supervisors from a large multi-field research-intensive university in Finland responded to the survey. The data were analysed with statistical methods. The results showed that the supervisors perceived their supervisory competence as being high. More experienced supervisors were more confident with their supervisory competence than less experienced ones. Support from the scholarly community was also related to positive perceptions of one’s supervisory competencies. The results further showed that engaging in supervisory development activities was also related to perceived supervisory competence. Some gendered and disciplinary differences in supervisors’ engagement in professional development activities were detected.

KEYWORDS:

• Doctoral supervisors
• competence
• research community support
• supervisory development
• supervisory relationship

Introduction

Quantity and quality of supervision is one of the core determinants of a positive doctoral experience and completion. High quality supervision, including frequent supervisory meetings (at least monthly; Li & Seale, 2007; Seeber & Horta, 2021; Shin et al., 2018), providing constructive feedback and pastoral care (Vekkaila et al., 2016), building shared expectations, and agreeing on supervisory practices (Ives & Rowley, 2005; Moxham et al., 2013; Pyhältö et al., 2015), engaging candidates in co-authoring and providing writing support (Kamler, 2008), and promoting research community integration and networking (Lee, 2008; Wisker et al., 2007), are associated with timely completion, positive doctoral experience and employment after completing the PhD degree. Considering the importance of doctoral education for universities, the systematic development of PhD supervisors, is a surprisingly underdeveloped area globally (Denis et al., 2019; Hanesova & Saari, 2019). A reason for this is research focusing on supervisory development is limited (Lee, 2018; McCulloch & Loeser, 2016; see also recent review by Huet & Casanova, 2022). This has resulted in a lack of grounding for developing effective research-informed practices for providing well-fitted support for supervisory development. Also, most studies on supervisory development have been small-scale exploratory qualitative studies (e.g. Halse & Malfroy, 2010; Henderson, 2018), whereas only few large-scale quantitative studies have been conducted on the topic (e.g. Lee, 2018; Motshoane & McKenna, 2021). We aim to bridge the gap in the literature on doctoral supervisory development by using a large-scale survey design to examine doctoral supervisors’ perceptions about their supervisory competence, the professional development activities they engage in, the interrelation between these two, and professional support from the research community.

Doctoral supervisory competence and supervisory development

There is an extensive body of evidence that supervision is one of the main determinants of a positive doctoral experience and degree completion (Ives & Rowley, 2005; McCallin & Nayar, 2012). A supervisor’s competence comprises knowledge, skills, and attitudes to act in the supervisory relationships (Korthagen, 2004; Westera, 2001) and plays a key role in enhancing high quality supervision in doctoral education. Supervisory competence involves cognitive, motivational, and behavioural dimensions. In a theoretical and descriptive perspective, supervisory competencies are cognitive structures that support certain supervisory behaviours, and in an operational sense, competencies cover a variety of skills, allowing a supervisor to act in complicated and evolving supervisory situations, and find solutions to them (Toom et al., 2021; Westera, 2001).

Characteristics of good supervision have been extensively described in the literature (see e.g. the review by Taylor, 2014). The attributes of high-quality supervision involve a supervisor’s knowledge, skills, and attitudes as well various supervisory behaviours. For example, having several supervisors (Cornér et al., 2017), a supervisor’s disciplinary and methodological expertise (Gube et al., 2017; Ives & Rowley, 2005), supervisory commitment and frequent meetings (Li & Seale, 2007; Seeber & Horta, 2021; Shin et al., 2018), providing informational and emotional support (Gatfield, 2005; Vekkaila et al., 2016), and matched expectations for the supervisory relationship (Moxham et al., 2013; Pyhältö et al., 2015) contribute to a positive PhD experience and study progress. As such, they reflect evidence-based or research-informed understanding of quality supervision, supervisory competencies, and current (development) needs identified in doctoral education. However, the landscape of the core competencies required by a doctoral supervisor still lacks coherence (Delany, 2009), and questions about whether and how development of supervisory competencies should be systematically supported and accredited have only just begun to rise (Hanesova & Saari, 2019; McCulloch & Loeser, 2016).

