Adapting voice-centred relational method to understand students’ experiences of synchronous online tuition

ABSTRACT This article presents the methodological approach taken to investigate students’ experiences of synchronous online tuition in health and social care at a large, distance learning university in the UK. Seeing the students as the ‘privileged knowers’ on the topic of their tutorial experiences, this study took an experience-centred narrative approach and used voice-centred relational method to analyse the data. This paper will explain and then discuss the two ways in which voice-centred relational method was adapted for the purposes of this research: firstly, the particular approach taken to the construction of the I poems in the second stage of the analysis and secondly, the addition of an extra listening, adding a deductive lens to seek out evidence of the different types of presence from the Community of Inquiry framework within the students’ narratives. These adaptations provided additional insights into students’ experiences and made aspects of these experiences more visible to educators.


Introduction
For almost 20 years, synchronous online tuition has played a key role within distance education and a developing one within other universities, usually as part of a blended approach. In 2020, with the arrival of the coronavirus pandemic, which made face-to-face teaching unfeasible for extended periods, how to teach effectively online became a pressing topic for educators. Online tuition is likely to continue to play a significant role, so it is essential to understand how students perceive their experiences of this medium. Students' experiences, including what they are thinking and feeling during tuition, affect their behaviour and their learning. It is important, therefore, to see students' lived experiences from their perspective. The significance of student voices has sometimes been overlooked in research in this area (Levy 2015) and this paper focusses on a method of analysis that can help researchers to hear students' voices more clearly. It explains how an experience-centred narrative approach and voice-centred relational method (VCRM) were used to listen to the online tutorial experiences of 10 female undergraduate students studying a health and social care module about children and young people's wellbeing at a large UK-based university that specializes in distance learning. The paper shares examples from five of the student narratives that have been chosen to illustrate the benefits and challenges of adapting VCRM. The findings of the study have been shared previously (Chandler 2022).

Voice-centred relational method
The diaries and interview transcripts were analysed using VCRM, an approach to narrative analysis that reflects a relational ontology (Brown andGilligan 1991, Mauthner andDoucet 1998), and recognizes that relationships are often central to experiences. It provides a specific framework with which to undertake narrative analysis in what can often feel like a broad field of possibilities (Bekaert 2014).
Researchers using VCRM carry out multiple listenings to the data. The term 'listenings', rather than readings, is used because the sharing of the narrative involves participation of both the teller or research participant and the listener or researcher (Gilligan et al. 2003). Typically, the first listening is to hear the broad story of the narrative and my first listening to the students narratives followed this pattern. As I listened to the data from the students' diaries and interviews, I focused on the context and the drama within them. I paid attention to repetitions, metaphors and emotional resonances, revisions and absences and changes of narrative position. I also considered how each narrative had been co-constructed and my own response to what each student had shared.
I then used one of the connecting strategies described by Maxwell and Miller (2008): the preparation of vignettes to reduce the data and identify key relationships. I prepared a vignette of each student's narrative to share with tutor colleagues via established forum communities and my research website. I reflected on their responses to the narratives, comparing them with my own. As I did so, I realized that the tutors' reflections were data in themselves. When analysing the student narratives, I copied relevant tutor data into each student narrative document, including tutors' responses to their vignettes, any subsequent discussion, and my own responses to both of these. This data prompted me to return to the transcripts to check details and consider alternative interpretations. It also enabled me to analyse tutors' responses to the narratives. Most selected just one aspect of a student's story that resonated with them and then used this or the subsequent discussion as a springboard to share stories about their own experiences in online tutorials, as tutors, students or even in the context of other roles.
The second listening within VCRM is to hear the 'self' speaking through the narrator. This involves constructing what are termed 'I poems', where each 'I' statement within the narrative is extracted along with the associated verb and placed on a separate line. The methodology of creating I poems has evolved in multiple directions (McKenzie 2021) and I developed my own approach, which will be discussed below.
