Reﬂecting on Ethical Processes and Dilemmas in Doctoral Research

: This paper discusses ethical dilemmas from an early-career researcher perspective, drawing upon doctoral research experiences—my own. The doctoral study involved life-history interviews with ﬁve primary-school-teacher mothers. During the study, ethical dilemmas arose that were not considered by me or in the ofﬁcial university ethical processes. This left me feeling vulnerable in the data-collection period and overwhelmed with concerns for the well-being of participants and for myself as researcher. This paper draws on my journal entries and reﬂections; detailed reﬂections of the pre-, during and post-ﬁeldwork stages were collected (totalling over 600 entries). The paper utilizes critical incidents analysis to explore two ethical dilemmas from the data collection phase. Findings include personal reﬂections on experiences of university processes and the mismatch between the metaphor of ethics as a ‘hurdle’ on a smooth track to completion, and the real-life incidents and dilemmas that followed ethical approval. Recommendations are made for a consideration of doctoral ethical dilemma support and the limitations of formal ethics processes in UK universities.


Introduction
This paper considers the ethical dilemmas experienced by a doctoral researcher, me, during the data-collection phase of my thesis, following approval by a university ethics review board (ERB). The paper addresses the following question, set by the editors, Busher and Fox, for this Special Issue: • How can institutional ethics committees effectively support educational researchers throughout the life of a study?
It also aims to highlight a personal account. The author challenges the narrative that the formal ethical approval process is a hurdle to be cleared, with no need for support with ethical issues afterwards. It also adds to a body of literature where ethical dilemmas and issues are shared, to avoid further normalization of 'hidden injuries' in the dissemination of research [1]. In this section, doctoral ethics processes are outlined as well as common metaphors of the doctoral experience.

Ethics Processes for Education in UK Universities
Each university has its own ethics protocols, supplemented by discipline-specific guidance and overseen and upheld by an ethics review board (ERB) or panel. These processes have been enacted to protect participants from exploitative research, addressing many notorious examples of malpractice (e.g., [2]) and to protect universities from reputational damage and legal action [3]. In education, the British Educational Research Association (BERA) guidelines (2018) [4] are one such supplementary set of guidelines used by researchers as standard. In all universities in the UK, ethical approval must be sought before collecting data from human participants, with approval dependent on coherent plans for research design, consent, anonymity, avoidance of harm, data storage and dissemination. As such, ERBs have become the 'gatekeepers of research' [5] (p. 2), as studies cannot go ahead without their approval. Critiques of ERBs point to differences between what is in the best interests of researchers and participants, and what can be neatly included in a review process [6]. Arguments for and against the work of ERBs demonstrate the difficult balance to be struck between institutional oversight, and researcher freedom and workload [7,8]. Indeed, it is suggested that researchers would benefit from collaboration with co-researchers/teams and ERBs, throughout the life of a project [3], leading to a more dialogic approach, which would benefit everyone involved. In the UK, ERBs do not commonly work in this way, but in some Scandinavian countries, this is normal practice in the social sciences [3].
However, as a doctoral candidate, it is usual for supervisors to support the ethical approval process, as the student plans their study and considers ethical issues before making an application for approval. Clearly, the support given to individual candidates is not homogenous and experiences differ [9][10][11], but the ethics approval process represents a turning point. As Rowley [12] puts it, 'the student then had the green light to proceed with their doctorate' (p. 20). What is clear is that ethical approval is often perceived as a hurdle to negotiate, so that the track to doctoral completion can then be smoothly followed.

