‘I Enjoy Having Someone to Rant to,I Feel Like Someone is Listening to Me’ : Exploring Emotion in the Use of Qualitative,Longitudinal Diary-Based Methods

Abstract

Emotions and the emotional labour of researchers have been increasingly recognised in social science disciplines, with many researchers providing personal and reflexive accounts. Such debates are less well recognised in health-related disciplines, particularly public health, who remain at earlier stages of valuing and understanding qualitative research. Drawing on personal experiences and methodological reflections gathered from a qualitative study, undertaken with young people over the course of 16 months during the COVID-19 pandemic in England, UK, the aim of this article is to offer further insight into the impact of researcher emotion, by specifically focussing on longitudinal, diary-based methods. My reflections are framed as three overlapping and intersecting themes. First, that qualitative longitudinal methods (and diary studies in particular) have enormous potential to curate rich emotional narratives. Second, that despite these positives, there are tensions or conflicting dynamics in using a method which helps to explore young people’s emotions but also involves emotional labour for the researcher. Third, that greater attention should be paid to ensuring ethical care for researchers, particularly those engaging with qualitative longitudinal and/or creative methods. Such strategies should not solely rely on self-care and must be considered at institutional or funding body level. To this end, my personal experiences and reflections, as well as those from previous offerings, are used here to underpin a framework for researchers or research teams embarking upon novel qualitative longitudinal methods: 1. Do not underestimate emotional burden. 2. Ensure meaningful debriefing is available. 3. Establish boundaries. 4. Make space for emotion throughout fieldwork as well as during analysis and writing (‘entering and exiting the field’).

Introduction

Emotion is a central feature of social life, emerging in response to, and embedded within, symbolic and power systems, and lived through everyday experiences and meaning-making practices (Burkitt, 2012). Along with a recognition that human (inter)actions are emotionally imbued (Olson, 2021), the importance of emotion, embodiment and feeling have become increasingly centralised within social and cultural research (Hanna, 2018). Described by some as the ‘affective turn’ (McGrath et al., 2020), this is framed by movements towards sensory, discursive and creative methodologies and attempts to explore ‘three-dimensional’ aspects of everyday life such as texture, colour, space, smell and taste. Nevertheless, such focus on emotion is usually concerned with insight into participants’ emotional realities (Hubbard et al., 2001; McGarrol, 2017; Ross, 2017). Whilst Bergman Blix and Wettergren (2014) argue that emotion work is required by social researchers, and should therefore be a central tenet of qualitative research, explicit focus on such emotional reflexivity remains a developing field. Thus, despite growing awareness that researchers may be emotionally affected by the work that they do, methodological and experiential focus on researcher emotion remains peripheral and even resisted by some disciplines (Dickson-Swift et al., 2009; Hanna, 2018; S.; Jackson et al., 2013). Indeed, Bergman Blix and Wettergren (2014) go on to suggest that researcher emotions tend to be treated as a bi-product to be ‘dealt with’ rather than engaged with as an important source of information or analytical tool.

Drawing on personal experiences and methodological reflections gathered from a qualitative study, undertaken by the author with young people over the course of 16 months during the COVID-19 pandemic in England, UK, the aim of this article is to offer further insight into the impact of researcher emotion, by specifically focussing on longitudinal, diary-based methods. To my knowledge, there are no published articles which explore researcher emotion from the perspective of longitudinal diary and/or correspondence methods (such as journaling and letter-writing), and no published articles which examine this in the context of a global crisis (the COVID-19 pandemic). Therefore, my intention is to add to the burgeoning body of literature about emotion in social research, as well as open up new debates within health-related disciplines. In the first section of the article, I briefly explore the concept of emotion in qualitative research and the value of longitudinal, diary-based methods. In the second part, I discuss the study context (young people and COVID-19). Third, my empirical reflections are presented and discussed in light of the literature. Here, I explore the tensions or dynamics experienced in using a method which helps to explore young people’s emotions but also involves emotional labour for the researcher. Finally, I use these reflections to examine what engagement with emotion means for ‘ethics in practice’ and suggest a practical framework for longitudinal qualitative methods that allows for further integration of researcher (and participant) emotion within both procedural (e.g. ethical guidelines) and analytical (transcribing, coding and writing) aspects of research.

Emotion within Research

There is a significant body of work and ethical processes built around the protection of research participants (Dickson-Swift, 2017; Olson, 2021). However, less emphasis is placed on researchers, who may expose themselves to acute emotional distress and challenging experiences, with potential impacts to their own professional identity and well-being (Jackson et al., 2013; Olson, 2021; Waters et al., 2020). Kumar and Cavallaro (2017, p. 648) define ‘emotionally demanding research’ as ‘research that demands a tremendous amount of mental, emotional, or physical energy and potentially affects or depletes the researcher’s health or well-being’. They suggest four possible types of emotionally demanding research experiences: (1) projects which focus on sensitive issues, (2) personal trauma previously experienced by the researcher, (3) experience of traumatic life events during research and (4) unexpected events that arise during research in what was previously not identified as a sensitive issue. Although reflexivity and the need to develop rapport are widely encouraged and emphasised, how this may invoke an emotional experience is a path less trodden, with little formal guidance available as to how researchers can or should harness or systematically incorporate emotion as data or to aid analysis (Dickson-Swift et al., 2009; Fitzpatrick & Olson, 2014). Further, I argue that emotion is complex and multi-faceted. My intention in this manuscript is to not only explore emotions associated with unhappiness, grief or sorrow, but to illustrate that other emotions such as joy, fear, anger, resentment, horror and disgust can be equally important and evocative, an argument I return to later within my personal reflections.

