Great to have these training issues debated. Here's a go at taking
further some of the beaut contributions. (Sorry it's long: a lot
involved.)

Odd's concern about non-theoretical use of software is one many of us
share. There were some really interesting answers from trainers to
Doug's question about training courses:
> Do they learn techniques to use the software by itself, or is this
embedded in some sort of training in the theory of qualitative methods. 

Seems to me there are layers of issues here.  One is what trainers
should do. Pity the trainer! As Susanne points out, courses have very
(unpredictably!) mixed membership, many participants not needing &/or
wanting such embedded theory. Moreover, as Clare says, enquirers "ask
for training in NUD*IST not for qualitative research with NUD*IST." Even
if we legislated that trainers must ensure participants knew their
theory, how are they meant to find out?  Ask the participants?  In my
experience, those who most need such a grounding are often least likely
to recognise that need (especially if they are in academic positions).
So does the trainer check qualifications? See below on courses in
universities.  Examine applicants? Hardly. As Susanne's stories
illustrate, you have to see who turns up and do your best to meet needs.
This, with Silvana's emphasis on self-assessment, is in my view the only
answer - but it's up to the trainer to make it clear that these are
questions researchers must ask. (I'd like to put in here that
qualitative computing is blessed by the most extraordinarily talented
trainer corps, and the very fact that there are so many trainers
concerned and involved in these issues is possibly the most hopeful
sign!)

Next issue: is this any different problem from that facing any technical
training area?  I agree with the several comments that computer courses
seem to make qualitative research accessible without methodological
training. I suspect that the numbers of untrained researchers using
qualitative computing are tiny compared to the number of users of
statistical software  who don't know what any of those 20 tests of
significance mean, just hang in there till the right number of zeros
appears after the decimal point after the p.  But the problem is
different.  Doug wrote: "Courses in multivariate statistics typically
require a sophisticated background in, and completion of other courses,
in statistics." Using SPSS, however, doesn't :-)  But the problem is
different because stats, unlike qualitative theory-building, is
presented as requiring special training.   People just using SPSS to
show patterns don't usually say they are doing a loglinear analysis.
That sounds like it needs understanding, whereas grounded theory (which
is far more difficult to do well!) seems more accessible.

Next issue: Could we anyway define requirements across a diverse
research base?  We need to be very clear what we are concerned about. I
think we should *not* foster elitist concern that everyone who touches
unstructured data has to have a course in qualitative method. (Some
Courses I Have Known foster no confidence that they would help! And I've
done software training with superb self-taught researchers who are
inspired by the literature and driven by the excitement of its
challenges.).  I also think we should not dismiss qualitative data
handling that is not theory-backed. A lot of research with unstructured
data has always been done theory-free. And it isn't bad, so long as it
names its shot. That's why I decided to give it a name - pattern
analysis. Doing a decent job of reading and reflecting on open endeds in
your survey means they are not wasted. Locating which doctors are giving
advice that isn't understood will help people. We have to be very clear
that qualitative computing assists this sort of pattern analysis, as I
argued in my last message, and helps people do it well. Such research is
a matter of concern mainly when it is presented as something else -
phenomenology or grounded theory etc.

OK, that gets me somewhere else. 
Doug wrote: "I take Lyn's points that neither the founders nor the
developers are making the problematic elision of grounded theory and QDA
software, but I think the interesting group here are the thronging
masses of followers who are  using qual methods and software in applied
situations. As Lyn rightly hints it is this group of users and
followers, and not the innovators, who are most likely to be only
semi-literate [my edit] in the theory of QDA." 
Mmmm...  Not sure that's what I want to say, Doug, or at least all I
want to say. For "thronging masses of followers" read "large numbers who
feel their research can be helped by these tools". I think we should ask
why so many are in applied areas, what they are doing and whether they
have particular sorts of methodology needs. But we also need to ask how
"literate" the university-trained researchers are: my experience
includes more university-based researchers than applied researchers who
didn't have any idea what they were trying to do or why. The major
differences seem to be that the university trained researcher is far
less likely to have any knowledge at all about qualitative computing,
let alone methodologically informed critique of it, and the applied
researcher is far less likely to call what they are doing grounded
theory!

PS
Intriguingly, QSR-Forum is debating the same issues currently. Lioness
Ayres wrote that a major problem is:

>  many people begin qualitative projects without a clear idea what
> they are about, and that this is an inherent hazard of qualitative
studies
Important. It seems you can start, wondering later what you are doing.
And the computer seems a way of dealing with the data you make as a
result. Methods training often doesn't deal with this assumption. The
special need for research design and special issues of qualitative
research design are given little emphasis in the literature or in
university courses. Inter alia, we should throw responsibility back to
grants bodies and research advisors, and particularly to supervisors, to
ensure that a novice researcher has asked what they are trying to do and
why before they turn up at a software training which may help them find
ways of doing it (if they knew what it was).

*****
(News item:  Anyone interested in this topic and currently wondering
what to give a paper on at Auckland Jan 2000 Advances in Qual Methods
conference?  There's a suggested symposium on qualitative research
design.  If you want to be part of this don't mail the list pleeeeease:
mail me personally)
*******

cheers
Lyn


Lyn Richards, 
Research Professor of Qualitative Methodology,  University of Western
Sydney,  
Director, Research Services, Qualitative Solutions and Research.
(email) [log in to unmask]   
(Ph) +61 3 9459 1699  (Fax) +61 3 9459 0435
(snail) Box 171, La Trobe University PO, Vic 3083, Australia.
http://www.qsr.com.au



%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%