The third listening is to hear contrapuntal voices or how different voices within the narrative relate to each other (Mauthner andDoucet 1998, Woodcock 2016). The term 'contrapuntal' arises from the musical term counterpoint, where there are two or more melodic lines in relationship. I wanted to gain insight into the extent to which students perceive themselves to be part of a learning community and how they perceive their relationships with peers, tutors, and others, so I considered the personal, the communal and the institutional contexts, as represented by 'I', 'we', 'you' and 'they' statements. I adopted the technique used by Jankowska (2014), separating out personal views (I statements) and placing them on the left of the page, collective opinions (you statements) in the centre and references to tutors and peers (they statements) on the right for comparison. I made two small modifications to this technique by also identifying 'we' statements, which gave an indication of where students had felt part of a group of learners and broadening out the inclusion of those participants referred to as 'they' to include not only the tutors and peers participating in the tutorial but also family members, friends, or colleagues.
Within VCRM, the number of listenings can be adapted to suit the purposes of the research and at this point, I introduced an additional listening, which will be discussed below.
To conclude the analysis, the researcher collates all that has been learnt through the different listenings, bringing the different voices back into relationship with each other and reflecting on the analysis process. None of the listenings are intended to stand-alone; only together can they represent someone's experience. Throughout my analysis, I upheld the commitment many narrative researchers attach to analysing each individual narrative in a case-centred way, rather than hurriedly looking for themes across cases (Riessman 2008). Only after completing the final stage of VCRM did I apply a categorizing analysis strategy that looks for similarity-based relationships (Maxwell and Miller 2008), constructing a conceptually clustered matrix. These order the display according to variables that reflect the research questions, so that the findings across all cases can be presented and meaning can be generated more easily (Miles and Huberman 1994).
The purpose of this paper is to explore two ways in which VCRM was adapted: the particular approach taken to the construction of I poem in the second stage and the addition of an extra listening, adding a deductive lens to seek out evidence of the different types of presence from the Community of Inquiry framework within the students' narratives. Adapting VCRM is not, in itself, unusual. Whilst still learning to use VCRM under Gilligan's guidance, Mauthner and Doucet were already both adapting the method in different ways to suit their own sociological research interests (Mauthner and Doucet 1998). The techniques used to analyse data are not neutral, so it is important that researchers demonstrate reflexivity in relation to our processes and document our understandings of the ways in which they influence our interpretation of the data.

I poems
The first adaptation of VCRM was the particular approach taken to the construction of what are termed 'I poems' in the second stage of the analysis. Within VCRM, listening for the self or the voice of the 'I' is a further example of a connecting analysis strategy (Maxwell and Miller 2008). It involves listening for the first-person voice speaking in the story, including its modulation and rhythms, and how the narrator speaks about themselves (Gilligan et al. 2003). This offers the possibility of working with voice as an emergent process, rather than a static entity (Chadwick 2021). I constructed I poems by removing every 'I' statement from the transcript of each narrative and placing each on a separate line, together with the associated verb and other significant words.
Applying this strategy is claimed to amplify the narrator's voice and has the potential to enhance the researcher's understanding the story on the narrator's terms (Brown and Gilligan 1991) but the outputs of this process when applied to my data were long, rather unwieldy lists of 'I' statements, which gave limited additional insight into the narrator's lived experience of tuition. Examining these lists, however, it became apparent that the majority of statements within the narratives could be divided into three broad types: (1) those revealing an aspect of identity, such as 'I do childminding', 'I'm a single mum' or 'I've got dyslexia' (2) those sharing feelings and preferences, such as 'I felt totally absorbed', 'I worried about saying the wrong thing' or 'I love to find the opinions of the other students' (3) those relating to actions taken before, during or after a tutorial, such as 'I signed up', 'I find someone there who also feels the same', or 'I'll go back and look at the recordings'.
I developed new I poems by separating the 'I' statements into these three types, omitting other I statements and keeping them in the order in which they appeared in the narrative. Punctuation from the original transcript was retained. Sometimes, it was hard to know how to classify a particular statement, so it was included in more than one poem or omitted. Whilst it felt important not to make changes to the vocabulary used, even when a word seemed grammatically incorrect or poorly chosen, relational ethics were prioritized over method and some details that might have identified participants were redacted.

Identity I poems
The identity poems provide insight into students' perceptions of themselves as adult learners. In the following I poem, Lisa identifies as 'a silent student'. Her narrative, when viewed through the lens of her poem, is one of struggle, not least because of her dyslexia. This aspect of Lisa's narrative was previously dispersed throughout the data that she shared and was only apparent after the identity I poem was created.