Emotions and the emotional labour of researchers have been increasingly recognised in social science disciplines, with many researchers providing personal and reflexive accounts, albeit less so in health research, particularly within public health, the discipline that I am primarily writing from (Kontos & Grigorovich, 2018; Mykhalovskiy et al., 2018). Further, whilst we can draw on examinations of emotional labour in other areas of allied health (such as nursing, physiotherapy or midwifery), this work tends to focus on participant or practitioner rather than researcher emotion (Brighton et al., 2019; Font-Jimenez et al., 2020; Kirby et al., 2014; Riley & Weiss, 2016; Wahlberg et al., 2020). Meanwhile, those grounded in feminist and criminological perspectives have suggested that researchers engage in various techniques, including maintaining empathy and rapport and/or paying particular attention to the way they dress or how they speak, in order to present appropriate emotional displays and engage in ‘continuous’ or prolonged emotion ‘work’ or ‘labour’ (Bergman Blix & Wettergren, 2014; Crewe, 2014; Hanna, 2018; Jewkes, 2011; McGarrol, 2017; Wakeman, 2014; Waters et al., 2020). Further, Jackson et al. (2013) have previously described the challenges of maintaining ‘emotional equilibrium’, and others have detailed the tension between open and free-flowing emotion as opposed to emotional suppression and/or detachment (Van Maanen, 2011). Indeed, Bergman Blix and Wettergren (2014) and others (Blackman, 2007; Dickson-Swift et al., 2009) have advocated for an explicit and active, rather than reactive, approach to emotions, suggesting that this will improve the quality of research, a position I will delve further into later in this article. Bergman Blix and Wettergren (2014) highlight three core ways in which researchers may use emotion within their research. First, strategic emotion work to gain and maintain access to the field. Second, emotional reflexivity, defined as an active process that requires researchers to invoke a reflexive agency throughout the researcher process. Finally, third, the emotion work performed to cope with the emotive dissonance between a researcher’s persona and sense of an ‘authentic’ self. Overall, in discussing how emotions may be harnessed by researchers, Bergman Blix and Wettergren (2014) argue that researcher ‘emotion management’ represents one type of ‘emotional labour’, required by qualitative researchers to successfully collect data.

Theoretical and empirical attention to the concept of emotional labour is not new and largely derived from seminal work by Hochschild (2012). Whilst the terms ‘emotional labour’ and ‘emotion work’ are often used interchangeably, initially these two concepts were developed by Hochschild to mean different things (Dickson-Swift et al., 2009). Thus, ‘emotional labour’ was used to refer to emotional management during work done for a wage, whilst ‘emotion work’ was used to refer to the work involved with dealing with other people’s emotions. Here, like Dickson-Swift et al. (2009) and others, I make no distinction between the terms ‘emotional labour’ and ‘emotion work’ in order to devote space and reflexivity towards the role of researcher emotion in diary-based methods and longitudinal qualitative research. Ultimately, it has been argued that researchers may invoke the analytical tools of surface and deep acting when managing their emotions (Hochschild, 2012), both of which are delivered according to ‘feeling rules’ (Hochschild, 1979). Some have likened this to the seminal concept of ‘impression management’ (Goffman, 1990) whereby people ‘perform’ to maintain credibility (Drake & Harvey, 2014; Leigh et al., 2020). Meanwhile, whilst not a methodological reflection, Addison (2017) has drawn on Bourdieu’s concept of habitus to expand on and critique Hochschild’s original framework of emotion work in order to draw out how it is that we can act differently in certain social settings by managing our emotions. Crewe (2014) describes such feelings as ‘field emotions’ whilst Jewkes (2011) acknowledges that emotional states are, to some degree, socially constructed and subject to the kinds of distortions we often guard against when assessing other forms of knowledge. Further, as Mounce (2018) suggests, human research, by its very nature, involves feelings and emotions that a researcher cannot help but be caught up in. She argues that what they experienced in their own research was authentic empathy rather than the socially acceptable management of emotion, concluding that such emotional ‘intelligence’, reflexivity or positionality requires a researcher to be emotionally attuned and sensitive to their own emotions and the needs of participants. Likewise, Carroll (2013, p. 548) states that ‘emotion is not simply bracketed off while doing sensitive research’ and that the emotional selves of researchers can be foregrounded (rather than estranged), sometimes unwillingly or sub-consciously. Although the experiences I discuss in this article took place during a unique pandemic context – where emotions were potentially heightened – I contend that many of my reflections and experiences would have been similar outside of a pandemic. Further, whilst there are articles focussing on emotion and research, and others exploring how we may capture participant emotion in diary studies (Cottingham & Erickson, 2020), no published work explores emotion work/researcher emotion in the context of longitudinal diary-based methods, and it is to this study design that I now turn.

Using Diary-Based Methods in Qualitative Longitudinal Research

Described by Plummer (2001) as ‘documents of life’, diaries have a number of unique characteristics. As a genre of writing, diaries are considered to be a type of private writing meaning that the assumed audience is ‘self’. Thus, by their very nature, diaries are associated with the foregrounding of less filtered emotional expression. To this end, Cottingham and Erickson (2020), Cao and Henderson (2020) and Karadzhov (2020) suggest that diaries provide access to intimate and spontaneous emotions. Indeed, diaries have the potential to capture a unique sense of authenticity, helping participants to think through thoughts and experiences that may be difficult to distil during a single interview (Alaszewski, 2006; Crozier & Cassell, 2016; Harvey, 2011). As a research method, a small, but established, evidence base suggests that qualitative diary methods can yield rich, complex layers of understanding, particularly with marginalised groups or where topics may be difficult to discuss face-to-face (Day & Thatcher, 2009; Karadzhov, 2020; Milligan et al., 2005). Diaries may also facilitate prolonged engagement with research participants and be of particular use where people may have difficulty articulating themselves through other types of research methods, enhancing rigour, quality and trustworthiness of data collected (Filep et al., 2017; Morrison, 2012). Thus, building on Wilkinson and Wilkinson (2018), for young people who participated in the study upon which my reflections are based (introduced above and below), diaries may have offered a way to express themselves, perhaps with less embarrassment, or fewer feelings of being judged, than in interview scenarios. Further, use of diary-based or correspondence methods, such as letter-writing, have been described to minimise the presence of the researcher, perhaps offering young people a greater degree of control in the research process (Alaszewski, 2006; Bartlett, 2012).