Also, the landscape of PhD supervisory environment has somewhat recently undergone significant changes such as the massification of doctoral education, new types of PhDs, accountability demands, online supervision, and an increase in diversity of PhD candidates (Andres et al., 2015; Huet & Casanova, 2020; Lee, 2018), suggesting the need for exposure to a broader repertoire of supervisory approaches. Respectively, supervisory models are transforming: the personalised dyadic model of supervision in which supervision is considered as private space between the supervisor and PhD candidate has given floor to various forms of co- and team supervision (Grant et al., 2014; Guerin et al., 2015; Lee, 2018; Robertson, 2017). Simultaneously increasing attention has been paid on supervision pedagogy (Bruce & Stoodley, 2013; Henderson, 2018). However, many supervisors still draw on a traditional supervisory discourse i.e. the apprenticeship model, and their personal experiences as PhD candidates (Fulgence, 2019; Raffing et al., 2017; Vehviläinen & Löfström, 2016). This is also reflected in supervisory competence development that heavily relies on the supervisor’s own experiences as PhD candidates, occasionally observing and discussing topics with their colleagues, being an examiner, and on ‘how to’ literature (Halse, 2011; Peelo, 2011; Wisker et al., 2014). Hence, the role of formal training and providing systematic support for supervisors is still of minor importance in cultivating supervisor’s competence, compared to informal learning (Amundsen & McAlpine, 2009; Hanesova & Saari, 2019; Lee, 2018; Turner, 2015). However, there is tentative evidence of a positive impact of engaging in supervisory development activities in improving PhD supervision (Taylor, 2014; Wichmann-Hansen et al., 2020). Developing a variety of well-fitted and functional (research informed) supervisory development activities and support practices might be particularly beneficial for early career supervisors’ competence development (Vereijken et al., 2018).

More extensive and systematic measures to improve the quality of doctoral supervision have been called for and implemented at many universities (e.g. Olson & Clark, 2009; Park, 2007; Powell & Green 2007). Such measures have involved the development of supervisors through courses, workshops, mentoring and awards, though providing these activities is a relatively new area of staff development in universities (Kiley, 2011; Taylor & McCulloch 2017). A recent survey study involving 311 institutions from 32 European countries showed that compulsory specialised supervisory training for PhD supervisors was required in only 17% of responding institutions, either in the majority or in all their doctoral programmes (Hasgall et al., 2019). Courses for PhD supervisors were offered by 43% of universities, either in the majority or in all their doctoral programmes (Hasgall et al., 2019). The findings imply that universities are seeking more systematic ways to engage PhD supervisors in professional development activities to cultivate their supervisory competencies.

Doctoral education at the University of Helsinki, Finland

The University of Helsinki is a large multidisciplinary research-intensive university that has been ranked within the top 100 universities in the world according to the Academic Ranking of World Universities (ARWU), the World University Ranking by Quacquarelli Symonds (QS) and the Times Higher Education Ranking (THE). It has 4900 PhD candidates, and 500 to 600 PhD degrees are awarded annually. Since 2014, all PhD candidates at the University of Helsinki belong to one of the four doctoral schools, and to one of the 33 doctoral programmes offered. Doctoral studies at the university (total 240 ECT) involve conducting doctoral research (80%; i.e. 200 ECT), and complementary coursework (20%; i.e. 40 ECT) based on a personal study plan (Andres et al., 2015). The PhD dissertation can be written as either a monograph or a set of articles, but most PhD candidates at the University of Helsinki write an article-based doctoral dissertation (over 70%; Pyhältö et al., 2022), comprising three or four peer-reviewed published articles and a summary. Eligibility criteria for a PhD include having a Finnish second-cycle master’s degree or comparable foreign degree and demonstrating academic language proficiency either in Finnish, Swedish or English. A research proposal and PhD study plan must be included in the application. One must also have a ‘commitment to supervise’ letter from one or two supervisors if accepted (University of Helsinki, 2022). Typically, at least one of the prospective supervisors is a full professor (permanent position). The prospective supervisor(s) reviews the research plan and the application prior to submission. The target time for undertaking a PhD is four years studying full-time, but the average time for degree completion is 5–6 years (Pyhältö et al., 2022). There are no tuition fees, but funding is not automatically provided for doctoral study by the universities, projects, or foundations. The thesis examination proceeds in three stages: first, two external examiners appointed by the faculty pre-examine the dissertation. Second, if approved, a public defence follows the pre-examination. Finally, the faculty awards the PhD degree.

