Interview analysis

The approach to interview analysis allots a fixed procedure to the treatment of the interviews. The following figure visualises the over-all modus operandi:

Figure 4: Interview analysis

As noted above, this analysis strategy has been developed with the help of Coffey &

Atkinson (1996), Kruse (2006) and Lucius-Hoene & Deppermann (2002). In the following, I want to describe the particular steps shown in Figure 4 in more detail:

1st Step:

(1) Content oriented coding

Coding constitutes the first step of analysis. The passages of the interview become segmented with the help of certain codes. The codes are developed both from the questions raised beforehand (see chapter 2.5.1 ) and from the text itself. The latter ensures that the analysis is open to subjects raised by the interviewees that possibly have not been considered before. Thus, the ”principle of openness” is taken into account again.

A practical example, illustrating very well how coding is used in this analysis strategy, is a code/topic – the code ”freelancing as a temporary situation” (Freiberuflichkeit als temporärer Zustand) from the case excerpt of Petra, one of the German freelancers. The description of freelancing as something temporary is prominent in Petra's narration. She already raises this topic in the very first passage of the interview. A look at the semantics in other parts of the interview confirms the importance of this topic – it thus becomes an independent topic in the case excerpt. This topic can even be described as crucial for the interpretation of the interview, as it relativises or at least challenges the predominant account of freelancing as very unproblematic in the interview. (see case excerpt 1, appendix 1, p. 16)

As this example also shows, it is clear that coding is not used to simplify or only shorten the data 'but to open them up in order to interrogate them further, to try to identify and speculate about further features' (Coffey & Atkinson 1996, p.30). In this case, the identification of ”freelancing as a temporary situation” as a code led my attention to that topic in the other steps of the analysis, leading to an interesting and even crucial interpretation in the end. Hence, the three steps of the analysis strategy go hand-in-hand and indeed become complementary.

Another very important function of coding is the structuring of data, thus making it easier to analyse the data in the next step. The respective sequences of the interview are subsumed under a thematic code in order to be analysed together in the next step. As the practice of analysis shows, several codes are often assigned to one part/sequence. It is of great importance to keep in mind that several subjects and narratives are often dependent

upon each other (as the above example illustrates very), even if they are analysed separately. Therewith, the principles of the second hermeneutical circle concerning the interplay between part and whole (see chapter 2.2) can be implemented within the practice of analysis.

(2) Analysis

This step constitutes the actual analysis in a narrower sense, which means that in this phase the by means of coding structured sequences of the text are examined by focussing on content, metaphors / figurative language, use of words, narrative figures and agency in order to come to interpretations.

The first aspect – content – is quite straightforward since the interviews have already been structured in regard to this category during the coding phase. Therewith, the context is always considered, and thus interpretation is always related to context.

The theoretical roots of the analysis of metaphors has been discussed in chapter 2.4.

Function and meaning of metaphors and figurative language use are thereby particularly of analytical interest. A good example for the application of the analysis of metaphorical language is the case of Elena: One of her most important strategies to cope with some of the hardships of freelance work is ”taking herself seriously”. In order to illustrate her point, she uses medical terms as metaphors: She describes her strategey of ”taking herself seriously” as ”growing an own skin in order to become less prone to infections⁷” (see case excerpt 7, appendix 1, p. 98 / interview transcript 7, appendix 2) This particular use of metaphorical language reveals a lot about Elena's view on freelance work as containing something she needs to be protected from and eventually led to interpretations concerning her views on customers and work in general (see chapter 7.3 for a discussion on the

”invasive” power of work).