Doctoral Researchers in Education
Doctoral researchers are an under-researched group [10], with scarce literature exploring the developmental stages before completion of the thesis [13]. Due to this, the experiences of doctoral students are not well understood [14]-this is despite their growing numbers; In 2020/2021, over 104,965 students were on postgraduate research programmes in UK universities, and roughly 5% are researching in the field of education [15]. Doctoral students in education departments may have a different experience of their doctorate compared to other disciplines, partly because education students are often part-time and self-funded [9], and may, therefore, be juggling employment, family and caring responsibilities. Part-time doctoral study is also under-researched [16], but given the commitment of time and finances involved, it is not unusual for candidates to pursue an area of personal importance [17,18]. My own doctoral project on teacher mothers was an 'insider' project; I was a teacher mother myself. All research, and particularly insider research, has the potential to be emotional [19].
To mitigate the impact of this emotional work, Sikes and Hall [19] recommend that we (as researchers) are aware of the risks and protect ourselves, but also that we should be aware that insider research may not seem potentially hazardous, but 'because they touch on our own and participants' lives' (p. 170) can have emotional impact. Sikes and Hall also share the benefits they found in working together and mutual emotional support, and suggest that this is helpful when working in areas that 'come too close for comfort' [19] (p. 170). What literature does exist on doctoral students suggests that support from supervisors and loved ones is key to well-being [9,20]. Mantai [13] points out the importance of feeling successful for doctoral students, yet notes that developing as a researcher is still framed as a series of concepts to be understood [13].

Metaphors of Doctoral Programmes
Part of the difficulty I experienced with institutional ethics came about because I had an unrealistic view of the doctoral journey, based on commonly shared metaphors. To embed ethical thinking into the doctoral experience in this paper and my own reflections necessitated the consideration of metaphors of the doctoral experience and, in particular, of the ethics approval process as a 'hurdle'. Gravett [21] discusses the common metaphors of the doctoral experience: a journey or pathway. The 'journey' is framed as a linear process where, if the rules are followed, the outcome is predictable [11,21]. Batchelor and Di Napoli [22] explain, 'traditionally, the experience of doctoral studies has been reified as a fairly compact process characterised by a set of aims, rules and expectations.' (p. 18).
The assumption underpinning such a framing is that the doctorate is a linear process, defined by transparent and fixed rules. This characterisation chimes with my conception of the doctoral journey in the early part of my studies, a race down a straight track, with occasional hurdles to jump, which are negotiable if the rules are followed; see Figure 1. Ethical approval is one hurdle and can become 'perfunctory, formulaic and procedural' [23] (p. 1) in the mind of the student researcher. Educ. Sci. 2022, 12, x FOR PEER REVIEW 3 of 13 fairly compact process characterised by a set of aims, rules and expectations.' (p. 18).
The assumption underpinning such a framing is that the doctorate is a linear process, defined by transparent and fixed rules. This characterisation chimes with my conception of the doctoral journey in the early part of my studies, a race down a straight track, with occasional hurdles to jump, which are negotiable if the rules are followed; see Figure 1. Ethical approval is one hurdle and can become 'perfunctory, formulaic and procedural' [23] (p. 1) in the mind of the student researcher. This paper seeks to define a more realistic metaphor of the experience of doctoral research for many researchers who do not recognize the smooth-track metaphor, as well as considering what universities could do to support doctoral students.

Materials and Methods
This paper focuses on my navigation of doctoral research. Therefore, it utilises research journaling and critical incidents analysis, which emerged from my reflexive work as part of my doctoral experience. These approaches are used together to identify pivotal and significant situations from the doctoral project data collection-here, two are selected but many incidents occurred, involving insider research, use of social media, how to report personal stories and researcher self-care [24]. Over 650 journal entries were made over the doctoral programme. The two selected entries represent two extremes of my experience-one is an incident that stayed with me for a long period, requiring discussions and changes to research design to address. The other put me in a short-term vulnerable position, which was quick to resolve but emotionally exhausting for that short period. These extremes were selected to maximise the potential for recommending support for doctoral candidates from universities and ERBs. Firstly, however, I outline the research design developed for my doctoral study to give context to this exploration of ethical complexity. Further information, including the full analysis of interview data and discussions of insider research, positionality and reflexivity, can be found in the thesis [24].

