Interview studies of studio-based learning

The semi-structured interview has become the most common method of data generation in the social sciences (Madill, Reference MADILL2011; Roulston, Reference ROULSTON2006; Peräkylä, Reference PERÄKYLÄ, Denzin and Lincoln2005; Potter & Hepburn, Reference POTTER and HEPBURN2005), corresponding with an increasing interest in ‘experience, meaning, life world, conversation, dialogue, narrative, and language’ (Kvale, Reference KVALE1996, p. 11). Through interviews more than, for example, observations, qualitative researchers can investigate the perceptions, intentions, and attitudes of participants, though importantly, it cannot be assumed that interviews, or any method of data collection, can offer direct and unproblematic access either to the respondent's views or to the subject that she may be discussing. Denzin and Lincoln (Reference DENZIN, LINCOLN, Denzin and Lincoln2005) explain:

While it is widely accepted that ‘reality’ is constructed, in the social if not the physical world (Pring, Reference PRING2000, p. 60), there is some debate about the extent to which research participants can be regarded as having stable, psychosocial characteristics that might explain their behaviour, and as being reliable sources of information about their social world (for example, Hammersley, Reference HAMMERSLEY2003; Potter, Reference POTTER2003). In some cases the limits on the reliability of respondents have been clear, and noted as a topic of interest in itself. For example, Hallam (Reference HALLAM2001) interviewed children about their practice behaviour, and noted differences between what they reported doing, and corresponding video evidence; she described the discrepancy as a ‘production deficit’ (p. 10). This was related to child development, and in particular, the development of linguistic competence; it is likely to be exposed when the task at hand – for example, practising the violin – ‘could be construed as reflecting or embodying certain intellectual competencies which have nothing intrinsically verbal about them’ (Flavell, Beach & Chinsky, Reference FLAVELL, BEACH and CHINSKY1966, p. 297).

Another mismatch between participant reports and observation evidence presented itself in a study conducted by Koopman et al. (Reference KOOPMAN, SMIT, DE VUGT, DENEER and DEN OUDEN2007), who compared written logs and questionnaires from conservatoire students, with video data of their lessons and practice. Discrepancies were found between student reports and the researchers’ analysis of the video data, concerning the number of initiatives students took in lesson interactions, and the clarity of both homework instruction and lesson aims. The researchers suggested that the students may have felt that these were ‘sufficiently manifest’ in what they were doing in lessons, and that with a focus on practical results, ‘the explicit verbalisation of aims was not considered as necessary’ (p. 388).

It seems important that Koopman et al. responded to their apparently inconsistent data, not by questioning the honesty or competence of student participants but by asking how the discrepancy might have come about, and by proposing a more richly textured explanation for the situation they were studying. A similar attitude perhaps was adopted by Gaunt who, having interviewed 20 conservatoire teachers (2008) and students (2010), compared their responses in a further study (2011). Aside from drawing out common themes, Gaunt noted that the two groups tended to describe the teacher-student relationship in different terms, with teachers referring to friendship or parental roles, and students tending to focus on the felt benefits of the relationship. The differing perspectives served to highlight ‘the complexity of the student-teacher relationship [and] the individuality of each relationship over a sustained period of time’ (p. 174).

Of course, some of the ambiguity apparent in interview data may arise from attitudes to the situation itself, particularly with an interviewer who may be an academic in the same institution, and who therefore might be regarded as representative of it. Gaunt notes that her position as an ‘insider teacher-researcher’ gave her access to interview participants, and an opportunity to cultivate a degree of understanding and trust (2010, p. 181); arguably, however, her position as a colleague of their current teachers might be associated with an apparent unwillingness of students to criticise them, in contrast with their remarks about past teacher-student relationships. This possibility might be linked to the reluctance of teachers and students in a Norwegian academy of music to engage in the formal evaluation of studio teaching, as required by the institution: according to Hanken (Reference HANKEN2011) this was partly because that formal mode of communication was perceived as unnatural and redundant:

Both Gaunt (Reference GAUNT2010) and Hanken (Reference HANKEN2011) draw links between students’ ability or willingness to articulate concerns about their studio lessons, and the perceived need to maintain a positive teacher-student relationship, founded on trust.

