stages of their fieldwork by the growing ‘pile’ of recordings and notes waiting to be transcribed. The only real advice here, albeit hard to follow, is, if you can’t pay someone to do it for you, keep transcribing ‘little and often’.
Transcription, especially of interviews, is a change of medium and that introduces issues of accuracy, fidelity and interpretation. Kvale (1988, p. 97) warns us to ‘beware of transcripts’. There are, he suggests, dangers, when moving from the spoken context of an interview to the typed transcript, such as superficial coding, decontextualization, missing what came before and after the respondent’s account, and missing what the larger conversation was about. As we shall see later, this change of medium is associated with certain kinds of errors that researchers must watch out for. One response here is to go back to the recording to check your interpretations in the transcript. You may find that hearing the voice makes the meaning clearer and even suggests different interpretations. Furthermore, most transcripts only capture the spoken aspects of the interview and miss out the setting, context, body language and general ‘feel’ of the session. Mishler (1991) suggests a parallel between a transcript and a photograph. A photograph is one, frozen, framed, printed and edited version of reality. The same is true of a transcript. The issue is not whether the transcript is, in a final sense, accurate, but rather whether it represents a good, careful attempt to capture some aspects of the interview. There is always an issue of how to convert speech into written text. Very few people speak in grammatical prose, so the researcher needs to decide how much of what is in the recording needs to be transcribed. As we shall see later there are several options here, though we have to recognize that the transcript will never be completely accurate.
A similar point can be made about the move from handwritten notes taken during interviews or during fieldwork. Transcription here usually involves a process of ‘writing up’ the notes. This is a creative activity and not just a mechanical reproduction. It involves expressing the notes as ideas, observations of certain kinds and so on, and represents the start of data analysis as well. I will discuss these issues in more detail in the next chapter.
Reasons for transcribing
It is not necessary to transcribe all or even any of the information you have collected in your project in order to analyze it. Some levels and forms of analysis can be done quite productively without any copy of the interviews, texts or observations you have collected or recorded. In fact, some researchers advocate analyzing directly from an audio or video recording. Of course, this will involve taking notes on the content you hear or see and probably going through the recordings several times. Doing it this way you are more likely to focus on the bigger picture and not get bogged down in the details of what people have said. This is fine for some types of analysis, such as policy research and evaluation research, where you are working with very clear theories about the situation and therefore have a good idea what you are looking for and can afford to be selective with the data. But for most approaches, and especially for discourse and conversation analysis, a detailed transcript is a necessity. It forces you to listen carefully to what was recorded and to look carefully at your notes, and it provides you with an easily readable version that can be copied as many times as necessary. Having a transcript also makes it easier to work in a team, where tasks have to be shared and there has to be good agreement about the interpretation of the data. A typescript means everyone can read the texts and everyone can have a copy.
Strategies for transcribing interviews
There are various strategies you can adopt when transcribing. You could, for example, only transcribe parts of the recording. For the rest, you could just take notes and use those for coding and analysis, or even code directly from the recording or handwritten notes. In some cases you may find that your memory of an interview or your research diary tells you that at certain points the respondent went off topic and so these parts can be ignored. Such an approach will clearly be quicker, and might also allow you to focus on the larger themes and not get bogged down in the particular words. But there are several drawbacks. You may find that the parts you have transcribed lose their context and you find it harder to interpret what they really mean. Moreover, the ideas you have at the start of analysis, which might lead you to decide what parts need transcribing, may well be different from those you develop later in the study.
Names
The convention is to put the name of the person speaking, in capitals, at the start of each time they speak (e.g. each answer to the interviewer’s questions or each turn in a focus group discussion). Putting the name in capitals makes it stand out on the page but also means you can use case-sensitive search and find in computer software to look for only what the interviewee said or just when their name is used by others elsewhere in the interviews. This is particularly useful when analyzing focus group discussions. Use the name that is easiest for you to remember the interviewees by. This is usually their first name. Then either type a colon and a tab before the actual text, or start the text on a new line. If you are dealing with a large number of respondents then you might opt to indicate the person’s surname with their first name and surname (or first letter of surname), as in ‘MARY C:’ to distinguish them from others with the same first name. Indicate the interviewer’s speech in the same way. Use ‘I:’ or ‘IV:’ or ‘INT:’ at the start, or if you have several interviewers and you want to distinguish them in the transcripts, then use ‘I-JOHN:’, ‘I-KATE:’, etc. Make sure that all names are spelled correctly and consistently. This means you can use search and find in your word processor to anonymize the text and find all the speech by the same person in CAQDAS programs.
Anonymization
As you will eventually be quoting from your transcripts in your write up of the research and you might even be depositing the data in a public archive so other researchers have access to it you will need to consider how you will ensure confidentiality. Do this by anonymizing the names of people and places to make it safe for participants (if their activities are illegal or illicit) and safe for the researcher (e.g. if you have been investigating covert operations or paramilitary groups). It is easiest to produce an anonymized copy immediately after transcription. However, you may find it is best to do your analysis using the unanonymized version, as familiarity with the real names and places can make it easier. However, many researchers report that if you anonymize early on in the analysis, you soon become just as familiar with the anonymized names.
Create a listing, in a separate file, kept somewhere safe, of all the names – people, places, organizations, companies, products – that you have changed and what you have substituted for them. Use search and find in your word processor to find each name and substitute it with the anonymized version. Make sure you search for both normal text versions of respondent’s names (‘Mary’) if these appear in other respondent’s talk as well as capitalized versions (‘MARY:’) if you have used that to identify speakers. It is usually best to use pseudonyms rather than crude blanks, asterisks, code numbers and so on. You will still need to read the transcript carefully to ensure that more subtle, but obvious clues to a person, place or institution are not evident. If you are going to deposit your data into a data archive, then remember you will need to retain and deposit the original, unanonymized versions alongside the accessible anonymized versions.
Level of transcription
I noted above that the act of transcription is a change of medium and therefore necessarily involves a transformation of the data. There are varying degrees to which you can capture what is in the sound recording (or your handwritten notes) and you need to decide what is appropriate for the purposes of your study. Sometimes just a draft version of what is said is sufficient. This is often the case in policy, organization and evaluation research, where the salient factual content of what people
