How do higher degree research students and supervisors navigate ethics- in-practice for educational research in sensitive or ‘fragile’ contexts?

The past decade has seen increased attention paid to the ethical complexities of educational research undertaken in sensitive or ‘fragile’ settings, where trauma, marginalisation and socio-political precarity are prevalent. Yet, despite increased awareness of micro-ethical issues encountered in the field, there is limited research that engages with these issues from the perspective of higher degree research (HDR) students, and few studies that focus on supervisory practices to promote micro-ethical reflexivity. Here, we draw on interviews with HDR students and supervisors researching in the fragile context of forced migration and related settings of conflict and crisis, exploring issues of gendered violence, sexuality, cultural and linguistic marginalisation, and mental and physical well-being, to explore their experiences with micro-ethical complexities in field-work. We consider student and supervisor sense of preparedness to engage reflexively with micro-ethical challenges and identify key supports for navigating ethics-related dilemmas. Importantly, in exploring gaps in extant supports, we consider issues of individual, collective and institutional responsibility regarding HDR student and supervisor engagement with micro-ethics, posing key questions about duty of care for novice researchers working in fragile or sensitive contexts


INTRODUCTION
The past decade has seen increased attention paid to the ethical complexities of research undertaken in what Fox et al. (2020) describe as 'fragile contexts', or settings where participants have experiences of past or ongoing trauma, marginalisation and socio-political precarity. 1In this paper, we focus on research in the fragile context of forced migration and related settings of conflict and crisis, addressing issues of mental and physical well-being, gendered violence, sexuality and marginalisation by language or cultural isolation, which carry particular ethical considerations for researchers.Complexities with research in these fields may range from issues regarding informed consent (Hugman et al., 2011), to questions of beneficial outcomes for participants (Darling, 2014) and the appropriacy of conducting research with people who are detained and/or undergoing legal proceedings (Gillam, 2013).Guillemin and Gillam (2004) captured the navigation of these everyday complexities encountered in the field-or what Doná (2007) describes as 'micro-ethics'-as a point of distinction from 'big issue' bioethics (p.265).
While the 2004 article that first marked out this distinction between these approaches to ethics is well renowned (with over 3750 citations at the time of writing), the nuance of these divergent approaches remains a largely academic, rather than a practical, distinction.Navigating complex micro-ethical decisions and dilemmas in the field can be particularly challenging for new researchers, such as higher degree research (HDR) students, and for researchers who are new to a particular fragile context (Fox et al., 2020).Moreover, depending on the nature of their own research, HDR supervisors may have limited experiential knowledge of how to anticipate and navigate micro-ethical moments (Astill, 2018;Krause, 2017).Yet, there is a marked paucity of research focusing on the preparation of HDR students to engage with key issues regarding the ethical conduct of research during and on leaving the field, and even less attention paid to these issues within fragile contexts, where complex and multi-faceted issues of precarity and marginalisation are prevalent.
This paper explores the findings of an Australian study of HDR students and supervisors undertaking research in fragile contexts, including topics related to displacement and resettlement of people with refugee and asylum seeker experiences, gendered violence, sexuality, cultural and linguistic marginalisation, and related issues of mental and physical well-being.The study emerges from our experiences and engagement with ethics-inpractice as former HDR students, and now as scholar-practitioners and supervisors guiding novice researchers through the complex ethical challenges they encounter when undertaking studies in fragile contexts.The project addresses the following research questions: 1. What micro-ethical issues do HDR students and supervisors encounter when researching in fragile contexts?2. How do supervisors encourage the development of micro-ethical reflexivity during HDR candidature?3. What key supports assist HDR students and supervisors to navigate micro-ethical complexities in research and what additional supports are required?

CONCEPTUALISING APPROACHES TO RESE ARCH ETHICS
Prior to presenting the research design and exploring the findings regarding HDR student and supervisor experiences with micro-ethical dilemmas, it is pertinent to consider some of the main ways in which research ethics are conceptualised in the literature.In their foundational article, Guillemin and Gillam (2004) propose two divergent approaches to research ethics; first, identifying the dominant 'procedural ethics' approach that aligns with gaining prior ethical approval from a Human Research Ethics Committee and second, 'ethics-inpractice' as the ethical decision-making regarding the most appropriate ways of proceeding with research whilst in the field.