In a number of studies, insight into student attitudes has been sought through interviews combined with video observation, using video-recall methods to embrace participant intentions and concerns in both data collection and research design (Johansson, Reference JOHANSSON2013; James et al., Reference JAMES, WISE and RINK2010; Nielsen, Reference NIELSEN2010; Hultberg, Reference HULTBERG2005). James, Wise and Rink (2010) explain that the procedure ‘has the potential both to allow insights into their experiences that could not be gained from viewing the footage alone, and to facilitate participants’ conscious access to processes that they may not often think about or articulate’ (p. 225). In the broader study from which the current paper was drawn, participants were interviewed about their work shortly after their lessons had been filmed, relying on fresh memories rather than video recall. Within that, sub-studies have used ‘rich transcription’ to explore a total of four studio lessons, with a close description of verbal, nonverbal and performance behaviours, and with teacher and student interviews casting further light on the observation analysis (Burwell, Reference BURWELL2010, Reference BURWELL2012, Reference BURWELL2016a, Reference BURWELL2016b). Rich transcription is a satisfying, but painstaking and time-consuming task, that would be difficult to sustain over a larger number of analyses. It highlights however the complexity that lies in both studio practices and participants’ representations of them, alongside the value of establishing a sense of context for individual cases.

Procedure

In the current study the ‘dissonant’ cases of Sandra and Gavin are contextualised by considering their interviews alongside those of their 25 colleagues. In the rich transcription of their cases, interviews were regarded not as reports about studio practices, but as accounts to be interpreted; even so, the main focus was on the analysis of video data, with interviews providing signposts and sidelights on lesson behaviour. Concentrating on the interview data now would allow, and indeed call for, a more reflexive approach to its analysis. Thus Potter and Hepburn (Reference POTTER and HEPBURN2005) draw attention to the presence of the interviewer in discussions, the interactive nature of the procedure, and the relative footing of interviewer and respondent; although the differences between interviews and naturally occurring talk are sometimes exaggerated (Potter, Reference POTTER and Silverman1998, p. 236), interviews share some of the asymmetrical features of ‘institutional’ talk (Madill, Reference MADILL2011). Patterns of verbal behaviour, and a sense of order, might be evident in interview data as much as lesson behaviour, though they may be more or less explicit for the participants. Through a free interplay of analytical techniques (Kvale, Reference KVALE1996) qualitative researchers can investigate how interview accounts are constructed, as well as casting light on the practices under discussion (Roulston, Reference ROULSTON2006; Hammersley, Reference HAMMERSLEY2003; Potter, Reference POTTER and Silverman1998).

Participants were identified for the broader study through a call for volunteers among instrumental and vocal teachers in a university music department; the teachers in turn helped to identify student participants. The final selection of participants was made with reference to the range of instruments and styles covered, and the convenience of scheduling interviews and the filming of lessons. The procedures were consistent with the ethics protocols of the host university: participants were not classified as being particularly vulnerable, were involved through informed consent, and were assured of anonymity in reporting, while the topics under investigation were not considered particularly sensitive. In quoting interview data, names are given to any participants who have been given pseudonyms already in previous research, or who are mentioned more than once in the Findings and Discussion to follow.

The data collection relevant to the current study involved filming three studio lessons for each of nine teachers, during the final weeks of the academic year, shortly before performance examinations took place. After each lesson was filmed, teachers and students were interviewed individually; it is important to note, in the light of the previous discussion and the nature of the study, that the interviewers were lecturers and researchers working in the same department. Students were asked for some biographical information, focusing on their experience in music teaching and learning, and to comment on their studio work, in the lesson just filmed and in more general terms. The interview schedule consisted in a series of prompts, allowing for further questions and the elaboration of topics as they arose.