Ultimately, however, I argue that the principal value of using qualitative diaries as a data collection tool –which I return to later when setting out the study context below – is how well suited they are to the collection of longitudinal data (Baker, 2021; Mendoza et al., 2021). Qualitative longitudinal research conducted over a prolonged period of time has the capacity to follow people as their lives unfold in ‘real-time’ (Neale, 2021). She describes this as intensively walking alongside people to gain a processual understanding of how experiences and perceptions are created, negotiated, lived and experienced. Thus, qualitative longitudinal research is particularly suited towards studies that investigate changes and adaptations to traumatic and historic events (such as the global pandemic), as well as pathways, transitions and trajectories over time (Dwyer & Patrick, 2021; Grossoehme & Lipstein, 2016; Makleff et al., 2021; Patrick et al., 2021; Treanor et al., 2021; Weller, 2012). Nevertheless, there are very few qualitative, diary-based studies with young people, longitudinal or otherwise. A small number of - as yet largely unpublished – UK research studies have used qualitative diaries to document COVID-19, such as ‘Lockdown Diaries of the Working Class’ (see https://www.kickstarter.com/projects/lockdowndiaries/lockdown-diaries-of-the-working-class-uk?ref=project_link), ‘CoronaDiaries’ (Ward, 2020) and ‘Covid Realities’ (see https://covidrealities.org), as well as the ‘Coronavirus Diaries’, implemented by Britain Thinks (see https://britainthinks.com/news/britainthinks-coronavirus-diaries). However, these studies have predominantly captured the experiences of adults or families. At time of writing, the project described here is one of only a small number of published studies that has collected diary data directly with young people during the COVID-19 pandemic (Percy-Smith, 2021).

Study Context – The Lockdown Life NE Study

The project upon which this article is based is a qualitative, longitudinal multi-method examination of the impact of COVID-19 on young people aged 13–17 in North East England (‘Lockdown Life NE’). In phase 1, diary and interview data were collected with 31 young people between July–November 2020 (Scott et al., 2021). The full ongoing longitudinal analysis comprises 26 young people who have kept a diary since June 2020 and who took part in serial interviews held every 2–3 months until September 2021. As the young people in this study resided in North East England, in addition to the first (March 2020–July 2020) and second (December 2020–March 2021) UK national lockdowns, they were subject to ‘stay home’ orders put into force for 4 weeks from October 2020 (though face-to-face school and college classes continued) as well as localised ‘tier’ restrictions which were in place from October 2021 up until England began to ease restrictions from April 2021. The restrictions that young people described included social distancing, both in a literal (maintaining 2 m distance) and figurative (ceasing or reduction in contact with friends and family members outside of the own household) sense, school closures and remote learning, and localised lockdowns of retail, community and hospitality spaces. Such restrictions interacted, amplified and overlayed in young people’s lives. During the pandemic, I was subject to the same restrictions as young people. As a 36-year-old female academic, I was balancing working from home with young children and home-schooling. Thus, despite having over 10 years of experience in conducting qualitative research with children and young people, it is worth noting that my own work/life balance, and emotional well-being, was very different to how it was prior to the pandemic.

Taking a qualitative longitudinal approach, and the curation of diaries, was a pragmatic and methodological decision. Primarily, the aim was to explore young people’s ‘journey’ in and out of lockdown, which a one-off interview would not adequately capture. Ontologically, use of diaries aligned both with the focus on exploring the unfolding nature of the crisis, as well as with a feminist ethics of care and justice, which emphasises the need to ensure that experiences can be sensitively captured and explored, and which suggests that diaries have the potential to be cathartic or therapeutic for participants (Tarrant & Hughes, 2019). Indeed, it may be that there is similar value for researchers, and I will unpack this further in the sections that follow. By taking this methodological stance, I also contend that revealing aspects of the self can break down power differentials typically associated with data collection and be enormously beneficial to the research process (Ward & Gahagan, 2010). Relationships with participants were strengthened and allowed for a safer research space because I shared my own experiences of the pandemic. A solicited, flexible approach was employed. Solicited diaries are defined as diaries curated specifically for the purpose of research, in which research participants are asked to record their activities, experiences, thoughts, feelings and/or observations for the duration of the study (Bartlett & Milligan, 2015; Karadzhov, 2020). Like Herron et al. (2018) and Breheny et al. (2020), a flexible approach to diary content was taken, allowing young people to choose how (and when) they wished to narrate their experiences. A key focus was the capture of non-verbal facets of the everyday lives of participants, conveyed through videos, photographs and audio recordings (see previous work by Bernays et al., 2019; Thompson & Oelker, 2013). Other studies have utilised multi-modal diaries, combining different response formats (for example, text and images) within a single diary research design (Bartlett & Milligan, 2015). Such alternative diary formats have been shown to augment the methodological and theoretical capabilities of pen-and-paper diaries, allowing for a richer, more dynamic, sensorial and embodied insight into respondents’ lives (Crozier & Cassell, 2016).

