• No results found

Interview process and topic guide

In document Health Literacy: (Page 150-154)

Chapter 4 Research Methodology and Method

4.6 Research method

4.6.5 Interview process and topic guide

The hermeneutic interview is a fluid dialogue between researcher and participant therefore, the researcher aimed to encourage a discussion that flowed as a conversation - the purpose of the interviews was to investigate what individuals experience and understand in terms of health literacy (Vandermause & Fleming, 2011). For all participants the recruitment process was a useful stage in the research design as it allowed time for the participants to think more deeply about their experiences. After an initial question, the researcher used probes and follow-up questions that were focussed on the participant’s experiences (Dinkins, 2005;

Sayre, 2001). In the interview participants were asked to tell a story that stood out for them regarding health literacy.

The interview topic guide was constructed reflecting the dimensions of Nutbeam’s health literacy with an emphasis on communicative and critical health literacy and also the dimensions proposed in the conceptual framework in Chapter 3. The interview topic guide and research purpose did not specifically focus on the details of individual health events, but rather the health literacy surrounding their experiences. Three pilot interviews were conducted comprising two baby boomer interviews (one male and one female) and one PHCP interview. Following the pilot interviews, one addition was made to the baby boomer interview topic guide adding a question on obstacles individuals experienced in accessing primary healthcare. Refer to Appendices 8 and 9 for the two interview topic guides.

For the baby boomer participants, each interview began with the participant answering seven demographic questions (refer Appendix 8), including a question on self-reported health status. Self-reported health is a simple measure which has consistently been shown to be valid indicator of health status (Franks, Gold, &

Fiscella, 2003), particularly in population samples similar to this research (Miilunpalo, Vuori, Oja, Pasanen, & Urponen, 1997, p. 517). The answers to these questions provided a demographic profile of the baby boomer participants. Each

136

participant then answered the Chinn and McCarthy (2013) 13-item All Aspects Health Literacy Scale questions (AAHLS, refer Appendix 8 for the questions, and Appendix 10 for the coding summary).29 This scale was chosen for a number of reasons. It included statements measuring both critical and evaluative aspects of health literacy as well as functional aspects. The parsimony of the questions meant it did not detract from the main purpose of the interview, which was to develop a conversational dialogue between the researcher and the participant; in the majority of the interviews the questions served as useful prompts for discussion around the main interview topics. The demographic information and the AAHLS scores provide participant profile data, summarised in Appendix 11.

Interviews were conducted over five months between August and December 2013.

For all participants, the research interviews occurred in four different locations:

the participant’s workplace, a quiet small room on a university campus, the participant’s home, and the home of the researcher. Three interviews were conducted via Skype. Each in-depth interview was recorded using two digital recorders and then transcribed. This resulted in 787 single-spaced pages of transcriptions. After each interview the researcher wrote herself memos, recording reflections regarding the interview data, ideas, or patterns that were revealed by the participants’ data. These memos proved useful in the initial iteration of textual interpretation (refer Appendix 12 for examples). The average length of the interviews with baby boomer participants was 43 minutes and 52 minutes for the PHCP participants. The interview details are summarised in Appendix 13.

This research uses verbatim transcripts. The transcription process when “we transform others’ words from spoken to written form” (Bucholtz, 2007, p. 802) is part of the research process involving a variety of choices and interpretations that are important to disclose. Transcription has been labelled a socio-political act (Bucholtz, 2007, p. 802) that differs with research perspective and purpose;

“…neither transcripts nor electronic recordings should be treated as data that are simply given, in an unmediated fashion…” (Hammersley, 2010, p. 556). Ricoeur similarly notes the problems posed by “the passage from oral to written discourse”

29 Email approval to use the AAHLS survey received from Dr D. Chinn, King’s College, London, 19 June, 2013.

137

(Ricoeur, 1981b, p. 37) and although “the text acquires its semantic autonomy…and is dissociated from the ostensive references peculiar to oral discourse…text implies inscription…of an experience to which it bears testimony”

(Ricoeur, 1981b, p. 37). In hermeneutics, the interview text is not the same as already written literary texts: “…a qualitative research interview involves both the generation and interpretation of the text” (Kvale, 1983, p. 187) and the interviewer is often the co-creator of the text that is subsequently interpreted.

For the purpose of the present hermeneutic analysis the interview transcripts are textual representations where the research focus is “less on the mechanics of speech and more on the informational content of the interview and the social or cultural meanings attached to this content” (Hennink & Weber, 2013, p. 700).

Therefore, a “naturalized” 30 approach (Bucholtz, 2000) is taken to the transcription which detechnologises31 the transcribed text so that it conforms to the conventions and practices of written discourse. Despite choosing to create a transcript that is writing-like (as opposed to speech-like) the researcher made no further alterations of the transcripts except to remove filler words (e.g., mm, agh, er). Colloquial dialogue, slang, elisions (including the omission of the end of one word or the beginning of another), non-standard grammar and syntax have not been corrected so retaining some links to the speech-like aspects of the interview dialogues (similar to the protocol recommended by McLellan, MacQueen, &

Neidig, 2003, p. 66). Only when the spoken English in the transcript excerpts was considered to prevent a clear understanding for the reader did the researcher construct a meaning of the text by adding words in square brackets. This follows Hammersley’s (2010) call to make changes to transcripts with circumspection since changes can lead to false inferences from the text. While editing the transcripts for slang, nonstandard grammar, and syntax could improve their readability or comprehension, these elements have been kept intact as the researcher preferred to keep the authenticity and variability of the participants’

dialogue for readers to consider.

30 The term retains its American spelling as in the original reference.

31 Transcripts produced by linguists and discourse analysts “can technologise a text”

(Charteris, 2014, p. 100) by including highly detailed conventions/symbols for example, recording intonation, emphasis, pronunciation, pauses, and overlaps.

138

Table 4.3 Composition of baby boomer sample - gender, age, and ethnicity

Male Female Total Participants

Leading edge

Trailing edge

Total Leading edge

Trailing edge

Total Leading edge

Trailing edge

Total

European 4 1 5 4 3 7 8 4 12

Māori 3 3 6 1 5 6 4 8 12

Pacific Peoples

1 4 5 1 5 6 2 9 11

Asian 3 2 5 1 5 6 4 7 11

Total 11 10 21 7 18 25 18 28 46

139

In document Health Literacy: (Page 150-154)