The role of situated talk in developing doctoral students’ researcher identities

ABSTRACT

It is widely recognised that an important aspect of doctoral study is the development of a researcher identity. However, little is known about how to support this. Although previous research has highlighted the importance of discursive engagement for researcher identity development, no studies examine talk or discuss how identities are constructed through interaction. This article examines how doctoral students’ researcher identities develop during tutorials on a professional doctorate in education. Analysis reveals how researcher identities are constructed and accomplished, turn by turn, during study-based talk. Researcher identities are co-constructed with and confirmed by peers, and verified by a tutor who validates students’ actions and describes experiencing similar, often difficult, processes herself. Knowledge and understanding facilitated by discussion prompts identity development and tutorial talk builds a sense of belonging and confidence, acculturating students into the research community. The article makes an original contribution to research by analysing situated talk to show identity accomplishment in action. The article also makes recommendations for both practice and further research which include setting up opportunities for doctoral students to talk and share experiences, and close analysis of doctoral interactional events such as supervision meetings and peer support groups.

KEYWORDS:

• Doctoral researcher identity
• tutorial
• professional doctorate
• situated talk
• interaction

Introduction

The doctoral journey involves developing knowledge and research skills, but equally, and crucially, it also involves identity formation (Austin and McDaniels 2006; Green 2005; Jazvac-Martek 2009; Sverdlik et al. 2018) and identity transitions such as professional (e.g. teacher) to researcher (Labaree 2003). Recognising both the difficulty of developing new identities (Hall and Burns 2009; Jazvac-Martek 2009; Leshem 2020) and the fact doctoral students’ ability to internalise and enact new identities may influence whether they continue or withdraw from their studies (Baker and Pifer 2011) as well as the time they take to complete (Austin and McDaniels 2006), scholars highlight the importance of helping doctoral students develop a researcher identity. While recognising that learning and identity development are closely connected (Baker and Lattuca 2010; Foot et al. 2014), few studies have examined empirically how identities are constituted and negotiated through doctoral study-based activities (Baker and Pifer 2011).

In this article, we conduct an empirically based examination of the development of researcher identities during tutorials within a professional doctorate in education (EdD) in the UK. Professional doctorates, focused on contributing to both knowledge and practice, are conceived as professional learning and undertaken for different reasons, ‘implicating various identities and satisfying a number of needs’ (Wellington and Sikes 2006, 728). The professional doctorate is distinguished from a traditional PhD by a focus on an individual’s practice and the inclusion of a modular phase, often undertaken as part of a cohort, with additional assessment requirements (Bourner, Bowden, and Laing 2001). This structure allows for interaction with peers, course tutors, guest lecturers and supervisors, and for a variety of learning events such as seminars and workshops as well as individual supervision meetings. Professional doctorates are frequently undertaken by experienced professionals for a complex mix of reasons, some seeking to affirm or challenge their identities (Wellington and Sikes 2006). They are often working full-time, their emerging identities as doctoral researchers supported through pedagogic practices and relations such as formative assessment and supervision (Crossouard and Pryor 2008). Professional doctorate candidates are typically mature students and as such may be more likely to have family and caring responsibilities in addition to managing demanding professional roles. For example, Webber (2017) notes that women professional doctorate candidates experience turbulence as they negotiate identity shifts alongside personal challenges and caring responsibilities.

Acknowledging the complex work, lives and environments of professional doctorate candidates, and the fact that deliberately affecting learning on a professional doctorate is challenging due to the different spaces (physical and social) that candidates interact in and the complexity of professionals’ learning (Pratt et al. 2015), the course team anticipated that tutorials would help improve retention, enable timely completion, and encourage greater participation in the research community. In this paper we analyse ways in which EdD students claim, ascribe, and verify their own and each other’s researcher identities during tutorial talk, demonstrating the significance of a focus on situated interaction. The analysis suggests that such study-based talk can contribute to researcher identity development and foster a sense of belonging.

We begin with an overview of our theoretical understanding of identity. We then review approaches to exploring doctoral researcher identities, noting the absence of studies focused on situated talk. Next, we situate the study, and describe methods of data generation and analysis. Drawing on an illustrative tutorial sequence, we demonstrate how doctoral researcher identities are manifested during tutorial talk and highlight the significant role of interactants (peers and tutors) in identity work. The article concludes with recommendations for practice and research.

Identity

Identity is defined by Gee (2000) as ‘being recognized as a certain “kind of person” in a given context’ (99) and by Bucholtz and Hall (2005) as ‘the social positioning of self and others’ (586). Both definitions highlight social context and engagement with other people as integral. Drawing on poststructuralist theories of language and meaning, we see identity as social (Morgan and Clarke 2011), active (Sarangi and Roberts 1999), situationally emergent (Bucholtz and Hall 2005), and influenced by context (Donaghue 2020a; Gee 2000; Olsen 2011). In this paper, we understand identity as something that we do in situated social practices (Sarangi and Roberts 1999) rather than an essentialist, fixed psychological attribute that we have.

People ‘perform’ (Block 2017) or construct identities for themselves and others in situated social practice. While identities can be constituted through the way we act, dress or move (Block 2017), talk is perhaps the most important resource in forming identities (Varghese et al. 2005). Identities emerge and develop in interaction (Benwell and Stokoe 2006; Bucholtz and Hall 2005): as people claim, ascribe, verify and contest identities, they are ‘talked into being’ (Heritage 1984). While language acts as a process in constituting identity, it is also a product in which identity is made visible (Olsen 2011). In this article, we argue that the analysis of naturally occurring interaction can offer insight into doctoral researcher identity development.