Building on these approaches, a range of media was collated in the current study including text, photos, audio and drawings, using a combination of email, text messaging and Instagram. Project pages with the study name ‘Lockdown Life North East’ were set up on both Instagram and Facebook that participants could choose to ‘follow’, with private diary extracts collected using Direct Messaging. Use of digital platforms such as Zoom and Teams, as well as extended use of telephone interviews and social media platforms, arguably represent ‘the new normal’, particularly during and post-pandemic (Archibald et al., 2019; Gray et al., 2020), with young people increasingly used to communicating in these spaces (Gibson, 2020). A growing literature base indicates that collecting data in this way provides rich, multi-media insights into the world of young people, not readily accessed through other methods (Weller, 2017; Wilkinson & Wilkinson, 2018). I was responsible for all data collection with the exception of a sub-set of interviews conducted by a Masters student at one study timepoint (June 2021) and analysed as part of their dissertation project. I did not set boundaries as to what times of the day/how often young people should engage with the study. Likewise, participants were not obligated to engage daily and, on average, participants sent 2–3 diary entries per week. I encouraged regular contact by sending targeted (for example, ‘how do you feel about the announcement today’) or open and neutral prompt questions, probes and posts on the project Instagram page (for example, ‘is there anything that you’d like to share or tell us about today’). Study methods are discussed in detail elsewhere (Scott et al., 2021). Diary submissions and serial interviews were considered primary data and analysed thematically in the first instance (at the end of Phase 1) in order to capture a particular snapshot in time (Scott et al., 2021). Longitudinal data will subsequently be explored through narrative storytelling, creative poetry and wordplay in order to fully capture the temporal nature of personal, social and interactional components of young people’s lives over the course of 16 months, at a time of huge change and transition for most. My intention in this article is to explore my own experience of emotion within the research encounter by drawing upon interview transcripts, diary entries and field notes as well as my own informal recollections collated over the course of the study.

Experiencing Emotion Within Diary-Based Methods – A Worked Example

The Potential of Longitudinal Qualitative Diary Methods to Yield Rich Emotional Narratives

In this project, curation of diaries resulted in rich, raw and candid data from young people, more so than imagined at the offset. Diaries also facilitated real-time, two-way dialogue, but dialogue that I was less able to control, giving young people more freedom to direct the course of their reflections. Indeed, it felt like I was following young people’s ‘stories’ as they unfolded, and at times this invoked heightened emotion. Although the content of other modes of qualitative data (such as interviews, focus groups, observation) is often emotive, I felt that use of diaries sharpened these feelings; by their very definition, diaries are personal and intimate, and young people appeared to treat their reflections as they would have a personal, private diary. Indeed, one participant compared diary writing to blog writing, commenting that diary writing was unfiltered and usually for personal consumption, and likening project diary reflections to writing in a ‘real’ diary, particularly as the presence of the researcher was removed:

‘I’ve really enjoyed having someone to rant to. I’m getting something out of it, you’re getting something out of it, I feel like someone is listening to me. It’s nice to just text someone and just get it all out, that’s definitely what I did. Calling it a diary though? Yeah definitely. If you’d called it a blog that might have been a bit different because a blog means you have to re-read it and proof read it whereas a diary you don’t, you just spill I guess’ – Female, aged 16.

Further, young people shared a great deal, such as about their sexuality, bereavement/loss, cancer diagnoses, educational transitions, mental health difficulties, and thoughts/fears about the future. I am an experienced qualitative researcher, having conducted a number of studies that involved the collection of sensitive data (see Scott et al., 2017, 2020). Yet, despite this experience, diary accounts moved me immensely, and I felt young people’s exasperation, isolation, loneliness and fear acutely:

‘The effects of social isolation are draining. I’m constantly tired and I’m lucky if I leave me front door once a week, my 7-year-old sister now has crippling separation anxiety and my mum’s physical health is not doing great at all’ – Female, aged 16

I had no biographical congruence with the young people in my study, and this is not intended to be an auto-ethnographic account, meaning that this emotional connection did not come from my own lived experience per se. It did, however, come from being moved as a human and from sharing a collective experience (the COVID-19 pandemic), and therefore sharing in the loneliness, isolation, powerlessness, and frustration this caused. Collective trauma and/or emotion has been explored in relation to COVID-19 and previously in response to events/crises including wars and natural disasters and has been defined as a group-level cataclysmic, tragic experience that is made sensible through co-constructed, collective discourse (Hirschberger, 2018; Stanley et al., 2021; Wettergren et al., 2020). Further, as Saul (2013) and Watson et al. (2020) suggest, collective trauma as distinct from individual memory due to its capacity to persist across space and time. Thus, using metaphors as an analytic tool, Stanley et al. (2021) have demonstrated that emotional experiences of COVID-19 converged around several deeply held negative emotions: (a) grief, (b) disgust, (c) anger and (d) fear. Meanwhile, Metzler et al. (2021) analysed tweets from 18 countries during the first 5 weeks of the outbreak and observed a strong early upsurge in anxiety-related terms. Further, sadness terms rose and anger terms decreased around 2 weeks later as social distancing measures were implemented.

