Transcription

1
Transcription Computer Assisted Qualitative
Data Analysis (CAQDAS)

• Stephen Gibson (York St John University)
• Sally Wiggins (University of Strathclyde)
• Colm Crowley (University of Greenwich)

2
Session outline

• Transcription (40 mins)
• CAQDAS (20 mins)

3
Transcription

• Transcription is theory (cf. Ochs, 1979)
• Different transcription conventions are
  appropriate for different analytic perspectives
• Transcription always takes longer than you think!

4
Questions from students

• Do I have to transcribe everything?
• How much detail do I need to include?
• I say um and er a lot do these matter?
• Do I have to transcribe my interview questions as
  well as the interviewees answers?
• I didnt ask the questions in exactly the way
  Ive written them on the interview schedule,
  which version should I use for the transcript?
• I know what they meant to say here, but theyve
  not said it should I tidy it up?
• How do I transcribe body language, gesture etc?

5
Answers?

• For many of these questions, the answers depend
  on what youre doing with the transcript
• General advice to students have a look at
  studies using your chosen method of analysis and
  see what they do

6
Transcription in the 4 approaches covered in the
workshop a (very) general comparison

• Interpretative (IPA GT)
• Focus on content meaning
• Easily readable to the uninitiated (similar to
  play script)
• Orthographic representation
• Standard spelling punctuation
• Limited transcription of paralinguistic (e.g.
  intonation, elongation) extralinguistic (e.g.
  gesture, body language) features

• Discursive (CA DA/DP)
• Focus on active language use
• Difficult to read at first
• More likely to use some form of non-orthographic
  representation (typically Jeffersonian)
• Lots of paralinguistic detail
• Limited extralinguistic features (see Charles
  Antakis CA web tutorial)

7
Teaching transcription

• Little substitute for experience
• Students typically underestimate the amount of
  time needed
• Smith Osborn (2003) 5-8 hours transcription
  time per hour of audio material (IPA)
• General guidelines for minimum amount of audio
  data for undergraduate dissertations
• IPA GT 5 hours
• DA 3-4 hours
• CA ( CA-influenced DA) 1-2 hours
• Gough, B., Lawton, R., Madill, A., Stratton, P.
  (2003). Guidelines for the supervision of
  undergraduate qualitative research in psychology.
  York LTSN Psychology.
• http//www.psychology.heacademy.ac.uk/docs/pdf/p20
  030626_ltsn_report_3_text.pdf

8
Exercises with published data

• Introducing the importance of careful
  transcription
• Comparison of two versions of the same transcript
  from published papers
• Ready-made example (e.g. McCrone et al 1998 vs
  Bechhofer et al 1999)

9
Exercises with published data

• Introducing the importance of careful
  transcription
• Comparison of two versions of the same transcript
  from published papers
• Ready-made example (e.g. McCrone et al 1998 vs
  Bechhofer et al 1999)
• Online materials (see URLs on handout)
• Particularly useful for Jeffersonian
  transcription
• Papers including different transcripts of the
  same recording (see refs on handout)

10
Exercises with primary data (1)

• Students transcribe the same piece of (short)
  audio/video material using different conventions
• Demonstrates differences btwn 2 transcripts based
  on the same audio
• Also the different length of time it takes to
  complete them

11
Exercises with primary data (2)

• Two students transcribe the same bit of
  audio/video using the same transcription
  conventions
• Likely to produce variations btwn transcripts
  esp. if material selected carefully
• Useful for highlighting the role of the
  transcriber in producing the transcript (i.e. the
  transcript is not an objective record of some bit
  of reality)

12
Exercises with primary data (3)

• Students record short stretches of
  conversation/interview etc transcribe it
• Then swap recording with someone else
  transcribe
• Compare transcripts
• Highlights the role of insider knowledge (both
  pos neg) in producing a transcript
• Transcribing interactions you have been involved
  in vs ones you havent

13
Practical issues

• Digital recording better quality than tape
  easier to transcribe (many students have
  phones/cameras that allow the recording of
  digital audio video files)
• Regular breaks
• Transcription machines
• Cost implications (c. 70 for a digital
  transcription machine)

