The role of ‘friendship as method’ with child co-researchers in the primary school environment

ABSTRACT

Within social science research the complex nature of relationship-making and ‘friendship as method’ has gained enthusiasm. However, there is still a significant lack of research on ‘friendship as method’ with children and young people in participatory studies. Drawing on empirical case studies, we ask: how does ‘friendship as method’ work in research with children? The paper considers the role of vulnerability and reservations, friendship facilitator, and discusses the ethical dimensions of creating and sustaining ‘friendships’ between researcher and participants (as co-researchers). We argue that friendships in research are not a hierarchical or linear continuum, but a spectrum: friendships often mean different things to different people at different times; they can be positive and negative, both liberating and restrictive, fleeting and sustained, energising and tiresome. We recommend that participatory research with children considers not only the participatory components of power and action, but the emotionality and relationality of participatory research with children.

KEYWORDS:

• Friendship
• friendship as method
• children
• creative methods
• participatory research
• practitioner research

Introduction to friendship

Friendships are often paradoxical in nature, as Diphoorn and van Roekel (2019) observe, ‘[friendships] can be fragile and potent, liberating and restrictive, stable and fickle, hopeful and discouraging, fulfilling and neglectful, enjoyable and frustrating’ (7). Similarly, friendships and relationships within the research environment can maintain this paradox by being both fleeting and sustained, positive and energising, but create exhaustion, meaningful yet somehow superficial. Friendships, by definition, require shared trust, emotionality, and negotiation (Edirisingha et al. 2017; Sexton and Sen 2018). These emotive aspects contain a sense of belonging, love, closeness, and familiarity – which become embodied and enacted in our posture, facial expressions, what we wear, how we speak, and so on. Rawlins (1992, 271, in Stevenson and Lawthom 2017) defines a friend as ‘somebody to talk to, to depend on and rely on for help, support, and caring, and to have fun and enjoy doing things with’. Thus, friendships often involve self-affirming properties, such as loyalty, appraisal, acceptance and feelings of security and protection. Indeed, a central theme in research on friendship is the notion that it operates at the interface between the self and other. The self is affected (created, even) by positive experiences and interactions with a friend, and through friendship one becomes more empathic, self-aware, and attentive to the needs of others (Bukowski and Sippola 2005). This echoes Aristotle’s ‘perfect friendship’ whereby friendship is about the recognition of ‘moral good’ (Cooper 1977). Aristotle also defines friendship as:

A free relationship of two or more people, (1) who bring into the friendship some personality strengths conducive to forming a relationship and attractive to the respective other, (2) who develop an attitude to, appreciation and understanding of each other through past interactions, and (3) who repeatedly act out their friendship with and towards each other through a variety of activities (Aristotle in Langkamp 2021, 8)

From this definition, friendship can be understood as a self-initiated performance which involves two actors repeating interactions in various contexts based upon their shared history.

Children and childhood

Developments in contemporary childhood studies points to the multidimensional, heterogeneous nature of children and young people’s lives, and the ‘socially constructed nature of childhood itself’ (Ní Laoire et al. 2010, 156). Children are not passive receivers of culture, learning, identity but agentic, subjective beings who actively shape their own social worlds (James 2007). Therefore, constructions of friendships and participation in the research process is similarly multidimensional and co-constructed, with children actively carving out their own relationships and daily lives.

Research has shown that friendships, play, and inclusion are important in children’s lives (Jones 2005; Spencer-Cavaliere and Watkinson 2010; Welply 2022). This is also the case for newly arrived migrant learners, Manzoni and Rolfe (2019) found that children commented that the fastest way to feeling happy in school and home is to make friends. For adults, perhaps constructions and daily navigating friendships may look different than for children. For example, for children, school plays an important role in supporting peer-relationships, friendships, inclusion, and positive wellbeing. Jones (2005, 65), in a study with young children in England, found that ‘strong messages emerged about the importance of effective behaviour management’ and the important role teachers have; children also commented on a strong wish to be included in school, friendships and relationships.

