Thanks Pat, that's the most sense I've heard so far and yes please I would
like more info about these techniques.
Georgina.
At 10:50 16/11/98 +1100, you wrote:
>Georgina,
>At the risk of earning the ire of those who only ever use specialised
>CAQDAS, my response to this would be simply to use Excel. Make a column for
>each 'theme' or issue of concern, allow a row for each interview. With
>interview in hand, work across the row, putting a summary comment or (very)
>brief quote, or even a summary 'rating' under each topic/issue. Include
>some columns for variables like who is responding etc. Then, to analyse,
>sort your data based on some of those variables, then attempt to summarise
>for subgroups on each topic of interest - you will end up with some Miles
>and Huberman style matrices. Try different sorts, or sorts within sorts, to
>see if the patterns are clearer.
>If you then decide you still need to look at your data using, say, NUD*IST,
>it is possible, using the merge functions on your word processor to prepare
>your documents, and command file searches in NUD*IST, to import from your
>spreadsheet and 'automatically' code each segment of data for the topic it
>was under and for what kind of person said it (contact me privately if you
>want more detail on these techniques).
>We used the technique above (just Excel) for analysing transcripts from
>heads of departments in universities, with regard to the opportunities
>presented for early career academics in their departments (the most boring
>transcripts I ever hope to analyse!). My daughter also used the same sort
>of technique - combining Excel and NUD*IST - for an ethnographic study of
>quality control in shearing sheds (i.e. for dealing with structured field
>notes). Having imported and auto-coded the notes from her spreadsheet, she
>was able to look at the relationships between different segments of
>information for each of her 12 sheds in some detail. It is reductionist,
>but I suspect in a policy environment your work is both structured and in
>need of a way of reducing the volume of info derived from a qual study. At
>least you can always get back to the original transcripts if you need to
>pursue a particular issue in more detail.
>Pat Bazeley.
>
>
>At 03:35 PM 13/11/98 +0000, you wrote:
>>Hi, does anybody have any experience of doing policy research using CAQDAS?
>>previously I have used straightforward interview write-ups and I am now
>>considering either:
>>
>>a) the manual framework approach which includes writing 'reports' on the
>>transcript text
>>b) coding transcripts directly in NU-DIST
>>c) combining the two methods ie: putting framework 'reports' of the
>>transcripts into NU-DIST for comparison
>>
>>My problem is that the interviews, which are with small business owners,
>>managing directors or technical managers, are really very straightforward
>>ie: not many hidden themes.
>>
>>Any comments or other suggestions or references that might be useful?
>>
>>Thanks, CAQDASERS,
>>Georgina
>
>Dr Patricia Bazeley
>Head, Research Development Unit
>University of Western Sydney Macarthur
>PO Box 555 Campbelltown 2560
>Australia
>ph: +61 2 4620 3268
>fax: +61 2 4627 2406
>email: [log in to unmask]
>btwebsh.macarthur.uws.edu.au/patB/
>
>
****************************************************************************
******
Dr Georgina Holt
Centre for Food Economics Research
Dept. of Agricultural and Food Economics
The University of Reading
PO Box 237
Reading RG6 6AR
***************************************************
Tel:0118 987 5123 ext.4190
Fax:0118 975 6467
email:[log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|