Skip to content ↓

Oct 2018 - Year 20 - Issue 5

ISSN 1755-9715

Interview Data in Action Research

Neil England is a Lecturer in the School of International Studies, University of Technology Sydney. His research interests are in language teacher and language teacher educator cognition, language teacher education, EAP curriculum development, and how qualitative educational researchers approach their work, especially in intercultural settings. Email:



Interviews are a common data collection method in action research. The interviewees are normally the researcher’s students, although depending on the focus of the study, the researcher’s teaching colleagues and people in authority in the researcher’s teaching context may also be interviewed.  Interviews produce data to do with ‘what people think, believe and perceive and also the way they explain their personal histories, experiences and actions’ (Burns, 2010, p. 74). As such, they can be an effective way of getting to what is below the surface of observable actions.

A standard perspective on interviews is that they are research instruments which produce reports on interviewees’ biographical, experiential and psychological worlds. As Hollway (2005, p. 312) puts it, the idea is that ‘you ask, they answer, and then you know’, in an unproblematic transparent process. Here I challenge this perspective and recommend that action researchers see interviews as a form of social practice (Talmy, 2010, 2011); that is, as fundamentally social encounters, where the data is co-constructed in complex ways by the interviewer and the interviewee(s) in specific social, cultural and physical contexts. I discuss the implications of this social practice perspective in terms of how action researchers might collect, analyse and display interview data.


Understanding the research interview as social practice perspective

Talmy (2010, 2011) was the first to use the term research interview as social practice. In reference to essentially the same construct, Holstein and Gubrium (1995, 2011) refer to the active interview, and Talmy and Richards (2011) to a discursive perspective on qualitative interviews. Talmy and Richards’ (2011) definition of their term provides a succinct overview of the common perspective:

What we mean by the term ‘discursive’ as it applies to theorizing interviews is that the interview is conceptualizing explicitly as a socially-situated ‘speech event’ (Mishler 1996) in which interviewer(s) and interviewee(s) make meaning, co-construct knowledge, and participate in social practices (Holstein and Gubrium 1995, inter alia). This contrasts with the more common place …perspective of the interview as a neutral technology, or research instrument (Talmy 2010), used to mine the attitudes, beliefs, and experiences of self-disclosing participants. A discursive perspective on interviews, in other words, aligns with what Holstein and Gubrium (1995, inter alia) have called the ‘active interview’…The active interview is a theory of interview that foregrounds not only the ‘content’ drawn from interviews – that is, the whats – but also the linguistics and interactional resources used to (co)construct it – or, the hows. (p. 2)

Table 1, adapted from Talmy (2010, p. 132), presents a summary contrast of the interview as research instrument and research interview as social practice perspectives.


Interview as research instrument

Research interview as social practice

Status of interview

A tool for gathering information.

A site for investigation itself.

Status of interview data

Data are reports which reveal truths, beliefs, attitudes, interior mental states of self-disclosing respondents.

Data are accounts of truths, beliefs, attitudes, interior mental states co-constructed between interviewer and interviewee.


Interviews ‘give voice’ to interviewees.

Voice is situationally contingent and discursively co-constructed between interviewer and interviewee.


Interviewers must strive to obviate data contamination.

Reflexive recognition that data are collaboratively produced; data cannot therefore be contaminated.

Analytical approaches

Content or thematic analysis, summaries of data and/or verbatim or abridged quotation; data ‘speak for themselves’.

Analysis centres on how meaning is negotiated, knowledge is co-constructed, how interview is locally accomplished.

Analytical focus

Product-oriented: what.

Process-oriented: what and how.

Table 1: Contrasting conceptualizations of the research interview (Adapted from Talmy, 2010, p. 132)


‘Showing the workings’ of a study

A focus on the hows is part of what Holliday (2007) refers to as ‘showing the workings’ of a qualitative study. This means the researcher is clear, explicit and honest in their public accounts of the choices they are considering or have made in the conduct of their study and, in particular, the circumstances and processes of data collection and analysis. This type of reporting - in a talk or in a publication – is a way of establishing the rigour of the research. It can also be extremely useful for novice researchers, particularly novice action researchers who may lack confidence in planning and managing the methodological aspects of their study. A researcher’s succinct in-context description and justification of the workings of their own study can provide novice researchers with useful methodological direction and, importantly, confirmation that their own choices are valid. This type of direction and confirmation is usually much more powerful than that available through the often idealistic guidelines presented in research manuals (Hobbs & Kubanyiova, 2008).

An action researcher’s interviews with their own EFL students are sites for a complex interplay of issues of language, power and role expectations. A rigorous and reflexive action researcher in this context needs to consider, and report on, how data are shaped by language choices and by interaction with the students as both teacher and researcher. This is a key part of ‘showing the workings’ of a study of this kind.