Against this backdrop of friendships and contemporary childhood studies the research question we hope to answer is ‘what does friendship mean in this context of participatory research with primary-aged children?’ Brady et al. (2018), highlighted the importance of engaging with marginalised children and young people in participatory ways, but were specific in the importance of ‘working with’ and ‘working alongside’ children and young people (Brady et al. 2018, 30). The researchers recognised that, when successful, this type of research work is often underpinned by flexibility, sensitivity, and critical reflection (Brady et al. 2018). However, good participatory research frequently relates to the practical requirements and implications, rather than emotional, spiritual, and relational connections that can be built with children in the context of participatory research. This paper does not detail the nature and expressions of friendship in society (see Langkamp 2021; Murphy 2019), and it is not an exploration of participatory research specifically (see Banks, Herrington, and Carter 2017), or how arts-based methods were utilised within the researchers’ respective projects. Instead, this paper focuses on the practicalities and dimensions of ‘friendship as method’ (Tillmann-Healy 2003) within participatory research with children, valuing pupils’ voices (James 2007) to consider possibilities, challenges, and spaces for ‘friendship as method’ as an approach with children to (re)examine and compare two research projects.

Context

For both of our individual PhD projects, we engaged with methodological approaches underpinned by participatory research onto-epistemologies; we involved primary-school pupils in creative workshops, in their respective schools in the north-east of England. Holly explored children’s perspectives and experiences of belonging and cultural, ethnic and linguistic identities, and Nikki worked with children who instigated harm within the home; both of us found that our embodied, creative approaches to collaborative work with children, which involved all forms of creativity, from dance to painting, was underpinned by our relational practice. This was potentially due to our related professional backgrounds (Holly [first author] is a qualified teacher, Nikki [second author] is a registered social worker).

Participatory research can be understood as ‘recursive cycles’ where findings cycle back into the research process, changes are made and further research is carried out (Banks, Herrington, and Carter 2017). Crucially, people being studied also carry out aspects of research; there is a desire for co-production, compassion and understanding, and democratic and community engaged action. As Cook (2009) reminds us, this process is inherently complex and ‘messy’. Educational action research can be understood as an approach to knowledge creation that centres on researchers working with schools and practitioners, bridging the gaps between theory and practice, and prompting change and educational improvement in practice and pedagogy (Huang 2010). Therefore, participatory research and action research and closely linked – in ideology and practical approaches. Participatory research includes participants in knowledge creation and understanding – with a focus on how the voices, engagement and choices of participants can be negotiated and respected throughout the research process. Action research, whilst participatory, focuses on the outcomes of research for practice, and the ‘desired change as a path to generating knowledge and empowering stakeholders’ (Huang 2010, 93). Within this context, and given both researchers’ engagement with participatory research, this paper considers how ‘friendships’ were negotiated and expressed within two research projects.

Method

Some academics have highlighted the limitations of researching with ‘friends’, acknowledging the so-called ‘strangers on a train’ phenomenon, whereby individuals involved in research are more likely to be open and honest with one another when disclosing sensitive experiences with strangers than with their friends, due to the level of anonymity provided (Derlega and Chaikin 1977; Sassi and Thomas 2012). Nevertheless, many feminist epistemological approaches highlight the importance of recognising the entanglement of self and other when researching topics relating to social justice, thus highlighting that women may respond and enact disclosures differently to men when engaging in the performance of friendship in their method (Sassi and Thomas 2012).

Friendship in participatory research

Children are often seen to be hidden voices in co-production, collaboration, and participatory work, and this is often compounded when they have other marginalised identities (L. M. Brady et al. 2018). Participatory research with adults has provided more evidence of how participatory methodological approaches can facilitate the development of friendship between researcher and researched, as found by Sexton and Sen (2018, 875) who sought ‘to examine the nature of the relational dynamic between [the researcher and researched] which moved from a collaborative partnership towards friendship.’

Friendship as method

A fundamental difference between participatory research and ‘friendship as method’ is that the former platforms power and utilises the research process to make change, whereas the latter platforms rich emotionality and shared lived experiences and is more closely aligned to interactive interviewing or collaborative witnessing (Tillmann 2015). Using friendship within a methodological approach has been defined as ‘a form of qualitative inquiry which involves researching with the practices, at the pace, in the natural contexts, with an ethic of friendship’ (Tillmann-Healy 2003, 730). Whereas when exploring friendship in general ethnographic research, there have been critiques that research is ultimately exploitative and instrumental, thus the blurring of friendliness, camaraderie and friendship requires further reflection (Ramírez-I-Ollé 2019). ‘friendship as method’ is about building relationships with participants, or co-researchers, in an organic and well-paced way, which is ethical, evocative, and pertains to an emotional connection. As to whether these relationships can be maintained has been debated:

It is, however, usually unrealistic for a mutual, close and lasting friendship to develop between researcher and every participant in her/his study, drawing upon all these elements, particularly if there are substantial numbers of research participants (Owton and Allen-Collinson 2014, 5)

Nevertheless, Tillmann-Healy (2003) says that researchers do not need to adopt the ‘whole vision’, but a ‘stance of friendship’ by adopting a philosophy of mutual respect, honouring people’s voices and stories, listening with empathy, and seeking deep understanding of the context. By taking this stance, there is an opportunity to reduce the hierarchical distinctions that often exist between researcher and researched (Tillmann-Healy 2003). Furthermore, by centring oneself in friendship, all aspects of the research will invoke an ethics of care, empathy, and emotion between all those involved in the research. The relationship will be the primary focus, with the data, or data elicitation approaches of secondary concern.