Lisaidentity I poem I'm not very good with my spelling I'm very much a face-to-face person I need to have a conversation with them I'm not very confident I always end up getting it wrong I need like facts I need just someone to … I am struggling I struggle with references I can get into my head I sort of panic I don't speak to anyone I don't talk to other students. I just need someone to … I don't go on the forums I get marked on my forum contribution I'm a silent student I struggle I've got dyslexia I'm a youth care I run a youth club I'm due to move I'm moving I really do sort of engage in a face-to-face situation I normally use YouTube to help me understand

Feelings I poems
The feelings I poems brought together how students felt about their tutorial experiences. Some poems showed the high value students place on tutorials, sharing their joy in learning and the sense of motivation gained. They include suggestions as to how tutorials might better reflect their preferences and meet their needs, highlighting particular aspects of students' lived experiences, such as the longing for face-to-face connection. Karen's experience takes place in the context of living in a remote rural location, too far away to attend face-to-face sessions usually, but also in the context of the first coronavirus lockdown in the UK, which was happening at the time of her interview. She has not had an opportunity to attend a tutorial with her own tutor and tutor group but would very much like the opportunity to do so. Webcams were not used in the online tutorials she describes and she feels a lack of face-to-face connection. This poem illustrates how I poem sections can highlight particular issues, the whole poem not being necessary to the reader's understanding.

Karenfeelings I poem section
I guess it would be nice to see faces I really would have liked to go to some of the face-to-face ones I really just have to go to the online tutorials. I guess to be able to see people might be quite nice. I guess there's just a lot of virtual life going on just now. I suppose the smaller the number, the more interaction I'm guessing there are large numbers. I guess if it was your own tutor group, I would have really liked that I think had I been face-to-face and actually met my tutor I do think they're really good. I do feel that is still a support. I'm really just quite happy working my way through the materials. I guess that's the beauty of [name of university] I guess a lot of it gives me a deeper understanding. I guess it just gives me a much deeper understanding

Action I poems
The action I poems gave me insight into what students were doing before, during or after a tutorial. In Amie's action I poem below, for example, she joins on time and attends the session, she listens and translates what she hears into her first language, she interacts (using the written chat, not the microphone), wonders how to apply what she has learned and plans to watch the session recording. These poems also highlighted students' interactions with others who were present but invisible to the tutor and other students. In Amie's household, her four children were told to watch television, so that she could take part in the tutorial session.
Amieaction I poem I joined in on time I could write on group chat I should attempt my assignment I was trying hard to capture I haven't attended too many tutorials. I mean, last year was my first year I attended one tutorial I did not bother attending any more tutorials until, until this year I attended I attended I did my assignment I went to that tutorial two or three times. I just wrote I did not use the microphone I just did not use the microphone I am talking I am listening to English, I am constantly translating it in my mind I have to say them in [my first language]. I do understand most of it, 90% of it I do have to tell them to just sit down and watch TV I can interact with my tutors I have got some good feedback I was just wondering how to follow it I definitely will, yeah … I will definitely watch it These different poems made it possible to analyse different aspects of the students' experiences in more detail.

Adapting the I poems: discussion
Within the narrative research process, there are five levels of representation or points at which meaning can shift: attending, telling, transcribing, analysing and reading (Riessman 1993). Therefore, it can be argued that participants' narratives, researchers' interpretations, and readers' understandings are all unique to them as individuals and to particular times and contexts. This means that a narrative study cannot be replicated, even with the same participants, researcher, and readers. Whilst positivist research is exercised by what is termed 'reliability' or the extent to which results can be consistently achieved, this term is not relevant within qualitative research. Qualitative researchers instead work towards dependability, making robust records of the decisions taken around data collection and analysis processes and ensuring that our methods of analysis are consistent with the accepted standards for the approach used (Korstjens and Moser 2018). The responsibility to do this takes on a particular significance when our methods of analysis take us in new directions.
The creation of I poems during analysis is one level of representation where there is the potential for meaning to shift. Whilst some researchers, including Gilligan, with whom the method originates, favour an approach that strips out all context and grammar, I decided against removing them entirely, as it can diminish the narrator's voice (McKenzie 2021). Even so, occasionally, it was important to be aware of some contextual information being lost during this part of the analysis. For example, when one student, Joanne, made the statement, 'I think that helps greatly' in relation to having a break during a tutorial, this was in the context of there not having been a break in the tutorial that she was describing. When creating the I poems, I noted such issues carefully and referred back to them in the final stage of VCRM when gathering the different voices back into relationship with each other and reflecting on the analysis process.
