From talk to text. Ethical representations of participant voice

The choice of interview as a research tool is usually made because of the particular affordances of an in-person conversation, in contrast to a questionnaire or written documentation. Each interview is a unique social interaction, and the goal of interview research tends to be particularity, contextualisation and nuance, rather than simplistic generalisations. Research interviews, therefore, allow for an in-depth exploration of participants’ perspectives on the research issue, offering of ‘an understanding of the world from the subjects’ points of view, to unfold the meaning of peoples’ experiences’ (Kvale,1996, p. 1). The direct interaction between interviewer and interviewee, occurring in real time, generates flexibility, allowing the interviewer to ‘press not only for complete answers but for responses about complex and deep issues’ (Cohen et al., 2018, p. 506). Crucially, interviews give voice to the participants: rather than being obliged to use closed responses, such as a scale or multiple-choice questions, participants respond ‘using their own words’ (Braun & Clarke, 2013, p. 78). So, for researchers keen to understand human situations, contexts and phenomena, interviews seem to be a wholly appropriate choice to fully represent the complexity of human thinking, behaviours and actions. However, it is important to think critically about the ethics of representation when presenting data from interviews. Many articles make claims about their interview data: for example, that they ‘offer a more accurate assessment of the impact of life events on any individual’ (Fallon, 2008, p .387). Whilst this may be true, there is an ethical responsibility on the researcher to ensure that participant voices are represented as accurately as possible, avoiding bias, partial representation or cherry-picking of the data. Drawing on interview data from an international comparative study funded by the Norwegian Research Council, which is comparing language use and instruction across contexts (‘LANGUAGES’), this presentation will examine the ethics of representation when using interview data among teachers and students in secondary school in three different countries: Norway, England and France. It will do this in two ways. Firstly, we will consider how the design of the interview schedule and the data collection process is intrinsically linked to ensuring the integrity of representation – particularly in relation to interviewing technique, transcription decisions, and methods of data analysis. Secondly, we will address how participants’ voices should be represented in publications, notably the clarity of the link between data analysis processes and form of data presentation; how quotations will be used and the contextualisation of quotations; and direct attention to contradictions in individual participants’ data, plural and sometimes contrasting voices, and alternative perspectives between participants (e.g. teachers and students in the same class).