‘Challenging from the start’: novice doctoral co-supervisors’ experiences of supervision culture and practice

ABSTRACT There is a widespread interest in doctoral supervision, yet novice doctoral co-supervisors’ experiences remain understudied. The aim of this qualitative study is to explore how supervision culture and practice are experienced by novices. This thematic analysis is based on focus groups and in-depth interviews with 23 novice co-supervisors from two Swedish research-intensive universities. The study reveals three dimensions related to the challenging experience of the supervision culture, namely closedness, dependence and competition. Moreover, the analysis proves that co-supervisors’ practices are embedded within the ‘hidden curriculum’, here defined as the unplanned and implicit support outside formal meetings and activities. Novice co-supervisors bridge the gap between main supervisors and doctoral students and make the path to the doctoral degree smoother when engaging in emotional, intellectual, practical and mediation support. This study has practical implications and suggests improvements for co-supervision that would also benefit the doctoral students.


Introduction
With the number of enrolled doctoral students increasing globally, novice co-supervisors recruited from early career academicsare one of the most rapidly growing groups of supervisors.For them, co-supervision is often part of the collegial activities that they are expected to undertake (Geschwind et al., 2022).For the novices, finding out what the cosupervisory role includes can be a challenge (Wilkin et al., 2022).In higher education institutions worldwide, the expectations about what supervision work includes, such as number of supervisees or involvement in co-authorship, are imprecise (Amundsen & McAlpine, 2011).Furthermore, how to distribute the actual workload between the supervisors in the team is not always clear (Kumar & Wald, 2022).Doctoral co-supervisors account for a considerable part of all supervisors, yet cosupervision, and specifically the novice co-supervisors, has received little attention in the literature.Therefore, the study aims to explore what it means to become a co-supervisor by addressing the following research questions: How do novice co-supervisors experience the supervision culture?What types of supervision practice do they engage in?By highlighting these questions, it is possible to enhance the support and training for the novices.

Review of the literature
There is a growing number of studies on doctoral supervision examining, for example, the framework of supervisory approaches (Lee, 2008), ethical dilemmas within doctoral supervision (Löfström & Pyhältö, 2020), doctoral students' and supervisors' well-being (Wisker & Robinson, 2016), and match (Pyhältö et al., 2015).Further, research has focused on intercultural supervision (Manathunga, 2007;Grant, 2010), and supervising doctoral students at a distance (Huet & Casanova, 2021), just to mention a few topics.
Studies addressing co-supervisors remain few.One of the first studies on co-supervision was undertaken by the sociologist Pole (1998) and concluded that having a team with one or more co-supervisors may be a safety net for the doctoral student, albeit not automatically.Moreover, some researchers have analysed the collaboration with other supervisors or tensions within the supervisory team (Watts, 2010;Manathunga, 2011Manathunga, , 2012;;Guerin & Green, 2015;Robertson, 2017aRobertson, , 2017b;;Lee, 2008).Further, recent research found that co-supervisors pointed out overwork as a problem while the doctoral students found conflicting advice a challenge (Olmos-López & Sunderland, 2017;Pole, 1998).Nevertheless, studies suggest that the roles within the team should be clarified from the start (Wald et al., 2022).
Some studies address novice supervisors, both main supervisors and co-supervisors.There are studies about the challenges to take on a new role when crossing the border from being a doctoral student to shoulder the role as a doctoral supervisor (Motshoane & McKenna, 2021;Henderson, 2018).Novice supervisors often drew upon their own experiences to understand the work as a supervisor (Amundsen & McAlpine, 2011).Moreover, novice supervisors interpreted their role broadly and many of their activities were outside of supervising the thesis (Amundsen & McAlpine, 2009).They also learned that compliance was needed in order to fit in (Maritz & Prinsloo, 2019).A study by Carter (2016) showed that novice supervisors found that supporting doctoral students with mental health issues was difficult.Further, the study revealed the importance of novices being diplomatic.Grossman and Crowther (2015) stressed that novice co-supervisors should raise their questions about what is expected from them before they accept the invitation to become a co-supervisor.However, becoming a supervisor is an ongoing process with challenges and where new supervisors seemed to accept existing guidelines, while experienced questioned the conditions (Halse, 2011).
Furthermore, the supervision culture has many layers.McAlpine and Norton (2006) defined the supervision culture as a nested context.Academic culture has also been discussed as having hierarchies and power relations influenced by a 'culture of silence', in which students and staff are hesitant to complain or violate cultural norms (De Welde & Stepnick, 2015;Nardone, 2018;Grubbström & Powell, 2020).Based on the literature above, it is of pivotal interest to further explore novice co-supervisors' experience of being novices, what type of supervision activities they are engaged in, and how supervision culture impacts on co-supervisors' thinking, feeling and behaviour.Moreover, given the power relationshighlighted by earlier researchit is particularly interesting to analyse how novice co-supervisors handle social relations within the supervisory team.