Within my study, the collation of the I statements in the narratives into different categories described above might be seen as an unnecessary fracturing of the data and as a departure from the narrative research tradition of keeping the thread of the story unbroken during analysis (Riessman 2008). Even so, this separation and the creation of these new I poems gave me further understanding of the narratives from the participants' perspectives. It was a new path but one worth exploring.
As well as being concerned with dependability, narrative researchers must also aim to achieve transferability, having a responsibility to share participants' narratives in a way that makes the relevance of the stories evident (Hollway and Jefferson 2000) and enables readers to decide whether the study's findings are transferable to their own contexts. The narratives may be useful to others, but it is the readers' responsibility to make a judgment about this.
The creation of poems within the process of VCRM generates powerful research outputs that enter the world of art-based literature (Gilligan 2015) but that does not necessarily mean that they are examples of 'good' poetry, nor should they be. Faulkner (2007), holds that researchers using poetic forms in their research must study the craft of poetry, including poetic traditions and techniques, and argues research poems are only 'good' when aesthetic concerns are prioritized. Piirto (2002) is similarly concerned with the 'quality' of poetry used within qualitative research, whilst for others, the literary merits are insignificant compared to the power that poems hold. In our use of poetry, narrative researchers are not usually aiming to be poets but are looking for a way to represent our data (Byrne 2017). For Leggo (2012), the key question is not, 'Is it a good poem?' but rather, 'What is the poem good for?' This suggests that the merits of poems are in the impact that they make. Research poetry can have impact beyond conveying the voices of the research participants. Richardson (1993), for example, uses the poem 'Louisa May's story of her life' and her observations of how it was received by academic colleagues to challenge epistemological assumptions, question traditional definitions of validity and provoke a debate about the androcentric nature of research.
Others argue that there should be a pragmatic space for 'good enough poetry' (Lahman and Richard 2014), including what are termed 'transcription poems' that are typically formed from a single interview as in the case of those shared here. It could be argued that transcription poems are unlike literary poems and perhaps it is not helpful to call them 'poems'. Their value is not primarily in their artistic merit at all and there is no poetic craft involved in their creation, just a systematic process that is rigorously followed. The purpose of the I poems is not aesthetic, even if they sometimes have engaging aesthetic elements similar to other types of 'poetry'. Neither are they created for emotional effect. Instead, the so-called 'poems' are found within a pre-existing narrative and their value is in providing an additional and sometimes powerful lens for analysing data and through which the voices of research participants can be heard, thus expanding the horizons of qualitative research. They allow the researcher and those with whom the research findings are subsequently shared to think with, rather than about, a research participant and their lived experience (Miller et al. 2015).
I questioned the ethical implications of listening for individual voices within my data. Although VCRM is most of all concerned with ensuring that participants' voices are not subjugated by the researcher (Gilligan et al.2003), there seems to be some tension created when a researcher takes steps beyond really listening to what participants know they want the researcher to discover and begins analysing what the participants share for hidden subtexts, whether conscious or unconscious. According to Gilligan (2015, p. 73) 'a good method is also an ethical method.' Gilligan frames this ethical dilemma as a relational one, which can only be resolved when researchers are prepared to question social and cultural norms and communicate their deliberations to the extent that participants can communicate what they really think. Typically, researchers using VCRM are in an ongoing relationship with their research participants, the participants having time to get to know the researcher before sharing their narratives, so that they feel comfortable doing so (Bright 2016). Brown and Gilligan had opportunities to build close relationships with participants over some years in their work as psychologists (Brown and Gilligan 1991). For my research, however, my relationships with the participants were, by necessity, of a shorter duration. This meant that the most important ethical considerations around analysis were deontological, as well as relational, maintaining a commitment to analysing the data thoroughly and consistently, but also to foregrounding the findings that contribute to a better understanding of how to provide students with positive tutorial experiences, consistent with the aims of the study as shared with the participants.
The creation of poems generates research outputs that can be presented separately from discussion of the narratives in which they originated or even performed. Reflecting on the ethical implications of these possibilities, I concluded that it would not be appropriate to do either. One reason was that the participants had not been consulted about my use of VCRM; my approach to analysis had been decided at a later stage in the study. Secondly, each poem only presents some of the voices that can be identified within one listening and these need to be seen in the context of the whole narrative and its analysis. I decided that the poems generated from the narratives should only be used within the context of presenting the research, again consistent with the aims of the study as shared with the participants.