Procedural ethics and ethics-in-practice
Procedural ethics assumes that researchers have predicted and pre-empted any ethical challenges and mitigated the risks identified in the way they plan to enter the field.Arguably, this leaves relatively little space for flexible, 'on-the-ground' decision-making when microethical dilemmas emerge, as they frequently do, in the complex messiness of human research (Fox et al., 2020).With procedural ethics, Human Research Ethics Committees are the holders of power-they sanction, surveil and control the research decisions.These committees are concerned with efficient processing, meaning they adhere closely to the remit of focusing on internal ethical aspects, and protecting the institution's reputation and limiting its risk to litigation (Adler & Adler, 2016;Gillam, 2013;Montero-Sieburth, 2020).Increasingly, scholar-practitioners are exploring the potential for Human Research Ethics Committee inprogress and annual reporting procedures to play a greater role in supporting researcher engagement with ethics-in-practice (see Fox & Busher, 2022;Fox & Mitchell, 2019).However, for logistical reasons, such committees cannot oversee all aspects of ethics-in-practice; but researchers can and should (Guillemin & Gillam, 2004).In this way, Hugman et al. (2011) contend that … social researchers [should] not regard ethics simply as a technical matter … nor should ethics be seen as something to be left to experts (although there may be scope for expertise to assist and support colleagues) but it must be regarded as everybody's business.(p.15) As Gillam (2013) notes, Human Research Ethics Committees are unlikely to have specific expertise regarding all possible fragile contexts, such as research with refugee and asylumseeker communities, meaning the burden is placed on the researcher to provide evidence of previous research practices or arguments for taking a specific approach to community engagement.As such, Gillam (2013) argues that 'HRECs [Human Research Ethics Committees] are not necessarily well equipped with the appropriate knowledge and experience to meaningfully carry out this interpretation for refugee or asylum-seeker research' (p.22) without being educated by those familiar with these contexts.
In contrast, Guillemin and Gillam's (2004) distinction between procedural ethics and ethicsin-practice recognises that researchers are capable of, and should be engaging in, ongoing ethical reflexivity (Block et al., 2012) to guide decision-making (in consultation with ethics committees where necessary).Ethics-in-practice approaches acknowledge the need for flexibility in the research design to leave space for researchers to respond to the daily complexities of researching with human participants, and the need for reciprocity to avoid veering into extractive and exploitative practices.Hugman et al. (2011) argue that such reciprocal research tries to facilitate more equal exchange of ideas and benefits to be gained from the research, is open to different methodologies and methods, and demands dialogue.Spaces for this dialogue need to be created (see Mackinlay & Barney, 2014), and supervisory dialogue has the potential to model engagement in the field between researchers, gatekeepers and participants (Fox & Mitchell, 2019).Similarly, ethics-in-practice approaches foreground the need to think carefully about ongoing, continual negotiation of consent and flexibility in temporal and design terms to be able to respond to micro-ethical concerns when 'entering', 'in' and on 'leaving' the field (Mackenzie et al., 2007).Indeed, Halilovich (2013) explains that researchers need to be able to develop 'customised' methodologies and methods to suit the needs of research communities.Such bespoke and flexible approaches require attention to the power dynamics of the interaction and 'the pre-empting of potential ethical problems before they take hold' (Guillemin & Gillam, 2004, p. 276), recognising the imbalance of power and, with this, the potential for the devaluing of knowledge in the research community (Wood, 2017).Ethical reflexivity is thus an extension of ethics-in-practice, requiring researchers to remain vigilant to power and status dynamics, to create conditions that permit flexibility in the research design and to develop an ethics of care, which involves compassion and empathy (Christie, 2005;Clark-Kazak, 2017;Doná, 2007).An ethics-in-practice approach expects researchers to 'engage with the ethical dimensions of ordinary, everyday research practice' as well as 'ethically important moments' (Guillemin & Gillam, 2004, p. 276).Further, research and knowledge production are undertaken with (not on) participants, who are viewed as agents rather than objects or subjects of research, with respect to their values, knowledge, expectations and needs.Importantly, research relationships are characterised by mutual respect, care and interdependence rather than distance or hierarchy (Block et al., 2012).The kinds of ontological transitions recommended for researchers-such as shifting to acknowledgement of differential and flexible ethical processes and taking an ethics-in-practice approach to entering, remaining in and leaving the field-are complex, unsettling and require support.It is therefore imperative that we investigate how HDR students learn to engage with ethics-in-practice in complex or fragile contexts of research and identify the resources and supports that are fundamental to their engagement.

METHODOLOGY
This study adopted a survey design following a desk-based review of relevant literature and Australian university websites pertaining to research ethics advice, particularly relating to support for HDR students researching in 'fragile' contexts.Following Human Research Ethics Committee approval (Protocol No. H-2020-0383), we conducted semi-structured interviews with 18 students and supervisors undertaking research in various fragile or sensitive contexts (see Table 1).Participants were recruited by Authors 1-4 via scholarly networks and professional associations and it was these authors who participated in the interview 14693518, 0, Downloaded from https://bera-journals.onlinelibrary.wiley.com/doi/10.1002/berj.3945 by The Open University, Wiley Online Library on [15/01/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License processes.Participants who were known to one or more of the researchers were given the option of being interviewed by another member of the team.
After careful consideration of the ethical issues that may arise when interviewing students and supervisors together, we invited participants to choose whether to be interviewed on a one-to-one basis or with their supervisor(s)/student.While there was no requirement for both the supervisor(s) and the student to participate in the research, we wanted to provide the opportunity for collaborative reflection, given the potential for such experiences to add to critical engagement with ethics-in-practice (see Fox & Mitchell, 2019).All semi-structured interviews were conducted on a one-to-one basis, except for student Amber and her supervisors Frankie and Delia, who requested a group interview to explore shared experiences.Table 1 provides an overview of the participants, including a broad description of their fields of research.Only one supervisor, Teresa, was in the first 5 years following doctoral completion.All names are pseudonyms, either chosen by the participants or the researchers.
Our interview questions focused on participants' experiences of navigating micro-ethical dilemmas, their sense of preparedness to engage with such complexities and existing resources, as well as gaps in the provision of support.These questions were informed by our research focus, our experiences as scholar-practitioners and the major themes identified within the literature, including students' descriptions of precarity and isolation as novice researchers (see Astill, 2018;Dickson-Swift et al., 2008) and the lack of explicit engagement with micro-ethical issues within institutional resources designed to support students and supervisors (see Bettez, 2015;Field et al., 2021;Krause, 2017).Throughout the interviews, we often recounted our own experiences managing emergent ethical dilemmas as a way of encouraging participants to share their stories freely.As a research team, we prepared for the interviews by interrogating our individual and collective experiences to ensure we provided opportunities for participants to discuss issues and perspectives that contrast with our own understandings and practices.We have all undertaken doctoral studies and work as researchers in Australian (Authors 1-4) and British (Author 5) universities.Authors 1, 2, 3 and 5 have supervised through to completion; as an early career researcher, Author 4 is starting her journey as a supervisor.Many of our HDR students have explored issues in fragile contexts.
Following professional transcription and member checking, inductive content analysis, including inter-researcher reliability checking, was undertaken to identify significant themes in the textual corpus (Berg, 2001).Due to the exploratory nature of the study, inductive content analysis was utilised, rather than employing pre-conceived themes informed by existing research.At the analysis stage, we met regularly to discuss and refine emerging themes, with the aim of strengthening the reliability of our findings.
Here, we present our findings regarding the key micro-ethical challenges that students and supervisors reported encountering when researching in fragile contexts.These findings are discussed in relation to key themes in the literature to reflect on the significance of the learning experiences and supports that students and supervisors identify as important.Finally, we reflect on the ways in which supervisors encourage the development of students' micro-ethical reflexivity and discuss potential avenues for better supporting HDR students and supervisors working in fragile contexts.