The analysis of the use of words is approached eclectically: It is mainly concerned with the usage or omission of words. Sometimes, a closer look on how a specific word or

7 The citation is translated from German to English by the author. In the German original the interviewee says: ”Also, das ist so ein bißchen wie eine eigene Haut zu bekommen, so dass man nicht infektionsanfällig ist,ne?” (see case excerpt no 7, appendix 1, p. 98)

term is defined or used by the interviewee leads to crucial interpretations. An example of this is the use of the term ”flexibility” by the interviewees. My analysis shows that a large number of the interviewed freelancers used the word but in different ways. This instance led to some very relevant insights into the different ways in which freelancers conceptualise and deal with work and life (see chapter 7.4 for a discussion on the ”flexible”

interpretations of ”flexibility”). This analysis tool is inextricably entwined with the over-all holistic approach towards interview analysis applied in this study. By viewing every interviewee as a case, an analysis of concepts in the context of the whole becomes possible.

Hence, the hermeneutical assumptions about how understanding is achieved are strongly accounted for. Another function of the analysis of the usage of words is closely connected to the analysis of narrative structure. The analysis of why certain words are used (or not used) in a certain context can help to identify narratives.

Concerning narration, the main focus is on the analysis of narrative figures. This has been inspired by Kruse (2006), who defines narrative figures as 'self-contained, recurrent figures of set-up and organisation of speech' (p.79). These are identified in order to gain an interpretation of how certain topics are conceptualised, set in context, and so forth. One examples of these narrative figures would be the modes of discoursivation like individualisation/normalisation or ”contrasting”. In fact, the latter has proved to be a very common narrative figure: The interviewees quite often defined certain “things,” events, situations, and so on, in a negative way, that is, by describing what (or how) they are not, and they did this by contrasting these things, events, situations, etc., with something else.

Therefore, an analysis of contrasting is a very useful and fruitful tool.

The last ”tool” applied in the analysis, namely agency analysis, also belongs to the field of narrative figures. However, agency analysis has proved to be such a useful tool in analysing the first interviews that I have included it as a constant heuristic and used it systematically. Agency analysis is an approach derived from positioning analysis, as mentioned above (chapter 2.4). The concept draws on the interviewees subjective understandings of who or what contributes to or causes certain events, and thereby

identifies whom or what the interviewees ascribe agency to in certain specific situations.

As this research project asks for similarities and differences in the freelancers' handling of work and life, an analysis tool that is able to identify who or what they think is in charge of the situation(s) they are in is extremely useful. Lucius-Hoene&Deppermann (2002) use the following examples to illustrate the concept:

(1) ”... and then I was brought to the hospital and got operated on”

(2) ”... and then they brought me to the hospital and operated on me”

(3) ”... and then I went to the hospital and got myself operated on” (Lucius-Hoene &

Deppermann 2002, p. 59, author's translation)

In the first sentence, the subject attributes agency both to an anonymous power; in the second sentence, the subject assigns agency to others that are in charge of the situation (they brought me to the hospital and operated on me); whilst in the third sentence, the subject ascribes agency entirely to himself or herself (I went to the hospital and got myself operated on).

As noted, this analysis apparatus has been applied to all interviews. However, one additional feature has been employed in the case of the foreign language (Danish) interviews. Besides consulting a Danish dictionary (Becker-Christensen et al., 2005) and an online idiom dictionary (Bergenholtz 2009), I also discussed certain matters with native speakers so as to clarify any obscurities. If I was not sure about certain understandings and interpretations, for instance, I noted down the respective sentences and the context in which they appeared, and then discussed them with fellow Ph.D. students, who are native Danish speakers. This happened either in a group discussion or in exchange with only one other person. The strategy proved to be very useful in the case of the Danish interviews and helped to diminish misinterpretations.

(3) Case excerpts

The third and last step of my analysis strategy is the creation of a case excerpt for every interview. These excerpts summarise the findings and interpretations of the respective interview analysis. They help to reflect the findings again in the context of the whole

interview (see the discussion on hermeneutical circle 2, chapter 2.2). In addition, the excerpts are devices for structuring the findings in a well-ordered manner, thus making them more easily accessible for further analysis. The case excerpts contain direct citations from the transcripts and references to the parts of the interviews that are relevant for the respective topics. Those case excerpts vary in length between 9 and 21 pages (see case excerpts, appendix, p. 15 - 177).