Conceptual framework
The three conceptsenvironment, culture and climateare sometimes used interchangeably.Environment is the broadest concept, while culture and climate both describe the internal environment of an institution (Peterson & Spencer, 1990).The theoretical underpinnings for this study capture the view of culture and delineate what we want to examine in the supervisors' practice.First, culture is endurable with 'deeply embedded patterns of organizational behaviour and the shared values, assumptions, beliefs, or ideologies that members have about their organization or its work' (Peterson & Spencer, 1990, p. 6).Second, we agree with the sociologists' approach that defines culture as socially constructed and that it 'serves as the "social glue" binding an organization together' (Cameron, 2008, p. 452).Finally, we are inspired by McDowell's (1999) definition of culture as social relations linking people and places and that describes how practices also can be sustained by power or exclusion.These social relations may exist in the supervisory team, being colleagues in different roles in the department, having collaborated in previous projects, or being involved in ongoing research.In this study, the term supervision culture will be used to describe a part of the socially constructed academic culture connected to research education that links people within the departments, the research groups and the supervisory teams.Supervision culture has an impact on co-supervisors' thinking, feeling and behaviour.
Already during the interviews, it became clear that the novice co-supervisors repeatedly talked about the unplanned and unofficial supervision they were involved in and which took up a large part of their time.In light of this, to understand this particular part of the supervision practice, we borrow the concept 'hidden curriculum' of doctoral supervision (Elliot et al., 2020).Here, the hidden curriculum is used in a slightly different way than usual.It is defined as supervision practice outside the planned and officially organised activities of the doctoral programme, such as seminars, supervisory meetings and conferences.Instead, these may be implicit, unplanned and informal activities, but valuable in the sense that it supports the doctoral student's progress (Elliot et al., 2020).We use the concept to highlight certain hidden parts of the co-supervisory work.

Study context and method
Participants in the study Our study includes co-supervisors at two Swedish research-intensive highly ranked universities.A large part of the research conducted in Sweden is externally financed, and supervisors at the universities spend a lot of time applying for funding in order to stay in academia.Besides their own research, they take on teaching and supervision work.Co-supervision is important to progress within the research career.However, the opportunities for supervising depend on the possibilities of funding, the department's strategies to include younger researchers and whether the younger researcher is considered a good match in relation to the doctoral project.In Sweden, national law requires each doctoral student to have at least one co-supervisor.The main supervisor has the main responsibility for the quality of the theses and for administration.The relationship between main and co-supervisors can be seen as hierarchical.In other parts of the world, co-supervision could also be 'joint supervision', with co-supervisors sharing more or less equal parts of the work, or could be mainly used when the topic is interdisciplinary (Spooner-Lane et al., 2007).It is important to bear in mind that the situation for the co-supervisors and how co-supervision is organised can vary considerably, not only between countries but also between and within universities.
To find interviewees for both the initial focus groups (FG) and the additional in-depthinterviews (IDI), we used a purposive sampling technique.Those who had co-supervised for at most four years were deemed eligible to take part.Moreover, this study was limited to the formally nominated co-supervisors.Following this, an invitation to participate in focus groups was sent by e-mail to former course participants in the three-week basic doctoral supervision course that in Sweden is a mandatory course for those who want to become a Docent.The invitees were self-selecting, therefore, biases could be present as regards those who wanted to tell their story, both positive and negative.However, the aim was to create an equal distribution of the sexes and a variation of disciplines (Table 1).