RQ1: What micro-ethical issues do HDR students and supervisors encounter when researching in fragile contexts?
While there is a paucity of research focused on HDR engagement with ethics-in-practice in fragile settings, the literature is replete with accounts of the challenges that engagement with ethics can create.Field et al.'s (2021) reflective discussion of Field's doctoral research highlights tensions for novice researchers in navigating paradigms of research rigour that emphasise the minimisation of 'bias '-what Field et al. (2021) describe as the 'positivist pull' (p.12)-and the 'messiness' of responsive engagement in person-centred research.Following Naveed et al. (2017), who suggest that HDR students tend to 'feel unable to admit their challenges in the field when such struggles seem to be so rarely acknowledged in the wider community' (p.785), Field et al. (2021) advocate for open engagement with the 'messiness' and call for greater support for the development of micro-ethical reflexivity among novice researchers through discussion of associated complexities.
Our semi-structured interviews with HDR students and supervisors produced richly detailed commentary regarding the complex micro-ethical challenges of research in fragile contexts.Many of the issues identified by our participants reflect challenges discussed within the literature regarding the ethical conduct of research in a range of HDR settings, such as the demands on students' metacognitive development, evolving researcher identity/ies and supervisor/student roles and relationships.However, working in fragile contexts intensifies and further complicates these issues.Here, we explore four main areas of micro-ethical challenge identified by our participants: navigating relationships, consent, expectations and reciprocity.
Navigating the micro-ethics of research relationships in fragile contexts HDR students encounter myriad challenges when navigating research ethics, with arguably more profound difficulties for those working in fragile contexts.A common theme within our student-participants' commentary was the description of a 'journey' from conceiving of research ethics as an 'either/or situation', where actions are undertaken 'completely accurately or not', to engaging with the nuances and fluidities of research in fragile contexts, where 'it becomes a bit more about the day-to-day decision-making' (Safiyah, student).For some participants, this epistemological transition required navigation of tensions between positivist expectations of researcher objectivity and 'unbiased' engagement, and more humanistic understandings of the relationships and rapport central to consultative and community-led research.While not particular to fragile contexts, these issues are heightened in these settings due to increased participant precarity, experiences of trauma and greater power imbalances.
Similar to Field et al.'s (2021) argument about the 'positivist pull', Joan, a supervisor, identified how the assumption of '[being] objective [with] separation between you and your participant' is at odds with the realities of researching in fragile contexts.To illustrate, Joan drew on her own research experience by referring to how a participant in her longitudinal study requested that she continue visiting 'as a support person' following the completion of the project.Joan discussed the challenges of responding to this participant's request, explaining 'it was very much outside the bounds of the project, and it was really hard, there wasn't very much guidance about what to do'.Joan eventually determined '… the ethical thing to do as a good person' was to 'honour that relationship and keep going… I couldn't just leave her'.Joan commented on how the learning from this ethical reflexivity was a resource that she could offer when supporting HDR students to prepare for entering similar research situations.This understanding of research relationships 'based on a process and not seen as a single event' (Hugman et al., 2011(Hugman et al., , p. 1280) ) was identified by supervisors and students as necessitating careful forethought from the outset of HDR candidature.As supervisor Fatima noted, this involves supporting students to 'Think what's going to happen long term with the relationships you build [in the field]'.
Likewise, Amber, a student, described how she has become a key contact for families participating in her research, who look to her as a friend and source of assistance regarding life in Australia.Frankie, one of Amber's supervisors, explained: 'Amber has been in Australia for longer than them, she carries how to survive the system.So, she's become their go-to'.Touching on the complexities that may occur when researchers and participants experience tensions between shared cultural understandings and the expectations and assumptions embedded within institutional ethics procedures, Amber noted: '… I can't tell them, "Look, don't associate with me, part of my ethics and all that… in my culture if I did that, I would be ostracized"'.Likewise, Delia, Amber's other supervisor, reinforced the importance of understanding that researcher responsibility in fragile contexts means: I can't just go in there and say, "I'm here, sign the consent form.I'm going to interview you for 3 hours".I'm gone… but I think there are a whole group of supervisors that are supervising in these kinds of areas that don't understand that and are supporting students to flit in and flit out; those kinds of extractivist approaches.
The complexities of navigating research relationships in contexts where participants are often without broader social and specialist supports were also identified by participants as particularly challenging.Supervisor Ria described how 'in a normal research relationship, I guess you would do the interview and say, "Look, I'll touch base with you in six months for the next interview, take it easy, see you later"'.However, in fragile contexts where participants may be experiencing extreme and protracted hardship, Ria described the sense of responsibility to 'touch base' with participants more regularly to ensure they 'had somewhere to stay and [were] alive'.However, Ria also discussed the importance of protecting both participant and researcher well-being, describing how she arranged referrals to appropriate specialist services while maintaining clear professional boundaries: 'I kind of implanted the idea of, it's probably good for you to chat to someone apart from me, and someone who is specialised in this area'.
For students who are new to a particular fragile context and are without supervisor guidance, there can be limited resources to support the navigation of reciprocity in research relationships while safeguarding both participant and researcher well-being (Fox et al., 2020).As Sam (student) explained: 'the administrative ethics side doesn't always… translate to how it works in practice… in working with humans and emotions and relationships and people'.