To guide our analysis of identity construction during situated interaction, we draw on identity research in the field of applied linguistics. According to Joseph (2013), identities are accomplished during interaction in three ways: identities are (1) signalled by categories and labels, (2) indexed by ways of speaking, and (3) interpreted by others. Types of people (or identities) can be labelled explicitly (e.g. children calling ‘teacher’, patients saying ‘doctor’) but can also be categorised by activities typical of or expected to be done by incumbents (Fitzgerald and Au-Yeung 2020). For example, talk show hosts can be identified by the actions of introducing topics for discussion, introducing callers, asking questions and making transitions between callers (Fitzgerald and Housely 2002). Interactants also index identities through ways of talking. For example, speakers perform powerful manager identities by claiming expert knowledge (Clifton 2012; Svennevig 2011), exercising the right to evaluate a conversational partner (Heritage and Raymond 2005), taking and keeping the floor (Donaghue 2020b; Heydon 2003), and giving directions for future action (Svennevig 2011). However, claiming an identity is only the first step in the process of identity negotiation. The next step (and Joseph’s (2013) third way of accomplishing identity through interaction) is the hearer’s reaction or interpretation. Claimed identities can be either verified or contested. In an example of contesting an identity, Riordan and Farr (2015), analysing face-to-face and online discussions between student teachers and tutors on a teacher education course, show how a mentor contests a novice teacher’s ‘inexperienced’ identity and re-constructs a more positive one involving knowledge and experience. In another example, Schnurr and Zayts (2011) show how a team leader identity claimed by a newly promoted employee in an international financial corporation is resisted and challenged by other team members as they signal a lack of support for her decisions and question her authority and decisions during a meeting.

These studies share a common underlying belief that identity is achieved through interaction with others in local contexts. They also share a similar methodology. All prioritise naturally occurring spoken data, employ micro analysis techniques to conduct a close turn-by-turn analysis of talk, and all illustrate their conclusions with data extracts. This article adopts the same belief and methodology.

Literature review

In a review of recent doctoral researcher identity literature, Choi, Bouwma-Gearhart, and Ermis (2021) conclude that doctoral students achieve researcher identities through legitimacy, i.e. recognition or confirmation, from self and others, of their competence, confidence, autonomy and agency in scholarly activities, products and communities. This legitimacy is achieved through engagement with activities and processes which McAlpine, Jazvac-Martek, and Hopwood (2009) categorised as formal e.g. submitting a thesis, completing coursework; semi-formal e.g. presentations to other doctoral students, meetings with a supervisor; and informal e.g. conversations with student peers and reading/knowing literature. This ‘doctoral doing’ (Mantai 2019) underlies the development of an academic identity (Paré, McAlpine, and Starke-Meyerring 2006): research activities overlap with identity formation (Jazvac-Martek 2009).

Development and recognition as a researcher also comes from reflection, supportive relationships and discursive engagement. Reflection which focuses on students’ awareness of researcher processes and of their own identity development is cited as impacting doctoral students’ achievement of researcher identities (Ai 2017; Foot et al. 2014). This kind of reflection is prompted by activities such as course projects, workshops, mentorship and collaborative relationships with tutors, and research and writing groups. Tutors being transparent and vulnerable about their own developmental processes also helps students reflect explicitly on their identity transitions. For example, students’ development of researcher identities is influenced by tutors sharing feelings of vulnerability about their writing and research (Lassig, Dillon, and Diezmann 2013) or sharing factors impacting work-life balance (Rockinson-Szapkiw, Spaulding, and Lunde 2017). It is also helpful for students to realise that the development of a researcher identity is ‘punctuated with tensions’ (Choi, Bouwma-Gearhart, and Ermis 2021) so is often not straightforward.

Supportive relationships contribute to the development of researcher identities. Relationships are formed with tutors and supervisors who nurture students’ growing knowledge and research skills, provide students with relevant research opportunities, and give practical advice and feedback (Ai 2017; Dollarhide, Gibson, and Moss 2013; Inouye and McAlpine 2017; Rockinson-Szapkiw, Spaulding, and Lunde 2017). Peers also form supportive relationships as students provide each other with social and emotional support and share scholarly products for critical feedback and peer review (Crossouard and Pryor 2008; Inouye and McAlpine 2017; Jazvac-Martek 2009; Lassig, Dillon, and Diezmann 2013; Maher et al. 2008; Murakami-Ramalho, Militello, and Piert 2013). These supportive relationships engender validation or legitimacy i.e. being seen as knowledgeable and capable of doing research. Validation from academics (e.g. through collaboration on research projects), peers (e.g. through peers’ perceptions of a student being ‘more knowledgeable, accomplished, or otherwise “closer” to being a scholar’ (Choi, Bouwma-Gearhart, and Ermis 2021, 107)) and the wider academic community (e.g. through acceptance of students’ conference proposals) increases students’ confidence which in turn contributes to the development of their researcher identities (Åkerlind 2008; Mantai 2017).

