> Dear Qualitative List,
> If one wanted to assess the "richness" of narrative reports using a
> qualitative data analysis program, which program would you recommend? To
> provide some context, I am conducting a study where participants will be
> randomly assigned to one of two conditions and asked to provide a
narrative
> report of an event that they have experienced. My research question
concerns
> whether there are differences in the "richness" of reports depending on
the
> condition to which the participants were assigned. Thank you in advance.
>
Jason - I agree with much that Harriet Meek said earlier in response to
this. Her questions and comments about definition and criteria of richness
being central to a helpful answer about software. It would also be helpful
to know in what context you 'imagined' a software program could help you
assess 'richness'. A couple of extra questions I'd want to ask - how big the
dataset ? what is your disciplinary background? what is the epistemological
background to your study? I only ask because I think its useful to be clear
about what you don't need as well as what you might need. I was once
involved very peripherally with someone who had a fairly small data set - 12
or so interviews based on interviewees' responses to vignettes concerning
moral dilemmas. She began (this was 5 years ago) using a software to 'code'
ideas, themes, issues observed in the data. Then midway thro her analysis
and after extensive use of Nud.ist, after advice from supervisors, she
started to rethink her approach. She realised that she actually wanted to
start all over again - looking at the language and style of expression from
a much more discourse analysis perspective. She stopped using any software
package at all except for 'Word', and simply broke down each sentence or
phrase with long, in depth written analysis of what 'work' was going on in
each statement. There was always going to be limit to how much a software
program like NUD*ISt could support that approach, but she wasn't clear
enough about what her approach was, before she started.
Some software programs are designed to help you 'manage 'and process large
volumes of text - but the trade off is, that you lose touchy feely contact
with how you can physically 'handle' and 'annotate' the text.
NOW... 5 years later, there are software packages that might have helped her
handle her data in more flexible ways. In any of the 'rich text format'
softwares, MAXqda, NVivo (parallel stream package to Nudist) she could have
had both the touchy feely ability to mark and colour her text and break it
down phrase by phrase, with embedded written analysis (or more hidden
annotations)...but also have used parallel, traditional coding/retrieval
devices if she had wanted, to help her 'manage' other aspects of her data.
Rather differently, ATLAS.ti, though not quite yet rich text format -
provides efficient ways to set up linked passages (hyperlinks) of say,
'trails thro the data' the sequence of which you want to preserve (when
re-examining them. This is the restriction with coding... this fragments
the text and tends to retrieve the text fragments in 'text' order - not so
helpful with narrative.
.... hope that helps to contextualise some of the issues you may need to
consider -
I suppose what I am saying also is that - the management of your data is
important - and this is what CAQDAS softwares tend to do well. But ask
yourself what you were hoping it could do for you?
All these packages are very dependent on you thought processes - nothing
happens automatically. There are some packages (investigate Diction 4 for
instance which analyses the tone, or confidence level - based on pattern
recognition etc - used in management contexts) ...but I'd hesitate though
before recommending it as a flexible data management tool.
cheers
Ann Lewins
CAQDAS Networking Project: http://www.soc.surrey.ac.uk/caqdas/
[log in to unmask]
Dept of Sociology
University of Surrey
GUILDFORD GU2 5XH
+44 (0)1 483 25 94 55
mobile +44 (0) 7966 541 518
|
