depicts how many of the 30 pupils confirmed or disconfirmed each of the eight subcategories of the dimension ‘Relationship with the teacher’, generally supporting the hypothesis that grades influences social relations in school (Kvale, 1980, 1996b). Page 6 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. FIGURE 9.1 Categorization of the teacher-pupil relationship. The number of the 30 pupils interviewed who confirmed occurrence of a grading behaviour and attitude is shown on the right, and the number who disconfirmed a grading behaviour and attitude is on the left. As several pupils had no statements, or vague statements, regarding a category, the sum of direct confirmations and disconfirmations is less than 30. Implications for interviewing With categorization involving either/or decisions, it is preferable with precise pre-interview definitions of the categories and careful probing during the interview to ascertain how the statements may be categorized. When the codes or categories are not to be developed until interviewing and analysis, it is important during the interviews to obtain rich descriptions of the specific phenomena to be coded or categorized. Meaning condensation Meaning condensation entails an abridgement of the meanings expressed by the interviewees into shorter formulations. Long statements are compressed into briefer statements in which the main sense of what is said is rephrased in a few words. I shall here exemplify one form of meaning condensation developed by Giorgi Page 7 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. (1975) on the basis of phenomenological philosophy. The thematic purpose of his study was to investigate what constitutes learning for ordinary people in their everyday activities. The methodological purpose was to demonstrate how one deals systematically with data that remain expressed in terms of ordinary language and how rigour and discipline can be applied in data analysis without necessarily transforming the data into quantitative expressions. Table 9.2 demonstrates how an interview about learning was subjected to meaning condensation. The analysis involves five steps. First, the complete interview is read through to get a sense of the whole. Then, the natural ‘meaning units’ of the text, such as they are expressed by the subjects, are determined by the researcher. Third, the theme that dominates a natural meaning unit is restated by the researcher as simply as possible, thematizing the statements from the subject’s viewpoint as understood by the researcher. Table 9.2 depicts this third step of analysis. The fourth step consists of interrogating the meaning units in terms of the specific purpose of the study. In the fifth step, the essential, nonredundant themes of the entire interview are tied together into a descriptive statement. This form of meaning condensation can serve to analyze extensive and often complex interview texts by looking for natural meaning units and explicating their main themes. These themes may thereafter be subject to more extensive interpretations and theoretical analyses. Giorgi thus points out the importance of interpersonal relations by learning, which emerged in this study – a phenomenon that was rather neglected in the theories of learning at the time. (For further developments of the method, see Fischer and Wertz, 1979, and Giorgi and Giorgi, 2003.) It should also be noted that meaning condensation is not confined to a phenomenological approach and is also applied in other qualitative studies (Tesch, 1990). Implications for interviewing For a phenomenologically based meaning condensation it becomes paramount to obtain rich and nuanced descriptions of the phenomena investigated in the subjects’ everyday language. The interviewer’s theories of the subject matter should be ‘put into brackets’ during the interviewing. Meaning interpretation The interpretation of the meaning of interview texts goes beyond a structuring of the manifest meanings of what is said to deeper and more critical interpretations of the text. Meaning interpretation is prevalent in the humanities, such as in a critic’s interpretations of a poem or a film, and in psychoanalytical interpretations of patients’ dreams. The interpreter goes beyond what is directly said to work out structures and relations of meaning not immediately apparent in a text. In contrast to the de-contextualization of statements by categorization, interpretation re-contextualizes the statements within broader frames of reference. As compared to the text reduction techniques of categorization and condensation, interpretations often lead to a text expansion, with the outcome formulated in far more words than the original statements interpreted. Page 8 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. One example of meaning interpretation was given by the evaluation of Hamlet’s interview in Chapter 7 (see Page 9 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. Box 7.1). Different readings of the interview lead to rather different meanings, such as whether Hamlet’s leading questions lead to unreliable or