Aim

The aim of this study was to gain a better understanding of supervisory development among PhD supervisors by examining how the supervisors perceive their supervisory competence and detecting the professional development activities they engage in to cultivate their supervisory competencies. Also, the interrelationship between their perceived competencies and engagement in the development activities, and the professional support from the research community, was explored. In addition, we examined the variation between the supervisors in the competencies experienced and professional development activity involvement. We addressed the following research questions:

1. How do doctoral supervisors estimate their supervisory competence? Are their perceptions related to

   1. Length of supervisory experience

   2. Number of PhD candidates supervised

   3. Professional support received from their scholarly community

   4. Engagement in supervisory development activities

2. What supervisory development activities did the doctoral supervisors engage in?

3. Are there differences in perceived supervisory competence and engagement in supervisory development activities between:

   1. Supervisors from different disciplines,

   2. Women and men,

   3. Those supervising mainly candidates writing a monograph cf. an article-based dissertation?

Method

Participants

Altogether 561 doctoral supervisors (275 women, 50%; 266 men, 49%; 6 non-binary, 1%) from the University of Helsinki, Finland completed the survey. The response rate was 16%. The supervisors were from all the four doctoral schools: Humanities and Social Sciences, Health Sciences, Environmental, Food and Biological Sciences and Natural Sciences. Of the supervisors, 44% were working as professors or research directors, 38% as university lecturers or university researchers, 9% as university instructors or post-doctoral researchers, and 9% as tenure track professors. The representation of the sample was good in terms of disciplines and supervisors’ positions and gender at the university. Half of the supervisors (51%) had been supervising PhD candidates for more than 12 years (mean 15 years; SD = 9.7 years). According to the reports, the average number of PhD candidates per supervisor at the time of the survey was four, while the number of doctoral degrees supervised to completion was seven, yet considerable variation between the supervisors occurred. The supervisors reported that their supervisees (i.e. PhD candidates under their supervision) typically wrote article-based dissertation (82%). Most of the supervisors (68%) reported supervising primarily full-time PhD candidates.

Supervisory experience -survey

The data were collected online between August and September 2021 by using a modified version of a supervisory experience survey validated in previous studies (Pyhältö et al., 2015). Participation in the study was voluntary. Information about the research and the link to the supervisory experience survey were sent via email to the participants by using the Doctoral Schools’ supervisor mailing lists. No identifying information was collected. No incentives were used. For the present study, the following measures were used: The supervisory competence scale (six items; Cronbach alpha = .728) (adapted from the UK Research Supervision survey, UK Council for Graduate Education, 2021) and the professional support from the research community scale (modified from Pietarinen et al., 2013; five items; Cronbach alpha = .878) was measured with a 1-to-7-point Likert scale. The scales were measured with (1 = strongly disagree … 7 = fully agree). The engagement in supervisory development activities was measured with a six-item supervisory development scale (yes/no) (adapted from UK Research Supervision survey, UK Council for Graduate Education, 2021) that covered both informal (e.g. discussing with colleagues) and formal development (e.g. participating in supervisory training) activities. In addition, the following background variables were used in this study: disciplinary field, gender, length of doctoral supervisory experience (years), number of completed doctoral dissertations supervised, and dissertation format typically supervised (monograph cf. article-based dissertation). According to national guidelines (Finnish National Board on Research Integrity, 2019) no ethics review was required.

Analysis

The means and standard deviations were calculated to get the overall perceptions of the doctoral supervisors of their supervisory competencies and experiences of professional support within the scholarly community. To analyse the associations between the supervisory competencies, experiences of social support, length of the supervisory experience, and the number of supervised dissertations, Pearson correlation coefficients were used. One-way analysis of variance (ANOVA) was used to analyse the differences in supervisory competencies between supervisors who were engaged in various numbers of supervisory development activities. The Games Howell test was used for post-hoc comparisons. One-way ANOVA was also used to analyse the differences in supervisory competencies between disciplines and independent samples t-tests for gendered differences and differences based on supervisees’ thesis format. Disciplinary differences, gendered differences and differences based on candidates’ thesis format in supervisory development activities were analysed with cross-tabulations and chi-square tests. The results of parametric tests (t-tests and ANOVA) were confirmed using non-parametric tests, Kruskal-Wallis test, and Mann-Whitney U-test. The analyses were carried out using IBS SPSS Statistics 28 programme.

Results

Supervisory competencies and professional support

On average, the supervisors perceived their supervisory competencies as being high (M = 5.35, SD = .79, 1–7 Likert scale with 7 indicating highest confidence in their own competence). They were confident with their skills to supervise candidates from diverse backgrounds, felt satisfied with their abilities in being efficient supervisors and reported knowing how to enact their institution’s policies and procedures. They also reported receiving sufficient support from the scholarly community (M = 5.18, SD = 1.15). Accordingly, the supervisors felt that their colleagues provided them with encouragement and support and appreciated their work. They also described their scholarly community as having a good atmosphere, and that any problems were dealt with in a constructive way.