Complexities of consent
While attaining participant consent is often conceptualised as a pre-study, once-only requirement in procedural ethics-a singular and linear process addressed through the provision of consent forms and plain language statements-this position has been criticised as failing to account for participants' complex and changing circumstances, particularly in settings of precarity and marginalisation (Darling, 2014).While procedural ethics protocols usually state that participants can withdraw from the research without providing a reason, the students and supervisors in our study emphasised that in fragile settings where participants' circumstances, visa status and family situations in countries of origin can change without warning, consent needs to be 'ever-present' (Miriam, supervisor).Miriam described the importance of regularly checking in with participants to determine: 'Are you still okay with that?You know what you have said there, is that still right… We thought we might use that [data] again, is that okay with you?', while also maintaining awareness of how global events, national policy changes or developments in participants' personal lives may impact the capacity to be involved with research.
The importance of considering appropriate channels for providing consent when working in fragile contexts was another key theme in our findings, with student Camille advising 'some people don't know what they're signing, it's not in their language.And I describe it, and I talk them through it, but again, you know, there's a lot lost'.Likewise, supervisor Joan advised that signed consent is 'not going to be appropriate for everyone, not everybody even has a signature, and it actually would do more harm than good'.In these instances, participants discussed the value of oral consent options and working with interpreters where possible.However, these supports may also introduce new complexities, including participant concerns with confidentiality, especially in the context of smaller communities where participants and interpreters may have mutual contacts.Further, oral discussion still does not guarantee that participants fully understand what is involved in research (Araali, 2011).Ultimately, issues of consent and communication in fragile research contexts raise the question: 'How much needs to really be understood, and how much is okay if [participants] don't understand?' (Camille, student).
The centrality of the relational in ethics-in-practice approaches is clear in our data.As student Safiyah advised: 'So we spend all this time crafting these carefully worded [participant information statements], and then it's just what I say to them when we're having coffee or chatting at an event.That really is the most significant thing.And it's based on the relationship they have with me'.These kinds of researcher experiences with procedural ethics documents-premised on a notion of unfamiliarity-call into question the effectiveness of such processes when coupled with sensitive research in fragile contexts where trust and relationships are key (Oyinloye, 2021).

Engaging with participants' expectations
Like the need for reiterative consent practices, students and supervisors researching in fragile contexts identified the importance of ongoing engagement with participant expectations.This involves researchers guarding against making assumptions and being prepared for 'unknown and unknowing possibilities' (Mackinlay & Barney, 2014) by being open to the lives and experiences of participants.As noted, the procedural ethics requirement for a participant information statement in plain language, translated and/or provided verbally, does not necessarily ensure that participants understand the potential impact of their involvement in research, and these issues may be heightened in settings where people are in situations of extreme precarity and are unfamiliar with government processes and social structures.For instance, ensuring that participants understand that their involvement in research will not impact their application for refugee status (positively or negatively) was identified as a common challenge, with student Camille recounting: I had one family who, when they were interviewing with me, were pleading with me to help them get their children to Australia, their kids are still in danger… I had to be saying, "Look, your involvement in this interview has got nothing to do with that.I can't do anything… yes, the research might contribute to policy changes down the track, but you're not going to get any immediate [outcome]… you could see their hearts drop….Camille's experiences, and those of participants in our study who recounted similar interactions, raise important questions regarding support for novice researchers, particularly those without supervisors who can anticipate such occurrences and prepare for 'in the moment' responses.As Camille emphasised, while Human Research Ethics Committee protocols require links to counselling services for all projects where there is potential for participants to be distressed by their involvement, the researcher may still find themselves required 'to get out of the situation you're in' and to ensure that participants are not 'even temporarily negatively affected from engaging in your research'.Currently, the responsibility for ensuring that students are prepared for such eventualities falls to supervisors, who themselves may also be unprepared to manage such micro-ethical moments (Butler-Rees & Robinson, 2020).