Discursive engagement is highlighted as an important resource for helping doctoral students feel like and become recognised as researchers (Choi, Bouwma-Gearhart, and Ermis 2021; Foot et al. 2014; Mantai 2017). This engagement can be with individuals or groups within the institution such as academics, doctoral programme cohorts, writing groups, or other academic and professional development support groups (Ai 2017; Jazvac-Martek 2009; Klenowski et al. 2011), or groups outside the institution e.g. scholarly communities via academic conferences (Ai 2017). Studies stress the importance of talking to others about research in both formal (Ai 2017; Jazvac-Martek 2009; Klenowski et al. 2011) and informal (Klenowski et al. 2011; Mantai 2017) settings to aid the development of researcher identities. They include a variety of discursive events such as collaborations with other academics (Jazvac-Martek 2009), conference symposium discussions (Ai 2017), interviews for academic positions (Jazvac-Martek 2009), interacting informally with a teaching team (Klenowski et al. 2011), or defending a point of view to an online writing group (Mantai 2017).

Approaches employed in identity research

While highlighting the importance of discursive engagement, none of the studies outlined above show instances of the talk involved in discursive events or examine how identities emerge and are constructed through interaction. There is a growing body of research in medical and business contexts examining how identities are negotiated during naturally occurring, work-based interaction (e.g. Angouri and Marra 2011; Schnurr and Zayts 2011). However, in the field of continuing education, few researchers have examined how identities are constituted and negotiated during naturally occurring talk (i.e. talk not elicited for research purposes). Instead, most studies favour research interviews, preferring, in Silverman’s (2013) terms, to ‘manufacture’ data from interviews and focus groups instead of examining naturally occurring interaction in the field. From the empirical studies investigating doctoral students’ researcher identity development reviewed in this article (n = 22), 17 use interviews. Ten studies employ only interviews, while seven others combine interviews with questionnaires and/or document analysis. The rest use questionnaire logs (McAlpine, Jazvac-Martek, and Hopwood 2009; Murakami-Ramalho, Militello, and Piert 2013), written reflections (Lassig, Dillon, and Diezmann 2013), journals (Foot et al. 2014) or diaries (Ai 2017). These can be viewed as written versions of the interview i.e. focused on asking participants about identity. We have found no studies which look at how researcher identities are constructed in situ during the talk that students engage in as part of the doctoral journey e.g. supervision meetings, seminars, tutorials, or research/writing groups. This is surprising as many studies highlight discursive engagement during these types of activities as an important resource for enabling identity development (Ai 2017; Foot et al. 2014; Klenowski et al. 2011; Lassig, Dillon, and Diezmann 2013; Mantai 2017).

Interviews have been increasingly interrogated and critiqued as a method for investigating identity (Alvesson, Ashcraft, and Thomas 2008; Donaghue 2018). Alvesson, Ashcraft, and Thomas (2008) maintain that interviews assume that an individual ‘houses’ an identity, a more essentialist view of identity which conflicts with the sociocultural stance claimed by most scholars researching doctoral identities. These scholars tend to view identity as situated, social, multiple and dynamic. In addition, few studies recognise that interview interaction can shape as well as/instead of reveal identities (Miller, Morgan, and Medina 2017). Conceptualisation of the research interview has shifted from an empirical data gathering tool to a situated interactional event with metacommunicative norms that can influence the form and content of what is said (Baker 2004; Holstein and Gubrium 2003). Interview participants construct knowledge together and draw on resources which may alter or develop during an interview:

Meaning is not merely elicited by apt questioning, nor simply transported through respondents’ replies; it is actively and communicatively assembled in the interview encounter. (Holstein and Gubrium 2003, 114)

As meaning and knowledge are actively assembled via communication, so too are identities. Identities are dependent on and emergent from a specific interactional context and produced through identity work jointly constructed and negotiated by both participants. Roles and relationships (including previous or personal ones) can affect the way talk develops and affect what is generated. As a socio-culturally loaded communicative activity, an interview can place moral demands on participants (Rapley 2001) and shape how participants conduct and portray themselves. Alvesson and Sveningsson (2003) cast ‘serious doubt on the value of interviews as “realistic” expressions of how people define themselves’ (20–21) and question whether research interviews are sufficient to understand identity processes, calling for methods such as observation (e.g. video or audio recordings) of situated interaction. In a similar vein, Silverman (2013) urges researchers to stop relying on ‘the accounts that subjects give of their behaviour’ and to instead ‘simply focus on what people do’ (43, original emphasis).

Drawing on the view of identity as embedded in situated practices and interactionally accomplished, we examine identities as they emerge during tutorial talk. By analysing naturally occurring, study-based interaction, we show identity negotiation ‘live’ i.e. doctoral students and tutors doing, rather than talking about, identity. This enables us to extend previous research by showing how doctoral researcher identities are accomplished as well as offering an original contribution to the methods used to investigate doctoral researcher identities.

Methods

Context

This article reports on a study carried out with tutors and students on an EdD programme. This particular programme was established over twenty years ago at a post-92 university in the UK and has been a significant route to a doctorate for academics in the university and local educators, with around 50 candidates enrolled at any one time. The EdD programme is part time and typically recruits HE lecturers in education and health science disciplines, and staff (e.g. teachers, head teachers) working in schools or education authorities in the local area. The programme follows a blended model with on-campus study days held on Saturdays and online twilight sessions and tutorials. In the initial phase of the doctorate, students work closely in cohort groups of 10–20 with the support of a supervisor and the EdD team. This early stage is scaffolded by four non-credit bearing modules, designed to support researchers to develop their knowledge, understanding and research skills through activities related to their individual research proposals. The four modules are designed to be completed in two years (see Table 1), followed by a further two to three years in the thesis phase. The first two modules focus on situating the proposed research through a critical examination of policy, literature and the students’ own positioning in relation to research and practice. The next two modules enable students to plan their study, considering methods and methodology, through a small-scale pilot project. On successful completion of these four modules, students proceed, via a confirmation process, to the thesis phase. In this final phase they develop their research, supported by a supervisory team and facilitated group study days and workshops.

