Karen - just finished a 15 month project that used PLA techniques to
identify women's concerns about 6 different policy issues in Ireland. It
was double complicated in that not only did it mix qual and quant as PLA
often does, but we were 'scaling up' local level pla projects to attempt to
provide a national picture.
I was taken on to do the analysis of these projects - as my first job after
graduating !!!! eeek eeek - the other members of this list will attest to
the frequent cries for help from my direction. Anyway, just as PLA/PRA etc
use triangulation in their methods, so did I in the analysis, deciding that
this was the best way to reflect the unique combination of techniques used
in PLA. Here's an example;
Pie charts - pie charts in PLA already come 'part analysed' in that the
research group have already decided on the categories and the percentages
before you come to look at them so I used Microsoft Access which is flexible
enough to let you count words as well as amounts. It's all set up for you
so you don't have to spend ages setting something up in NUDIST or suchlike.
Questionnaires - SPSS which can be linked to NUDIST or NVivo or so on -
simple descriptive statistics
Observation notes and interviews - NUDIST
Matrices - these are difficult for two reasons a) because they're often done
'wrong' or not according to the way we had envisaged them - talk to Bernardo
Turnbull (hi Bernardo!) about strategic compliance in participatory
research.
Seasonal Calendars - keyed into SPSS or Excel whichever you prefer and
converted into line graphs which illustrate the peaks and troughs over the
course of a life.
My guiding philosophy would be to let the techniques dictate the approach to
analysis (including choice of software) rather than the other way around.
However on a general note - quant and qual methods are combined quite often
- it's not that difficult if it is the appropriate thing to do! For
example, the findings of a qual. study can be used to inform the design of a
questionnaire, or results from a survey may throw up questions that can be
answered through a qualitative study to find out the 'why', and so on....
Sarah Delaney
Research Officer
Health Services Research Centre
Department of Psychology
Royal College of Surgeons in Ireland
The Mercer Building
Mercer Street Lower
Dublin 2
00-353-1-4022121
[log in to unmask]
> ----------
> From: Karen Hampson
> Reply To: qual-software
> Sent: Tuesday, June 19, 2001 2:30 pm
> To: [log in to unmask]
> Subject: Qual & quant
>
> Has anyone used a CAQDAS package and tried to integrate outputs into
> quantitative formats or to further quantitative research?
>
> I have been carrying out surveys with farmers in southern Africa, using
> participatory and sometimes visual (PRA/PLA) research techniques, and
> using Nudist to deal with all the qualitative information. We were trying
> to get some 'quantities' out of the data, in terms of ordinal numbers to
> fit into a computer simulation of crop growth, and also to identify
> preferences for farmers to do their own trials eg 90% of farmers want to
> do trials on fertiliser, so we plan for fertiliser trials.
>
> I wondered if anyone has experience of the 'blending' of qualitative and
> quantitative methods - in terms of research, analysis, presentation of
> results and the use of results - and whether this was found to be useful,
> feasible or just too complex in trying to mix 'hard science' with
> qualitative enquiry.
>
> We are just trying to give under-resourced farmers a voice in the
> scientific research process! and I have found myself struggling with two
> seemingly opposing schools of thought. Ideas, suggestions, opinion on this
> debate most welcome.
>
> Thanks
>
> PS you can buy toothpaste with flip-top lids now......
>
>
> Karen Hampson
> Research Assistant
> University of Nottingham,
> Tropical Crops Research Unit,
> School of Biosciences
> Sutton Bonington Campus,
> Leics, LE12 5RD.
> ++ 44 (0) 115 9515151
> extn 18865
>
|