For the current study the interview data were analysed again, repeatedly interrogating the responses that Sandra and Gavin gave to the question about their teachers’ approaches, and considering those in the context of the whole body of interview data. This involved the adoption of ‘an analytic mentality which is sensitive to a range of features of discourse’ (Potter, Reference POTTER and Silverman1998, p. 239), with some reference to features of the organisation of everyday language (p. 240). The procedure was used to establish whether and how the dissonant cases might be distinct from the rest: firstly in their accounts of their teachers’ approaches; secondly in the occurrence of two salient hallmarks, feel and think; and finally in remarks related to the order of lesson interactions.

Findings

The point of departure for this case study was the interview question asking students whether their teacher's approach was appropriate for them, at their current stage of development. Of the 27 students interviewed, only two answered this question negatively:

• SANDRA: I don't really think it is actually. I just don't feel comfortable sometimes with it. I feel a bit let down. I asked her last week ‘How am I doing in comparison to everyone else?’ because I just feel that I am not as good as what she is used to. She doesn't encourage me as much as I want her to. Her style is to make me bring out a big sound but it's not very appropriate because the piece I am doing at the moment is very light and this big sound doesn't really go with it, I find.

• GAVIN: It is not really. I want more self-discovery, exploring different ways of doing things. I don‘t want to be told that there is a definitive way. Today he was talking about playing vibrato on a particular note with three fingers. I was playing it with my little finger. I couldn't hear the difference. I can't see the benefit of putting my whole hand out of position to play it.

How distinct were these responses from the remaining 25? Five students (19%) answered the question positively but almost automatically, in that they provided no immediate elaboration – for example, ‘Yeah: I am very happy’ and ‘Just what I wanted’. 17 students (63%) answered positively and added some explanatory information, referring for example to choice of repertoire, clarity of instruction, flexibility and devolving responsibility.

Not all of the students answered plainly: three students (11%) gave answers that were positive but included some ambivalence. One of these was Beth (discussed in detail in Burwell, Reference BURWELL2012), who demonstrated an emotional attachment to a previous teacher, saying ‘Well, I would rather have my old teacher but he's great. . . if she didn't exist then he would be perfect’. The next student indicated that all was well, but this had not always been so: ‘When I first came, because I suppose I am quite shy, and to play in front of her was quite strange. . . but I think we have got used to one another now’. The remaining student, known as Rebecca, answered positively about her teacher's approach but went on to remark that today’s lesson had been frustrating: with performance exams imminent, she would like to have spent less time on detail.

Sandra and Gavin stood out for their frankly negative answers, particularly as their teachers each had two other students involved in the study, and all of them answered entirely positively. While Gavin emphasised professional issues in explaining his response, however, Sandra's response emphasised the more affective aspects of her learning, indicating how she felt about her situation before attempting to explain it. In the response quoted above she used the word feel three times; the third of these introduced an opinion, but the first two referred to Sandra's personal feelings – she felt uncomfortable, let down – and she wanted more encouragement from her teacher.

How distinct was this reference to personal feelings? In the interview schedule there were two consecutive items referring directly to the teacher, inviting students to characterise their teachers before proceeding to the ‘appropriate’ question. In responses to these questions, the words feel, feeling, feels, or felt were used 22 times. Eight of these references were to personal feelings: four students used the word feel once each, in positive descriptions of their lessons, while the remaining – negative – references all came from Sandra and Gavin. Gavin reported leaving lessons ‘sometimes feeling quite disappointed’, and later, he left a statement about his feelings incomplete:

How distinct was the reference to feeling, across whole interviews? In all, the words feel, feeling, feels, or felt were used 72 times; 29 of these referred to affect, and 9 of those (31%) were negative. Three students referred to negative emotions in this way, once each: Beth recalled that her previous teacher ‘had this amazing way of making me feel really bad if I didn't do any practice’; Rebecca, as previously noted, reported that she felt ‘completely frustrated’ – and ‘I never want to see those four bars again!’; the third student, a singer, felt ‘depressed’ today because she was not performing well. Sandra and Gavin referred to negative emotions three times each, and were the only students to link negative emotions to their teacher's approach.