Examining Marina de Regt’s article on her friendship with Noura from Yemen, Diphoorn and van Roekel (2019) highlight a different way of utilising friendship in research, stating:

Friendship does not necessarily function as method, but is a crucial way of accessing and collecting data, especially over longer periods of time and when fieldwork sites transform into war zones. Furthermore, she calls for more honest discussions about the benefits conjured from intimate relationships and the need to debunk a reigning demonishing gaze in anthropology about the role of money (and other financial dimensions) in maintaining friendships. (Diphoorn and van Roekel 2019, 9)

The relationship-building aspect of the work resonated with both Holly and Nikki. Within community-based research, ‘friendship as a method’ has gained enthusiasm (Cherry 1996; De Regt 2015; Heron 2020; Stevenson and Lawthom 2017; Tillmann 2015; Tillmann-Healy 2003). However, there is a paucity of literature and empirical studies on ‘friendship as method’ with children and youth, as such, Holly posed the idea of using our respective doctoral projects as empirical case studies and applying friendship as our unit of analysis. For the remainder of this paper, we will consider the complexities of friendship and relationship-making in participatory research with children. When we refer to children’s participation in this paper, we are referring to their involvement in our respective doctoral studies. ‘friendship as method’ is utilised as the lens to re-understand our reflections on our relationships with them and the embodied performances of friendship.

Ethical considerations

Within the complex and potentially ‘sticky’ ethical area of ‘friendship’ in research, we draw on relational ethics, namely Carolyn Ellis’ (2007) theory of relational ethics which requires researchers to act from our ‘hearts and minds’. Ellis (2007) sees relational ethics as the values of ‘mutual respect, dignity, and connectedness between researcher and researched, and between researchers and the communities in which they live and work’ (4). This requires us to deal with the reality and practicalities of changing relationships and dynamics with our research participants over time, for instance, if our participants become friends with each other, or if participants become friends with the researcher.

Within relational ethics, researchers such as Tillmann-Healy (2003, see also Brooks 2006) have studied friendships as an approach to research, where ‘researcher’ and ‘friend’ roles should weave together, enriching, complicating, expanding each other. As participatory researchers, we acknowledge the exciting possibilities and challenges of ‘friendship as method’, supported by relational ethics (Ellis 2007), yet there is a distinct lack of work focusing on ‘friendship as method’ with children and young people – between children as co-researchers and between co-researcher and adult researcher. This is despite some of the particular ‘sticky issues’ of friendships between adults and children, not least due to the differing power dynamics.

As we were both conducting research within school spaces, this provided in-built safeguarding policies and procedures for us both to follow if there were any concerns brought to light during our respective projects. In both cases, we received ethical approval from our respective University departments, and gained consent from those with parental responsibility, and assent from the children in each case. There were also unexpected ethical issues which arose during both cases. For Holly, all the children were recruited from year six, the final year of primary school, and the project explored children’s perspectives, experiences and imagined stories of identity, inclusion and belonging. Similarly, for Nikki, 21 children were involved, but 13 of these were recruited from years five and six, with the focus of the work being about emotions; the impact the social world has on our emotions, and how our emotions change the way we respond to the world. Years five and six have other challenges for children: it is a period of transience as children move from primary to secondary school or prepare to do so. Thus, friendship is frequently a core feature of their experience, as they may become aware of the future and the challenges associated with maintaining and leaving friends or making new friendships. This can be a period of perceived ‘growing up’, influencing perceptions of the self, friendships, and the relationships with the wider world.

Reflections on case study one (Holly)

Case study one is based on Holly’s doctoral study, which explored children’s experiences and perspectives of belonging and inclusion in school and their wider lives, including narratives of identity, ‘difference’ and ‘otherness’. The project included two multicultural primary schools in north-east England over an eight-month period, and 27 children engaged in weekly creative and performative arts-based workshops (including paintings, drawings, storyboards, focus groups, and dance/drama). There were 13 workshops in school one and 17 workshops in school two. Inspired by critical approaches to children’s voice in research (Couldry 2009; Fairey 2018) and participatory research (Banks, Herrington, and Carter 2017; Clark 2017), the participants were invited to collaborate as ‘co-researchers’. Specifically, they were invited to help co-analyse the data through interactive pinboards (identifying themes, reading and discussing each other’s comments, checking for clarity and meanings etc.), and they helped disseminate the findings through dance and drama performances to their school.