Doctoral Study Research Design
The doctoral study data collection was carried out in 2018/2019, before the COVID-19 pandemic and the methods reflect the orthodoxy of face-to-face interviewing. The This paper seeks to define a more realistic metaphor of the experience of doctoral research for many researchers who do not recognize the smooth-track metaphor, as well as considering what universities could do to support doctoral students.

Materials and Methods
This paper focuses on my navigation of doctoral research. Therefore, it utilises research journaling and critical incidents analysis, which emerged from my reflexive work as part of my doctoral experience. These approaches are used together to identify pivotal and significant situations from the doctoral project data collection-here, two are selected but many incidents occurred, involving insider research, use of social media, how to report personal stories and researcher self-care [24]. Over 650 journal entries were made over the doctoral programme. The two selected entries represent two extremes of my experienceone is an incident that stayed with me for a long period, requiring discussions and changes to research design to address. The other put me in a short-term vulnerable position, which was quick to resolve but emotionally exhausting for that short period. These extremes were selected to maximise the potential for recommending support for doctoral candidates from universities and ERBs. Firstly, however, I outline the research design developed for my doctoral study to give context to this exploration of ethical complexity. Further information, including the full analysis of interview data and discussions of insider research, positionality and reflexivity, can be found in the thesis [24].

Doctoral Study Research Design
The doctoral study data collection was carried out in 2018/2019, before the COVID-19 pandemic and the methods reflect the orthodoxy of face-to-face interviewing. The original study took a postmodern feminist approach epistemologically, and the research was qualitative, utilizing a narrative methodology and a Foucauldian discourse analysis [24]. The ethical approval process is also outlined, as this is significant to the results and discussion of doctoral experiences.

Ethical Approval
Ethical approval was sought for a pilot study, which was managed internally on the doctoral programme, followed by full ethical approval for the main study by the ERB. The university used an online ethics platform, with sections for all areas under consideration in terms of ethics (e.g., consent, anonymity, data storage) and allowed for the ERB and supervisors to assess the application remotely and to give feedback, request additional information and reject/accept applications. Initially, ethical approval was not granted due to lack of information given about how participants could withdraw their interview transcript from the data. These issues were resolved in a second application (for further information see [24]. A further amendment to the ethics approval, requesting the use of social media for participant recruitment, was granted before the data-collection period began.

Participants
Five participants were recruited through social media. An advert was posted on Twitter with contact information and a summary of the study [24]. Recruitment via social media can engage participants who are hard to reach and unlikely to take part through other recruitment methods [25]. This method of recruitment may introduce demographic selection bias [26], particularly when the researcher's profile gives a strong indication of key indicators (such as motherhood).

Methods
Five unstructured life-history interviews began with one question; 'what is it like to be a teacher-mother?'. The question was developed after engagement with the literature on unstructured interview techniques [19,24], to give participants as much freedom on how to tell their story as possible. Locations of interviews were arranged before meeting, depending on the preferences of each participant, following their written consent and access to the research information sheet. My only stipulation was that it should be a public place, which was part of the ethics approval for the study to support safe researcher/participant relationships [4].

Reflection on Researcher Experiences
The process of reflecting on the doctoral project began in 2016, with a particular focus on the ethical dilemmas and considerations emerging once data collection had commenced. Reflection was used both as a tool for keeping track of my own thoughts and feelings, but also in an effort to be reflexive, a valuable practice in ensuring work is ethical [27][28][29]. Writing the journal allowed for me to consider more deeply the process I was going through; 'writing is a way of allowing meaning to emerge. In this way it allows embodied, emotional and intuitive knowledge' [30] (p. 173).