Following through on the theory

Mann (2011) makes a number of points about how researchers can follow through on a theorization of research interviews as ‘active’ and as a form of social practice. In this section I take up some of these points and present them as recommendations for action researchers working in an EFL context. The recommendations are classified under two headings: (1) Taking the full interactional context into account, and (2) ‘Seeing’ co-construction of knowledge.  The recommendations under the first heading relate to the notion of situated knowledge, grounded in physical, temporal and social circumstances and specific events. The recommendations under the second heading relate to making the hows of the co-construction of knowledge as explicit as possible.

  1. Taking the full interactional context into account
  • Take into account researcher-interviewee relationships developed through events that took place prior to the start of official data collection.
    These relationships and events can have a significant impact on what happens in an interview and what data is co-constructed in it. For example, the length of time the action researcher has known and taught the students, and specific events that have taken place with the class and with individual students, will shape rapport and trust and therefore what is said and not said in an interview.
  • Take note of how you as an action researcher informed your students of the purpose of your research project and the purpose and format of any interview or parts within it.
    The content of the talk with your students will be determined in part by their ‘task understanding’ (Mann, 2011, p. 10). This type of understanding is often fragile in contexts in which issues of language grading are particularly relevant.
  • Take into account the physical and social circumstances of interviewing.
    This recommendation relates to matters such as the possibly threatening presence of a recording device, and whether the interview is/was with an individual student, students in pairs, or students in a group. Personality, confidence, trust, respect and rapport issues may be relevant to the nature of the student-student interaction in pairs or groups during the interview.


  1. ‘Seeing’ co-construction of knowledge
  • Present a transcribed excerpt from the interview data in its interactional context.
    Many stretches of talk are only meaningful if considered in the context of the preceding stretches of talk. When displaying transcribed excerpts from the interview data, action researchers should therefore consider providing readers, where appropriate, with a summary of the preceding discourse and how the talk in the excerpt fits in to that discourse.
  • Include the interviewer’s turns when presenting excerpts from the transcribed interview data.
    In reports on a completed study or a study in progress, the reader is very often presented, as ‘evidence’, transcribed statements from just the interviewee. The interviewer’s turns - that is, what the interviewers actually asked or said to elicit the interviewee’s response – are sometimes summarized and at other times not mentioned at all. This method of data display does not show how meaning was actually co-constructed.
  • Consider the use of mediational tools.
    Interview talk in language teaching and learning contexts is often mediated by tools such a published text, activities from a course book, or a piece of student writing. Action researchers can usefully note, analyse and report on how these mediational tools were introduced to the interviewees, kept in focus, deviated from, and even in some cases abandoned.


Some concluding remarks

The recommendations I have presented here demand a high level of researcher reflexivity and afford documented reflexivity the status of valid data in a carefully conceptualised and conducted action research project. As noted in the section ‘Showing the workings’ of a study, by systematically presenting the choices they have made, action researchers establish the rigour of their study and give other action researchers useful methodological direction and confirmation. Researcher reflexivity can be documented in a research journal (Borg, 2001).

A final note is recognition of word limit constraints on published reports on completed action research projects. In enacting the recommendations presented here, particularly those under the heading ‘Seeing’ construction of knowledge, writers will face challenges in meeting word limits for the whole text if they wish to focus equally on the whats and hows. There is no easy answer to this problem other than employing principles of conciseness and being highly selective in the display of interview data.



Burns. A. (2010). Doing action research in English language teaching. New York: Routledge.

Borg, S. (2001). The research journal: A tool for promoting and understanding researcher          development. Language Teacher Research, 5, 156-177.

Hobbs, V., & Kubanyiova, M. (2008). The challenges of researching language teachers: What   research manuals don’t tell us. Language Teaching Research, 12(4), 495-513.

Holliday, A. (2007). Doing and writing qualitative research (2nd ed.). London: Sage.

Hollway, W. (2005). Commentary. Qualitative Research in Psychology, 2, 312-314.

Holstein, J. & Gubrium, J. (1995). The active interview (2nd ed.). London: Sage.

Holstein, J. & Gubrium, J. (2011). Animating interview narratives. In D. Silverman (Ed.), Qualitative    Research (3rd ed., pp. 149-167). London: Sage.

Mann, S. (2011). A critical review of qualitative interviews in applied linguistics. Applied Linguistics,   32(1), 6-24.

Talmy, S. (2010). Qualitative interviews in applied linguistics: From research instrument to social practice. Annual Review of Applied Linguistics, 30, 128-148.

Talmy, S. (2011). The interview as collaborative achievement: Interaction, identity, and ideology in a speech event. Applied Linguistics, 32(1), 25-42.

Talmy, S. & Richards, K. (2011). Theorizing qualitative research interviews in applied linguistics. Applied Linguistics, 32(1), 1-5.


Please check the How to be a Teacher Trainer course at Pilgrims website.

  • Interview Data in Action Research
    Neil England, Australia