From the I poems, I gained insight into the factors that affected each participant's tutorial experiences, as well as their individual tuition needs. They helped me to see what proved effective for them and what could be improved. I was also able to compare the findings from this stage of the analysis with those identified in the other listenings carried out within the other stages of VCRM. One of these was listening for evidence of the different types of presence within the Community of Inquiry framework.

Listening for evidence of the different types of presence
The second way in which VCRM was adapted for this research was via the addition of an extra listening, adding a deductive lens to seek out evidence of the different types of presence from the Community of Inquiry framework within the students' narratives.
Originally a term used in a scientific context (Pardales and Girod 2006) and usually attributed to Lipman (2003), a 'community of inquiry' in the context of online learning has been defined as a group of individuals who collaboratively engage in purposeful critical discourse and reflection to construct personal meaning and confirm mutual understanding. (Garrison 2011, p. 2) The Community of Inquiry framework adopts Dewey's philosophy of education, seeing education as a social, interactive process, that involves practical inquiry or critical thinking. It articulates the relationship between the three elements required for a worthwhile educational experience: social presence, teaching presence and cognitive presence. It has often been applied to asynchronous computer-mediated higher education (Garrison et al. 1999). This is surprising given that one of Lipman's features of a community of inquiry is face-to-face relationship, a characteristic often absent in asynchronous online contexts.
Within a community of inquiry, students contribute and further develop their cognitive presence or, in other words, 'construct meaning through sustained communication' (Garrison et al. 1999, p. 89). This sustained communication requires social presence, the definition of which has been much debated and revised. Garrison's most recent definition conceptualizes social presence as the ability of participants to identify with the community, communicate purposefully in a trusting environment and develop inter-personal relationships by way of projecting their individual personalities. (Garrison 2009, p. 352) Also necessary to the community of inquiry is teaching presence, described by the framework's creators as the 'binding element' (Garrison et al. 1999, p. 96), as without input to set the climate and select the content for the interaction, social presence and cognitive presence cannot be generated and there will consequently be no community of inquiry. Students themselves, as well as teachers, can contribute to teaching presence (Anderson et al. 2001) and the Community of Inquiry Survey Instrument recognizes three types of teaching presence: design and organization, facilitation, and direct instruction (Athabasca University 2018).
Additional types of presence proposed to exist within a Community of Inquiry include emotional presence, defined as the outward expression of emotion, affect, and feeling by individuals and among individuals in a Community of Inquiry, as they relate to and interact with the learning technology, course content, students, and the instructor. (Cleveland-Innes and Campbell 2012, p. 283) Garrison disagrees with this addition, arguing that there is insufficient evidence to warrant the inclusion of emotional presence separate from social presence (Garrison 2017).
A further proposed type of presence proposed to exist within a Community of Inquiry is learning (or regulatory) presence, which reflects the proactive stance adopted by students who marshal thoughts, emotions, motivations, behaviours and strategies in the service of successful online learning' and 'indicates the exercise of agency and control rather than compliance and passivity. (Shea et al. 2012, p. 90, Kilis and Yildirim 2018, Blaine 2019 Garrison also disagrees with the suggestion of this addition, arguing that learning presence is incongruent with the collaborative constructivist principles of the framework and that separation between teaching and learning is artificial (Garrison 2017).
The Community of Inquiry framework gives a clear account of the mechanisms involved in learning (Levine 2010) and has been used as a theoretical framework for multiple studies of students experiences of online learning (Kear 2010, Gallagher 2015, Murphy 2015, Fayram 2016, Lowe et al. 2016, Ozaydın Ozkara and Cakir 2018. A systematic review of the ways in which technology use in education is evaluated found that four of the six studies that looked at types of presence in online environments used the Community of Inquiry framework (Lai and Bower 2019). The Community of Inquiry framework has also been adapted for use as a learning design model (Nolan-Grant 2019) and its use in this context suggests that it continues to be seen as a robust framework. Whilst the framework has limitations, including the extent to which it acknowledges the reciprocal nature of learning (Edwards et al. 2011), the way in which it conceptualizes learning processes, including critical thinking and the role of a sense of identity within a group of learners, give it significant advantages, as well as its recognition of interconnectedness between the different elements involved in learning.