Vulnerabilities and reservations

Friendship characteristics can look very different for adults and children (Heshmati et al. 2021), informed by social norms, power dynamics, cognitive development, and past experiences, not to mention the complex and shifting ethical considerations and positionality the ‘adult’ researcher must consider. I wanted to pick up on the notion of ‘positionality’ then, in other words, how did the child co-researchers ‘see me’? Did they see me as a ‘friend’, a ‘peer’, a school visitor, a researcher-artist, a teaching assistant? Sometimes participants would refer to me as their ‘bestie’ or ‘bestie pops’, demonstrating the positive relationships fostered between myself and some of the co-researchers:

Yeva: [Holly] is my bestie pops!

Furthermore, recognising vulnerabilities is a key aspect in ‘friendship as method’. While the children were engaging in their art activities, there were voice recorders on the tables, and at the start of each session I reminded them about how the voice recorders would be used and why. The fluidity of conversations often came to an abrupt end when the participants remembered or noticed the recorders on the table. In these cases, they stopped what they were going to stay, they lost their train of thought, they reminded each other that I will hear it, or they told each other to stop talking as the ‘teachers might hear it’. There was an expression of vulnerability here as their stories and ‘secrets’ were threatened with discovery.

Aminah: I don’t play with them [her classmate]

Aasab: Don’t say anything, they’ll come after you!

Was this a risk to the trust in the relationships I had created with my co-researchers, as they thought what they said was not going to be confidential? By withholding voice because they were aware of the research project, demonstrates that the children did not see the conversations as just friends talking, but a surveillance activity which required behavioural modification. Some children explained how they often ‘forgot’ they were being recorded, and this surveillance impacted them differently at different points.

Fareeha: When you, like, put it on the table, I didn’t even notice it was there, but it felt weird.

Teresa: It felt kind of weird.

Asman: At first, when I erm [pause] when you mentioned it, I was like [facial expression, confusion] why is she interviewing us with this thing?! But, as time goes on, you kind of forgot it’s there, as if it’s invisible, it has superpowers, it just wants to listen to our conversations.

Sarah: It listens to all our voices, it’s a creep, it think it’s creepy.

These participants associated the recorder with negative connotations, ‘creepy’, ‘weird’, ‘invisible’, and Fareeha and Asman sometimes ‘forgot’ the recorder was there. However, interestingly, some participants enjoyed the experience of being recorded and called the voice recorder their ‘bestie’.

Furthermore, just as the co-researchers expressed vulnerabilities and reservations, in my role as the ‘adult-researcher’ I was reserved at times, withholding my views and vulnerabilities when discussing belonging and inclusion, as I did not want to influence the children’s stories by presenting ‘definitions’ of what I thought these concepts meant. I was interested in hearing the children’s voices, but for Tillmann-Healy (2003), friendship-as-method requires researchers to move from ‘studying them to studying us’ (735). Thus, fostering friendships with my participants, perhaps would have required me to put myself out there, to open up and share my stories and views on belonging, cultural identity and inclusion, just as the participants did.

Friendship facilitator

I saw the co-researchers on ‘good days’ and ‘bad days’, and my relationships evolved and devolved: sometimes they were seemingly bored and non-talkative, other times they were excited and full of ideas. I reflect on how participatory research approaches are a vehicle through which friendships are both facilitated and reduced. For instance, while the children had choices and over some aspects of the project, collaborating on more equal grounds much like the negotiations within an organic friendship (Tillmann-Healy 2003) and acting as critical friends offering feedback (i.e. using interactive pinboards), sometimes I felt the need to ‘step in’ as the adult researcher to keep the project on track. Sometimes I had to wear my ‘teacher- hat’ to stop arguments and keep everyone safe, an ethical obligation as a researcher. Consequently, I wonder if this took me out of the friendship sphere and into a teacher-pupil dynamic. Utilising friendship as a tool can facilitate a sense of openness and trust within the research relationships, generating a deeper level of reflective data (Cann and DeMeulenaere 2012), there was also the question of meaningful versus transactional interactions and when the children moved beyond the interactions and behaviours required of the research process, the ‘teacher hat’ came on. This may undermine the friendship approach, but arguably the research felt meaningful and transactional. If friendship can be understood as existing on a spectrum then the power dynamics were disruptive, as I had reservations that trying to be their ‘friend’ – as in, trying to establish a relaxed and more ‘equal’ power dynamics – may not always yield the ‘best’ results. Nevertheless, the care and sense of commitment to the children was something which developed at the ethical level and emotional level. At times, increased participation felt risky; juggling the promises made to participants, schools, the ethics committee, the spatial/time constrictions, alongside the desire for productive relationships between researcher and co-researchers were all in play.