Table 1. Doctorate in education programme overview.

Module title	Tutorials recorded
1. Developing your research focus	Tutorial 1 Group A (n = 5; 60 mins) Group B (n = 5; 88 mins) Tutorial 2 Group A (n = 5; 78 mins) Group B (n = 5; 61 mins)*
2. Developing the literature review	Tutorial 1 Group A (n = 5; 58 mins) Tutorial 2 Group A (n = 5; 56 mins)

*The illustrative example analysed in this paper comes from Module 1, Group B (highlighted in bold).

This study involved doctoral researchers from one cohort (n = 10) studying on the first two modules of their course. All research participants were university lecturers. Further details such as gender, research area and age have been withheld to protect anonymity.

Data generation

In each of the first four modules, doctoral students participate in two tutorials designed to create spaces for sense-making. Short tasks and suggested discussion prompts are provided by a module tutor who also facilitates tutorial discussions. For this project, six tutorials from the first two were audio recorded (see Table 1) with the aim of examining study-based talk. Despite limited evidence on attrition in doctoral programmes (Berry, Niven, and Hazell 2022) we were interested in focusing on support in this early stage of the doctorate to help retention and influenced by Mantai’s (2017) assertion that researcher development starts early in the doctoral journey.

Data analysis

Data analysis combined coding and micro analysis of interaction. The first stage of analysis aimed at familiarising ourselves with the data. This stage began with an iterative process of listening to audio recordings, checking transcripts, and noting. We then carried out initial coding on a subset of the data. In keeping with our understanding of coding as an interpretive endeavour, we relied on intensive discussion (Saldaña 2021), comparing and discussing codes to establish an initial coding framework with the understanding that this could be changed or expanded as analysis progressed. During a second round of coding of the full data set (six tutorials), regular research meetings deepened our understanding as we explained and justified decisions, working towards clarification and shared meaning.

The second stage involved segmenting the transcripts according to what participants were doing in terms of speech (e.g. asking or answering questions), or talking about, using aspects such as intonation, discourse markers (for example ‘Ok, let’s move on’) and topic shifts to identify boundaries. See Table 2 as an example.

Table 2. Segments from Module 1, Tutorial 2, Group B.

1	Outline of tutorial
2	Supervision
2a	How things are working with supervisors
2b	Comparing masters and EdD supervision
2c	Pleasure of supervision discussions
3	Admin: whether or not to have breakout rooms (decided not to)
4	Module 1 assignment examples from previous cohort
4a	Differences between the examples
4b	How much of the module assignment might end up in the final thesis
4c	Different writing styles and relating examples to own writing
5	Refining the research questions/focus
6	Admin: notice about a networking event for EdD cohorts
7	Covid affecting study
8	Admin: notice about postgraduate forum
9	Covid affecting study
10	Covid affecting school pupils’ grades
11	Closing

The final stage involved conducting a close, turn-by-turn micro analysis of interaction. Within a segment we looked at talk across turns, paying particular attention to how tutorial participants claimed and constituted identities and how others verified or challenged these identities. For this article, we have chosen one segment from the second tutorial in Module 1 with Group B (highlighted in bold in Table 2) as an illustrative example. We chose this example as it was coded with three of the most prominent codes (researcher practices, affect, interaction with others) and because it is a compact illustration of the processes of identity co-construction that are evident throughout the data. Below we analyse the illustrative example to show how identities are constructed and verified during tutorial talk.

Ethical considerations

From the initial planning stage, we were attentive to ethical issues and aware of potential risks to participants as we researched the early stages of their doctoral journeys. These included the possibility of those not wishing to participate being disadvantaged and the risk of extracts being attributable to individuals. We addressed these issues in our successful application for institutional ethical approval by recording only those tutorials where all participants had given consent with anyone not wishing to participate in the research accessing parallel tutorials. We discussed with participants the possibility of them being identified through their comments, taking care to promote opportunities for them to redact parts they might have concerns about. Those consenting were given a further opportunity to ask questions and either affirm their consent or withdraw prior to each tutorial, with a further week after the tutorials and interviews to withdraw or redact any comments. We anticipated that the project would benefit the cohort of doctoral researchers, participating or not, in part through sharing our research processes and ethical considerations with them. For participants, the opportunity to reflect on their engagement with our research as they began to design their own pilot studies was additionally beneficial. While the tutorial participants did not make explicit any orientations to the fact that their talk was being recorded, we nonetheless recognise that recording may have affected or constrained their interaction.

Author positionality

A significant consideration throughout was the complex power relations resulting from our multiple roles and responsibilities as researchers, tutors on the EdD, doctoral supervisors and colleagues of those doctoral researchers who were also university staff. We discussed the challenge that these roles posed, considering how we might both inform and consult participants about the study, seek their consent and stay alert to possible issues. Building on existing policy and practice that minimised power relations, for example by ensuring that colleagues in the same team or who had close personal or line management relationships are not in supervisory relationships, we allocated tutors and participants to tutorial groups adding the stipulation that doctoral researchers would not be allocated to a group led by their supervisor. As tutors and researchers we maintained transparency about access to and use of the data; this was complemented by reciprocity and tutors disclosing aspects of their own doctoral experiences, a common feature of our practice.