This framework has often been used for gathering quantitative data using the Community of Inquiry Survey Instrument (Athabasca University 2018), providing information about students' perceptions but not the reasons behind them or considering the impact of these perceptions on their studies. In contrast, my study uses the Community of Inquiry framework in the context of qualitative research. To do this, taking account of the descriptions within the survey instrument, I developed descriptions of what these presences might look like within online health and social care tutorials (see Figure 1). Whilst following Garrison's rationale for rejecting the inclusion of learning presence and emotional presence within the Community of Inquiry framework (Garrison 2017), I chose to consider the possibility that the proposed additional types of presence might be relevant for the purposes of my study and included them within my descriptions. I viewed cognitive, social and teaching presence as the essential elements needed for learning to take place and emotional and learning presence as the outcomes of the learning experience.
To conduct the additional listening as part of the VCRM process, I coded the presences that I identified within the diary sheet and interview transcript. In some narratives, particularly when students described tutorials in which they were asked to take a more active role, teaching presence (TP), social presence (SP) and cognitive presence (CP) frequently occurred together. An example is shown in the following extract, where Vicky describes how the tutors showed a video about a child whose family was evicted from their home and invited the students to discuss questions about it: But doing that and listening to how the tutors (TP) and how the students in the chat box were tackling it (SP), it did throw up some ideas that I hadn't thought of, which made me think about it in a slightly different way … not in a way to like get the answer but to add onto your own thoughts about it and maybe put a different perspective on things (CP) and I know that video threw up quite a few interesting conversations in the Facebook group, which again is obviously just another platform on which people are sharing ideas (SP), but I think the more ideas you get on the same subject helps you build that sort of opposing argument in your head and in your essay (CP). In our tutorial, it threw up a conversation (SP) about the fact that the child's mum was relying on the benefits and that's never going to be directly the child's fault or … It wasn't impacting on her fully directly but it impacts on her mother therefore, it will have an impact on her (CP) and just listening to all the different views (SP) on that sort of subject just makes you think about it in a slightly different way (CP) so I think that sort of activity (TP) is helpful. (Vicky, interview) Some issues arose when coding the data. Sometimes the different communities in which students participated overlapped, as in the section of Vicky's narrative above where she describes how a video was discussed both in the tutorial she attended and within a Facebook group. I then had to identify the instances of the different types of presence in the narrative that could be fully or partially attributed to involvement in the tutorial. Having the learning network tables that the students had completed to indicate which individuals and groups they usually discuss their studies with helped with this.
Another issue that arose during the analysis was the sense in which instances of emotional presence seemed one step removed from the tutorial itself. Whilst my description of what emotional presence might look like in a health and social care tutorial involved students responding emotionally to ideas or activities and feeling comfortable expressing the emotions within the tutorials (see Figure 1), the emotions that I identified within the narratives were often those expressed outside or after the tutorials, when completing the research diary and interview, rather than them having been expressed in the tutorials that the students were describing. For example, when one of the participants, Joanne, repeats the word 'happy', it is in describing how she feels about joining in and contributing, rather than having expressed this emotion within the tutorial itself: I'm always happy to … Again, it's through the chat box. But I'm always happy to join in and contribute in answer to the questions that are phrased to us. I do quite enjoy the tutorials [slight laughter] (Joanne, interview) As such, these instances did not meet Cleveland-Innes and Campbell's definition of emotional presence as 'outward expression' (Cleveland-Innes and Campbell 2012); the emotions might not have been expressed at all had the students not taken part in the research. Their expression within this context, however, could be understood as representative of an outcome of the tutorial in terms of how the students felt about their experience and so I recorded them.

Listening for evidence of the different types of presence: discussion
Whilst the number of listenings and the focus of each one can be adapted to suit the purposes of the research, the application of the Community of Inquiry framework within a qualitative study is unusual. It was, however, insightful. Finding evidence of the different types of presence confirmed that synchronous online tutorials in my context could be considered as communities of inquiry. Variation in the numbers of instances of the different types of presence between narratives, as well as variation in the extent to which the types of presence overlapped, provided an additional, deductive lens through which to view the narratives and compare with the findings of the other listenings.