Consequently, perhaps my role, at times, was a ‘friendship mediator’ or ‘friendship facilitator’ on the spectrum of friendship. As mentioned above, sometimes I mediated aspects of the decision-making and direction of conversations, and I negotiated tensions and conflicting views between the co-researchers, facilitating (new and existing) relationships/friendships between the co-researchers. My role was fluid and dynamic, moving around a spectrum of friendship, sometimes this was a positive experience and other times I needed to be the ‘bad guy’, or a ‘critical friend’ (Tillmann-Healy 2003).

Reflections on case study two (Nikki)

The second case study was a small component of a much larger doctoral research project into ‘child-to-parent violence’, and was conducted within a specialist primary school for children with social, emotional, and mental health (SEMH) needs. This means all children had an education, health and care plan (EHCP; a plan identifying children who have specialist educational needs), specifying that each of the children had a primary need of SEMH. This school was chosen for three reasons: I had a pre-existing, on-going relationship with the school, which reduced the concerns of gatekeepers; parents of children within the school had been requesting additional support for ‘child-to-parent violence’; the school had a pre-existing interest in this topic and had already conducted their own research relating to its relevance to their cohort of pupils.

This case study involved observations (ethnographic data) over a period of four months, and weekly arts-based workshops with 21 children for a period of two months. The arts-based workshops ran every Wednesday, with five workshops per-day, hosting four or five children in each workshop. The sessions were led by the children, but all started with a ‘check in’ and ended in a body-mapping exercise. Sessions could involve play, painting, dance, drawing, movement and construction. Children joined timetabled sessions each week and were recruited via referral from their teacher, who had identified that the referred child struggled with their emotions, and parents had reported ‘child-to-parent violence’. Rather than focusing upon ‘child-to-parent violence’, the workshops were participatory and focused upon emotions and emotional expression rather than violence or harms, thus the work was for the children, rather than about them, as they were able to focus upon themselves, their emotional states, and their own wellbeing (Bradbury-Jones and Taylor 2015; Honkanen, Poikolainen, and Karlsson 2018). Thus, the focus on emotionality met the first of many criteria for friendship-as-method; immersion in emotions, in comparison to participatory action research, which is more interested in power and change.

In each workshop, children knew one another, but there was variability regarding established friendships. Some groups had a core group of four children with three established friends, others were a group without clear friendship dynamics. As such, conceptualising friendship as an existing phenomenon was complex in this environment and differed between groups. It also provided opportunities for my relationships with them as individuals, and as a group, to evolve in an organic way. Thus, in some groups, I was the ‘outsider’ and I needed to build trust through care and time; in others there were interactions I perceived as the processes typically experienced in group forming, storming, norming, performing. The participatory approach to the workshops meant that, like case study one, children were co-researchers and not participants. The child-centric nature of each individual workshop means that they were not comparable across activities, but there were patterns in the way relationships were approached and interpreted, and, on reflection of the process, I understand this as sitting on a ‘spectrum of friendship’. My relationship with each child was different.

Vulnerabilities and reservations

A key component of ‘friendship as method’ is in the expression of vulnerability, and this was something I did not share with the child co-researchers. This was primarily a protective strategy as I did not wish to share my own emotions or challenges which could burden the children unnecessarily. However, this could also be said of some adult friendships, and the difficulty some adults can have in expressing vulnerabilities to others. Nevertheless, the children were able to be vulnerable with one another, and felt safe to do this. In their workshops, young people explored their emotions and the impact these emotions had on their behaviour through art, and frequently articulated their own lived experiences of victimisation and loss. These expressions of victimisation and loss were navigated by other young people in the group. This was observed through shared disclosures, presentations of empathy, and advice given between each of the group members. For instance, as one child explained why he lived with a foster carer after living with domestic abuse, other children explored their experiences of special guardianships and being removed from their parents after similar experiences.