Methods
Reflection on experiences began at the start of the doctoral programme. It was written in an informal diary style, with some sections that referred to experiences from beyond the doctoral programme and theorising more widely than the situation at the time. Journal writing occurred sporadically; during data collection, there were sometimes multiple entries in one day, while at other points in the doctorate, weeks passed with only one to two entries. Over 650 research journal entries in total were amassed over four years. The aim of the journal was initially to aide reflexivity [27]; it also became a record of more practical issues. Selected journal entries are reproduced verbatim in the Results section of this paper.
Reflexivity will not 'automatically strengthen the credibility of an account' [31] (p. 9). However, as Pillow [27] discusses, there is no harm in attempting to be reflexive, and there is potential to improve our understanding of a situation. Folkes [28] urges us to reject the 'shopping list' of reflexive activities (e.g., listing commonalities between the researcher and researched) and to look to a 'kitchen table' reflexivity that includes informal talk and reflections between activities. Pillow [27] advocates an uncomfortable reflexivity and this became part of my experience of the research. Uncomfortable reflexivity refers to is a continued process of reflecting on ones' biases, limitations and oversights, in an effort to avoid reflexivity becoming a cosy process of self-congratulation. What follows is an extract (Figure 2) regarding the journaling experience: issues. Selected journal entries are reproduced verbatim in the Results section of this paper.
Reflexivity will not 'automatically strengthen the credibility of an account' [31] (p. 9). However, as Pillow [27] discusses, there is no harm in attempting to be reflexive, and there is potential to improve our understanding of a situation. Folkes [28] urges us to reject the 'shopping list' of reflexive activities (e.g., listing commonalities between the researcher and researched) and to look to a 'kitchen table' reflexivity that includes informal talk and reflections between activities. Pillow [27] advocates an uncomfortable reflexivity and this became part of my experience of the research. Uncomfortable reflexivity refers to is a continued process of reflecting on ones' biases, limitations and oversights, in an effort to avoid reflexivity becoming a cosy process of self-congratulation. What follows is an extract (Figure 2) regarding the journaling experience:

Critical Incidents Analysis
Critical incidents analysis was used when considering my research journal entries and experiences and was not used with the original study participants or interview transcripts. This technique evolved from World War 2 strategies for learning from RAF pilot 'near misses' [32]. However, in education, this strategy has evolved to be used in the classroom: Critical incidents, therefore, are not necessarily sensational events involving a lot of tension. Rather they may be minor incidents, small everyday events that happen in every school and in every classroom. Their criticality is based on the justification, the significance, and the meaning given to them. [33] (p. 431) Critical incident analysis (CIA) begins with the identification of a key incident, however insignificant it may seem to an outside observer. Following Tripp's process, the researcher considers the incident using a variety of questioning and thinking strategies: dilemma identification, personal theory analysis and ideology critique [34]. These strategiess, respectively, help to identify the dilemma, reflect on personal theories that underpin the reaction to the dilemma, and then the ideologies that shape the way a person views the world around them, as evidenced by the incident [35]. There may be a number of personal theories that underpin an incident, and Tripp acknowledges that theories can be complex-it may not be possible to identify all of the theories that are contributing to the reflections on the incident. Critique of CIA tends to centre on the feasibility of use for busy practitioners and difficulties in the definition of 'critical' [36]. However, the thinking strategies employed here are dependent on personal reflections that were identified as critical at the time.
Journal extract: Keeping a research journal Many of these (journal entries) included 'meta-entries' about the business of keeping a research journal. Keeping a research journal has been like picking a scab, over and over and over and over again, until it seems certain that a scar is unavoidable. The worst feeling is that some of the ethical scabs that I keep picking will never heal and I will just have to learn to live with them, in the work, in the thesis, in whatever I do next. Why did I think keeping a journal would solve every issue?