Collection of empirical material
Before commencing the interviews, an interview guide was developed.Five FGs were then conducted between January 2019 and June 2020.The group discussion revolved around the following topics: a) initial support; b) main challenges; c) team collaboration; d) group members' own doctoral supervision experiences; and e) future needs.However, to deepen the analysis, we needed complementary IDIs in which novices could explain their feelings and relations more thoroughly.Moreover, the topics in the FGs sometimes developed in a sensitive direction and we felt that using IDIs was an opportunity to hear more about these issues one-to-one.The in-depth interviews were collected between August and December 2021, and lasted approximately 1-1.5 h.The topics brought up in the IDI were: a) co-supervision culture; b) co-supervision practice; c) co-supervisors' wellbeing; and d) emotional challenges.Anonymity was guaranteed, and the participants in the FG and in the IDI were informed about the possibility of withdrawing their participation at any time.Oral consent was given at the beginning of each FG and IDI, according to the research ethics at the universities.With permission from the participants an audio recording was kept for all, collected via the web tool Zoom during the pandemic, or face-to-face in five pre-pandemic cases.

Method of analysis
The reflexive thematic analysis began with verbatim transcription.Elements reflecting the 'supervision culture' and the 'co-supervision practice' were identified and clustered.All themes were seen as interpretative stories that were actively created by the researchers (Braun & Clarke, 2006, 2019).The following themes were found for the supervision culture: Since both authors are involved in doctoral supervision training programmes at one of the universities and engaged as course leaders in the supervision course that the interviewees have taken, this might have influenced the participants' willingness to participate and share experiences.Therefore, it was important for the authors to reflect on their own potential biases.However, the impression was that having met beforehand created trust and contributed to an open discussion climate.

Findings
In this section, we analyse the novice co-supervisors' experience when they supervise a doctoral student for the first time, how they experience the internal supervision culture and what supervision tasks they undertake.

The supervision culture
Most notable themes about supervision culture were cultures of closedness, dependence and competition.The analysis below illustrates the three themes in detail by quotes from the interviewees.

Closedness
One of the novice co-supervisors, Elin, described her positive experience of her first day.She was invited to the supervisory team to have an open conversation about workload allocation and how the co-supervision expectations should be met.However, her story was one of the exceptions.Talking about the introduction, Andrew shared his negative first experiences of co-supervising: There was no clear indication from either the student or the main supervisor, especially on the expectations for me as a co-supervisor.I had no idea what was expected from me … I tried to raise the issue for discussion with the main supervisor on two occasions.
Another co-supervisor, Renee, also asked for clarifications about her role and said: 'I was like, I cannot get answers from this man.I give my email.He doesn't respond.It's super strange.'The poor communication or silence between the co-and main supervisor is something unexpected and creates frustration when co-supervisors want to know more about what is expected of them.Further, some women had difficulties in understanding if they really were expected to supervise.Local policies recommend having at least one female supervisor in the team.However, women were not always informed that they were on the list of supervisors in the team.Renee was surprised: 'I was never asked actually the student came to me (after eight months) and said: "Did you know that you're my co-supervisor" [laughing] and I had no idea.'The silence and the poor communication was part of the exclusion in this closed culture.
Notwithstanding, some co-supervisors seemed to choose to remain silent because of the closed culture, as Alonso explained: 'In our department you won't be able to discuss things openly.Even when there are event of a more acute character, one avoids bringing it up.'The shared assumptions and beliefs were apparently reproduced by the novices, who learned how to behave and perform accordingly in order to fit in.Dependence Some co-supervisors described well-functioning relationships.Alfie said: 'We had a really good process … everyone was involved in every meeting.'Yet, another distinct and contradictory picture turned up in parallel: co-supervisors who chose to be amenable and accept all kinds of tasks in order to fit in and be accepted as a group member.Maria experienced this as frustrating: 'and then you try to say "no" but you … then you stand there and have to [if you refuse] you're not supportive.Then you're not, then you're inflexible and not helpful.'Sam's quote illustrated his feeling of dependence as loyalty: 'So then it's a weird situation.You are part of a closed environment and you had to be kind of faithful to that [laughing], to that, to your group of supervisors.'Moreover, interviewees explained that they had to nurture good relationships and be loyal, since they were dependent on senior colleagues for academic advancement.They chose to avoid conflicts by being peripheral.As Melissa explained: 'but I need this for my qualification that yes, I'm a bit low … peripheral.' One reason behind these strong feelings of dependence on seniors might be the relationship between supervisors and the fact that some worked with their own former supervisors.Additionally, some had been invited as co-authors and therefore felt a debt of gratitude and most of them feel dependence related to their future career.