Further investigation showed that the length of the supervisory experience was related to increased satisfaction with their supervisory competencies (r = .213, p < .001). Also, the number of doctoral dissertations completed under one’s supervision was also related to perceived supervisory skills (r = .178, p < .001). Those who had supervised a higher number of completed doctoral dissertations were more confident with their skills. Professional support from the scholarly community was related to more positive perceptions of their supervisory competencies (r = .340, p < .001). The more positively the supervisors perceived the climate of the scholarly community and the higher the levels of support and encouragement they received from their colleagues, the more satisfied they were with their supervisory skills.

Supervisory development activities

In general, most of the supervisors (79%) reported being engaged in at least two supervisory development activities. The most common professional development activities that the supervisors reported were informal ones. The most typical activity was discussing with colleagues about their supervisory practices (78%). About half of the supervisors reported that they consulted the university’s policies on doctoral education (60%) and evaluated their supervisory practices with the candidates (54%). Reading scholarly literature about supervisory practice (27%) were less commonly employed of all professional development activities. Participating in formal supervisory training was reported by more than third of the supervisors (38%).

To analyse, whether the supervisors’ engagement in supervisory development activities was associated with their perceptions of competencies, six groups based on the quantity of supervisory development activities (from 0 to 5 activities) that the supervisors had engaged in, were formed. The supervisors who reported being engaged in several development activities were more satisfied with their level of supervisory competence (F(5, 553) = 7.51, p < .001). More specifically, the supervisors who reported engaging in three (m = 5.46, sd = .67, p < .001), four (m = 5.52, sd = .72, p < .01) or five (m = 5.65, sd = .90, p < .001) supervisory development activities perceived their supervisory competencies as being significantly higher than those who reported engaging in only one supervisory development activity (m = 5.03, sd = .79).

Differences in supervisory competencies and supervisory development activities

No disciplinary differences were detected in supervisory competencies (F(3, 467) = .152, p = .93) or between women and men (t(526) = −.071, p = .94). There were no differences in the supervisors’ perceptions of their supervisory competencies either regarding their supervisees’ thesis format (i.e. monograph cf. summary of articles) (t(548) = −.042, p = .97).

However, disciplinary differences were detected in how typically the supervisors read the scholarly literature about supervisory practice (χ²(3) = 17.53, p < .01), consulted university policies on doctoral education (χ²(3) = 17.49, p < .01), and participated in supervisory training/workshops/forums/seminars (χ²(3) = 14.04, p < .01). More specifically, supervisors in health sciences, humanities and social sciences, and environmental, food and biological sciences reported reading the scholarly literature to improve their supervisory practices more often than supervisors in natural sciences. The supervisors in humanities and social sciences and health sciences reported more often that they consulted university policies regarding doctoral supervision than supervisors in environmental, food and biological sciences and natural sciences. Furthermore, the supervisors in humanities and social sciences had participated in supervisory training/workshops/forums or seminars more often than supervisors in all other disciplines. No disciplinary differences were detected in how typically the supervisors evaluated supervisory practices with their supervisees (χ²(3) = 2.50, p = .48), or discussed the practices with colleagues (χ²(3) = 4.66, p = .20).

Gendered differences were detected in the participation in the supervisory development activities. Women reported that they evaluated their supervisory practices with supervisees (χ²(1) = 6.63, p < .05), read the scholarly literature about supervisory practice (χ²(1) = 21.24, p < .001), consulted the university’s policies and on doctoral education (χ²(1) = 12.02, p < .001), as well as participated in supervisory training/workshops/forums/seminars (χ²(1) = 36.54, p < .001) more often than men did. No gendered differences were detected in discussing the practices with their colleagues (χ²(1) = 3.13, p = .08). No differences were detected in the supervisors’ participation in any of the supervisory development activities based on supervisees’ thesis format.

Discussion

Methodological considerations

We used a supervisory experience survey data collected from all the supervisors from a research-intensive highly globally ranked multidisciplinary university to analyse the interrelationship between the perceived supervisory competence, engagement in various supervisory development activities, and professional support received from the colleagues. The response rate was somewhat low, but the representativeness of the sample was adequate in terms of disciplines, age and gender of the supervisors at the given institution. The supervisors also represented all the career stages after PhD completion. The reliability of the scales used can be considered satisfactory (α > .72 in all scales). Due to the cross-sectional design, conclusions about causal relationships cannot be made based on the results. It is also important to keep in mind that the supervisors’ responses might have been affected by the fact that the access to supervisory development activities was limited by the COVID-19 pandemic. Moreover, one should be careful in generalising results between countries that have different doctoral education systems.