It has been noted that sometimes the word feel was used not to describe feelings but to introduce an opinion; in such cases, think could be substituted for feel and the statement would still make good sense. Across whole interviews, 39 student opinions were introduced by the words feel, feeling, feels, or felt; of these, allowing for some overlapping, 18 (46%) referred to the teacher's approach, 13 (33%) referred to the self, and eight (21%) referred to the subject. Of the 39 opinions, 22 (56%) were critical, and of these, 15 (68%) concerned the teacher's approach, six (26%) involved self-appraisal, and one (13%) concerned the subject. The 22 negative opinions introduced by words based on feel came from only six students, and two students were responsible for most of them. Five negative opinions came from Rebecca, who – while acknowledging the good intentions of her teacher, and her own responsibility for taking initiatives in lessons – criticised the time-management in today's lesson, for example:

Gavin however stood out for introducing 11 negative opinions in terms of feeling. Six of these referred to his teacher's approach, and five to self-appraisal. Interestingly perhaps, Gavin introduced opinions in this way in only two other places, both positive, and both concerned with his own work, whether playing (‘I felt I had really got to grips with this piece of music’) or teaching (‘I feel [my pupils] need to hear good things about their work’). Three other students used the word feel to introduce opinions that were effectively self-compliments, about their capabilities or progress.

To what extent might the use of words based on feel depend on the initiative of the student? Gavin stood out by using feel, feels, feeling or felt, in whatever sense, no less than 19 times; he was followed by Rebecca (12) and Sandra (seven). The interviewer used those words far more rarely, with only 32 usages across the 27 interviews, and yet it was the interviewer who used them first, in the interviews with Gavin and Sandra. The following extract comes from an early stage in Gavin's interview. In all such extracts, the speech of the interviewer is shown in italics; here, the word feel is highlighted for the convenience of the reader.

In this extract, each use of feel by the interviewer is matched shortly afterward, by Gavin; and with the last usage, Gavin begins to apply the word in discussing his lessons. He goes on to use it more or less regularly during his interview. A similar pattern occurs in Sandra's lesson, when she uses the word feel shortly after it is introduced by the interviewer; a moment later, she uses it three times more in asserting that her teacher's approach is not appropriate for her – as previously quoted. In contrast, Rebecca uses words based on feel 12 times before the interviewer uses it at all, and the last seven of these are embedded in her discussion of her lessons and her teacher's approach.

It has been remarked that in expressing opinions, think could be substituted for feel without loss of sense. How distinct were they, in the student interviews? The words think, thinks, thinking, or thought were more than three times as common as words based on feel, with 232 student usages altogether. Of these, 53 (23%) did not signal opinions at all, but referred to thinking as an activity, often elicited by the teacher: for example, ‘[He will] get me to think of at least three different ways that I can interpret it’. In the 171 opinions introduced by words based on think, students referred to the subject (44%), the teacher's approach (36%), the self (16%), or the camera (8%). The remaining nine usages of words based on think were reports of the teacher's views.

Among student opinions introduced by words based on think, only five (3%) were critical. These were identified by the intention, rather than the grammatical function. Thus for example ‘I don't think so’ was used by eight students when asked if they had anything to add at the end of the interview, but while the structure of the expression was negative, there was no criticism evident in the context. All of the opinions that were taken to be critical concerned the teacher's approach. Three of these came from Gavin, who has already been quoted (‘I thought, I wouldn't do that to my students’, ‘I don't think that is necessary’, ‘I think the instrumental lessons are too laid back’). The two remaining negative opinions were less directly expressed. One came from Pamela, who in discussing her teacher's approach, offered a tentative reservation – ‘I think it is maybe, sometimes, with some people it seems it might be better for younger children’: eschewing all modification from this statement would have given a starker impression – ‘It would be better for younger children’. The other negative opinion introduced by think came from Sandra's response, again already quoted, when she was asked whether her teacher's approach was appropriate for her, at her current stage of development. Without the verbs think and feel, and without the adjectives really, actually, just, sometimes, and a bit, that response would have been ‘It is not; I am not comfortable with it; I have been let down’.