The children also trusted that if there were aspects of their experiences shared which they did not want included in the research, they trusted me to not include it, despite the differences in power; they understood (as did I), that the relationship and the trust was more important than the research. To ensure child co-researchers were always aware of the voice recorder, I ensured it was visible, on a table in the centre of the room at all times. On a couple of occasions, a child would share an emotionally evocative experience and ask for it to ‘not be on that’ (referring to the voice recorder). On these occasions, I interpreted this as children asking for what they had shared to not be included in the research; they had withdrawn their assent for this example, and so their vulnerability was not included in the general data set. Thus, there was an additional element of trust, as the co-researchers could ‘trust that I will honor their disclosures and try to use them in ways that promote liberation’ (Tillmann-Healy 2003, 739); in this case it was to honour their disclosure by maintaining privacy and dignity. Whilst recognising the importance of assent/consent in research is fundamental to good ethical practice, young children are not as aware of this (hence why their parents give consent on their behalf), thus these requests evidence that there was trust that some conversations, some vulnerabilities were so intimate, so private, that they wanted them to be sacred and shared only within group.

Friendship facilitator

Whilst ‘friendship as method’ is typically an approach whereby researchers use the research process to build and develop friendships, consequently finding the research and the process become intertwined as they navigate social justice issues. In this research I found that rather than becoming part of a friendship with the child co-researchers, my role was as a friendship facilitator. For instance, in the workshops, child co-researchers worked together to identify their own and one another’s emotions, identify and work-through challenging experiences, and documented what was important to them as children, individuals, and researchers. Whilst I worked with them to collaborate on this work, and to navigate the ethical, emotional, and practical challenges to do research which was evocative, challenging, and new for the children, primarily the relationships which developed was more egalitarian within the workshop groups than between myself and the co-researchers.

Children with SEMH needs are more likely to have internalised their experiences of isolation, believing they are not deserving of friendships or positive relationships (Newton, Taylor, and Wilson 1996). As such, developing and maintaining friendships is a difficult task, with my findings indicative of children instigating self-destructive behaviour which reinforces that they are not ‘deserving’ of love or friendship. This challenge was compounded when all of the children in this case study had primary SEMH needs. As such, my position as an adult was one which emphasised power over children who were already disenfranchised and had experienced multiple losses through their experiences of education.

Through the workshops, children were able to have space in which to explore some of their more challenging emotions and life experiences; expressing and exposing their vulnerabilities and having other children respond to them. These interactions between the children, and supported by me, evoked but contained those vulnerabilities so that children could share very painful histories in a cathartic way. For instance, sharing experiences of domestic abuse was a common occurrence in the workshops, as were narratives of being taken into care. More rarely, but just as importantly, I needed to hold the space as a mediator between children who had experienced conflict outside of the workshops. For instance, one group of children opted to cover the emotion of ‘anger’, which evoked particular feelings with one of the children:

Stacey: I know a time when I got angry … it was just the other day and you [other child in the room]… you wouldn’t let me play … I was talking and you said ‘ignore her’ and you wouldn’t let them play with me again. Jonathan: Yeah, but that time you were just being annoying.

Annie: Yeah … I mean you were a bit.

Stacey: [appears physically distressed]

Researcher: Can you all see why [Stacey] might feel angry about that though … how it must feel to be left out?

Annie: Oh, I hate being left out … it makes me cry.

Jonathan: Yeah, I’d be angry.

Stacey: I didn’t cry though.

Researcher: But it would have been ok if you did cry; that was upsetting, and you’re allowed to be upset when your feelings are hurt.

In the above example, the three children had an existing friendship, but tensions had arisen over the course of the week between the two sessions. Here my role was not to be a friend to the group, but rather help them in navigating the conflict that can arise in friendships. I was facilitating a dialogue, with the potential to be framed as the ‘bad guy’, by talking through the difficult emotions experienced by all three children.

Discussion of the findings

It is important to note that, whilst we are exploring the relevance of friendship in research in schools, we are doing so as adults, and children are not adults. Indeed, since children experience the world in different ways to adults (James 2007), we can assume that children experience and conceptualise friendships in different ways to adults too (Heshmati et al. 2021). Friendships in research require critical, reflective, and reflexive negotiation, and participatory research in particular can assist in the opening of these relationships (Sexton and Sen 2018). Nevertheless, this was difficult in the context of research with children; particularly due the different opportunities to navigate and negotiate the relationships, and the unequal power dynamics. Furthermore, children may have very different understandings of friendship to us as adult researchers. Not only were we managing the ethical and power-dynamics of researcher-researched, but we were also navigating the social roles of adult and child. In each of our respective projects we were able to meet with pupil-researchers weekly, laugh together, listen to one another, debate, discuss, problem-solve, and create artwork together. For both of us, our research ‘procedures [were] those we use to build and sustain friendship: conversation, everyday involvement, compassion, generosity, and vulnerability’ (Tillmann-Healy 2003, 6).