Critical Incidents Analysis
Critical incidents analysis was used when considering my research journal entries and experiences and was not used with the original study participants or interview transcripts. This technique evolved from World War 2 strategies for learning from RAF pilot 'near misses' [32]. However, in education, this strategy has evolved to be used in the classroom: Critical incidents, therefore, are not necessarily sensational events involving a lot of tension. Rather they may be minor incidents, small everyday events that happen in every school and in every classroom. Their criticality is based on the justification, the significance, and the meaning given to them. [33] (p. 431) Critical incident analysis (CIA) begins with the identification of a key incident, however insignificant it may seem to an outside observer. Following Tripp's process, the researcher considers the incident using a variety of questioning and thinking strategies: dilemma identification, personal theory analysis and ideology critique [34]. These strategiess, respectively, help to identify the dilemma, reflect on personal theories that underpin the reaction to the dilemma, and then the ideologies that shape the way a person views the world around them, as evidenced by the incident [35]. There may be a number of personal theories that underpin an incident, and Tripp acknowledges that theories can be complex-it may not be possible to identify all of the theories that are contributing to the reflections on the incident. Critique of CIA tends to centre on the feasibility of use for busy practitioners and difficulties in the definition of 'critical' [36]. However, the thinking strategies employed here are dependent on personal reflections that were identified as critical at the time.

Results and Discussion
The results shared here are the outcomes of reflections of researcher experience and use journal extracts to explain ethical dilemmas that occurred during the data collection and analysis for the doctoral study. Tripp's analysis approaches have been applied to two ethical dilemmas that arose during my data collection, which had not been anticipated at the ethics approval stage. Using Tripp's critical incident analysis, the same incident will be reflected on using different tools, refined by Tripp [35] with a view to understanding the dilemma in a wider perspective [37]. Due to word limit constraints, this analysis has been condensed in the Results section; however, for clarity, dilemma identification, personal theory and ideology critique have been applied to both incidents (for further explanation of critical analysis thinking strategies, refer to [34]).

Critical Incident 1: Social Media 'Likes'
The critical incident described below (see Figure 3) pertains to an ethical dilemma that had not been addressed during ethical review or raised by the research ethics panel, nor had I anticipated this or the other specific issues that subsequently occurred with social media use [38]. Whilst ethical review may consider participant recruitment issues, such as under and over-subscription, participant suitability and sharing of information [39], the specific and unique considerations that are introduced by social media have not been sufficiently understood by ethics review boards [40].
the dilemma in a wider perspective [37]. Due to word limit constraints, this analysis has been condensed in the Results section; however, for clarity, dilemma identification, personal theory and ideology critique have been applied to both incidents (for further explanation of critical analysis thinking strategies, refer to [34]).