reliable knowledge, and whether it is Polonius or Hamlet who is fooled in the interview. No systematic method of meaning interpretation was in play here. Within a hermeneutic tradition of text interpretation in the humanities, principles have been sought for arriving at valid interpretations of religious, legal and literary texts (see Palmer, 1969). Hermeneutics does not involve any step-by-step method, but is an explication of general principles found useful in a long tradition of interpreting texts. Thus interpretation of a text is characterized by a hermeneutical circle, where the meaning of a text is established through a process in which the meanings of the separate passages are determined by the global meaning of the text as it is anticipated. Re-reading of the single passages may again change the first anticipated global meaning of the text, which again alters the meaning of the single passages, and so on. In principle, such a hermeneutic text interpretation is an infinite process, whereas in practice it ends when a sensible coherent meaning has been arrived at. Interpretations of meaning are sometimes steeped in mistrust of what is said. Hamlet’s interview was thus read as expressing a pervasive distrust of the words and acts of the other players, leading to conversations of ‘per indirections find directions out’. Within a ‘hermeneutics of suspicion’, statements are critically interpreted as meaning something else than what is manifestly said, such as when a psychoanalytic interpreter looks for unconscious forces beneath what is said, or Marxist interpreters look for ideological class interests behind political statements. Implications for interviewing and transcription For deep and critical interpretations of meaning, rich and nuanced descriptions in the interviews are advantageous, as well as critical interpretative questions during the interview. For some types of interpretation, detailed verbatim descriptions may be necessary, such as when critically reading of a pupil’s many denials of competition in Chapter 8. Interview analyses focusing on language A good craftsman is familiar with the material he works with and the tools for working with it. The medium, or the material, with which interviewers work is language. The interview process occurs through speech, and the interview products are presented in words. During the last few decades, qualitative social science researchers have been influenced by the linguistic turn in philosophy, and they have started to use linguistic tools developed in the humanities to analyze their linguistic material. These include linguistic analysis, narrative analysis, conversation analysis, discourse analysis and deconstruction. Linguistic analysis Interviewing is linguistic interaction, and the product of the interview is a language text. A linguistic analysis addresses the characteristic uses of language in an interview, the use of grammar and linguistic forms. A linguistic analysis may thus study an interviewee’s use of an active and a passive voice, of personal and Page 10 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. impersonal pronouns, the temporal and spatial references, the implied speaker and listener positions and the use of metaphors. An example from the grading study may indicate the importance of linguistic form. The analysis did not follow from any linguistic competence of the researcher, but arose as a practical problem by categorizing the pupils’ statements. While most grading behaviours were commonly described in a first-person form, such as ‘I find the grades unfair’, ‘I bluffed the teacher’, a few activities, such as wheedling, were always described in a third-person form as ‘They wheedled’ or ‘One wheedled’. If I had then been more sensitive to the differential use of personal pronouns, we might, in the interviews, have probed more into such vague expressions as ‘one wheedles’ and clarified whether it referred to the speaker or to other pupils. While a method problem when categorizing the statements as referring to the interviewee or to the other pupils, the diverging use of personal pronouns was of importance to the research topic as one of the many indications of the contrasting social acceptance of bluffing versus wheedling among the Danish high school pupils. Attention to the linguistic features of an interview may contribute to both generating and verifying the meaning of statements. While the significance of the different use of grammatical forms such as the above example of personal pronouns may follow from common sense, a linguistically trained reader would immediately look for the linguistic expressions and be able to bring out nuances that may be important for interpreting the meaning of a statement. Arguments in favour of applying the techniques of linguistics as a ‘statistics’ of qualitative research have even been put forward (Jensen, 1989). With more attention to the linguistic medium of interview research, we may perhaps see social researchers use linguistics as consultants when faced with interview texts, corresponding to the commonplace use of