170 from Text to Theory influence is increased if the researcher is present in the research situation for technical reasons- The greater the effort in videotaping and the more comprehensive the insight it permits into the everyday life under study, the greater may be the possible scepticism and reservations on the- part of participants in the study. This makes the integration of the recording procedure in the daily life under study more complicated. Scepticism about the naturalness of recordings Correspondingly, thoughtful reflections on the use of recording technology in qualitative research can be found. These forms of recording have replaced the interviewers' or observers' notes - which were the dominant medium in earlier times. For Hopf they provide 'increased options for an intersibjecrive assessment of interpretations ... for taking into account interviewer and observer effects in the interpretation .. . and for theoretical flexibility' compared with 'the necessarily more selective memory protocols' (Hopf 1985, pp. 93-4). This new. flexibility leads 'to a new type of "qualitative data hoarding* owing to the delays in decisions about research questions and theoretical assumptions which are now possible'. New questions concerning research ethics, changes in the studied situations caused by the form of recording1 and a loss of anonymity for the interviewees {see Bergold and Flick 1987, pp. 13-15) are linked to this. The ambivalence Hopf (1985) expresses against the new options for recording qualitative data suggests that it is important to treat this point not as a problem of technical detail but rather in the sense of a detailed 'qualitative technology assessment'; and also, in the considerations about the appropriate method for documentation, to include 'out-of-date' alternatives which were displaced by the new technologies. FIELD NOTES The classic medium for documentation in qualitative research has been the researcher's notes (Lofland and Lofland 19&4; Sanjek 1990). The notes taken in interviews should contain the essentials of the interviewee's answers and information about the proceeding of the interview. The participant observer repeatedly interrupts his or her participation to note important observations, as the description in Box 14.1 of the classic documentation technique, its problems and the chosen solutions to them makes dear. Lofland and Lofland (1984) formulate as a general rule that such notes should be made as immediately as possible. The withdrawal necessary for this may introduce a certain artificiality in the relation to interaction partners in the field. Especially in action research, when the researcher takes part in tr# events in the field and does not merely observe them, it is Documentation of Data ' _' 171 Box»i^-1 Field notes in practice Our usual practice was to spend limited periods of time in the field, perhaps two or three hours. When we could appropriately leave the field, we headed immediately for a typewriter or Dictaphone. If leaving was impossible, we took brief memory-refreshing notes whenever lulls occurred and recorded them fully as soon thereafter as possible. The recording of field holes presented a number of problems involving discrimination among events seen and heard, as well as an interviewer's impressions or interpretations. As professionals, all or us were mindful of the pitfalls attending recall and the all-toc-easy blurring of (act and fancy. We attempted therefore to make these discriminations clearly, either by staring them unmistakably or by developing a notaboňal system for ensuring them. Verbal material recorded within quotations signified exact recall; verbal material tvithin apostrophes indicated a lesser degree of certainty or paraphrasing; and verbal material with iio markings meant reasonable recall but not quotation. Finally, the interviewer's impressions or inferences could be separated from actual observations by the use of single or double parentheses. Although this notational system was much used, none of us was constrained always to use it. Source: Strauss et al. 1964, pp. 28-9 additionally difficult to maintain this freedom for the researcher (see Decker 1979). An alternative is to note impressions after ending the individual field contact. Lofland and Lofland (1984, p. 64) recommend that researchers use a 'cloistered rigor' in following the commandment to make notes immediately after the field contact, and furthermore that researchers estimate the same amount of time for carefully noting the observations as for spending on the observation itself. It should be ensured that (maybe much) later a distinction can still be made between what has been observed and what has been condensed by the observer in his or her interpretation or summary of events (see Chapter 18 on procedural reliability of protocols). On the other hand, researchers may develop a personal style of writing notes after a while and with increasing experience. All in all, the production of reality in texts starts with the taking of field notes. Tliis production is essentially marked by the researcher's selective perception and presentation. This selectivity concerns not only the aspects that are left out but above all those which find their way into the notes. It is only the notation that raises an occurrence out of its everyday course and transitoriness and makes it into an event to which the researcher, interpreter and reader can turn their attention repeatedly. One way of reducing or at least qualifying this selectivity of the documentation is to complement 172 From Text to Theory the notes by diaries or day protocols wntten by the subjects under study in parallel with the researcher's note taking. Thus, their subjective view» may be included in the data and become accessible to analysis, Such documents from the subject's point of view can be analysed and contrasted with the researcher's notes. Another way b to add photos, drawings, maps and other visual material to the notes. A third possibility is to use an electronic notebook, a dictating machine or similar for recording the notes. Correspondingly, Spndley 11980, pp. 69-72) suggests four form* of field notes for documentation: • condensed accounts in