Critical Incident 1: Social Media 'Likes'
The critical incident described below (see Figure 3) pertains to an ethical dilemma that had not been addressed during ethical review or raised by the research ethics panel, nor had I anticipated this or the other specific issues that subsequently occurred with social media use [38]. Whilst ethical review may consider participant recruitment issues, such as under and over-subscription, participant suitability and sharing of information [39], the specific and unique considerations that are introduced by social media have not been sufficiently understood by ethics review boards [40].
Guidelines may suggest that researchers use official email addresses and university contact details, but social media has made researcher 'distance' difficult to maintain. In analysing this critical incident, I am transported back to the time and the very real sensation of having part of my life revealed to someone who I felt should be kept at a professional distance. Personally, this was a relatively small incident in the scheme of things, but it was the first time something had happened that had not been considered at all in the ethical approval process and this in itself felt shocking. I began asking questions about why this had happened, as the journal extract reflects, and more substantial analysis using Tripp's thinking strategies follows.
Dilemma Identification is clear-should I continue to recruit this prospective participant, if so, how to I negotiate this situation? Should I go back to the ERB to gain approval for however I decide to tackle this dilemma? In short, the dilemmas identified in this incident make clear some of the ethical challenges inherent in recruiting participants online, which had not been clear in my ethics approval documentation, or in the feedback from Guidelines may suggest that researchers use official email addresses and university contact details, but social media has made researcher 'distance' difficult to maintain.
In analysing this critical incident, I am transported back to the time and the very real sensation of having part of my life revealed to someone who I felt should be kept at a professional distance. Personally, this was a relatively small incident in the scheme of things, but it was the first time something had happened that had not been considered at all in the ethical approval process and this in itself felt shocking. I began asking questions about why this had happened, as the journal extract reflects, and more substantial analysis using Tripp's thinking strategies follows.
Dilemma Identification is clear-should I continue to recruit this prospective participant, if so, how to I negotiate this situation? Should I go back to the ERB to gain approval for however I decide to tackle this dilemma? In short, the dilemmas identified in this incident make clear some of the ethical challenges inherent in recruiting participants online, which had not been clear in my ethics approval documentation, or in the feedback from the ERB. I had not considered the implications of using social media accounts for research purposes, nor had it occurred to me that I would have an emotional reaction to a participant accessing my social media content. Through the ERB process, practical matters such as how to contact participants, distribute project information and ensure consent was gained in an informed manner were interrogated, but the specific issues around social media were not.
Personal Theory in this dilemma has several aspects. First, there is a personal theory at work about how researchers and participants are meant to behave; this incident has disrupted that theory; the prospective participant has 'crossed a line' in finding out more about the researcher than was intended, but all the information they gathered about me was publicly available. Another personal theory, and the one that clearly has been shaken the most by this dilemma, is the theory that the ethical review process and approval to begin data collection meant that no further ethical thinking would be required or, at least, if ethical issues arose, they would be documented and planned for in the ethical approval documentation.
My perception of risk was entirely focused on my participants-and it is likely that ERBs, supervisors and students themselves may not consider studies like mine as ethically 'risky'; the risks may not be apparent on first consideration and when every effort is being made to protect the participants, the researcher's safety can be easily overlooked in the process. In my own experience of ethical approval, the ERB were justifiably concerned about participants' ability to withdraw their data. Beyond considering meeting places and physical safety, the vulnerability of the researcher was not the priority; the emotional well-being and potential harms for the researcher are perhaps considered beyond the remit of the ERB in this situation.
At times, the ethical consideration of 'do no harm' was not an idea I was applying to my own well-being and emotional load. University and programme policies could support self-care; as Sikes and Hall [19] attest, there is 'no shame in looking after our own well-being as we would that of participants when the personal costs do become too harmful' (p. 169). However, as they acknowledge, our passion as researchers to complete our work and improve conditions for the groups we are working with means that often we see the pain as part of the work and a price that is worth paying. This is where university processes could counter the narratives of the selfless researcher, and particularly where projects constitute a 'labour of love'. ERBs could work with students and supervisors to identify where researcher safety should be discussed in further detail, not as a barrier, but to ensure that researcher mental health is considered as well as physical safety.
The resolution of this dilemma came about through 'kitchen table' [28] discussions with my supervisors, who calmly talked me through the options, and allowed for me to choose the route I was most comfortable with. We considered how social media recruitment does have different ethical considerations, which went against some of the guidance in the literature at the time [38]. The bigger issue, of an ideology about academia, took longer to resolve, and writing this paper is perhaps a part of that process.

Critical Incident 2: Researcher Breakdown
The second critical incident is another situation that had not been anticipated in advance, either by the ERB or me. It differs from the first in that it potentially put me in physical danger, an issue which is well covered in the ethics guidelines [4] and was a section of my ethics approval documentation. I had not considered the emotional impact of insider research-this was the last of five interviews and whilst I had felt moved by all of the participant stories, I can only describe this as having 'the wind knocked out of me', as the following extract (Figure 4)   Reflecting on this incident, this was significant in terms of my safety as a researcher. I was far from home, at night, feeling unsafe to drive my car. Fortunately, this feeling passed, and I managed to get home safely, but the fear I experienced and the feeling that I had embarked on something foolishly and without consideration of the possibilities, is still fresh in my mind. My dilemma at that moment was whether I could get home safely