Reciprocity of research outcomes
Pursuing co-constructed, participatory approaches to research in fragile contexts also raised important ethical questions for students and supervisors in our study, particularly regarding the need to properly compensate co-researchers and participants with lived experience.Frequently, HDR projects are unfunded and/or students and supervisors may not foresee the need to budget for these provisions or understand the associated ethical issues.For instance, Tara (student) recounted how: … based on my readings, I thought financial incentives [for participants] were coercive.But my supervisor had a very strongly different opinion and she said, "You have to compensate them for their time, because they are disadvantaged groups, and they're also low-income backgrounds".
How to recognise participation (as opposed to incentivising) can be very difficult to prefigure; as Bailey and Williams (2018) note: '[i]t is not until the researcher is in the field that they can accurately assess the participants and establish what incentives would suit, and in what manner they should be provided' (p.364).Human Research Ethics Committees are unlikely to provide clear guidelines on what is appropriate, leaving it to the individual researcher to decide, which can be particularly daunting for students if they do not have a supervisor with experience of working in fragile/participatory contexts.Financial incentives are not the only form of reciprocity-other forms can include co-design of outputs or advocacy-but all are rife with complexity and difficult to predict accurately.
Ensuring reciprocity of beneficial outcomes from research for both participants and researchers was a common concern for our participants.As evident in student Sam's reflexive engagement with her own investment and motivation in pursuing her research, these concerns are heightened by the deeply personal and sensitive nature of the topics under investigation: … why am I doing this research?… am I asking people to give a really difficult part of their lives… so I can get a PhD?It's such an important topic, and people are sharing such sensitive and difficult times of their lives that I want to make it worthwhile… But I don't know what the greater benefit is, but if there's a piece of research around the impact of offshore detention, then I'm hoping that's a good thing.
Having a supportive and collegial space in which to critically explore responses to these and other questions regarding research outcomes was identified as an important resource for students and supervisors.At the same time, our participants highlighted the need to balance the requirement to ensure participation is not overly burdensome for co-researchers with lived experience, with the need to question assumptions about capacity.Describing her initial experiences undertaking participatory research in the fragile context of forced migration, Miriam (supervisor) explained: 'I learned a lot about people's capacity for agency and how I had to respect that, and to unsettle, I guess, my lens… How might we best work together?What assumptions do I make about their capacity… I need to be very clear [and] careful not to make assumptions on what they may or may not want to do'.Our participants' experiences challenge us to reconsider reciprocity beyond the transactional to acknowledge the relational nature of researcher-participant interactions, leaving researchers to carefully consider the impact of the relationships they have built, which they then withdraw on leaving the field (Zhang, 2017).This consideration should acknowledge the emotional work involved-including for supervisors (see also Johansson et al., 2014)-and realistic management of what can and cannot be offered to continue to respect these relationships once the researcher is no longer present in the setting.

RQ2: HOW DO SUPERVISORS ENCOURAGE THE DEVELOPMENT OF MICRO-ETHICAL REFLEXIVITY?
A key intention of our research was to gain greater insight into supervisory practices that may assist students to develop the ontological understandings and micro-ethical reflexivity associated with an ethics-in-practice approach.As Dickson-Swift et al. ( 2008) outline, paying attention to issues of risk to HDR students is part of a 'duty of care' that all stakeholders (institutions, Human Research Ethics Committees, supervisors and students) share.This is particularly important for students working in fragile contexts; Astill (2018), writing of her experience of studying natural disasters, notes that a key challenge for students engaging in risky or sensitive research 'is finding a supervisor with this level of understanding' (p.51).
Likewise, Krause (2017) poses key questions regarding boundaries between individual, collective and institutional duty of care, enquiring: … should students be allowed and/or expected to carry out research on highly sensitive issues in insecure regions?Do they have enough training?Do lecturers have sufficient capacity to supervise all projects appropriately and to support students?Or are students responsible for their actions and for acquiring skills?(p.28)  2020) call for greater mentoring and peer support for postgraduate students undertaking fieldwork in fragile contexts.Yet, as Darling (2014) has argued, there are also 'elements in fieldwork which may elude the grasp of doctoral training schemes and ethical audits as they only emerge through the practice of research itself ' (p. 203).Issues pertaining to student and supervisor navigation of microethical challenges in fragile contexts of research and overarching questions of individual/institutional responsibility are therefore multi-faceted and complex.We suggest that taking an ethics-in-practice approach in one's own research does not guarantee capacity for supporting others to develop micro-ethical reflexivity, although clearly the two are connected.Supervisor accounts of seeking to support student engagement with ethics-inpractice suggest that, as with other aspects of HDR supervision, having undertaken a doctorate or currently conducting research is not sufficient preparation for guiding a novice researcher to develop the required ethical reflexivity.Importantly, supervisors also emphasise the significance of the relational element: '… mentoring is the key … and having a space where you've got a senior academic who does perhaps similar work, and you've got that open-door policy … that is absolutely crucial' (Miriam, supervisor).