Tensions or Conflicting Dynamics in the Use of Longitudinal Qualitative Diary Methods

Moving back to the emotion that researchers may experience, Jamie O’Quinn has recently reflected (via Twitter) on the secondary trauma qualitative researchers may experience during fieldwork (O'Quinn, 2021):

‘I have entire 3-hour long interviews that I have absolutely no memory of a few days later… This is a protective measure that our brains often take automatically when we experience prolonged exposure to others' trauma…’

In the current study, I experienced different emotions at particular points of the study. This may, in part, be linked to the methods undertaken – phases overlapped and accumulated as opposed to the reality of most qualitative interview studies or even an ethnography where most of analysis is done after data is collected. The emotional workload of the study also increased for me as data collection and analysis accumulated (like many researchers this was not the only project I was working on) and as the pressures of managing home and working took their toll, particularly where my own school or nursery aged child’s COVID ‘bubble’ burst, meaning a return to home-schooling and full-time childcare. Importantly, however, the emotions experienced and shared were by no means always negative. For example, over the course of the study, the participants and I celebrated a number of lockdown birthdays, young people articulated relief and joy at the respite lockdown provided and they shared other causes of celebration, such as exam results, getting to visit family and the hope that came with the new school year:

‘I’m now a proud year 10 student starting to study for his GCSEs, something I had not really thought about up until now’ – Male, aged 14

‘Hi Steph! Happy birthday for last week! I hope you had a great day (at least the best it could be in the given circumstances). There have been a few strange celebrations this year…’ – Female, aged 16

Significantly, even outside of direct contact with participants, I felt emotional responses (positive and negative) during coding, re-reading and analysis, described previously by Woodby et al. (2011) as ‘a sequence of exposures’. Similarly, Johnson and Clarke (2003) describe these feelings as ‘cumulative distress’ or burden and – certainly – like Woodby et al. (2011), McKenzie et al. (2017) and Mounce (2018), I found transcription, coding and analysis to be emotional despite having experienced young people’s narratives and reacting to them in the moment. Woodby et al. (2011) describe encountering anxious feelings during the coding process whilst Mounce (2018) discusses the physicality of transcribing, becoming immersed in narratives and hearing stories recounted over and over again, as invoking feelings of sadness.

In the current study, this took its toll on me both emotionally (feeling overwhelmed and anxious) and physically, which manifested as fatigue. Data collection involved sustaining a relationship rather than maintaining a sample, which meant that I had to imbue or share quite a lot of me at times – potentially more than within other qualitative methods that are not longitudinal or as intimate in nature as a diary. For example, I wrote to the young people and noted:

‘It feels like everyone is isolating right now, including my 2-year-old!!! How many of you are in isolation?? How are you feeling?’

Boundaries and Ethical Care for Researchers Engaging in Qualitative Longitudinal Methods

Curating diaries was a huge investment of time for both my participants and I. I experienced distinct blurring of boundaries between work and home, particularly as social distancing requirements mandated me to work at home, leading me to joke frequently about the ‘24/7 project phone’:

‘It is Saturday morning and I’m dressing a two-year-old to go on his bike. I’m still not dressed and I’m breaking up an argument between said two-year-old and his older brother, who is also not dressed. My project phone flashes and I check it. I immediately feel guilty for checking it, parental guilt. It’s [name] sharing her mum’s cancer news. I’m juggling two boys and its Saturday morning. Can I not answer this? You cannot read a message steeped in such emotion and not reply. I leave the shouts and walk to the kitchen for some space to compose my response’ – extract from fieldnotes.

The decision to curate flexible diaries promoted this lack of boundary of course, this was deliberate and rewarding. Nevertheless, it is important to acknowledge the resultant (emotional and physical) workload. Based on ethnographic work conducted in Australia, Watson et al. (2021) articulate this as ‘leaky boundaries’ between workspaces and family/domestic life that are brought about by, through, and beyond the digital. This was an immersive but lonely experience for me – for large parts of the study, I was the sole data collector and would increasingly find myself wanting to discuss the stories of participants with colleagues as a way of understanding, processing and debriefing. During the pandemic, such coping mechanisms were not available to me to the extent they had been prior to the pandemic; gone were office ‘chit chat’, informal discussions had at the beginning of meetings and whilst making a cup of coffee in the kitchen, and at times my partner needed to fulfil this role for me. Other debriefs had to be pre-planned and organised, usually via Zoom or telephone. Similarly, unlike Hanna (2018), I did not ration how much time I spent looking at the data, though retrospectively, I perhaps should have done so as a mechanism of self-care, and a topic to which I will return to in the discussion of this article.

Such immersion was advantageous too. My original rationale for the work (early into the pandemic) was steeped in the idea that there was very little prolonged qualitative research being undertaken with children and young people. I felt that this data could make a difference and that the study could give young people an outlet or ‘voice’ (see Liebenberg et al., 2020, p. for a detailed exploration of extending youth voice). The benefits for me were that I created stronger connections with participants and immersion facilitated familiarisation with the data. Over time, I felt increasingly attuned to the feelings of participants, which is (in my view) a natural and direct consequence of the longitudinal nature of the work undertaken. Writing about diaries, Hanna (2018) describes this as ‘distanced empathy’ whereby the usual strategies of offering participants a break or offering more physical or visceral demonstrations of empathy are not available. Overall, I felt a sense of protection and responsibility towards participants. Like Hanna (2018), I was not prepared for the depth of feeling and extent of emotional labour, and assumed researching online, and at a distance, would possibly be less rather than more emotive. Perhaps young people’s narratives felt so emotive due to the unfolding crisis or perhaps it was due to my other role in life, as a parent of young children, whom were also experiencing isolation, home-schooling and lack of interaction with friends and family. Most likely it was a combination of the two. This meant that I frequently felt a need to amplify their voices and a sense of urgency as time passed, particularly as their accounts focused on the pandemic, a time-critical global event; I also felt guilty if I did not respond quickly to diary entries received. Similar points have been recognised by Coles et al. (2014) and Waters et al. (2020); the latter of which suggest that interviewers derived a sense of purpose from giving participants a voice, and whose sense of responsibility often led to increased productivity and motivation in an attempt to ‘give back’ to their participants. Although this often re-invigorated researchers, it was equally burdensome. I felt this acutely in my own work – young people had given up so much time and shared so much over the course of 16 months that I wanted to ‘do their words justice’ (and still do). In turn, this makes disengaging from the field (and ceasing emotional labour) a methodological and emotional challenge, a challenge which has been recognised in previous work (Abbott & Scott, 2018; Gobo & Molle, 2017) and potentially heightened in the case of longitudinal research (Batty, 2020; Smit et al., 2021; Treanor et al., 2021). One of my own practical strategies or ‘coping mechanisms’ for beginning to exit the field was to ensure young people felt valued and knew how their narratives would be utilised, reassuring them that co-produced booklets and materials would be sent to them for their input first and foremost. On a more personal level, I found it cathartic to ‘let out’ emotion within writing, both within my fieldwork diary and when drafting manuscripts based upon the project. Likewise, Crewe (2014) and Jewkes (2011) both acknowledge that the documentation of emotion illuminates the shape and findings of our studies. Nevertheless, ceasing emotional labour is perhaps a challenge I am yet to fully face. Having only recently completed the final set of serial interviews, I am still ‘in the field’ so to speak (and immersed in data analysis).

