Ann Lewins has broadened this discussion to include the Qual-software list -
the original subscriber had posted his question to both the QSR and Atlas
forums so I am copying them in as well. Apologies for cross-postings to
those who belong to more than one of these lists (I belong to all 3.)

I think it is very difficult for someone new to qualitative software to
choose a package that is best for them.  There are only a handful of us who
know several packages well and we are mostly trainers which is why we invest
the time to learn several packages.  Several of them have already commented
on one or the other of the lists (I felt that Suzanne Friese's comments on
the Atlas list were very fair - although you can have interactive codes in
NVIVO clicking on the code in the coder rather than the margin. I also agree
with Ann's best aspects of
both copied below.)  Some people have suggested to look at the demos of the
software.  I agree that is a starting point but if you are new to this it is
hard to know where to begin, what you should be looking for, etc.  Also from
the demos it is hard to understand how you would use them in practice.  I
think the best way is to talk to someone who knows a range of packages about
what kind of analysis you want to do.  That is why I have a 'Which software
is best for you' page on my web-site where you write about your specific
project and I write back giving you the pros and cons of various packages
according to your needs.  The CAQDAS project offers workshops in the UK
which compares a number of packages and I will be doing the same in the US.

All the packages have their strengths and weaknesses so it really depends on
your approach to analysis about which is better suited to your needs.  As
far as Atlas and NVIVO are concerned they are both good packages so even if
you think you made the 'wrong' choice, there are workrounds to get at what
you want to do.

There are other considerations you should take into account.  First of all
what are other people in your institution or organisation using?  It is
important to have good local support.  In the university context look
broader than your own department.  I have had people tell me that noone in
their university is using a particular package but I happen to know through
my training that there are people in their university using it - sometimes
even in their own department.  So I often put users together.  You should
also look at the support offered by the developers of the packages.  Thomas
Muhr provides excellent support via the Atlas forum - so does QSR via their
forum and their wonderful Ted who is leaving at the end of this month after
5 years at QSR but is handing over to Fiona who I am sure will do as good a
job. (And Ted will still be around on an independent basis offering
support.)

To get on to strengths
Atlas's strength is its network tool.  This is important if you  are
interested in the links between your codes.  It is also important if you
want to look at links within the text.  I like the way you can look at how
one
statement contradicts another statement later in an interview which leads to
a particular action.  You can take those statements out of the linearity of
the text and display them in the network - together with links which you can
label according to your needs.

NVIVO's strength is its search tool.  As Ann has already mentioned below the
matrix tables it produces gives a quick cut at the data so you can see if
patterns are emerging between the different groups in your study or across
different themes.  I particularly like that it gives the result not only as
a report which you can print out but as part of the database - so you can
ask further questions based on previous searches.  I also like the
hyperlinks you can insert within a document in NVIVO to another document or
to a precise point in a document.  This is invaluable for memo writing.  You
can write your journal or memos directly in NVIVO and have links to text
that support the argument you are constructing.  This technique is also
valuable when using NVIVO to analyse literature reviews.

There is a lot more to both packages.  They are both quite powerful and it
is worthwhile to get some training so you can make sure that you are using
whichever package effectively.

Silvana

Dr. Silvana di Gregorio
SdG Associates
Research & Training Consultants
Tel/Fax: +44-(0)20-8806-1001
E-mail: [log in to unmask]
Web-site: www.sdgassociates.com




----- Original Message -----
From: "Ann Lewins" <[log in to unmask]>
To: <[log in to unmask]>
Sent: Friday, February 14, 2003 11:24 AM
Subject: Re: comparing Atlas TI and NVivo


> I'm broadening this discussion which has appeared on both ATLAS.ti and QSR
> user-group lists to include 'qual-software' discussion list.  There are
many
> excellent aspects to each software; subjectively speaking if I was to pick
> my one best thing for each software this is what they would be:
>
> NVivo - at later stages of work - after 'coding up' - simply brilliant at
> 'searching' and  producing resources to write up from... e.g output tables
> giving at same time frequency tables which are interactively connected to
> qualitative data sitting behind each cell of the table. Thus, at its
> simplest level, great for comparative work across say, different types of
> respondent at the end of coding process. I'd expect to produce masses of
> these matrices during the writing up stages. Its not that frequency is
> important, qualitatively speaking, - its just that those tables provide
> instant individual access to many different groups of respondents talking
> about multiple themes (each theme appearing separately in the table)
>
> ATLAS.ti - simply brilliant at allowing you to work in all sorts of ways
> with data, keeping you 'grounded' or located at the whole context (if
thats
> what you want) - while at the same time allowing you to flick quickly
> through e.g. your thematically coded data for a particular theme (the
coded
> data is HIGHLIGHTED in SURROUNDING context). Thus great for where 'context
> is all' - for instance with focus groups - notoriously difficult to code
up
> because of the interaction and dynamics of the group - one can relax and
say
> to oneself - thats Ok if I haven't selected the perfect bit of text - I
> don't EVER have to lift coded text out of its context at all (although of
> course I can if I want).
>
>
> To just focus on those aspects barely scratches the surface of
> course....sorry!
>
> cheers
> Ann Lewins
>
> Ann Lewins
> Manager CAQDAS Networking Project
> Dept of Sociology
> University of Surrey
> GUILDFORD  GU2 5XH
> email:   [log in to unmask]
> CAQDAS web site:  http://www.soc.surrey.ac.uk/caqdas/
> Tel +44 (0)1 483 68 94 55
> Fax +44 (0)1 483 68 95 51