Analysis

In this section, we show how researcher identities are achieved during tutorial talk by analysing an illustrative sequence from the second of two tutorials on the first module of the EdD programme (the entire sequence is presented in Appendix). This sequence comes around half way through the tutorial. In this sequence, the tutorial participants talk about refining their research questions.

Our tutorial sequence begins with a question from the tutor:

Extract 1

Table

1 2 3 4

Tutor

Thinking about where your research focus was and perhaps where your research questions were at the beginning of the module – do you feel like [your reading] is starting to help you think about refining those questions or refining your focus in any way?

In response, Fay, one of the doctoral students, talks about various texts she has been guided to read on the module:

Extract 2

Table

1 2 3 4 5 6 7 8 9 10 11 12 13

Fay

I found [Module 1] so helpful because of the ways in that you've given us to, you know, for example so many different ways to look at policy and, in a way, you could argue oh well that makes the reading even broader and perhaps I might like to take that perspective on my research question … for example you know, the discourse analysis and then, yeah, the text piece we did, we read by Emery on the wellbeing and that sort of dissonance … I'm really grateful for having these readings signposted and I think there is something for everyone there and not everyone will gravitate to the same piece. And of course you read that and that takes you down other roads. So the depth I would say, looking back to January from now, is – I sound like I'm really showing off – but I feel like it’s quite phenomenal really. And that keeps me going if I'm feeling oh when am I going to do it – I think wow my head is like buzzing with ideas and I love that. It feels just really creative

Fay describes engaging in practices typically undertaken by researchers in her field: looking at policy (2), ‘reading even broader’ (3), taking different perspectives on her research question (3–4), going ‘down other roads’ (9) and finding ‘phenomenal’ (11) depth. She also describes engaging with community members – the tutors who have given her reading suggestions and the text authors who have helped her see different perspectives (1–7). This engagement has resulted in ‘phenomenal’ (11) learning – her head is ‘buzzing with ideas’ (12–13) and she feels ‘really creative’ (13). As Fay engages in common research activities, she claims membership of the research community.

The next turn in the tutorial conversation is from another student, Paul, who responds to Fay by telling the group about how his supervisor has guided him to accept uncertainty as a natural part of doing research:

Extract 3

Table

1 2 3 4 5 6 7 8 9 10

Paul

For me it goes back to where we started in terms of supervision because I’ve not been entirely sure about my research area and my supervisor’s been really reassuring and saying well it probably won't come until the end of this module or the next and so I'm thinking well I’ll stick with this for the time being but it still doesn't feel an entirely comfortable fit and so – but I'm not worried about it because my supervisor’s been there and done it all, seen it all. So I'm thinking well I'm going to stick with this and already I'm reading around things which I wouldn't have even thought about before, so it feels like the learning is happening and that’s fantastic. I’ll see where that takes me. I’m not forcing myself into changing or staying at the moment.

Like Fay, Paul narrates processes familiar to many researchers – being uncertain: ‘I’ve not been entirely sure about my research area’ (1–2); experiencing discomfort: ‘it still doesn't feel an entirely comfortable fit’ (4–5); and accepting uncertainty: ‘I’m not forcing myself into changing or staying at the moment’ (9–10). Like Fay, he also describes the research activity of reading with a clear sense of learning: ‘I'm reading around things which I wouldn't have even thought about before’ (7–8); ‘the learning is happening’ (8–9). Like Fay, Paul also describes engaging with a more expert community member – in his case his supervisor. Through sharing her experiences: ‘but I'm not worried about it because my supervisor’s been there and done it all, seen it all’ (5–6), Paul and his supervisor have built a trusting relationship, enabling Paul to embrace processes of uncertainty and wait before making important decisions (5–6).

The next turn comes from the tutor who provides reassurance by emphasising the unique and individual aspect of doctoral study and recognising the transitory nature of knowledge as she comments ‘it’s always going to be changing and evolving and re-drafting and I think that’s part of the journey’ (see Appendix for a full transcript of the tutor’s turn). By recognising Paul’s actions and choices as valid, the tutor verifies Paul’s researcher identity.

The next turn comes from Lily, another doctoral student:

Extract 4

Table

1 2 3 4 5 6 7 8

Lily

But I think in my very first supervision meeting I had a conversation and I said I'm so stressed out about all the different directions and, you know, I could do this and I could do my research question, methodology there’s too many things, there’s too much and [my supervisor] was like ‘that’s OK, it’s fine, don't worry’. And I think I am getting okay with the fact that, you know, you can change and things can evolve and think I'm okay with that now, but I remember in my first supervision meeting that that really was stressing me, the idea that it could just go off in different directions.

Like Paul, Lily is coming to terms with uncertainty, moving from ‘stressed out’ (2) to ‘okay with that now’ (6). Lily repeats Paul’s points and builds on them by introducing the idea that change is possible (6), not only performing her own researcher identity but also verifying Paul’s by confirming similar thoughts and experiences and backing this up with reference to her own supervisor. Thus, as the conversation progresses, students explore researcher practices and build a shared repertoire of norms, thereby legitimising each other’s opinions and practices in a process of constructing and verifying researcher identities.