statistical consultants when analyzing numbers. Implications for interviewing and transcription Attention to linguistic form may improve the preciseness of interview questions, as suggested in Box 5.4, and further sensitivity in listening to the subjects’ use of language. To carry out systematic linguistic analyses of the interview interaction, detailed verbatim transcription and also linguistic training is necessary. Conversation analysis Conversation analysis is a method for studying talk in interaction. It investigates the structure and the process of linguistic interaction whereby intersubjective understanding is created and maintained. Inspired by ethnomethodology, conversation analysis implies a pragmatic theory of language, it is about what words and sentences do; the meaning of a statement is the role it plays in a specific social practice. Conversation analysis started with studies of telephone conversations by Sacks and co-workers in the 1960s, and has since been used for a wide variety of talk in action, such as doctor-patient interactions, therapy sessions and news interviews. Conversation analysis examines the minute details of talk-in-interaction, which has become widely accessible with the advent of tape recorders. It focuses on the sequencing of talk, in particular upon turn-taking sequences and repair of turn-taking errors. The centre of attention is not the speakers’ intentions in a Page 11 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. statement, but on what a specific speech segment accomplishes. Consequently, the outcome of the conversation analysis of the telephone conversation transcribed in Box 8-1 was: E apparently has called M after having visited her. She provides a series of ‘assessments’ of the occasion, and M’s friends who were present. E’s assessments are relatively intense and produced in a sort of staccato manner. The first two, on the occasion and the friends in general are accepted with Oh-prefaced short utterances, cut-off when E continues … The assessments of Pat are endorsed by M with ‘yeh’, followed by a somewhat lower level assessment. ‘a do:ll?’ with ‘Yeh isn’t she pretty’ and ‘Oh: she’s a beautiful girl.’, with ‘Yeh I think she’s a pretty girl.’ …The ‘work’ that is done with these assessments and receipts can be glossed as ‘showing and receiving gratitude and appreciation, gracefully’. (ten Have, 1999, pp. 4–5) Conversational analysis thus sticks rather close to the verbal interaction of the talkers, forgoing interpretations in depth. The labour-intensive transcription and the minute analyses of the speech sequences rules out conversation analysis as a general method for analysis of large amounts of interview material. Conversation analysis may, however, be relevant for selected significant parts of an interview, and it may also be useful in the training of interviewers, to make them aware of the subtleties of the interaction in the interviews. Implications for transcription Here there are no specific requirements to interviews since any verbal exchange can be made the subject of conversational analysis. As will have appeared from the transcription in Box 8.1, there are, nevertheless, very specific and elaborate requirements for how interviews are to be transcribed in order to be amenable to conversational analysis (see also Rapley, 2007). Narrative analysis A narrative is a story. Narrative analyses focus on the meaning and the linguistic form of texts, they address the temporal and social structures and the plots of interview stories. The narrative structures of stories people tell have been worked out in the humanities, starting with Propp’s analysis of the structures of Russian fairy tales in the 1920s, and followed up decades later by Greimas and Labov. In the structure of a fairy tale, the main subject position may be taken by the prince as the actor, who seeks the object in the form of the princess. On his way, the prince encounters opponents as well as helpers, and after overcoming many obstacles, the prince receives from the king the princess and half his kingdom. Greimas used this structure to work out an actant model pertaining to narrative structures in a variety of genres (see also Gibbs, 2007). The analysis of an interview can take the form of narration, as a continuation of the story told by the interviewee. Narrative analysis focuses on the stories told during an interview and works out their structures and their plots. If no stories are told spontaneously, a coherent narrative may be constructed from the many episodes spread throughout an interview. The analysis may also be a reconstruction of the many tales told by the different subjects into a ‘typical’ narrative as a richer, more condensed and coherent story than the scattered stories of single interviews. As with meaning condensation, narrative analysis will tend to stay within the vernacular. Page 12 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. A narrative sequence from Mishler’s interview with a furniture craftsman-artist dropping out of the woodworking program at school was presented in Box 6.2. We may also