Competition
The research culture is characterised by a focus on performance.For externally financed researchers without permanent positions, competition is part of their academic life.Their proposals must be successful if they are going to stay in academia.Including women in applications can increase the chances for funding.However, women were sometimes openly informed about this strategy.Lena experienced this: This was the third day of my position when I started it all three years ago.The third day this guy came to my office and said: Hey, I'm submitting a proposal to this private company to have a PhD.Would you like to be a co-supervisor?And with no hesitation he said: I need a female on this to be more successful and it's either you or Mary.And Mary is not here today so.I was like: Oh OK.And I asked him: Can I at least read the proposal?And he is like: Sure, but the deadline is today midnight so I can't change anything.So I was just like: So you're literally just putting my name?And at the time I guess you know, I just didn't really know what his proposal was about.
We identified several cases in which this gendered behaviour was deeply embedded, a norm developed among the colleagues and a seemingly natural part of the application process, particularly in fields of research with few women with a doctoral degree.These women found it difficult to know whether they were expected to co-supervise or if they were just a name on a piece of paper to make the proposal successful.In most cases they were ready to take on the role.The espoused values, such as equality, are widespread and present an ideal image of the department.However, the embedded values and the individual desire to stay in academia through competition guide everyday action.
Furthermore, our interviewees discussed the complexity of competition and how this affected their own supervision.Sam reported that: At the end the main supervisor you are helping is your competitor.He is a direct competitor.In my supervisory team it's like three of us are competing for the next (like) professor position … If we support you in supervising your PhD, this means you are helping, you get better than us.
The sense of rivalry between the supervisors also forced novices to step back from an active role in order not to threaten the main supervisor.As Melissa reflected: 'But I think it's this bit of rivalry here and now.It's just to drop things right now.I'm just laying low a bit.I think that's what the thing is right now.'This lack of involvement may affect the quality of supervision negatively.In a culture of competition, social relations take on a different meaning.Shared values and behaviours can be sustained by power and exclusion.

The co-supervision practice
In our interviews, the novice co-supervisors repeatedly returned to the unofficial supervision practice they were engaged in.It was also clear that this practice was surrounded by the three aggravating sides of the supervision culture: closedness, dependence and competition.The culture informs the co-supervision practice and at the same time the co-supervisors' practice shapes the culture.The relations between the supervision culture and practice seemed to interact and fit into each other like two cogwheels, which is illustrated in Figure 1.Furthermore, the figure shows that the novice co-supervisor has a 'dual challenge': first, trying to figure out what co-supervision practice is without knowing what the expectations are and, second, trying to understand a supervision culture that is not always open and supportive.