Doctoral supervisors’ perceptions on their supervisory competence

Our aim with the study was to contribute to the body of knowledge on doctoral supervisory development by exploring the interrelationship between the doctoral supervisors’ perceptions about their supervisory competence, the professional development activities they engage in, and support they experienced from the research community. On average, the supervisors perceived their supervisory competence to be high. Yet, more experienced supervisors reported their competence to be higher than those having less experience in supervising PhD candidates. This is in line with findings from previous studies suggesting that new supervisors are often relatively ill repaired to take up their role as PhD supervisors (Motshoane & McKenna, 2021; Turner, 2015). This is likely to result from several complementary factors: while much of the supervisory development still occurs through ‘learning by doing’, supervisors’ access to formal professional development activities such as courses and workshops, provided by their institution is often random or limited (Hanesova & Saari, 2019; Lee, 2018). Also, the institutional structures for supporting supervisors’ engagement in informal supervisory development activities are often missing, leaving particularly, new supervisors without sufficient support for the supervisory development (Amundsen & McAlpine, 2009; Turner, 2015; Vereijken et al., 2018). Accordingly, gaining more experience in supervising PhD candidates was reflected in the elevated levels of perceived supervisory competence. Furthermore, the high levels of support and encouragement from the professional community were related to perceiving one’s supervisory competence as being good. This indicates that supervisory development is not an individualised process. Instead, the findings support the view on supervisory development as a co-regulated learning process (Hadwin et al., 2011) proceeding from interindividual behaviours to intraindividual competencies (Vygotsky, 1978).

Supervisors’ engagement in supervisory development activities

In line with previous findings (e.g. Wichmann-Hansen et al., 2020), our results showed that engaging in supervisory development activities was related to perceived supervisory competence. More precisely, supervisors engaging in three or more development activities perceived their competence to be higher than those who reported only one activity. This might indicate that engagement in supervisor development increases a supervisor’s confidence in their competence (see also, Motshoane & McKenna, 2021). On the other hand, it might also be that the supervisors who felt confident with their supervisory competencies, were more active in seeking opportunities for professional development or they might have been more interested in investing in supervision in general. Accordingly, in the future, longitudinal studies are needed to explore the causality between engaging in development activities and supervisory competencies. The result also indicates that being involved in several development activities from different spheres of doctoral education was of benefit in supervisory development. The activities the supervisors engaged in ranged from feedback from the PhD candidates under their supervision to taking supervisory courses and to consulting the institutional policies, reflecting different spheres of doctoral education. This implies that a supervisory development programme should not only provide a variety of professional development activities, but that a nested systemic approach, involving going beyond individual supervisor’s competences and engaging with culture and structures of doctoral education (Motshoane & Mckenna, 2014) should be adopted in orchestrating the activities

Differences in perceived supervisory competence and engagement in supervisory development activities

Gendered and disciplinary differences in supervisors’ engagement in professional development activities were detected. Our results showed that women were more active in engaging in supervisory development than men. The result is partly contradictory with results of prior studies suggesting that women scholars often have fewer career development opportunities compared to men (Barnes & Beaulieu, 2017; Gibbs et al., (2017). Also, the supervisors in the fields of humanities and social sciences were more active in engaging in supervisory development activities than their peers from other fields. During previous years, article-based dissertations with co-authored articles, have become the dominant dissertation format also in humanities and social sciences in the case university (Pyhältö et al., 2022), implying an emergence of more collective forms of supervision. Accordingly, a potential reason why the humanities and social sciences supervisors were more active in engaging in supervisory development compared to their peers in STEM fields might be that they are to a greater extent challenged by the transition from dyadic model of supervision to more collective ones, and hence actively seek support in making the transition (e.g. UK Council for Graduate Education, 2021; Mason & Merga, 2018). The result is also aligned with the gendered differences detected since humanities and social science is dominated by women while STEM is still predominantly male dominated. Moreover, considering their disciplinary expertise it might be that the supervisors in humanities and social sciences were more sensitive in recognising potential positive influence of the engagement in supervisory development activities for their PhD candidates compared to their colleagues in STEM. In general, the result indicates that individual variation both in supervisor’s willingness to engage in supervisory development activities and in professional learning is likely to occur depending on the individual and disciplinary attributes. There is also tentative evidence that varied supervisory development trajectories can result an increase in supervisory competencies (Turner, 2015) suggesting that one should be cautious with the generic notion of ‘best practise’ of supervisory development (Motshoane & McKenna, 2021). Accordingly, this indicates that to engage diverse groups of supervisors working in varied supervisory environments, development activities must be engineered by considering their needs in the given environment. The results further imply that the effectiveness of supervisory development programmes might be at least partly dependent on their ability to consider supervisors’ individualised needs.