note that the interview on learning interior decorating (Table 9.2) had the spontaneous form of a narrative, which Giorgi did not address by his meaning condensation of the interview. Chapter 4 on designing an interview study was introduced by a constructed narrative of successive emotional deterioration, followed by an idealized story of a linear progression through seven stages of an interview inquiry. Implications for interviewing and transcription Interviewing for narratives was described in Chapter 6, where questioning for concrete episodes and the following up of the subjects’ spontaneous stories, elaborating their temporal and social structures and plots, were emphasized. In transcription one may experiment with the textual layout in ways to make the narrative form accessible, such as with the stanzas in Mishler’s craftsman story (see Gibbs, 2007; Rapley, 2007, for more details). Discourse analysis Discourse analysis focuses on how truth effects are created within discourses, which are neither true nor false. Foucault’s (1972) analysis of the power relations of discourses has inspired later forms of discourse analysis. Discourses are discontinuous practices, which cross each other and sometimes touch, while just as often ignoring or excluding each other. In Chapter 7, an interview sequence was presented, which, inspired by discourse analysis, was analyzed as a crossing of swords between the diverging discourses of learning by the interviewee and the electronics pupil. In discourse analysis the talk itself has primacy, the focus is on how the talk is constructed and what the social consequences are of the different discursive presentations of a social situation: Participant’s discourse or social texts are approached in their own right and not as a secondary route ‘beyond’ the text like attitudes, events or cognitive processes. Discourse is treated as a potent, action-oriented medium, not a transparent information channel. Crucial questions for traditional social psychological research thus cease to be relevant. For example, we are not asking whether a sample of people are revealing their ‘genuine’ attitudes to ethnic minorities, or whether fan’s descriptions of what happens on the soccer terraces are ‘accurate’. (Potter and Wetherell, 1987, p. 160) From a discourse-analytic perspective, some common objections to the validity of research interviewing thus dissolve. This concerns the question of authentic personal meanings – ‘How do you know you get to know what the interviewee really means?’ – as well as the objective reality question – ‘How do you know that your interviewee gives a true description of the objective situation?’ A persistent objection concerning the reliability of interviewing has been that different interviewers get different results. If subjects present themselves differently to different interviewers, and also change their opinions during the interchange, then interviews do not produce reliable, objective, knowledge. Page 13 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. These objections may be based upon conceptions of the research topic, such as attitudes or presentation of the self, as expressions of an essential stable core person. In contrast, a discursive understanding treats attitudes and the self as interrelationally constituted. These phenomena may vary in different situations with different interviewers, and interviewing is a sensitive method to investigate the varying social presentations of the self. Thus, according to the differing epistemological conceptions of attitudes and the self – as stable authentic essences, or as socially constituted and more or less fluid ones – the interview appears either as a highly unreliable or as a finely tuned valid method. Implications for interviewing While discourse analysis may be applied to common interviews, such as the one described as discourses ‘crossing swords’, a specific discursive interviewing will focus on variation and diversity, and on the active participation of the interviewer in the discourse, as shown in Chapter 6. The search for real inner meanings and objective presentations of external reality dissolve in wild goose chases. Deconstruction The concept of ‘deconstruction’ was introduced by Derrida as a combination of ‘destruction’ and ‘construction’. Deconstruction involves destructing one understanding of a text and opening it for construction of other understandings. The focus is not on what the person who uses a concept means, but on what the concept says and does not say. It is affiliated with a critical ‘hermeneutics of suspicion’, but in line with conversational and discursive analysis, it does not search for any underlying genuine or stable meaning hidden beneath a text. Meaning is understood in relation to an infinite network of other words in a language. A deconstructive reading tears a text apart, unsettling the concepts it takes for granted; it concentrates on the tensions and breaks off a text, on what a text purports to say and what it comes to say, as well as what is not said in the text, on what is excluded by the use of