This developing understanding is recognised by Sam (a student), the final contributor in the sequence. However, Sam feels he hasn’t made the same progress:

Extract 5

Table

1 2 3 4

Sam

It sounds like other people have had quite a significant journey between January and now. I don't feel I have, but I'm not sure whether I ought to be worried about that or not, or whether I actually ought to be pleased or not, or sad or not, I don't really know what it means

Sam compares himself unfavourably to the other doctoral students and talks about feelings of uncertainty. However, as his turn unfolds, we see that Sam is also engaged in the valid and expected researcher practice of refining, sharpening and focusing his research questions:

Extract 6

Table

1 2 3 4 5 6 7 8 9 10 11 12 13 14

Sam

But what I do know is that the journey I have experienced, or the only part of it I can kind of capture, get hold of, seems to me to fit very well with what I understand this module’s main aim to be, i.e. refining your research questions, because although I don't feel that I've got anything – I was going to say not very positive – that’s probably the wrong phrase – I don't feel I’ve gained something – I feel I've lost something in a useful way. I feel that I’m beginning to see ever more rapidly the more things I look at, which bits of the research are simply going to be impractical or impossible, so I'm crossing bits off my list, like one of the questions that was written into my original proposal was to look at A, B and C and if C is proven is it conscious or unconscious – well I've crossed that off because I've realised that it’s going to be virtually impossible to be able to ascertain whether it’s conscious or unconscious, so forget that question, no point in setting myself up for something I can't answer. So I don't feel like I've had a particularly big distance travelled

Sam expresses confidence that the ‘crossing off’ (8, 11) work he has done aligns with the module objectives (‘refining your research questions’ (3)) and has been useful: ‘I feel I've lost something in a useful way’ (6). Although he doesn’t feel like he’s ‘had a particularly big distance travelled’ (14), his gain has been significant which is then recognised by the tutor:

Extract 7

Table

1 2	Tutor	What you're actually explaining there is about you going through a journey of discovery in terms of what’s feasible and not feasible
3	Sam	Mm
4 5 6	Tutor	And I think as you read things and as you have discussions with your supervisor, you can't actually cross things off until you’ve come a certain way and you recognise that actually that won't work. So even that in itself is a journey.
7 8 9 10 11	Sam	Yeah I suppose it is. I suppose it’s got to be a journey, hasn’t it, but listening to other people talk, it sounds as though their journeys have been a lot more transformative than mine, whereas mine I do find has been helpful but it doesn't sound to me as though it’s been that exciting, it’s just helped me to knock some things off and say ‘well let’s not waste time on those’!

The tutor provides confirmation and verifies Sam’s identity as a researcher by recognising the valid processes of narrowing and refining his research focus. Sam accepts this reluctantly: he prefaces his acceptance with ‘I suppose’ twice (7), adds a ‘but’ clause (7) and again compares his journey unfavourably with others’ (less transformative [8–9] and exciting [10]). The tutor seems to recognise Sam’s reluctance because she then repeats her message:

Extract 8

Table

1 2 3 4 5 6 7 8

Tutor

I think that’s really important though, [Sam], because, you know, I remember thinking of them as blind alleys you know, so going through that process, but I had to investigate them and have a look what was out there in relation to that, just so I could kind of discount it or at least know where what I was doing sat in relation to that. So it was actually like that is – you think oh gosh I've just spent two days looking at that and it’s not relevant! But actually it’s really important to know that that’s not relevant and know why I think as you shape things up. So it’s not really lost, it just takes time, it’s like sifting things.

She strengthens her point, using an amplifying adverb, ‘really important’ (1, 6–7), and adding emphasis by using Sam’s name (1). She also describes herself going through the same processes, again validating Sam’s researcher identity.

Relevant to institutional talk and identity formation is power (Donaghue 2020b). Power is exercised within relationships and social practice (Foucault 1977; 1980) and can be instantiated, activated and resisted in talk (Thornborrow 2014). At the beginning of the illustrative extract, the tutor indexes power by claiming the right to open the discussion, choose the topic, and ask questions (Extract 1). In institutional interaction, there are accepted roles and expectations which determine what is said and by whom (Drew and Heritage 1992) and participants rarely have equal rights i.e. some utterances are seen as legitimate for some speakers but not for others. As a higher status speaker, the tutor has a legitimate right to ask questions, while the lower status speakers (the students) collude by assuming the less powerful role of answerer. However, while the tutor indexes authority and power at the beginning of the sequence, her question elicits students’ opinions and gives them space to talk. She is not interactionally dominant – the doctoral students have equal access to the floor and have extended turns between themselves. Although the tutor again indexes a powerful identity by claiming the right to evaluate Sam (Extract 8), this is tempered by stories of her own vulnerability (Extract 8). She also uses her position of power to recast Sam’s negative identity claim (that he has progressed less than the others) into a more positive one of having gone through important processes: ‘So it’s not really lost, it just takes time, it’s like sifting things’ (Extract 8, line 8), similar to her own. Thus, while power asymmetry operates in the dialogic relationship established between the doctoral students and tutor, this does not impede students’ identity development.

Analysis of this short tutorial extract has revealed how tutorial participants co-construct doctoral researcher identities through interaction, highlighting the importance of study-based talk in building and verifying doctoral researcher identities. This analysis also shows the methodological warrant of micro analysis of naturally occurring interaction to investigate identity.