Journal Extract: Carpark
Another unexpected emotional event. The interview tonight was too much; it went really well, I think -but I walked away feeling a massive weight on me. When I got into the car, with a two hour journey (at least) to get home, I just cried and cried -I couldn't start the car until I stopped (crying). It was overwhelming that this story is now in my bag, on a recorder -I have been handed it, this unique telling, with an understanding that I will do something worthwhile with it. But I can't help feeling that I don't deserve it. It's too much. I had no idea I would feel like this about an interview with a fellow teacher mum…I wasn't prepared at all. Reflecting on this incident, this was significant in terms of my safety as a researcher. I was far from home, at night, feeling unsafe to drive my car. Fortunately, this feeling passed, and I managed to get home safely, but the fear I experienced and the feeling that I had embarked on something foolishly and without consideration of the possibilities, is still fresh in my mind. My dilemma at that moment was whether I could get home safely or not, and whether I could manage the weight of responsibility for handling the data in the weeks that followed. Whilst all ethical procedures and approval required by the university ethical review board (ERB) had been met for this doctoral study, the data collection threw up dilemmas like this one, which left me quite literally stranded. The situation was intensified for me by my own expectations of the doctoral journey-the ethical approval hurdle had been negotiated, and, hence, data collection should not pose any problems. This suggests that for me at least, the view of ethical approval as a minor hurdle is problematic.
The formal ethics processes did not constitute the overcoming of a hurdle; they were a subtle foreshadowing of the much bigger obstacles that would need to be overcome in the years that followed-see Figure 5. The conception of ethics as a minor hurdle was part of a wider fallacy, of the doctoral journey being a smooth race-track [21]. For me, the journey had unexpected turns and events, multiple crossroads and decisions to be made when the way through to an endpoint was not clearly defined. It felt like trying to find a destination in a strange city, when the destination was not named, the city was upside down and my feet were chained together. Expecting ethics review processes to account for this complexity was unrealistic, but the process of applying for and being granted ethical approval perhaps feeds this theory of a hurdle; there is a 'sign off' that has great meaning for the future of the project, rather than a sense of an ongoing negotiation or re-evaluation throughout the project [3]. It seems likely that a different ethics process (see section 4.1) for studies that are not biomedical controlled trials would help to counter the narrative that the doctorate is a straightforward track [21}.
Again, my supervisors were pivotal in steering me through this experience, but in the immediacy of the carpark, I turned to another 'insider'-my mum. Her calm insistence that my well-being was the priority, as well as her reassurance that I would find a way to do justice to the stories collected, meant that I could get home safely. Expecting ethics review processes to account for this complexity was unrealistic, but the process of applying for and being granted ethical approval perhaps feeds this theory of a hurdle; there is a 'sign off' that has great meaning for the future of the project, rather than a sense of an ongoing negotiation or re-evaluation throughout the project [3]. It seems likely that a different ethics process (see Section 4.1) for studies that are not biomedical controlled trials would help to counter the narrative that the doctorate is a straightforward track [21].

Conclusions
Again, my supervisors were pivotal in steering me through this experience, but in the immediacy of the carpark, I turned to another 'insider'-my mum. Her calm insistence that my well-being was the priority, as well as her reassurance that I would find a way to do justice to the stories collected, meant that I could get home safely.

Conclusions
Key recommendations from this personal reflection on doctoral dilemma experiences are:

•
ERBs and approval processes provide reassurance, but they are not designed to consider every eventuality and should be regarded as a safeguard, not a guarantee.
Other processes are needed to support doctoral researchers post-approval.

•
Normalising the occurrence of ethical dilemmas in doctoral research could bring about a research culture of shared dilemmas, around an 'uncomfortable kitchen table'.
It should be noted that conclusions and recommendations are based on the personal reflections of one researcher.