the text’s concepts. A deconstructive reading reveals the presuppositions and internal hierarchies of a text and lays open the binary oppositions built into modern thought and language, such as true/false, real/unreal, subjective/objective. Deconstruction does not only decompose a text, but also leads to a re-description of the text. A deconstructive reading could, for example, focus on selected interview passages and phrases and work out the meanings expressed, as well as meanings concealed and excluded by the terms chosen. Rather than deconstructing an interview text, I shall here attempt to deconstruct a phrase recurrently evoked in interview literature, also in my own earlier writings – ‘the interview dialogue’. We may start by wondering why the two similar terms ‘interview’ and ‘dialogue’ are often added together, and not uncommonly bolstered with embellishing words, such as ‘authentic’, ‘real’, ‘genuine’, ‘egalitarian’ and ‘trusting’. When ‘dialogue’ is used in current interview research it is seldom in the severe Socratic form, but more commonly as warm caring dialogues. ‘Dialogue’ exists in a binary opposition to ‘monologue’, which today may connote an old-fashioned, authoritarian form of communication. ‘Dialogical interviewing’ can imply a warm empathetic caring, in contrast to alienated and objectifying forms of social research, such as experiments and questionnaires. When the interview is conceived as a dialogue, the researcher Page 14 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. and the subject are then implied as egalitarian partners in a close mutually beneficial personal relation. The expression ‘interview dialogue’ here glosses over the asymmetrical power relationships of the interview interaction, where the interviewer initiates and terminates the interview, poses the questions, and usually retains a monopoly of interpreting the meaning of what the interviewer says (see Box 2.2). We further note that the term ‘dialogue’ is used today in texts from a variety of fields such as management and education, when advocating ‘dialogue between managers and workers’ and ‘dialogical education’ (Kvale, 2006). In these contexts, with obvious power differences and often conflicts, the term ‘dialogue’ may provide an impression of equality and harmonious consensus. We shall conclude this brief deconstruction of the phrase ‘interview dialogue’ by asking whether the term ‘dialogical interviews’ used about research interviews may, corresponding to dialogical management and dialogical education, serve to embellish the power asymmetry and cover up potential conflicts of research interviewers and their interviewees. Implications for interviewing As any kind of text may be made a subject of deconstruction, there are no specific requirements for interviewing. If a deconstruction of interview texts is considered, the interviewer may, however, address the use of key terms in a variety of contexts, from a multiplicity of perspectives, thereby providing a multi-faceted material for deconstruction. Interview analysis as bricolage Many analyses of interviews are conducted without following any specific analytic method. The researchers may then freely change between different techniques and approaches. Bricolage refers to mixed technical discourses where the interpreter moves freely between different analytic techniques. This eclectic form of generating meaning – through a multiplicity of ad hoc methods and conceptual approaches – is a common mode of interview analysis. In contrast to systematic analytic modes such as categorization and conversation analysis, bricolage implies a free interplay of techniques during the analysis. The researcher may here read the interviews through and get an overall impression, then go back to specific interesting passages, perhaps count statements indicating different attitudes to a phenomenon, cast parts of the interview into a narrative, work out metaphors to capture key understandings, attempt to visualize findings in flow diagrams, and so on. Such tactics of meaning generation may, for interviews lacking an overall sense at the first reading, bring out connections and structures significant to a research project. The outcome of this form of meaning generation can be in words, in numbers, in figures and flow charts, and in a combination of these. Box 9.2 Ad hoc techniques of interview analysis Noting patterns, themes (1), seeing plausibility (2) and clustering (3) help the analyst see ‘what goes with what’. Making metaphors(4), like the preceding three tactics, is a way to Page 15 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. achieve more integration among diverse pieces of data. Counting (5) is also a familiar way to see ‘what’s there’. Making contrasts/comparisons (6) is a pervasive tactic that sharpens understanding. Differentiation sometimes is needed, too, as in partitioning variables (7). We also need tactics for seeing things and their relationships more abstractly. These include subsuming particulars under the general (8); factoring (9), an analogue to a familiar quantitative technique; noting