The value of dialogue
Dialogue was identified as one of the most important ways in which supervisors work to assist student development of micro-ethical reflexivity.As supervisor Ria advised: 'I can talk through those things with him [the HDR student], and I have to say, with him, there are millions of ethical issues all the time… I can be his sounding board as his supervisor'.Likewise, Joan identified the importance of dialogue with students, explaining: 'we just do lots of talking.The [Human Research Ethics Committee] forms are a really useful starting-off point, but I just definitely find they don't cover [what] people will encounter [in the field]'.Dialogue within supervision sessions to anticipate ethical complexities and for debriefing was therefore identified as essential.
The importance of asking effective questions to prompt ethical reflexivity-rather than simply recounting or describing events-was also identified as vital.Supervisor Uma noted: 'I give them a fair amount of supervision in terms of assisting them with those particular questions… It does mean that you have to talk about the risks involved.And this is where the doctoral students require a fair amount of support from the supervisors'.These conversations with students were described as requiring time, care and significant emotional investment on the part of the supervisor.For instance, Miriam advised: 'My role here is to help that student navigate … what benefit is there to the people you're going to interview?What are they going to get out of this? … Why are you assuming that this is what that group of people need at this time?… throwing those questions in.It's a lot of dialogue, it's a lot of talking'.Likewise, Joan described the importance of discussing boundaries and self-care to protect participant and researcher well-being: 'I do also do quite a lot of talking to my students … I know that they could be really overwhelmed by some of it and making sure that they've thought about all of this before they take on a project like this.But that if they do, that they're aware there might be some additional things, or responsibilities they owe their participants, which I'll help with, too', explaining 'I don't leave them on their own…'.The importance of initiating dialogue regarding the micro-ethical complexities in students' research, with supervisors in a facilitation role seeking to actively encourage engagement with ethics-in-practice, was therefore a key approach to supporting micro-ethical reflexivity.

Supervisors as models of ethical practice
A second key role for supervisors is to model ethical practice.Delia discussed the importance of having a supervisor who can draw on their own experiences and 'understand the micro ethics and the culturally responsive ethics that are required within these contexts'.In recounting a conversation with Amber, her HDR student, Delia explained: 'Amber came to our supervisory meetings and said, "Oh, I didn't take food [on fieldwork], can I take food?Ethics says I can't".And I said, "Yeah, of course you can".' Here, the value of Delia and Frankie (Amber's co-supervisor) creating a safe space in which Amber can raise and critically discuss micro-ethical issues from various cultural and philosophical perspectives is evident.Importantly, Frankie also reflected on the relativist nature of these discussions, noting: 'We haven't got solutions.There's lots of thinking to go around all that kind of stuff.And hopefully, we can pick up where there are ethical things that we need to consider in more complexity'.
All but one of the supervisors interviewed stated that they did not receive professional development regarding how to actively scaffold HDR students' micro-ethical reflexivity when researching in fragile contexts.The one supervisor who was satisfied with the level of professional development provided by their university was Ajay, a researcher working in cultural studies in education, who described the 'very strong early career researcher program within the faculty and university wide' that supported HDR supervisors.Other supervisors noted a lack of institutional support, with Miriam explaining: 'I didn't really get much mentoring at all when I was new to the game.And I wish I had because there were some students who fell off along the way.And we shouldn't have accepted them in the first place'.Miriam identified the importance of having colleagues with 'experience in human rights-related supervision' to shape her own supervision practices, including the value of 'being able to talk through lots of ethical issues'.Supervisors therefore need support regarding a range of important modelling behaviours, from drawing on skills and resources they bring from their own research to facilitating sustained dialogue to elicit what the HDR student identifies as key issues.
As the supervisors reflected on their experiences, Krause's (2017) question regarding the appropriacy of HDR students researching in fragile contexts was raised.The duty of care that supervisors and institutions carry towards students working in such settings was central to these discussions.For instance, Joan (supervisor) commented that HDR students are often unable to see the potential complexities of their proposed research, explaining that students often 'want to save the world … they want to do these massive projects, and they really don't know what it will be like to do'.From the supervisor's perspective, Joan suggested that 'maybe it is about ethics sometimes saying, "Look, maybe, you know, if the self-care strategies aren't in place, maybe you need to do a low-risk topic that won't be distressing"'.In addition to issues of potential harm to students, Miriam noted that the pressures of HDR candidature may render work in fragile contexts, and the associated investment in careful community consultation and engagement, unfeasible, advising: 'I'm incredibly wary now of taking on students who engage in work in this area … ultimately, no matter what goodwill you might have, ultimately, it's more likely just going to benefit you and your career.Because you may not have the space to advocate, even if you have, you know, a desire or a willingness'.

RQ3: What key supports assist students and supervisors to navigate micro-ethical complexities in research and what additional supports are required?
As with other research from fragile contexts (e.g., Fox et al., 2020;Shanks & Paulson, 2022), the participants in our study noted that some ethical complexities cannot be anticipated.Reflecting on her own doctoral candidature, Teresa (supervisor) advised the importance of students commencing their studies 'knowing that something might happen that you're unprepared for … be prepared to be under-prepared for some of this stuff'.Likewise, supervisor Miriam advised: 'You can't have a manual to cover everything, can you?'.However, our participants identified a range of supports that were important to their navigation of microethical dilemmas in the field.As noted previously, the centrality of the relational and having 'supporting supervisors' (Sam, student) who offer 'empathy' (Amber, student) and guidance was a recurring theme.

Reflecting on micro-ethical complexities
While students and supervisors referenced the importance of experiential learning, with Teresa (supervisor) noting the importance of 'learning on the job' and Camille (student) referring to 'learning mid-research', a key component of such experiences was having the opportunity to critically reflect on the associated ethical complexities with experienced researchers.For instance, supervisor Miriam explained that she occasionally consults with a member of the institution's Human Research Ethics Committee but more frequently seeks advice from research partners with lived experience of displacement and 'other researchers or others who I know work in a deeply, ethically aware way'.
As was evident in supervisors' reflections regarding the approaches they use to encourage ethical reflexivity, students identified provocative questioning, rather than dialogue alone, as a key support for talking through ethical dilemmas.Camille (student) recounted the value of her supervisor, advising: 'Really think about why you need to do interviews.Do you really need to sit [participants] down and sort of rehash … their experiences … or is there some other way that you could do it that doesn't require the person to always tell their story?'.Likewise, Teresa (supervisor) recounted her experiences as a student benefiting from peers who: 'were really good at kind of asking me questions about the approach and what I was thinking and trying to get me to justify things…'.For student Sam, practicing an interview with another HDR student and talking through possible issues and responses was valuable.