For each of the 22 tutorials attended by the students, I created a diagram to assist in the analysis of the evidence of the different types of presence suggested by each narrative. An example is shown in Figure 2. These diagrams helped me to examine the relationships between the different types of presence suggested by each narrative, teaching presence, social presence and cognitive presence coming together to influence each student's experience as a whole. In Vicky's narrative of one of the tutorial experiences that she describes (see Figure 2), some teaching presence is apparent in that the tutors supported the students to focus on the session and designed two activities to enable dialogue and help students learn: a discussion activity based on a video and a referencing activity. Social presence, which I defined as the extent to which students felt comfortable interacting, were able to build relationships and felt that their point of view was recognized, played a role in Vicky's experience. Students shared how they were tackling their assignment via the chat box, but social presence was limited in that one tutor was talking during this written conversation, so it was hard for Vicky to focus. Cognitive presence or in other words, the extent to which Vicky was able to construct meaning, also played a role. She 'switched off' for the section of the tutorial that covered referencing, a skill in which she already felt competent, but the combination of the other activity and discussion made her appreciate different perspectives and she was able to make links between theory and practice.
I wondered if there might be some overlap between cognitive presence and what is termed 'learning presence', as both types of presence are concerned with what students do but I thought it might be possible to distinguish between students' behaviour in relation to constructing meaning (cognitive presence) and their behaviour in terms of how they exercised agency in relation to their learning (learning presence). I looked for evidence of learning presence by recording instances of self or co-regulation, where students' narratives mentioned planning, monitoring or use of strategies that were triggered by the tutorial activities. Instances of learning presence identified in Vicky's narrative (see Figure 2) included using ideas offered by other students in the tutorial for her assignment and contacting her tutor to ask if tutorials could include more examples. The instances of learning presence identified often also appeared within the action I poems created in the second stage of the analysis, which highlighted what students had done differently, if anything, as a result of their participation in the tutorial. In this way, searching for learning presence separately did prove helpful. As such, instances of learning presence could be seen as an outcome of tutorial attendance, and I chose to include it in an 'outcomes' column on the diagrams that I created.
Examples of emotional presence identified in Vicky's narrative (see Figure 2) included feeling 'fine' but also 'not overly comfortable.' Just as instances of learning presence identified often also appeared within the action I poems, instances of emotional presence could be helpfully compared with the feelings I poems created in the second stage of the analysis. In this way, the search for emotional presence proved useful in contributing a more complete picture of each student's tutorial experiences.
Overall, the addition of this extra listening through a more deductive lens and the construction of diagrams to illustrate what I learned proved helpful. It allowed consideration of the relationship between different types of presence but also provided an additional source of information that could be compared with that gained from earlier stages of the analysis.

Conclusion
I conclude that VCRM is not only a helpful specific approach to assist in the analysis of narrative accounts but one that can be usefully developed and creatively adapted in multiple directions. The use of a narrative approach and my choice of VCRM as a method for analysing the data has facilitated a different approach to studying students' experiences of synchronous online tuition, which has added to the insight available from previous research, much of which took a quantitative approach. Through applying the different listenings to students' narratives of their experiences, different voices could be heard and amplified that might otherwise be hidden or inaccessible, even from the students themselves. The research has contributed to the ideas around the theory and practice of VCRM by adapting it in two ways.
My first adaptation was to create I poems that spoke of three different parts of the student self in terms of identity, feelings, and actions. These highlighted some key aspects of students' experiences and heightened their visibility.
My second adaptation of this method was to add an additional listening to identify the different presences from the Community of Inquiry framework within the narratives. The addition of this deductive lens proved to be a valuable way of understanding each student's experience and the relationships between the different types of presence in the online tutorial sessions that they attended. It added an additional layer of evidence for the analysis that would not have been available otherwise.
Through choosing VCRM and adapting and adding to its different stages, I have learned that it is possible to interrogate the data from different perspectives and then bring these perspectives together, comparing and reflecting on the information gained to give a more detailed picture of participants' experience. Researchers should not feel limited to following the methods of analysis of those who have gone before them or restricted by the arguments of those who hold that researchers using poetic forms in their research must study the craft of poetry. I recommend that researchers choosing VCRM should embrace the flexibility that this method of analysis offers and creatively experiment within the different stages of their analysis to maximize their understanding of participants' experiences.