Emotional support
Part of the supervision practice was to exercise care and encourage students.Novice cosupervisors were fully engaged; however, this could be challenging in a number of ways.The co-supervisor Yasemin explained how her doctoral student often contacted her for support in stressful situations: All of a sudden she [doctoral student] thinks that nothing is good enough.And she was in the field and she had no competence and everything.And then we [doctoral student and cosupervisor] talked over the phone for an hour and then she [doctoral student] realised that this is ok.
The co-supervisors mentioned that they could be contacted and were expected to be accessible outside office hours at any timefor example via social mediaby doctoral students with mental health issues.John described this: It was earlier this year … she sent me a message on-on Facebook eh basically saying that she had a lot of panic attacks and had not been doing so well.I think we had a meeting planned and that she wanted to reschedule it or something like that … How should I respond to this?
In many cases, co-supervisors tried to handle the situation alone.In acute situations, cosupervisors did not turn to other peers for advice; instead some talked with their partner at home.In all probability, the novice co-supervisors' unofficial work or informal responsibilities helped the doctoral students to navigate their doctoral work and made it less stressful.Moreover, the doctoral student certainly showed trust and even approached them with private matters.They might be less formal, close in age and more recently qualified and therefore less strained.However, doctoral students also seemed to take severe issues to co-supervisors.For example, Leo talked about his feeling of inadequacy and discomfort when he learned that his doctoral student had suicidal thoughts.He adapted his schedule for them to work more closely on a daily basis: He was a little introverted for a while and then it appeared several weeks months later that he had even had suicide thoughts … that I had not understood at all … And so yes but should I actually report this somewhere else now?This can be very serious.But then we had quite frequent meetings in the greenhouse where we would grow.It was purely physical.We could of course have daily control so to say … But still it was a real feeling of unease.
This emotional support could be considered part of the official curriculum; however, when it happened often and outside formal meetings and was of an ad hoc, acute and private character, it slipped into the unofficial hidden curriculum.Sometimes, the support was even outside the blurry boundaries of the hidden curriculum, on the fringes of supervision work and beyond what might be a co-supervisory task, such as supporting students with severe mental health issues.However, it seemed that the co-supervisors wanted to keep up a profile as capable, independent and responsible supervisors.Their coping strategies were to be more accessible, adapt their working hours, monitor sick students by working closely with them, change their schedule to manage the intensive support, and ask for advice from partners or other family members outside academia.The social relations and the emotional support of the co-supervisor were important.However, novice co-supervisors engage a lot and are not trained for severe cases.

Intellectual support
Despite being overloaded with work, providing expert feedback seemed to be prioritised.Since novice co-supervisors were invited from the start because of their expert knowledge, they had expected this to be one of their main tasks.Unexpectedly, some co-supervisors identified themselves as an unused resource.Being a non-consulted expert created a feeling of frustration, as Elin expressed, 'Trying to say that yes.Hey, I exist!You can use me.'The novice supervisors felt excluded since the doctoral student turned to someone more senior for intellectual support.However, one exception seems to be those who had applied for their own research projects but were not eligible to become main supervisors yet but acted as main supervisors.For intellectual support, the significance of the rivalry and competition appeared to be more explicit.The co-supervisors wanted to participate in research-oriented support to gain new knowledge, be invited to co-author, or expand their network.However, only few of them were asked to be involved as much as they wanted and expected.

Practical support
The practical support included both thesis-related and other day-to-day support.The practical work, such as helping the doctoral student with contacting authorities, following fieldwork, or introducing the doctoral student to, for example, the lab, was rewarding.Collaborating in relaxed situations outside the meeting room created trust and a friendly and dialogue-based relationship with the doctoral student.They were looking forward to 'hanging out' as Melissa expressed: 'I like to go out for data sampling as well.I like to hang out at some point but I have not received that question [from her doctoral student].It is a pity.'However, seniors were often asked for support when it was thesis-related.The novice co-supervisors seemed to avoid intruding but were at the same time frustrated and also wanted to be asked to join in.Instead, the novice co-supervisors were more involved in other private practical matters, support that made life easier for the doctoral student.