Discussion

Although young people are less likely to experience severe symptomology or hospitalisation with COVID-19 (Bhopal et al., 2021), there has been recent, growing concern about increased transmission and what the long-term implications of the virus are for young people’s physical health, particularly those with underlying vulnerabilities or conditions (Fanner & Maxwell, 2021). Nevertheless, what we do know with much greater certainty is that COVID-19, and restrictions imposed in response such as social distancing and lockdowns, have resulted in adverse psychological consequences and a rapid decline in young people’s mental health and associated well-being (Branje & Morris, 2021; Dewa et al., 2021; Scott et al., 2021). Over the course of this 16-month study, I suggest that young people shared emotive, raw and sensitive data. Further, I contend that this study garnered particularly emotional narratives from young people (and a reciprocal emotive response from me) for two key reasons. First, the longitudinal, sustained nature of qualitative data collection provoked this response. Here, I draw on Lois’ (2010) and Somaiah and Yeoh’s (2021) conceptualisations of ‘temporal emotion work’ to suggest that both I and the young people involved in this study experienced ‘temporal emotions’. Lois (2010, p. 440) describes temporal emotions as ‘…a subclass of emotions … can only be felt by crossing timeframes … While all emotions can be felt in the present, remembered in the past, or anticipated in the future … few – such as nostalgia, regret, disillusionment, ambition, hope, optimism, and dread – … cannot be felt without bridging the present to either the past or the future’. Lois’ original framework in 2010 uses two key analytical tools – sequencing and savouring – to theorise time and emotion. ‘Sequencing’ refers to the intentional ordering of time and emotions by young people whilst ‘savouring’ refers to a focus on the present. Somaiah and Yeoh (2021) add a third – suppressing – which they define as a silencing of certain emotions to be able to function in appropriate ways.

At a more practical level, emotional recognition and reflexivity is important in longitudinal research where rapport and relationships will be developed over a considerable period of time. Like Gibson et al. (2013), I, and young people, found value in multiple contact points (serial interviews/follow up points), which provided key moments for collecting observational data, building trust and rapport, and negotiating data content and interpretation. There has been little research conducted examining the various contributions of different methods and media to understanding of emotion in health and illness (Smit et al., 2021). Indeed, in terms of young people specifically, Chakraborty et al. (2018) have called for more research on youth emotions (which they label ‘textured emoscapes’), and I am keen to combine creative, qualitative longitudinal research with a focus on emotion in future research (see McGrath et al., 2020 for work on emotion mapping). Of key importance in the current study, was the degree of flexibility that multiple, longitudinal methods afforded, particularly in a time of crisis. Nevertheless, this must be caveated with what a qualitative longitudinal project of this nature could ‘take’ from the researcher, which I consider in more detail below. As emotionally draining and time-consuming work, a researcher using qualitative longitudinal diary methods will not simply and cleanly get ‘in and out of the field’ (see Smit et al., 2021).

Second, the assemblage of diaries shifted the focus to participants as data creators, giving them autonomy over what they shared and how often. I believe this was a powerful data collection tool, allowing young people to share more openly and intimately, as illustrated earlier by a 16-year-old participant who recognised how cathartic the diary writing process was for her. Of course, it is important to recognise that diaries were specifically solicited for the purpose of a research study. Thus, drawing on Baker (2021), participants may have been selective over what to disclose in their diaries due to an awareness that they would be read by me, the researcher, and that this may have influenced how they presented their own emotion. Young people may therefore have withheld thoughts or events that they regarded as too personal to share or experiences that they deemed irrelevant to the research project. Similarly, as a researcher, the characteristics of the diary may have shaped a heightened awareness of my own emotions, imbued within my reflections and approach to analysis.