Research has shown that, in childhood, friendships largely involve play, mutual enjoyment, and learning which is crucial for socioemotional and cognitive development (Lu and Posada 2011). It is through sustained and repeated encounters that friendships develop, which we achieved in both case studies through an active process of talking, listening, exchanging material, co-creating artwork and sharing our experiences, in line with the friendship as method approach (Tillmann-Healy 2003). There was also an element of playfulness between ourselves and the pupil-researchers (such as when Holly was called ‘bestie pops’). The creative methods in both case studies enabled us to ‘play’ with ideas, emotions and stories together in creative ways. Nonetheless, reflecting upon ‘friendship as method’, was the question whether the relationships we fostered were a ‘true’ or ‘meaningful’ friendship? In our initial, informal discussions on friendship in our respective research projects, we considered friendship as being on a continuum. However, upon further reflection of our experiences, it is better understood as a spectrum. The spectrum of friendship does not pertain to a hierarchy of friendship, nor does it quantify or categorise it; rather it recognises that friendships provide different things to different people at different times. The spectrum of friendship that we posit, does not undervalue the relationships built within a research environment, but friendship is as measurable as the colour blue; it is embodied as it is emotional.

In each case study, we were able to identify different positions of friendship at different times and with different children. In the example of case one, some of the co-researchers were more excited and invested in the dance and drama workshops, bonded by a desire to create a good performance for the school, and this is demonstrative of the importance of action in ‘friendship as method’. Some of the co-researchers were less willing to participate at certain times, disengaging with the idea of creating dance/drama (‘I do not want a main part’, ‘drama isn’t really my thing!’). This perhaps exemplifies the honest, working relationships we were able to establish with the children – they felt autonomy over decision-making and felt they could share their reservations and reluctance at times.

Vulnerabilities and reservations

Creative and arts-based methodological approaches can provide catharsis for participants, who are able to explore otherwise challenging and evocative experiences, but this approach also provides a space for vulnerability (Bird 2018; G. Brady and Brown 2013). As we have already acknowledged, vulnerability is a key part of developing friendship, but this should be mutual vulnerability (Tillmann-Healy 2003). This has been supported in participatory research reflecting on friendship in the research process, which identified both researcher and researched as holding positions of ‘truster’ and ‘trusted’ and this can be supported through shared expressions of vulnerability, including the sharing of painful biographies (Sexton and Sen 2018). In both case examples, we constructed a space for the children to be vulnerable, but it was not always appropriate for us as researchers to be vulnerable too. How could we burden children with our experiences when they had already experienced so many difficulties in their short lives? How could we share our personal experiences and understandings of abstract concepts (such as identity, belonging) without influencing the children’s views, in a way that does not harm which is fundamental to research ethics? Nevertheless, ‘a one-sided friendship can also be stable if the sole nominator does not feel an imbalance between giving and taking in the relationship’ (Block 2015, 164), meaning that reciprocity is not necessarily fundamental to stable, positive relationships, as long as both parties are comfortable with the arrangement. Thus, on the spectrum of friendship, if the central friendship is a balanced, reciprocal relationship, ours diverged from this as a one-sided approach to intimate disclosure agreed within the interaction.

Scholars have documented the challenges of the blurring lines between friend and researcher when they are friends but also still the ‘subjects’ of the research. Murphy, in her ethnographic study, recalls how her participants had conflicted feelings when reading Murphy’s book and recalling her comments:

She read the passage aloud to me in the middle of the otherwise jovial book launch with a slight tremor in her voice and stated firmly: ‘I can’t believe you were there remembering my every word, writing about my every move. I just forgot you were a researcher too. I mean, I knew you were doing this, writing this book, but you became my friend, one of my main supports during this campaign’. (Murphy 2019, 21)

This was also evident in the work we conducted with children, as they were being recorded, and this resulted in a variety of responses. That, for some children, the voice recorders became ‘invisible’ reflects the idea that when there is deep immersion in the field, and the lines between friend and researcher becomes blurred, participants can forget the researcher is ‘researching’. Children often met the voice recorders with suspicion, sometimes reminding each other not to say certain things because ‘It can hear you’, or requesting certain conversations not be included in the data-set. In both of our case study examples, child-participants felt comfortable enough to express their preference. However, both showed examples of vulnerabilities in the research and how some participants chose to express a sense of reservation. Being recorded and then recalling the purpose of the workshops increased the sense of vulnerability in the room; whereby the vulnerability was induced by the research process. This created a tension in the researcher-researched relationship, as it was clear who was being recorded, who was most vulnerable, and it was not us or our private conversations being utilised as data.