Conclusions

We showed that on average the supervisors perceived their supervisory competence as being high. Most supervisors were engaged in at least one supervisory development activity, and the most common activities were informal ones. The positive perceptions of supervisory competencies were related to having more experience as supervisor, receiving support from scholarly community, and engagement in a variety of supervisory development activities. Disciplinary and gendered differences occurred in supervisors’ engagement in supervisory development activities.

Implications and recommendations for enhancing supervisory development

The study showed that perceived supervisory competence was positively associated with supervisory experience, professional support from the scholarly community and engagement in supervisory development activities. Accordingly, both informal and formal supervisory development activities seem to play a distinctive but complementary role in supervisory development (Huet & Casanova, 2022). The findings indicate that the supervisors benefitted from engaging in a variety of informal and formal supervisory development activities, and support. Particularly, peers seem to provide a central resource for both informal and formal institutional supervisory development activities. This implies that systematic supervisory development should draw on this resource for diversity of forms, including encouraging, orchestrating, and creating arenas for peer learning. Supervisors’ confidence in their supervisory competence was positively related to supervisory experience, indicating that a significant mode of supervisory development takes is ‘learning by doing’. This implies that measures to support supervisory development should be at least partly ingrained in their everyday work as supervisors. For example, peer mentoring and co-supervision can provide powerful tools for this (Olmos-López & Sunderland, 2017; Turner, 2015). However, guidelines and training for implementing such activities are needed to enhance the spreading of best practice.

The results further suggested that engaging in a variety of supervisory development activities was related to an increased sense of supervisory competence. This indicates that different development activities serve different purposes, and hence are needed to cultivate supervisory competence. In enhancing supervisory development at the institutional levels, this potentially has three main implications: first, to be effective the development activities and support needs to be aligned with a supervisor’s needs, suggesting that individual professional development needs must be identified. Second, the development activity or support provided should be aligned with the need identified. Finally, we claim that developing a sustainable and systematic supervisory development programme calls for institutional commitment to the research-informed development work, from three complementary perspectives in identifying development needs, most potential development activities fitting the needs, and in detecting impact of engaging in such activities.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Notes on contributors

Kirsi Pyhältö

Kirsi Pyhältö, PhD, is a professor of higher education in the Faculty of Educational Sciences, University of Helsinki. She is also an extraordinary professor at the Department of Curriculum Studies, University of Stellenbosch, and South-Africa. She is an expert in the field of researcher education and careers. Her research interests include doctoral education, supervision, researcher communities and post-doctoral careers.

Lotta Tikkanen

Lotta Tikkanen, PhD, is a senior lecturer in university pedagogy at the Centre for University Teaching and Learning (HYPE), Faculty of Educational Sciences, University of Helsinki, Finland. Her research interests include teachers and students’ well-being and learning in various educational settings ranging from primary school to doctoral education.

Henrika Anttila

Henrika Anttila, PhD, is a senior lecturer in university pedagogy at the Centre for University Teaching and Learning (HYPE), Faculty of Educational Sciences, University of Helsinki, Finland. Her research interests include doctoral students’ academic emotions, learning and well-being.

Appendix 1

Table

Scales and items
Supervisory competence (7-point Likert scale)
I am confident I have the skills to supervise doctoral candidates from diverse backgrounds
I receive support from my university/institution to be a good supervisor
I am satisfied with my ability to be an effective supervisor
I know how to enact my institution’s policies and procedures surrounding supervision of candidates
I enjoy being a supervisor
Being a supervisor increases the quality of my own research
Professional support (7-point Likert scale)
My colleagues provide me with encouragement and support
I feel my colleagues appreciate the work I do
My scholarly community has a good atmosphere
My scholarly community deals with problems constructively
The work assignments are divided fairly among the staff
Supervisory development activities (yes/no)
Which of the following do you do to enhance your supervisory practice?
Evaluate my supervisory practice with my candidates
Discuss my practice with my colleagues
Read the scholarly literature surrounding supervisory practice
Consult my university’s policies on doctoral education
Participate in supervisory training/workshops/forums/seminars