Engaging with key literature
Engagement with literature focused on micro-ethical complexities in fragile contexts of research was the other key support identified, and a chief means of exploring divergent views and perspectives.Again, HDR student participants reinforced the importance of a supervisor or mentor to help guide their engagement with these ideas.Reflecting on the evolution of her micro-ethical reflexivity throughout HDR candidature, Teresa (supervisor) commented: It just felt like some of the kind of predominant literature that I was reading about different health issues just felt really western-centric and just uncomfortable for me, which I think is probably what started me interrogating, like, what am I doing?Is this even appropriate?
Teresa advised that access to a reading list of literature with diverse perspectives on issues of ethical engagement was useful.Likewise, supervisor Miriam described how she recommends 'various articles and books that reflect on the ethics of doing work with people from a refugee background, particularly when you don't have that lived experience', noting the importance of having 'fairly deep conversations about it'.Attention to supervisor support to facilitate student reflection regarding key literature was therefore emphasised as vital.Targeted training and collective engagement When asked about the value of research ethics workshops offered through the university, most participants described these as focused on procedural ethics, with ethics-in-practice approaches rarely addressed.Consequently, many of our participants did not find these workshops to be useful in preparing them for the micro-ethical dilemmas they face in fieldwork.Teresa (supervisor), reflecting on her preparedness to enter the field as a student, noted: 'I was so naïve; I didn't even think there could be any issue.So, I think having those conversations and perhaps a training workshop where … you've got teaching cases or something where you can say this happened to a researcher, "how would you react, what would you do?" would be useful'.Likewise, supervisor Frankie recommended: 'I think we need some really messy, messy situations that confront potential supervisors and students in some of this training'.
Supporting students to navigate ethical processes and develop ethics-in-practice chimes with growing attention to the need for researcher care and protection (albeit not diminishing the need to focus on protection of participants), and this is arguably most urgent for the newest researchers: HDR students (Astill, 2018;Dickson-Swift et al., 2008).As Butler-Rees and Robinson (2020) argue-writing on experiences of 'failure' as two doctoral students in the field of Geography-limited engagement with ethics (both procedural ethics and ethicsin-practice) ill-prepares HDR students for the emotional risks (anxiety, disappointment or disillusionment) of conducting fieldwork: This rise in anxiety may be exacerbated by the lack of space created within academia to talk through emotions, such as anxiety, stress, fatigue and feelings of personal failure.At present, most postgraduate students' source of support comes largely through their supervisors, with students often not made aware of the wider University counselling and support systems available to them.(p.4) The conditions for anxiety are amplified if students experience their studies as a 'lone path', where 'you do the PhD alone, you suffer alone' (Collins & Brown, 2021, p. 1398).Gready (2014) qualifies this by noting that, while research is not a 'lone or autonomous project … it can be a lonely one ' (p. 197), and that students need to be reminded to draw on the support of others, such as family, friends, peers and supervisors.Engaging with the emotional work of research in these contexts is also an important keystone to developing a study with ethical integrity.Meloni (2020), writing about a study undertaken in a marginalised, undocumented community, describes how 'difficult emotions can particularly serve us as signals of ethical tensions and dilemmas in the politics of knowledge' (p.3), helping connect researchers with those in the field and with epistemological values.Supervisors and students also noted the value of collective approaches to education regarding micro-ethical reflexivity, where researchers at all career stages could come together to discuss specific ethical challenges from a range of perspectives.Supervisor Ria commented on the present lack of an appropriate forum to discuss such matters, noting: Sometimes, if there's a work in progress kind of thing at uni where you talk about your research and get feedback on it, sometimes that is useful in some aspects.But I'm not about to share the really kind of difficult things in that kind of forum … there needs to definitely be something greater.
A safe space for anonymous discussion of micro-ethical dilemmas was therefore considered a useful idea by participants in our study.As student Tara noted: 'you always feel like you're alone in this until, like, you meet other researchers, and you're, like … they are also going through the same thing.And then you get to hear how they overcame that challenge, what strategy they used'.