14
Bird, C. M. (2005). How I stopped dreading and
learned to love transcription. Qualitative
Inquiry, 11 (2), 226-248

• To transcribe my first cassette tape, I used a
  tabletop-sized, dual speaker, single-cassette
  player with a pause button but no counter. Not
  until later in the story of my transcription
  experience did I learn what a transcription
  machine was and how to use it. Thus, in the
  beginning, expending well more than 40 hours to
  transcribe a 1-hour tape did much to build my
  dread of transcription.
• After I had complained to my fellow researchers,
  one of them finally told me about transcription
  machines, how they worked, and where I could
  borrow one. (p. 233)

15
Ethical issues for students using transcripts

• Anonymity
• Not only of the speaker, but of people, places,
  institutions that are referred to.
• Removal of info which may potentially identify
  them
• Uses of the transcript
• Participant approval of transcript?
• Who will see the transcript?
• Will it be archived?

16

• transcription is theory laden the choices that
  researchers make about transcription enact the
  theories they hold and constrain the
  interpretations they can draw from their data
• Lapadat, J. C. Lindsay, A. C. (1999).
  Transcription in research and practice From
  standardization of technique to interpretive
  positionings. Qualitative Inquiry, 5 (1), 64-86.

17
Some more links

• http//onlineqda.hud.ac.uk/Intro_QDA/preparing_dat
  a.php
• general advice on transcription
• http//onlineqda.hud.ac.uk/resources.phpT
• Transcription refs
• http//onlineqda.hud.ac.uk/Step_by_step_software/C
  AQDAS_data_prep/index.php
• Preparing transcripts for CAQDAS packages

18
Qualitative data analysis software

• Whats available?
• In the most recent review of qualitative
  software, Lewins and Silver (2007) say its
  impossible to answer the question Which is the
  best? of the many software packages now
  available. They comparatively review 3 leading
  packages in depth (Atlas.ti5, MAXqda2 and NVivo7)
  and 4 others more briefly.
• In many departments, it tends to be a matter of
  working with whatever there is
• In this session we will take a brief look at
  NVivo7, as an example

19
Qualitative data analysis software

• Misconceptions include
• Analysis is done by the software rather than the
  researcher (can be a hope or a critique!)
• Features of the software constrain analysis (e.g.
  coding unavoidable, hierarchical classifications
  unavoidable)
• Software offers no advantage over manual
  methods
• The researcher is more distant from the data (see
  Crowley, C., Harré, R., Tagg, C. (2002).
  Qualitative research and computing
  methodological issues and practices in using QSR
  NVivo and NUDIST. International Journal of
  Social Research Methodology Theory Practice,
  5(3) 193-197.)

20
Qualitative data analysis software

• Questions for undergraduate settings
• Would qualitative software help or hinder
  students at this level?
• How widely is qualitative software used at
  undergraduate level?
• How can it be featured? Some examples
• - Demonstration in lecture
• - Hands-on taster lab session with data
  provided for analysis and clearly prescribed
  steps
• -Tutorials for 3rd year project students using
  their own data

21
Qualitative data analysis software

• Resources
• For colleagues comparatively assessing available
  packages Lewins, A. Silver, C. (2007). Using
  software in qualitative research A step by step
  guide. London Sage.
• Token chapters on qualitative software in
  otherwise good books on qualitative methods can
  be very misleading Often written by authors
  themselves inexperienced with the software, who
  pass on misconceptions from similarly uninformed
  authors, and/or descriptions of packages
  functions that are very outdated (even at the
  time of writing).

22
Qualitative data analysis software

• Resources
• While reading given here is the most up-to-date
  for 2007, continuing web updates are crucial.
• For a reliable overview of training, advice and
  other resources on a variety of software
• http//caqdas.soc.surrey.ac.uk
• For a user-friendly online resource on
  qualitative analysis with expert contributions on
  software
• http//onlineqda.hud.ac.uk
• For resource material (including teaching
  resource material) on NVivo from its makers
• www.qsrinternational.com

23
Qualitative data analysis software

• Resources
• For developing skills in using NVivo7 in an
  actual research project (applying your chosen
  methodology) Bazeley, P. (2007). Qualitative
  data analysis with NVivo. London Sage.
• There are no comparable books for the other
  qualitative packages (other than their training
  manuals). Books re earlier versions of NVivo are
  inadequate for the fundamentally redesigned
  current version 7.
• For a methodology conference and training
  workshops on using NVivo software, Durham 2008 (
  some papers from past conferences)
• www.qual-strategies.org