The tendency to modify such claims implies a constraint on what students were prepared to say about their lessons. The interview evidence suggested that some of the students were more, or less, conscious of the constraints that might apply, in their lessons. For example, when Gavin remarked that he could not ‘see the benefit’ of his teacher's suggestion about vibrato, the interviewer asked whether he had said as much, to his teacher:

Sandra, on the other hand, claimed that she was likely to assert herself with her teacher:

This claim was not borne out by the transcript of Sandra's filmed lesson, in which she made no verbal objection to the proceedings; and although this particular lesson may have been atypical, Sandra said in her interview that it was not: ‘That's what we normally do, yes’. Explicit objections to the course of a lesson, however, might not be easy to make. The interview evidence suggests that of all the students, Rebecca was the one who objected most, if not to her teacher's approach in general, then to the particular course of her filmed lesson, in which she had felt frustrated about the approach to time management, in the light of her impending examination. Yet Rebecca too was frank about not taking initiatives of this kind: asked whether she had discussed the issue with her teacher, she replied ‘No – well maybe at one stage, but no I haven't: you can't just come out with it.’

Implications

The close investigation of interview data confirms that ‘dissonance’ was present and distinct, not only in what Sandra and Gavin said about their studio lessons but in the fabric of their interview discussions. Their cases proved distinct, in producing directly negative accounts of their current teachers’ approaches, and in referring to those approaches in terms of their own negative feelings. At the same time, they framed those accounts by using the markers feel and think, or other modifying language, to introduce remarks that seemed to be delicate. In this their behaviour stands out, but has some support nevertheless, from the interviews of other students. Think is often used, across the sample, to modify opinions, and feel is used to modify opinions on particularly sensitive topics. While there is no evidence here to suggest that student dissatisfaction is common, it does seem clear that criticising one's teacher is a particularly sensitive topic, and not easily done; hence the efforts made by Rebecca and Pamela, to modify their criticisms of any aspect of their lessons. While it may not be clear that students are conscious of it, criticising one's teacher is evidently – even literally – out of order.

That there is an implicit order regarding studio-based learning is supported by Hanken (Reference HANKEN2011) who draws on Institutional theory to investigate attitudes to formal student evaluations; she explains that members of an institution are ‘influenced by collectively anchored values, norms and ways of thinking and acting that permeate the organisation and influence the way in which they will interpret the demands and, consequently, how they will choose to act’ (p. 246). This relies on a conception of institutions as ‘given, static, and constraining’, explaining continuity more than change; and yet discourse theorists have begun to seek sources of change within institutions themselves (Schmidt, Reference SCHMIDT2010, p. 2).

In previous reports about the case studies of Sandra and Gavin (Burwell, Reference BURWELL2016a, Reference BURWELL2016b), I have cited Jørgensen's (Reference JØRGENSEN2000) suggestion that institutions of higher education music should be taking a more active role in supporting the development of studio lessons, and argued that if students are able to articulate their concerns to an interviewer, the institution should be able to create a space in which they can articulate them to their teachers: a space in which teachers and students can engage in purposeful and objective discussions about their collaborative practice, working through any ‘incidents of studio dissonance’ (Reference BURWELL2016b, p. 511). The argument is complicated if the interview is regarded as a unique interactional form, in which students feel able to complain as if to friends, even without friendly commiseration – a view supported by the current study. Teachers and students often talk of their relationship in terms of friendship (Burwell, Reference BURWELL2016b, Gaunt, Reference GAUNT2011) though the professional setting of the studio must give it, also, an institutional character (Hanken, Reference HANKEN2011, Nerland, Reference NERLAND2007). But whether students can complain to their teachers about the approach taken in their studio lessons, is not the point: if objective discussions can be developed between them, mediated by a formal institutional framework, then student concerns, questions or anxieties about studio practices – ‘dissonance’, rather than impasse – can be addressed by participants before they can develop into less fruitful impulses. The evidence suggests that teachers and students are working within cultural norms that may be more or less explicit, more or less static; but it does not suggest that they would be averse to engaging with those in new ways, perhaps with the assistance that can be offered by theory and research; nor that institutions cannot support, in this sense, the evolving dynamics of the studio lesson.