DISCUSSION
These findings clearly illustrate both the complexities of research in fragile contexts and the potential for significant harm to both students and supervisors who are often required to navigate these pressures with limited external supports.Supervisors are seen to be the major support for students and provide many complex and nuanced resources such as reflexive dialogue, modelling ethical conduct and emotional support.This level of emotional work can place supervisors at risk of burnout and leads to our recommendation to offer clear institution-wide approaches to support.
Firstly, the supervisor(s) alone should not be the sole source of student input.Rather, it is important to seek out research experience in settings similar to the contexts in which students will be working.A useful starting point is desk-based research to identify articles that include researcher reflexivity on micro-ethical decision-making, alongside drawing on or expanding networks to identify those with experience of the research setting who can offer reflections and advice.Supervisors can also support, if practically and financially viable, exploratory observational and trust-building visits by the students to the research context to offer a basis on which negotiation and relationship building can develop.Supervisors can facilitate reflection on all three sources of knowledge (published research, researcher accounts and in-person experience) for a mutual identification of anticipatable issues, opportunities to enter the field and important local values and cultural expectations to be respected by the researcher.In these ways a culture or 'learning alliance' between students and supervisors (Halse & Bansel, 2012) is created to work out, together, how best to undertake ethical decision-making in the particular research setting.
Yet, the duty of care towards students, supervisors and participants should not be confined to the supervisory context.Student participants highlighted the value of critical reflection beyond supervision.Collective spaces that bring together other researchers, including HDR students and community representatives, particularly those with forced migration experiences, might facilitate important opportunities for dialogue.Likewise, there is a need for institutional valuing of the kinds of collective dialogue and whole-community investment needed to support students as they conduct research in fragile contexts.This would involve Human Research Ethics Committees inviting students to present the basis for their ethical decision-making in ways that empower students to become experts in culturally respectful approaches to research, especially in contexts not experienced by the Ethics Committee members.There is also value in cross-institutional connections through special interest groups in research associations, conferences and seminar series to provide opportunities for peer support and collective upskilling.As students themselves gain confidence regarding ethics-in-practice in the context of their research setting, their expertise should be drawn upon, both within their own project and to drive review of formal ethics-related policies and practices.In this way, HDR students can help build a resource about the needs, values and cultural expectations of research in fragile contexts for Human Research Ethics Committees and other researchers, including supervisors, new to these settings.As students and supervisors in this study advocate, institutional processes should support the goal of seeking epistemic justice in research across contexts and cultures.This work requires the acceptance that disruption of 'usual' ways of working in institutions remote from fragile contexts might be needed if they are not fit for the purpose of supporting ethical research.

CONCLUSIONS
The importance of engaging in ethics-in-practice in fragile research contexts is clearly established; however, we know significantly less about how students and their supervisors navigate micro-ethical dilemmas during fieldwork, or how they develop the requisite knowledge and practices.This paper has provided insights into the kinds of ethical challenges students and supervisors in Australian universities experience in the context of undertaking or supervising research in fragile contexts, as well as accounts of support accessed and/ or desired.While the multi-faceted and frequently unforeseen micro-ethical challenges described by the participants in this study may emerge in all research settings, they may be particularly fraught in fragile contexts.
Primarily, participants in this study described tensions in the ways that they navigate the expectations of the institution regarding research engagement-characterised many times as an assumption of objectivity that was divergent from their lived experience of undertaking and supervising research.Our findings suggest a greater need for institutional acknowledgement and support of the micro-ethical complexities of engaging in research in fragile contexts, and of the associated care work undertaken in supervision.While the supervisors who volunteered to participate in our study are all active researchers in the fragile contexts in which their students are working, their reflections regarding the complexities of supervision highlight the need for institutions to provide professional development in this area.Further, our findings raise urgent questions regarding support for supervisors with limited experiential knowledge of navigating micro-ethical dilemmas.
Drawing on our participants' experiences and key themes within the existing literature, we have suggested potential avenues for creating greater institution-wide support for students and supervisors to engage with the micro-ethical complexities that can arise when undertaking research in fragile contexts.Such supports would need to consider the dual elements of the precarious contexts of research-which can present particular tensions and complexitiesand the pressures and factors related to HDR enrolment.Our suggestions are offered as a means of contributing to collective debate on this topic; we fully recognise that different institutional settings, programme requirements and models of supervision may necessitate distinct forms of support.However, by seeking to acknowledge and identify the demands on students and supervisors, and the supports that are presently in place and that need to be introduced, we can better respond to individual, collective and institutional responsibility regarding HDR student and supervisor engagement with micro-ethics in fragile or sensitive contexts.

A C K N O W L E D G E M E N T S
The authors acknowledge the Traditional Custodians of the Lands where Australian universities now stand, and pay their respects to the Elders past, present and emerging.These Lands have always been places of knowledge, teaching and learning, and sovereignty was never ceded.The authors would also like to acknowledge the HDR students and supervisors who have contributed their experiences and insights to this research.

F U N D I N G I N F O R M AT I O N
This project was partially funded by the Australian Association for Research in Education (AARE).

C O N F L I C T O F I N T E R E S T S TAT E M E N T
There is no conflict of interest.

D ATA AVA I L A B I L I T Y S TAT E M E N T
Research data are not shared.

E T H I C S S TAT E M E N T
The research for this paper was pursued in accordance with Human Research Ethical Clearance (HREC) approval H-2020-0383. 1 While research undertaken with participants from refugee and asylum seeker backgrounds requires careful planning for and reacting to complex and specific ethical concerns, we argue against a reductive blanket assumption of 'vulnerability' of such cohorts (Gillam, 2013;Mackenzie et al., 2007;Perry, 2011).We follow arguments against the p/maternalistic ascription of notions of vulnerability that are assumed on the basis of cultural or linguistic diversity or life experiences (Block et al., 2012;Perry, 2011).

14693518, 0 ,
Downloaded from https://bera-journals.onlinelibrary.wiley.com/doi/10.1002/berj.3945 by The Open University, Wiley Online Library on [15/01/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Krause (2017) determines that 'research should only be conducted if it is in fact required to respond to a research question and if researchers -regardless of their position -have the knowledge, training and capacities to do so' (p.28).Likewise, Shaw et al. (

T A B L E 1 Overview of participants. Participant Alias Role Field of research
Downloaded from https://bera-journals.onlinelibrary.wiley.com/doi/10.1002/berj.3945 by The Open University, Wiley Online Library on [15/01/2024].See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions)on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License