Just as an aside - Trevors' original question to Fiona Wiltshire (a long time member of QSR, the company which makes NVivo)- 'which is best NVivo or ATLAS.ti?' - was a bit like asking Bill Gates which he thought the better platform - Windows or Apple Mac. Fantastic as she is as an exponent of NVivo - she took the tactful route and didn't respond publicly - unless I missed it.
There are some general... and possibly sweeping statements below. focusing mainly on the handling of text - (with a bit of quant) and I echo everything that's been said so far by Ross, Silvana and Paul...
ATLAS.ti - is a 'QUALITATIVE' tool - its user interface is tidy and spacious and you feel like it's a more open interface than the others somehow - it may not have been created with grounded theory in mind (I used to think it was but apparently this is not so) - but even so it keeps you very close to the data if you want to be - and all the lists you want to be visible can be visible at the same time - along with the text. Its codes margin prints out easily next to the text. It's incredibly easy to get started but later on things might get a little more difficult because of the architecture of the software. Its not designed to provide maximum flexibility when handling more quantitative data - e.g. open ended questions have to be organized into documents which represent a single respondent or case. Then you can integrate quant data OK and use query tool in this context later, but the quant info is very much in the background. As my colleague Graham Hughes pointed out to me in an illuminating moment - you can't for instance view the quant style data in a handy table inside the software - I'd never even noticed (this says something about how immersed you become in what a software CAN do rather than what it can't) - and Silvana Paul and Ross have each said something of the sort in different ways... and you don't necessarily need a software to do absolutely everything.
MAXqda is a nice easy qualitative instrument - the visibility simultaneously of all the working panes with the important lists (code lists, sets, documents lists, document text, segments and codes flexibly showing in margin (& all printable) makes for an intuitive tool. The software is geared up to handle the quantitative aspects of qualitative data with the minimum of fuss (if you've got that sort of info). Instant comparative textual info is available across subsets and variables at a couple of clicks - not replicated as easily in any of the others. You might not actually need a Query tool to ask further complicated questions, but it's there if you want it. but its not endlessly flexible. Its got some great ways of giving you overviews of what's going on in your work with two clicks (e.g. interactive lists of codings and memos)- altogether very slick indeed.
As Paul Dempster suggested NVIVo follows the Microsoft route. and as with Outlook it breaks up functions and lists into different windows. moving about the functions is intuitive if you are an Outlook user, but in itself that's a difficult aspect when you are still a bit uncomfortable with the software because you never seem to have the right list on view for the thing you want to do NOW. Though the codes margin display is interestingly flexible in what you can choose to see - you have to keep switching it on - and amazingly you can't very effectively print it out on the same sheet of paper as the text. But its query tool IS endlessly flexible - though slightly difficult to familiarize with. If you have quant data about qual data theres a lot more fuss and decisions which have to be made about data preparation, and how data is organized and laid out (especially for open ended questions) but if you get all that right - its pretty hot stuff when you get as far as the Query tool.
Under the umbrella of the CAQDAS Networking project at the moment we have another project underway - Qualitative Innovations in CAQDAS (QUIC) - which is looking at how various packages handle the integration of various types of data - quant - visual and geo 'stuff'. A lot of step by step stuff is being produced and will eventually be put online
(see what is coming and watch this space http://caqdas.soc.surrey.ac.uk ). Also necessarily a lot more comparative detail will be included in the online resource... Graham Hughes for instance is producing a huge amount of discussion and step by step papers which will help researchers make decisions about the integration of quant data into CAQDAS and I have him to thank for making me 'see' things from different perspectives.
Regards
Ann
Ann Lewins
Qualitative Innovations in CAQDAS (QUIC)
CAQDAS Networking Project
Dept of Sociology, University of Surrey
GUILDFORD GU2 7XH, UK
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
The QUIC project is funded by the UK ESRC under the National Centre for Research Methods programme
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
[log in to unmask]
[log in to unmask]
http://caqdas.soc.surrey.ac.uk
________________________________________
From: qual-software [mailto:[log in to unmask]] On Behalf Of Ross Perkins
Sent: 01 February 2010 20:58
To: [log in to unmask]
Subject: Re: NVivo 8 Training and Consulting
Any claim of "intuitiveness" is highly subjective .. If one has previously used NVivo (say, v. 7), then its interface is more "intutive" than Transana's or any other. Personally, I find Atlas.ti's UI to be convoluted, but I am sure it is simply because I have not spent enough time with it.
To Trevor's original question - and what many have hinted at already - is that the answer to which CAQDAS offering is "better" is, in all honestly, "It depends." It depends on type of data, format, time you have, qualifications of the research time, budget, anticipated results, and so on. If half
Regarding Atlas.ti vs. NVivo, I will say that if one is doing much work with PDF documents, then Atlas is the way to go. NVivo chokes on even modest image-based PDF files, and its attempt to convert text-based PDF files results in a total mess.
Otherwise, I like NVivo quite a bit and once one gets used to the terminology, its fine.
Many lists of qualitative analysis software exist, but here's mine should anyone find it useful...
https://sites.google.com/site/caqdasplant/varieties
Cheers,
- Ross
--
Ross A. Perkins, Ph.D.
Assistant Professor, Educational Technology
Department of Educational Technology
Boise State University
1910 University Dr. MS 1747
Boise, ID 83275
Phone: 208-426-4875
Skype: ross.perkins.bsu
|