Nevertheless, I contend that a movement away from a ‘traditional’ qualitative method potentially facilitated greater depth of feeling, a point recognised in creative and/or participatory methods (Askins, 2018; Pettinger et al., 2018; Tickle, 2020). Qualitative diaries, however, remain an under-used method, especially within longitudinal research, and have enormous potential where research focuses on sensitive topics or where there may be levels of underlying trauma. There were a number of specific ways that diaries facilitated access to emotion in this study. First, by allowing for near real-time, two-way interaction with participants, I was able to follow up entries with questions, probes and further dialogue. Second, I was able to frame correspondence around time-critical events such as school closures, easing of restrictions, lockdowns, events which sparked anger, frustration, sadness and sometimes joy. Each of these have parallels with other research where longitudinal methods are suitable and are not necessarily unique to the Covid-19 pandemic. Third, it allowed me opportunities to check on participants, especially any who began to respond more infrequently, aligning with a feminist ethos of care and social justice. By using this approach, I sought to break down disparities and centre participant voices as actors in their own lives (Askins, 2018; Liebenberg et al., 2020; Rogers & Kelly, 2011; Tarrant & Hughes, 2019), but doing so also involved supporting and caring for the young people involved. Although I asked young people how they felt about being part of a longitudinal diary study, I cannot ‘know’ young people’s emotional responses to being involved. Nevertheless, I contend that taking this methodological approach deepened rapport, trust and connection, leading to richer data (see Batlle & Carr, 2021).

Although there was a tangible focus on how the pandemic and associated restrictions had impacted upon young people’s emotional well-being, this was not a piece of research with trauma or grief at its core. Therefore, in many ways, I was not prepared for the toll the study would have upon my own emotions, and the complexity of these emotions, and this took me by surprise. Though attention to emotions, emotion work and emotional labour has gained traction in qualitative academic disciplines recently, it remains firmly at the margins, particularly within health research. Detailed accounts of ethics in practice from researchers are vital in order to highlight the ethical deliberations managed in the spaces outside of the procedural ethical review (Jackson, 2021). Moreover, as Waters et al. (2020) suggest, understanding these emotional reactions is paramount to then interpreting that data. Such points are reinforced by Woodby et al. (2011) who recognise that emotional reflexivity is particularly crucial during the process of transforming participants’ words into analytic codes, to assure faithful reproduction of their experiences. As the authors go on to acknowledge, few articles explore the experiential and emotive phenomenon of coding beyond the pragmatics/practicalities. Moreover, some authors suggest that acknowledgement of emotion (for participants and researchers) within the research process should be seen as a positive. For example, Wakeman (2014) contends that greater consideration of the emotive self, both past and present, has much to offer. Nevertheless, returning to previous points on burden and emotional labour, this involves delicately balancing the need to elicit quality data against the burdening of participants and researchers (Mendoza et al., 2021). Indeed, a counter argument has been put forward that long-term methods of data collection could be intrusive to participants (see Treanor et al., 2021). Further, as Woodby et al. (2011) and Waters et al. (2020) contend there is an expectation that researchers are skilled in emotional reflexivity, yet practical training to develop these skills is rarely provided to researchers. Reflexivity is a defining feature of qualitative research but has been critiqued by some as simplistic and one-dimensional (see Pillow, 2003). Moreover, emotion has traditionally been neglected as a dimension of reflexivity, perhaps because emotion could be seen as ‘unhygienic’ knowledge, because emotional reflexivity could result in discomfort or because the burden of emotional labour may not weigh so heavily within applied research where there can be a tendency to sanitise and reject messy narratives (Blackman, 2007; Holmes, 2010). Therefore, the blossoming body of literature which acknowledges the importance of the emotive self within research is emboldening. Indeed, Bergman Blix and Wettergren (2014) argue that ignoring emotions serves only to mystify dimensions of the research process, whilst Benoot and Bilsen (2015) contend that doing research in an embodied and reflexive way will enhance the quality of research, described by them as ‘embodied reflexivity’. Others have highlighted that ‘making space for emotion’ and/or feelings may help researchers to negotiate the challenges of research and uphold their personal safety (Boden et al., 2015), expressed by Warin (2011) and Bowtell et al. (2013) as ‘ethical mindfulness’ and McKenzie et al. (2017) as ‘harm minimisation’.

In recognition of such complexities, Dickson-Swift (2017) has advocated for a framework of emotional safety. Meanwhile, Coles et al. (2014) and Watson et al. (2021) have identified that researchers employ a number of protective strategies or coping mechanisms including self-care, space creation and communities of coping. However, as highlighted in my methodological reflections, I found it difficult to implement these traditional mechanisms during a crisis (the COVID-19 pandemic) where my workspace became my home space and vice versa. For working women in general, the pandemic has highlighted a large gap in household work between men and women (Collins et al., 2021). Meanwhile, within the academy, Bell and Fong (2020) found that, while journal submissions from male academics in public health disciplines increased as universities worked remotely, those from women significantly decreased (Guy & Arthur, 2020). Of course, although COVID-19 exacerbated such experiences, there are many people, particularly students, early career and marginalised researchers, who have been living with the reality and repercussions of significant inequalities in UK Higher Education Institutions (such as limited social support, single parenthood, increased family responsibilities and burden associated with childcare) all along (Bhopal & Henderson, 2021; Watermeyer et al., 2021). Further, whilst we took an equitable approach to caring responsibilities, even in our own household (heteronormative cisgendered couple), this did not remove the daily emotional burden, uncertainty and responsibility associated with being a working woman with childcare responsibilities during COVID-19 lockdowns. This has been recognised by Hjálmsdóttir and Bjarnadóttir (2021) in a diary-based study conducted in Iceland, and who found that mothers took on greater mental work than before. They also described intense gendered emotional labour, as they tried to keep everyone calm and safe.