single words, sentences, quotations from conversations etc.; • an expanded account oi the impressions from interview* and field • n íleldwork journal, which like a diary 'will contain .. . experience». Ideu, fears, mistakes, confusions, breakthroughs, and problems that arise during fieldwork' (1980, p. 71); • notes about analyses and interpretations, which start immediately alter the field contacts and extend until fúúslúng the study RESEARCH DIARY Especially if more thin one researcher is involved, there is a need for documentation of and reflection on the ongoing research process In order to increase the comparability of the empirical proceedings and focuses m the individual notes. One method of documentation is lo use continually updated research diaries written by all participants. These should document the process of approaching a field, and the experiences and problem* in the contact with the lield or with interviewees and in applying the methods. Important facts and matters of minor importance or lost facts in the interpretation, generalization, assessment or presentation of the results, seen from the perspectives of the individual researcher, should also be incorporated. Comparing »uch documentations and the dllferent views expressed in diem makes the research process more intcrsubjective and explicit. Furthermore, they may be used as 'memos' in the sense of Strauss (1987, in particular Chapter 5) for developing a grounded theory. Strauss recommends making memos during the whole research process which will contribute to the process of building a theory. Documentation of this kind is not only an end in itself or additional knowledge but also serves m the reflection on the research process. Several methods have been outlined for 'catching' the interesting «vents and processes, statements and pnxvedings. In the noting of interventions in Documentation oi Data 173 the everyday life under Sjtydv. the researcher should be led in his or her decisions by die following rule tf economy; to record only as much as is definitely necessary (or answering the research question. He or she should avoid any 'technical presence' in the situation of the data collection that is not absolutely necessary for his or her theoretical interests. Reducing the presence of recording equipment, and informing the research partners as much as possible about the sense and purpose of the chosen form of recording, make it more likely thai the researcher will 'catch' truly everyday behaviour in natural situations. In (he case of research questions where 'out-of-date' forms of documentation such as preparing a protocol or answers and observations are sufficient, it is highly recommended to use these forms. But these protocol» should be made as immediate and comprehensive as possible in order mainly to record impressions of the field and resulting questions. • DO! ! Ml VľATION SHEETS For interviews, it has proved to be helprV. to i-r I. i i. i documenting the context and the situation of data collection (for this see Hick 1996; Wit/.'l 1985). Wh.it information Ihcy should concretely include depends on the design of the study, eg. if several interviewers are involved or if interviews are conducted at changing locations, which supposedly might have influenced the interview. In addition, the research questions determine what should concretely be noted on these sheets. The example in Box 14.2 comes from die study of technological change in everyday life (Flick 1996), in which several intervl i'Wi'ľi Kiľ.diii'led interview« w;lh pt. r. .'.;un.il.s ir different work situations on the influences of technology on childhood, children's education in one's own family or in general, and so on Therefore, the documentation sheet needed to contain explicit additional contextual information. TRANSCRIPTION If data have been recorded using technical media, their transcription is a necessary step on the way to their interpretation. Different transcription systems are available which vary in their degree of exactness (for overview see FJ\lichand Switalla 1976; O'Connell and Kowall 1995). A standard has not yet been established. In language analyses the interest often focuses on attaining the maximum exactness in classifying and presenting statements, breaks etc. Here also the question of the appropriateness of the procedure may be asked. Not only does this contribute to the natural science ideals of 174 From Text to Theory Bon 14.2 fxample of a doťumentatlon ihn«! Information .»Nm! llw Inlrrvk-w and Ihc lnterviewi-e •IUI' I )ata of Ihe interview: ................... Place oť the interview: ................... Duration of Ihe interview; ................... Interviewer ................... Identifier for ihe interviewee: ................... Gender or the intrrvtewiv ....... Age of the MflfVtM ......... . . Probation of thľ Interviewet! ................... Working in tins [thIi- mou since: ................. I'rofvssaonal field ................... Raised (countryside/city): ................... Number of children ................... Ay,*.- nf the «hildren ................ I .«•!<.!■-? o) ihr ■ hildren ........ ..... Peculiarities of ihe Lnitrvlow! ................... Stwce: Rick 199* exactness in measurement wteaking into intcrprrtive social science through ihe back door, bul abo the formulation of rules for transcription tempts one into some kind of fetishism that no longer has a reasonable relation to the question and the product» of Ihe research. Where linguistic and conversation analytic studies focus on the organization of language, tlüs kind of .■..i.IU.--S may V justified In more j-y.vhi'V.1.:.-ii H MX ICtOgksÚ question! however, where linguist »-..lunge is a medium for studying certain tontmls, exaggerated standards -1 exactness in trans, nphons an* justified Documentation of Data •' 175 ordv to SMtptiOml CHSCS> Kf*vm*t Don reasonable to transcribe only u , much and only as exactly as is required by the research question (BtnuSB 1987). PbH, .i" »verexact transcription of data absorbs time and energy wluili OOUld N invested more reasonably in their interpret a I ion. !