Mediation
Interestingly, several female co-supervisors instructed doctoral students about the main supervisor's preferences and how to approach seniors with questions, a phenomenon that we call 'mediation'.The co-supervisor Maria felt responsible for successful communication between the student and the main supervisor and mediated: I know what good and bad sides my supervisor has so I can say: Ok, but there is no point in asking him about this because he will not answer then.But if you ask the question in this way, then you will get an answer within 5 minutes.So in that way I can help the doctoral student.
Moreover, both males and females seemed to encourage the student after they had received critical comments, or clarified what the main supervisor brought up during the supervision meetings.For example, Joseph found his student stressed after a meeting with the main supervisor: I've had scenarios where a student has called me quite stressed eh from a meeting with a principal supervisor.And what I found to be quite interesting is that from my perspective the stress in that situation wasn't needed … As a co supervisor it's, it becomes, it's such an easy fix.Eh, we have a conversation.We clarify and the problem is solved.Furthermore, doctoral students often turned to the co-supervisor when they believed that the main supervisor was dissatisfied with their efforts.The co-supervisor then comforted them and explained the main supervisor's complaint was just a matter of conversational style, as Ali explained: She was sometimes telling me: Oh, the main supervisor said this and this.What does it mean?Like, am I doing bad?Or, is it because I'm not good enough for instance?I said: No, no, no.But don't think like that because I know him like, eh.This is his style.
In cases where the main supervisor was their own former supervisor, co-supervisors could become even more specific in their advice and act like experts in the nature of a working relationship with that particular person.At the same time, the culture of silence and the closed culture reinforced the importance of communicating in a certain way, to be flexible, not to disturb, and not to cause tension between members of the group.Here, this analysis shows that the co-supervisors' mediation work helped doctoral students to bridge the chasm between them and senior researchers.Despite the fact that the supervision culture was challenging, the co-supervisors' experience was of being a social glue between team members.