Supporting Doctoral Students-ERBs and Supervisors
The ERB at my university followed guidelines for ethical conduct, but my expectation that their approval denoted a guarantee of ethical safety, with no unexpected incidents ahead, was unrealistic. The ERB achieved what they set out to do, in protecting the university and ensuring that the classic elements of biomedical research ethics were covered-but, as previously discussed, the ERB process was not enacted to identify future dilemmas or to support me through these. This is where my supervisors, friends and family became essential.
Being able to discuss issues as they arose was invaluable for my emotional well-being and my supervisors spent much more time working with me on ethical dilemmas than they were allocated. Whilst it may be tempting to recommend that doctoral supervisors spend more time with students on ethics beyond the application process, this time must be included in the supervisor workload to avoid the problem simply moving to another group of people. Ongoing discussion and an 'open door' policy on ethical issues and review was provided to me unofficially, partly because of the ethic of care my supervisors had for me, but making these benefits available to all researchers, and properly costed and allocated, would go some way to alleviating the mental health and well-being crisis amongst doctoral students and the wider research community. Busher and Fox [3] suggest practical ways that ERBs could be part of this process, leading to a much deeper understanding for all concerned, including the suggestion of an apprenticeship model of collaboration between researchers and ERBs. This paper concerns a personal account of ethical approval in one institution; the collaboration of ERBs is dependent on the systems that facilitate this, and concerns around methodological and disciplinary clashes are beyond the scope of this paper, but discussed comprehensively elsewhere [41][42][43].
I have constructed an ongoing offer of an 'uncomfortable kitchen table' discussion, from the combined work of Pillow and Folkes [27,28] whereby ethical incidents and issues are aired openly, and discussed and hopefully resolved as a functioning family unit would tackle a crisis. Painful ideas can be shared in safety, leading to acceptance of the issues, pragmatic advice being shared, and possible solutions rejected, fostered and discussed further. This takes time and strong relationships with the people who are 'around the table' (see Figure 6). The 'uncomfortable' aspects of the discussion should, in my opinion, not be shied away from; therefore, the name of the process also needs to be honest, even if this may initially sound unappealing. For the doctoral student facing ethical dilemmas, the situation is already potentially very uncomfortable.

Using Critical Incidents as Discussion Points
Critical incidents analysis and other reflective tools could also be shared with doctoral students in the ethics training offered by their university, to support their dilemma identification skills and provide a deeper understanding of how their ethical challenges can be resolved. In my personal experience, the ethics training offered was practical and referred to the processes and requirements of the university, whilst reflection on ethical dilemmas and tools for reflection were left for supervisors and students to address in individual contexts. I was lucky to have the supervisors and external support that I needed; offering some practical tools to deploy in critical situations may help those who have not.
The limitations of this study include my being restricted to how many dilemmas could be shared, and the personal, subjective view of an ERB and doctoral programme in one UK institution, which obviously could be very different for other students. Further areas for research, following this study, include the use of critical incidents analysis to support other doctoral students with their ethical dilemmas, and exploration of new metaphors for doctoral programme experience. Further research to inform wider doctoral debates could collect data from doctoral cohorts on their experiences of ERBs, supervisory relationships and research cultures.

Using Critical Incidents as Discussion Points
Critical incidents analysis and other reflective tools could also be shared with doctoral students in the ethics training offered by their university, to support their dilemma identification skills and provide a deeper understanding of how their ethical challenges can be resolved. In my personal experience, the ethics training offered was practical and referred to the processes and requirements of the university, whilst reflection on ethical dilemmas and tools for reflection were left for supervisors and students to address in individual contexts. I was lucky to have the supervisors and external support that I needed; offering some practical tools to deploy in critical situations may help those who have not.
The limitations of this study include my being restricted to how many dilemmas could be shared, and the personal, subjective view of an ERB and doctoral programme in one UK institution, which obviously could be very different for other students. Further areas for research, following this study, include the use of critical incidents analysis to support other doctoral students with their ethical dilemmas, and exploration of new metaphors for doctoral programme experience. Further research to inform wider doctoral debates could collect data from doctoral cohorts on their experiences of ERBs, supervisory relationships and research cultures.