Bringing in a wider gendered lens, a substantial body of literature and research attests to the persistence of gender inequalities in academic and research careers and a gendered division of labour within universities (Maxwell et al., 2019; Toffoletti & Starr, 2016; Van den Brink & Benschop, 2012). This has been described as ‘academic housework’ or ‘emotional service work’, with various administrative, pastoral and teaching related activities more likely to be taken on by women researchers as the primary caregiver role traditionally associated with being a woman blurs into a work (university) space (Heijstra et al., 2017). Thus, speculatively, it is perhaps –unfairly – expected that women researchers are simply better able to handle the emotional aspects of research labour, with more women thought to be employed as qualitative health researchers, and that calls for support could be seen as a gendered move and further stigmatised. Moving forward, as flexible modes of work increasingly become the norm in academia, ensuring ethical care for researchers will take careful and considered thought. Further, with creative and new methodologies come new expectations of researchers (see Wiles & Boddy, 2013). Therefore, I contend that what is discussed in this article (such as the potential for data collection to be isolating, lack of boundaries between work and home and the need for emotional reflexivity) remains highly relevant to post-pandemic contexts. Thus, based on my own reflections, I outline a self-care framework for researchers and/or research teams embarking upon novel qualitative longitudinal methods: 1. Do not underestimate emotional burden. 2. Ensure meaningful debriefing is available. 3. Establish boundaries. 4. Make space for emotion throughout fieldwork as well as during analysis and writing (‘entering and exiting the field’).

I have discussed the boundaries we set ourselves – as researchers but also human beings – and the pressure this can create in a long-term research project. However, this highlights several important, linked considerations for boundaries set between researcher and participant. First, that boundaries established and relationships developed through the research process can result in an ‘expectation’ on the part of participants. Second, in turn, this can exacerbate ‘leaky’ boundaries and volume of work (physical and emotional) undertaken. Nevertheless, in some ways this argument is cyclic - taking on this emotional labour can often be beneficial to the research process and, in this study, resulted in a safer, open research space for participants. You give a lot of yourself in the process (it is tiring, involves vast amounts of emotional labour) but this is also rewarding (richer data, giving a voice to those less heard, rapport with participants). In other words, qualitative longitudinal data collection ‘can giveth but can taketh away’ – and a core aspect of any ethical care framework for qualitative longitudinal researchers should relate to greater focus on preparation for entering and exiting the field, part of which links to the establishment of boundaries, upfront or otherwise. Consent and debrief processes could be suitable points to address this.

Although the focus of this manuscript is the researcher or research teams themselves, use of a sole self-care model has the potential to perpetuate the existing inequalities within academia outlined above. The implementation of self-care models or frameworks must therefore be accompanied by meaningful commitment from institutions, ethics committees and funding bodies, who each have a responsibility to ensure the emotional safety and security of researchers, both at early stages in research design and by building in time throughout the lifespan of a project. Future work should explore models or frameworks operational at such strategic levels, and which go beyond broader mental health support for staff and students, with the view to expanding or adapting existing models or initiatives to account for emotional safety and security during fieldwork. Different institutions will have health and safety policies, as may some funding bodies and professional organisations. Similarly, there are ideas from whole campus care models, recovery syllabus, to pastoral care and Equality Diversity and Inclusion. For example, the UK Research and Innovation Preventing harm (safeguarding) in research and innovation policy (see http://www.ukri.org/our-work/supporting-healthy-research-and-innovation-culture/preventing-harm), the framing of which currently aligns more towards exploitation, bullying and harm, could be expanded to account for emotional safety and security. A second pathway through which institutions and funding bodies could demonstrate meaningful commitment would be the provision of formal, practical training for researchers in emotional reflexivity. Finally, institutions, funding bodies and ethics committees should consider more carefully research from a researcher ethical perspective (in other words that mitigating harm extends to the researcher as well as the participant). Practically, this could be demonstrated through expanding current policies beyond one-dimensional safety and risk or by offering training as per the above. At a deeper level, due to funder expectations and milestones, sometimes there is very little ‘time’ to take the emotional space needed within a research project. Yet, having time and space for emotions within research is important, especially where we are unable to anticipate what the emotional responses might be and the consequences it might have for our well-being. This could be recognised by institutional embodiment of a feminist ethics of care position to ensure that care extends to researchers as well as participants (Haddow, 2021).

Conclusion

This article builds on the approach of previous methodological offerings which explore ethics in practice. As Hanna (2018, p. 527) articulates, ‘making the private public appears methodological advantageous, and it is in this spirit that I share my reflections’. There remains a need within academia to both engage with and be comfortable with emotion, which this article addresses in the context of a longitudinal diary method study. Drawing on Bergman Blix and Wettergren (2014), I argue here that both research quality and researcher well-being benefit from raised awareness of emotion and emotional labour, and that reflexive research involves emotional care for researchers as well as research participants. To this end, like Malacrida (2007) and Benoot and Bilsen (2015), I call for greater appreciation of ‘embodied reflexivity’ in order to ensure that qualitative health and social research is ‘characterised by intimate social relations and embodied interactions’ (Malacrida, 2007, p. 1338). Some applied health-related disciplines remain at earlier stages of valuing and understanding qualitative research and this may not necessarily be easy – I have outlined some recommendations based on identifying this gap and raised questions which will need to be considered in other studies. By way of conclusion, however, I return to the use (or under-use) of diaries and contend that longitudinal diary-based methods, particularly when combined with other qualitative methods, have the potential to be powerful and impactful to the emotionally engaged researcher. Yet, they can be equally burdensome and I refer qualitative longitudinal researchers of the future back to the 4-point framework I set out above. Young people (as well as researchers) are not homogeneous and there needs to be critical reflections on using diaries with young people in oppressed, marginalised or disenfranchised populations such as those who are justice-involved, those at risk of homelessness or those living in households with substance abuse or domestic violence. Such voices have been under-represented throughout the pandemic response, with a lack of data collection on protected characteristics across the board, preventing a truly intersectional approach.