Discussion

In the analysis above, doctoral students reflect on engaging in research practices (e.g. reading, refining their research focus/questions), interacting with community members (e.g. supervisors, authors), and the emotions these activities engender (e.g. confidence, discomfort). Through exploring these topics and sharing experiences, the tutorial participants make meaning and develop understandings of what it means to be a doctoral researcher. Their talk is dialogic (Alexander 2020): collective, supportive, reciprocal and cumulative. As participants build on each other’s contributions and chain them into coherent lines of thinking and understanding, they create a shared repertoire of norms and methods. This sense making is linked to identity. Some researchers use these terms interchangeably (e.g. Conroy and O’Leary-Kelly 2014), some see identity as part of the sense making process (e.g. Ojha 2005), and some see sense making as part of identity work (e.g. Tracy, Myers, and Scott 2006). Most agree that sensemaking and identity are intertwined, with Vough et al. (2020) suggesting they have a recursive effect on one another. In our illustrative extract, the doctoral students discuss and make sense of activities, challenges and relationships, all the while constructing researcher identities in a recursive process:

the self is the fundamental target of sensemaking: an equivocal problem being defined and redefined through interaction with others. Identity is produced through talk and action, and its legitimacy is negotiated as individuals work out, in different social contexts, who they [are]. (Vough et al. 2020)

Through their talk, students constitute and strengthen their membership of the research community and develop their own researcher identities.

These findings resonate with the literature on doctoral researcher identity. This body of work has identified reflection (Ai 2017; Foot et al. 2014), supportive relationships (Dollarhide, Gibson, and Moss 2013; Inouye and McAlpine 2017), engagement with research activities and processes (Jazvac-Martek 2009; Mantai 2019; Paré, McAlpine, and Starke-Meyerring 2006), and discursive engagement (Ai 2017; Klenowski et al. 2011; Mantai 2017) as important to identity development. Our study adds to this literature by examining these processes in action. We show tutorial participants doing identity, co-constructing and accomplishing identities during situated talk. We show identity to be social (Morgan and Clarke 2011), active (Sarangi and Roberts 1999), situationally emergent (Bucholtz and Hall 2005), and constructed during interaction (Varghese et al. 2005). These important aspects of identity have tended to be overlooked in doctoral identity research. The literature also identifies legitimacy as an important aspect of doctoral researcher identity development (Choi, Bouwma-Gearhart, and Ermis 2021). We extend this idea by highlighting the role of an interactant in situated talk in sustaining and verifying a speaker’s claimed identity. In our tutorial sequence, peers and tutor endorse students’ thoughts and actions by narrating similar (often difficult) experiences themselves. Through recognising these activities as legitimate (often necessary) researcher actions, they legitimise the speaker’s identity claim.

Conclusion and recommendations

This article reveals that tutorial talk enables the development of researcher identities. The analysis of an illustrative sequence shows identity construction as it happens, turn by turn. Identities are performed and displayed as students narrate practices common to the community, they are co-constructed and verified by peers who tell similar stories, accepting and building on previous students’ talk, and they are verified by the tutor who explicitly recognises the importance and validity of students’ actions and describes experiencing similar processes herself. While other studies have highlighted reflection, supportive relationships, discursive engagement and legitimisation as important resources for helping doctoral students develop researcher identities (Choi, Bouwma-Gearhart, and Ermis 2021), no other research has shown these processes in action.

The results of this study have important implications for practice. To facilitate identity development, we recommend doctoral programmes set up opportunities for small group discussions (e.g. tutorials) to give doctoral students the space to talk and share experiences. Extended data extracts (for example, Appendix) may provide a focus for sense making, adding dialogue from study-based talk to existing student experience stories (Adams 2018). We also recommend tutors have opportunities to examine their practice and raise their awareness of identity development. To help tutors become more critically aware of their professional talk, we suggest a guided analysis of discussion extracts (such as the one featured in this article), looking at how identities are manifested and their role in this.

This study also has implications for research. More research is needed into how doctoral research identities are developed in discussion, looking at, for example, how different pedagogic and linguistic resources (e.g. questioning, task design, giving students space to talk, tutors sharing their own doctoral struggles and experiences) might enhance or limit discussion and dialogue and prompt/limit identity development. This indicates the importance of conducting close analysis of talk during other doctoral interactional events such as supervisory meetings and peer support groups to see if/how identities are developed. This paper focuses on data from one tutorial early in the EdD aiming to show how tutorial talk facilitates learning and belonging. While it is beyond the scope of the paper to explore how doctoral researchers navigate belonging to multiple communities and how identification with these communities shift throughout the doctorate, these would be interesting avenues to explore in future research. For example, one potential research project would be to look at tutorial talk from later in the doctoral journey when students are carrying out their research and interpreting resultant data. Our focus on students at an early stage of the doctorate may be one reason we found few more difficult conversations where participants report struggles or feelings of stickiness. Analysing ‘troubles talk’ (Mewburn 2011) and the identity work done when students talk about more difficult aspects of doctoral study would enhance knowledge of identity negotiation during naturally occurring talk.

We end with a broader methodological plea. While continuing education research is increasingly interested in identity, few studies employ in-depth empirical analysis of the ‘live’ processes of identity construction and negotiation. For example, to our knowledge, no other research has looked at how doctoral researcher identities are constituted through study-based talk. While we realise that it is not always possible or appropriate to make recordings, and that awareness of recording can affect participants’ talk, we nonetheless argue that a focus on identity achievement in naturally occurring, work-based interaction enables researchers to identify practices that engender identity development.

Acknowledgements

The authors wish to express thanks to the doctoral researchers and colleagues in the EdD team who supported this research. For the purpose of open access, Gill Adams has applied a Creative Commons Attribution (CC BY) licence to any Author Accepted Manuscript version arising from this submission.