**iond, Uie message and Ihe meaning of what was iranscribed are sometimes concealed rather than revealed in the dilleren nation of the transcription and the resulting obscurity oť the protocols produced. Thus Bruce (1992, p. 145; quoted in O'Conneli and Kowall 1995. p. 96) holds; IV following very general criteria can be used as a starting petal In Ihe evaluation ti( a Iranschption system for -|»ifc.-n discourse: manageability (for the iraiucnoer), readability, leamabuity, and mtcrpn-übiliiy (fur ihr «nalyal ■id !>•! <)•>' ><'innuirt) It.is rr.uonaMr to think thai a transcription -v-trm ■hmild bo rany to write,..easy to ie.«l. easy to learn, and easy to uranii Beyond the > l.-ar rules of how lo transcribe statements, tum taking» breaks, ends of sentences etc., a second check of Ihe transcript against the recording and the anonynuzarion of dala (names, «real and temporal references) are central feature* of die procedure of Iranscriplion. Transcription in conversation UUÜysil (ato Chapter 16) has often baen Ih«-* model for Irans* rip I ions ::i si« ial .. Itnct, Drew (i995, p. 78) fives i 'a^ossary of transcripium i on venbora', which may be used after the criteria with regard to the í»"*mh h question mentioned above have been applied (Box 143). Box 14.3 Transcription conventions Overlapping speech: ihe precise point al which one person begins speaking whilst Ihe other is still talking, or at which both ''"gin »peaking simultaneously, resulting: in overlapping speech. (0.2) Pauset: wiihin and between speaker turns, in sto i ■! 'Aw ' I * I ended sounds miiiix) stretches shown by colons. W ptoporOon lo tlie length ol Ihe sirech. Word Underlining show, stress or emphasis, fishl A hyphen indicalfi lh.it a woid/sound is broken ofl ' hhhh' Audible intakes of breath are transenbed as '.lihhh' (Ihe number of h's w proportional to the length of the breath). WORD; Increase in amplitude is shown by capital letters, (words. . .): Parentheses bound uncertain Iranscription, including the transcriber's 'best gucW - mte» adapted bom Drew 1995, p 9V 176 from Text (o Theory REALITY AS TEXT: TEXT AS NEW REALITY Recording the data, making additional notes and the transcription of the recordings transform the interesting realities into text, and tales from the field result (van Maanen 1988). At the very least, the documentation of processes and the transcription of statements lead to a differenl version of events. Each form of documentation leads to a specific organization of what is documented. Every transcription of social realities is subject lo technical and textual strucrurabons and limitations, which make accessible what was transcribed in a specific way. The documentation detaches the evenis from their transience (Bergmann 1985). The researcher's personal style of noting things makes the field a presented field; the degree of exactness of the transcription dissolves the gcstalt of the events inlo a multitude of specific details' The consequence of the following process of interpretation is that 'Reality only presents itself to the scientist in substantiated form, as text -or in technical terms - as protocol- Beyond texts, science has lost its rights, because a scientific statement can only be formulated when and in so far as events have found a deposit or left a trace and these again have undergone an interpretation' (Garz and Kraimer 1994, p. 8). This substantiation of reality in the form of texts is valid in two respects: as a process which opens access to a field and, as a result of this process, as a reconstruction of the reality which has been texlualized. The construction of .i new reality in the text lias already begun at the level of the field notes and at the level of the transcript and (his is the only (version of) reality available to the researcher during his or her following interpretations. This should be taken into account in the more or less meticulous handling of the text which is suggested by each method of interpretation. The more or less comprehensive recording of the case, the documentation of the context of origination and the transcription organizes the material in a specific way. The epistemological principle of understanding may be realized by being able to analyse the presentations or the proceeding of situations from the inside as far as possible. The documentation therefore has to be exact enough to reveal structures in those materials and it has to permit approaches from different perspectives. The organization of the data has the main aim of documenting the case in its specificity and structure. This allows the researcher to reconstruct it in its gestalt and to analyse and break it down for its structure - the rules according to which it functions, the meaning underlying it, the parts that characterize it- Texts produced in this way construct the studied reality in a specific way and make it accessible as empirical material for interpretive procedures. NOTE 1 According to Ueigmann, 'an audiovisual recording of a social event Li by no means the purely de-scriptivc representation which it may seem lo be at first. Ovingfo its time-manipulative structure it has rather a con-struclivc moment in Documentation of Data 177 if (1985, p- 317). Thus after Stflrecording, a conversation can be cut off from lis unique, self-contained temporal course and monitored over and over again. Then it may be dissected into specific components fe-R. participants' non-verbal signals) ut a way which goes beyond the everyday perceptions of the participants. Bui tliis not only allows new forms ol knowledge but also constructs a new version of the events. From a certain moment, the perception of thcM ■•vents is no longer determined by their original or natural occurrence but b>* their artificially detailed display. FURTHER RCADÍNG The second text gives .an overview and some critical reflections 1 about transcription, a'pd the"others give an orientation for how to Work with field notes? Lofland, J., Lofland, LH. (1984)/Analyzing Social Settings (2nd edn). Belmont, CA: Wadswdrth; > O'Contiell.'p., Köwall,*S. (1995), Basic Principles of Transcription. In: Smith;;j;Ä'Harré,*R:, Langenhove, L.v. (eds). Rethinking Methods in Psychology, pp. 93-104. London': Sage. Sanjčk^R.;,(ed.) (1990),'.Fieldnotcs: The Making of Anthropology. Albaňy;-NY-'StátotUniverslty of New York Press, -