Discussion
When novice co-supervisors talk about their first experiences of supervision culture and practices, two things stand out.With respect to the first research question, it was found that the supervision culture consists of specific challenges for a novice and that this constrains how they work.The second question sought to examine the co-supervision practice.It was found that a considerable part of the practice takes place within the hidden curriculum.However, this implicit support seemed to lack distinct contours and boundaries and make the novice co-supervisors uncertain and stressed.
The novices try to perform at their peak right from the start.This study found that if co-supervisors are asked to help with something even outside of research or supervision, they often do not dare to say no.These results are in agreement with Amundsen and McAlpine's (2009) findings, which showed that novice supervisors interpreted their role broadly and many of their activities were outside of supervising the thesis.In our study this is taken even further; the co-supervisors mention that in some cases they even feel used.Some co-supervisors point out that they feel both honoured but also taken advantage of when they are asked to carry out a main supervisor's assignment without being nominated as a main supervisor.Firstly, this supportive or submissive behaviour can be explained in part by the norms of supervision culture.The positive side of this culture allows for supporting each other, 'co-assessing' and peer reviewing without always getting the time back.A downside is when the co-supervisors are expected to help and have a feeling of being used.Secondly, they may have been invited by seniors as co-supervisors, experts, co-authors or partners in projects and feel a debt of gratitude.Thirdly, the novices are also dependent on the seniors in power for their future academic career and therefore co-supervisors fear exclusion from the inner circle of researchers.
The novice co-supervisors also try to understand where their responsibilities begin and end.Most describe an initial period without introduction or discussions about what the role entails.Earlier studies confirm the new supervisors' challenges by 'experiencing a lack of clarity regarding role expectations in general' (Amundsen & McAlpine, 2009, p. 341).There are guidelines; however, these may not easily be translated into practice when supervision is difficult (Carter, 2016).In our study, this is further stressed.Novice co-supervisors are team members, but in order to show themselves as capable they tend to handle difficult supervision without using the other team members for advice.Therefore, co-supervisors are alone with their decisions and feel unsure of how to deal with, for example, the doctoral student's health issues.
Apparently, many of the co-supervisors are available all the time to provide the emotional support the doctoral students' need, a segment of the hidden curriculum.The co-supervisors seem to take a lot of responsibility for this part of the supervision, even in severe cases of mental health issues such as suicidal thoughts, panic attacks, and depression.Research indicates that the need for emotional support seems to be increasing (Wisker & Robinson, 2016), but is hard for a novice to handle (Carter, 2016).Our study has shown not only that emotional support is needed but also how novice co-supervisors think and act when they are in the middle of it.They feel committed and engaged but at the same time stressed and insecure.They seem to become the 'first-line-support' Perhaps the decisive thing for a novice is whether the supervision culture made them feel included or excluded.One finding is that co-supervisors encounter a culture that is not always open.Therefore, they often seem to choose to conform and remain silent to adapt.The novice co-supervisors' silence may partly be explained by fear of social exclusion (see Grubbström & Powell, 2020).In order to manage the closedness, the co-supervisors also need to develop their diplomatic skills, as shown in the study by Carter (2016).
Our study highlights that in addition to the diplomatic skills novices have developed, they also make sure to train their doctoral students in the same ability, through mediation.In this way, the novice co-supervisor takes responsibility for ensuring that the doctoral student develops a sensitivity in interactions with other colleagues and adapts to the culture.
In this study, novice co-supervisors seem to like and look forward to supervising.Yet they experience discomfort.Carter (2016) interestingly highlights the benefits of learning through challenges and Elliot et al. (2020) the importance of supporting the PhD within the hidden curriculum, the supervision outside formal meetings and planned activities.However, novice co-supervisors feel discomfort when they are left alone with the responsibility to act as main supervisors or handling students' severe mental health issues.In addition, sometimes they feel discomfort when they accept tasks outside their responsibilities out of fear of threatening their relations with senior researchers.As a novice, you quickly develop strategies that do not interfere or question those in power if you want to stay in academia.
This study analyses co-supervisors in two Swedish universities but can inform supervision practice in an international context as we address general questions about novice co-supervisors' thinking, feelings and behaviour.The co-supervisory role is underexplored, specifically that of the novice co-supervisor.Supervision culture has an impact on novice co-supervisors' social relations: how they think about their relations to the doctoral students and other supervisors when there are no clear boundaries and expectations; how they feel when they fear to be excluded from the group of supervisors at the department; and how they act in order to be seen as capable by others in the supervisory team and in the department.The findings lead us to some suggestions for measures (see Table 2) as a basis for further discussion.
Several questions remain unanswered.This study highlights an avenue for further research on novice co-supervisors' experiences.For example, how novice co-supervisors become the social glue in supervisory teams, and how their role in emotional support affects their own well-being.by 23 novice co-supervisors at two Swedish research-intensive universities.The study contributes to the field of research on doctoral supervision, particularly the unexplored co-supervisory role.The findings clearly indicate that the supervision culture of closedness, dependence and competition was challenging for the novices from the start, a culture that need a shift.The research has also shown that novice co-supervisors spend a lot of time supervising within the hidden curriculum.This was beneficial for the students but sometimes challenging for the co-supervisors, who were unprepared for handling students' mental health issues or tensions within the team.
Given the universal nature of co-supervision, we argue that our findings are valuable to consider and form a basis for commencing a discussion about the nature of co-supervision.The present study suggests that novice co-supervisors should be offered an introduction, have a discussion about workload allocation and possibly have a mentor in order to be useful.Nevertheless, beside all challenges, we could also discern signs of change when novices with confidence talked about their supervision philosophy, when they without hesitation asked for clarification, and when they actively supported the relationships in the team through mediation.Since good supervision seems to be a decisive factor for the doctoral student's progress (Raffing et al., 2017), it should be beneficial to support novice co-supervisors and use their competences right from the start.

Figure 1 .
Figure 1.The four types of supervision practice embedded in the supervision culture.

Table 1 .
Interviewees' disciplinary background, sex, pseudonyms, total numbers of PhDs supervised and type of material.