relations between variables (10); and finding intervening variables (11). Finally, how can we systematically assemble a coherent understanding of data? The tactics discussed are building a logical chain of evidence (12) and making conceptual/ theoretical coherence (13). Source: Miles and Huberman (1994, pp. 245–6). In line with a bricolage approach, Box 9.2 presents some useful ad hoc tactics for generating meaning in qualitative texts, arranged roughly from the descriptive to the explanatory, and from the concrete to the more conceptual and abstract. The box brings a summary from the book Qualitative Data Analysis by Miles and Huberman (1994), who also outline a variety of analytic techniques. To take one example, metaphor means to understand one kind of thing by means of another, thereby highlighting perhaps new aspects of a kind. A metaphor does not entail a definite conceptual structure to which a statement may be unequivocally categorized as belonging or not belonging. In the present text, ‘making metaphors’, such as the miner and the traveller, seeks to achieve more integration among the diverse conceptions of interview research by portraying and contrasting different understandings of interview knowledge. In my grade study a bricolage of mixed methods was applied to pursue a connection between talkativity and grades, postulated by a pupil (Chapter 1). According to a questionnaire follow-up (Table 4.1), 82 percent of the pupils believed that high grades were often a question of how much one talks in class. When reading through the 30 pupils’ interviews I had been struck by how significantly they varied in length, even though one school hour had been set aside for each interview. Following a hunch, I ranked the interviews according to number of pages, and then correlated the page numbers with the pupils’ grade point averages. The resulting correlation was 0.65, with a chance probability of p < 0.001. There is thus a statistically significant connection between how much the pupils talked during the interviews and their grade point averages. The correlation is open to several interpretations: Do the pupils get high grades because they generally talk a great deal? Or are pupils who get high grades more reflected about grading, and more at ease with talking at length with an interviewer about grades? This example demonstrates how it is possible to use a variety of techniques to investigate a hypothesis of Page 16 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. a connection between grades and talkativity: by discussing the truth value of a pupil’s interview statement in the interview itself, testing the generality of the belief in a questionnaire, and finding potential indirect statistical evidence in the length of interviews about grades. In this bricolage of mixed methods there is no epistemological primacy accorded to any of the methods and techniques, they are different means of investigating a provoking statement about grading. Interview analysis as theoretical reading A researcher may read through his or her interviews again and again, reflect theoretically on specific themes of interest, write out interpretation and not follow any systematic method or combination of techniques. We shall note that in several influential interview studies of the last few decades, leading to new knowledge in their discipline, no specific systematic analytic tools were used to analyse the interviews. This applies to the work of Bellah et al., Hargreaves, Bourdieu et al. and Sennett mentioned earlier. These investigations were based on an extensive and theoretically reflected knowledge of the subject matter, and in the studies by Bourdieu and Bellah also on a confronting Socratic interview form. No elaborate analytic techniques were applied at the theoretically reflected reading of the interviews to develop their rich meanings. This may perhaps suggest that recourse to specific analytic tools becomes less paramount with an extensive and theoretical knowledge of the subject matter of an investigation, and with a theoretically informed interview questioning. In Hargreaves’s study, ‘Changing Teachers, Changing Times (1994), the interviews with 40 teachers and principals had generated almost 1,000 pages of transcripts, which were read and re-read in order to establish a close familiarity with the data. Summary reports of each interview were written according to the key themes. Themes appearing in the text were registered, classified and re-classified on the basis of an active search for confirming and disconfirming evidence in the interviews. Hargreaves interrogated the data in a consciously eclectic approach, drawing in different concepts and theories. He describes his analytic approach as listening to the teachers’ voices telling about their work and comparing their descriptions with claims about their work from literature. ‘Throughout the study, I have attempted to sustain a creative dialogue between different theories and the data, in a quest not to validate any presumed perspective, but simply to understand the problems in their social