Friendship facilitator?

As to where on the spectrum of friendship our co-researchers sat, we also need to consider alternative roles that may sit within or alongside friendship. In their study, Stevenson and Lawthom (2017) stated ‘in further interrogating the nature of the relationship between researcher and participant, we argue that whilst all participants may not be considered as friends, we cannot afford to treat them as distant others either.’ (5). Attempting to frame our own interactions with child co-researchers as child-led, respectful and empathic is not enough to subscribe fully to ‘friendship as method’ but is more representative of good researcher-researched relationships and can have valuable aspects when considered as an aspect of a contextual spectrum.

Friendship facilitation has been acknowledged as an important consideration in the development of inclusive classrooms, particularly when the cohort is diverse or may otherwise have difficulty developing or sustaining friendship (Bergen 1993; Salend 1999). All of which was relevant to our co-researchers. Holding friendship through play and playfulness has been associated with improved wellbeing in childhood, and we both attempted to facilitate environments where children could play and be playful with one another, as well as ourselves (Rose et al. 2022). Thus, whether it was the friendship, or the playfulness that was most important in the research is an important question beyond the scope of this paper but deserving of further exploration.

When friendships are not to be facilitated, but moderated, has also been considered in previous research, with schools identified as key sites of friendship constraint, as school environments can help moderate, mediate, and manage escalating peer emotional and social tensions (Juvonen 2018). Similarly, the creation of each of our workshops facilitated a space where friendships could form and develop, thus we provided a space for friendship and creativity, but we also had a responsibility to the child co-researchers to ensure their emotional wellbeing was supported by managing and reducing opportunity for conflict (Juvonen 2018; Rose et al. 2022). Thus, we were less part of the reciprocal friendships, and more founders and sustainers as an active part of a one-sided friendship, through our researcher role.

Final remarks

For this paper we attempted to answer the research question: How does ‘friendship as method’ work in research with children? The answer is not simple; just as friendships are paradoxical, so are researcher relationships, and the two became entangled in both of our projects. We reflect that friendship in research is not a linear continuum, but a spectrum. It was both liberating and restrictive, as we found that our creative, participatory workshops provided a space where children had the freedom to lead the research and create space for emotive connection; however, the restrictive time-frames, and conducting both projects within school environments limited some of the opportunities and possibilities. There was a fragility, as our encounters were short-lived and could not continue beyond the length of the research projects; but they were also potent, as there was a depth to the intimate narratives shared by many of the children. These relationships were enjoyable, as we learned about one another, learning about each other’s lives, and finding joy in observing others’ friendships; but also frustrating when we needed to ‘wear the teacher hat’ and create boundaries. As such, there are certain dynamics of ‘friendships’ that can play out in participatory research when exploring participants’ complex insights into the research topics and when considering data collection and dissemination and the impact on practice as a result of research that require future exploration.

Whilst a huge amount of the joy we encountered when researching with children came from ‘play’; the possibility of play-as-method has potential for further exploration, but our experience of ‘friendship as method’ is more fluid and nuanced than play. We found that we could be a friend, facilitator or moderator, at different times, with different people, sometimes over the course of one workshop. Nevertheless, we believe it is important that future work explores the importance of play in research and its relationship to friendship in research with children. Furthermore, there is space to explore the spectrum of friendship, the researcher’s role, and unpack what friendships mean to different groups of children in various research contexts. As this paper was a retrospective analysis, children were not involved in producing, reflecting on, or analysing the material of this paper. This provides opportunity for future research where children could consider their own position on the spectrum of friendship in research with adults and children.

Researchers may negotiate relationships which are fulfilling and supportive; they can promote an ethics of care and genuine interest in the lives of their participants or co-researchers. Friendship develops organically, with time, respect, and mutuality, and this should always be the basis for good relationship-based research (Tillmann-Healy 2003). We recommend that future research with children within the school environment should begin with the lens of friendship, and particularly participatory research should understand the importance of friendship in the interactions, this is particularly important when working with children from minorised or with additional vulnerabilities, when their relationships with others are often entangled with experiences of loss. As to whether friendship can improve continued engagement in participatory and action research is beyond the scope of this paper, but this must be explored and examined in future work.

Acknowledgments

The authors would like to thank the individual children who gave their time and stories, and to the schools involved. Thank you to the anonymous reviewers for their helpful insights. Thank you to our PhD supervisors.

Disclosure statement

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

Additional information

Funding

This work was supported by the Economic and Social Research Council [Grant Number: ES/P000762/1].