Disclosure statement

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

Correction Statement

This article has been corrected with minor changes. These changes do not impact the academic content of the article.

Additional information

Funding

We are grateful for funding from Sheffield Hallam University in support of this study.

Appendix. Tutorial sequence

Table

1 2 3 4	Tutor	Thinking about where your research focus was and perhaps where your research questions were at the beginning of the module – do you feel like your reading is starting to help you think about refining those questions or refining your focus in any way?
5 6 7 8 9 10 11 12 13 14 15 16 17	Fay	I found [Module 1] so helpful because of the ways in that you've given us to, you know, for example so many different ways to look at policy and, in a way, you could argue oh well that makes the reading even broader and perhaps I might like to take that perspective on my research question … and I just, for example you know, the discourse analysis and then, yeah, the text piece we did, we read by Emery on the wellbeing and that sort of dissonance … I'm really grateful for having these readings signposted and I think there is something for everyone there and not everyone will gravitate to the same piece. And of course you read that and that takes you down other roads. So the depth I would say, looking back to January from now, is – I sound like I'm really showing off – but I feel like it’s quite phenomenal really. And that keeps me going if I'm feeling oh when am I going to do it – I think wow my head is like buzzing with ideas and I love that. It feels jus really creative
18 19 20 21 2223 24 25 26 27	Paul	For me it goes back to where we started in terms of supervision because I’ve not been entirely sure about my research area and my supervisor’s been really reassuring and saying well it probably won't come until the end of this module or the next and so I'm thinking well I’ll stick with this for the time being but it still doesn't feel an entirely comfortable fit and so – but I'm not worried about it because my supervisor’s been there and done it all, seen it all. So I'm thinking well I'm going to stick with this and already I'm reading around things which I wouldn't have even thought about before, so it feels like the learning is happening and that’s fantastic. I’ll see where that takes me. I’m not forcing myself into changing or staying at the moment.
28 29 30 31 32 33 34 35 36 37 38 39	Tutor	I do think you have to go with your own journey and I think if you're one of those people that needs that time to sort of sift through things and think about what you really want and, you know, because it’s no good making a decision too early and then not being happy with it or wanting to change it later on. So I think you have to sort of go with the sort of person you are but also it’s your own personal journey, isn’t it, it’s your EdD. I suppose it’s a little bit like your writing, isn't it, I think when you hand in a draft to a supervisor of something, it’s never going to be perfect, is it, it’s always going to be changing and evolving and re-drafting and I think that’s part of the journey, isn't it, really, accepting that there’s going to be that change and that movement, you know. I think all you can say really is this is where I am in my thinking right now, isn't it, and that’s OK! [Laughs]
40 41 42 43 44 45 46 47	Lily	But I think in my very first supervision meeting I had a conversation and I said I'm so stressed out about all the different directions and, you know, I could do this and I could do my research question, methodology there’s too many things, there’s too much and [my supervisor] was like ‘that’s OK, it’s fine, don't worry’. And I think I am getting okay with the fact that, you know, you can change and things can evolve and think I'm okay with that now, but I remember in my first supervision meeting that that really was stressing me, the idea that it could just go off in different directions.
48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66	Sam	It sounds like other people have had quite a significant journey between January and now. I don't feel I have, but I'm not sure whether I ought to be worried about that or not, or whether I actually ought to be pleased or not, or sad or not, I don't really know what it means but what I do know is that the journey I have experienced, or the only part of it I can kind of capture, get hold of, seems to me to fit very well with what I understand this module’s main aim to be, i.e. refining your research questions, because although I don't feel that I've got anything – I was going to say not very positive – that’s probably the wrong phrase – I don't feel I’ve gained something – I feel I've lost something in a useful way. I feel that I’m beginning to see ever more rapidly the more things I look at, which bits of the research are simply going to be impractical or impossible, so I'm crossing bits off my list, like one of the questions that was written into my original proposal was to look at A, B and C and if C is proven is it conscious or unconscious – well I've crossed that off because I've realised that it’s going to be virtually impossible to be able to ascertain whether it’s conscious or unconscious, so forget that question, no point in setting myself up for something I can't answer. So I don't feel like I've had a particularly big distance travelled
67 68	Tutor	What you're actually explaining there is about you going through a journey of discovery in terms of what’s feasible and no feasible
69	Sam	Mm
70 71 72 73	Tutor	And I think as you read things and as you have discussions with your supervisor, you can't actually cross things off until you’ve come a certain way and you recognise that actually that won't work. So even that in itself is a journey.
74 75 76 77 78	Sam	Yeah I suppose it is. I suppose it’s got to be a journey, hasn’t it, but listening to other people talk, it sounds as though their journeys have been a lot more transformative than mine, whereas mine I do find has been helpful but it doesn't sound to me as though it’s been that exciting, it’s just helped me to knock some things off and say ‘well let’s not waste time on those’!
79 80 81 82 83 84 85 86	Tutor	I think that’s really important though, Sam, because, you know, I remember thinking of them as blind alleys you know, so going through that process, but I had to investigate them and have a look what was out there in relation to that, just so I could kind of discount it or at least know where what I was doing sat in relation to that. So it was actually like that is – you think oh gosh I've just spent two days looking at that and it’s not relevant! But actually it’s really important to know that that’s not relevant and know why I think as you shape things up. So it’s not really lost, it just takes time, it’s like sifting things.