context, as experienced by teachers’ (Hargreaves, 1994, p. 122). No tables or quantified categorizations of themes are presented in the book. The findings are reported in a continuous interpretative text, with interview passages interspersed, such as the sequence leading to the concept of ‘contrived collegiality’ (Box 1.2), which appeared unexpectedly from reading of the transcripts; ‘In contrived collegiality, collaboration among teachers was compulsory, not voluntary; bounded and fixed in time and space; implementation-rather than development-oriented; and meant to be predictable rather than unpredictable in its outcomes’ (Hargreaves, 1994, p. 208). In the book, interview passages are integrated in theoretically informed reflections on teacher work, drawing on management literature and postmodern analysis of culture. The results are striking descriptions of the Canadian teachers’ work situation in a Page 17 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. postmodern culture, in particular regarding increasing pressures on time and collegiality, descriptions that are well recognizable for teachers in Denmark. Bourdieu et al. (1999) give little explicit textual interpretation of their interviews in The Weight of the World. Although elsewhere writing extensively and theoretically on the situation of the downtrodden in France, in this book Bourdieu mainly lets the many interviews reproduced in the text speak for themselves. However, the reader is aided in several ways. In an Appendix to the book – ‘Understanding’ – the approach to interviewing is described: … it seems to me imperative to make explicit the intentions and procedural principles that we have put into practice in the research project whose findings we present here. The reader will thus be able to reproduce in the reading of the texts the work of both construction and understanding that produced them. (Bourdieu et al., 1999, p. 607) When it is possible to let the interviews speak for themselves to a large extent, this may be because the interviews are preceded by a presentation of the social situation of the interviewees, allowing the reader to interpret their statements in relation to their life situation. Further, much of the analysis was already built into the interviews through a Socratic maieutic method of aiding explanations, such as formulating suggestions for open-ended continuations of interviewee statements, leading to an induced and accompanied self-analysis. Bourdieu also presents some key themes and analyses from an interview, such as the emotional impact upon himself of the interview with the two young men (Box 1.3): I did not have to force myself to share in the feeling, inscribed in every word, every sentence, and more especially in the tone of their voices, their facial expressions or body languages, of the obviousness of this form of collective bad luck that attaches itself, like a fate, to all those that have been put together in those sites of social relegation, where the personal suffering of each is augmented by all the suffering that comes from coexisting and living with so many suffering people together – and, perhaps more importantly, of the destiny effect from belonging to a stigmatized group. (Bourdieu et al., 1999, p. 64) Key points • The analysis of the interviews should be given thought from the beginning of an interview inquiry. The analysis starts when thematizing and designing the study, and the modes of analysis ought to be taken into account during the interviewing and transcription. • The more the analysis is undertaken in the early stages of an interview investigation, the easier and the more qualified the later analysis will be. • Interview analyses focusing on meaning include meaning condensation, meaning categorization and meaning interpretation. • Interview analyses focusing on language include linguistic analysis, conversation analysis, narrative Page 18 of 19 Analyzing Interviews SAGE SAGE Research Methods 2007 SAGE Publications, Ltd. All Rights Reserved. analysis, discursive analysis and deconstruction. • Analysis as bricolage and as theoretical reflection goes beyond following specific techniques or approaches to interview analysis and draws in a variety of techniques and theoretical concepts. Further reading The following books will help you to go deeper into issues of analyzing interviews: Gibbs, G.R.(2007) Analyzing Qualitative Data (Book 6 of The SAGE Qualitative Research Kit). London: Sage. Miles, M.B. and Huberman, A.M.(1994) Qualitative Data Analysis. Thousand Oaks, CA: Sage. Mishler, E.G.(1986) Research Interviewing – Context and Narrative. Cambridge, MA: Harvard University Press. Potter, J. and Wetherell, M.(1987) Discourse and Social Psychology.London: Sage. Rapley, T.(2007) Doing Conversation, Discourse and Document Analysis (Book 7 of The SAGE Qualitative Research Kit). London: Sa

Do you have a similar assignment and would want someone to complete it for you? Click on the ORDER NOW option to get instant services at We assure you of a well written and plagiarism free papers delivered within your specified deadline.