Have people noted Atlas.ti, Nudist, NVivo? There are several webpages comparing these (industrial standards). I just bought the newest Atlas.ti 7 but very disappointed. They're still old school in my opinion (no real advantage over Atlas.ti 6 in my opinion). I did find a convincing screencast of Tinderbox and the qual research process here: http://brandsavant.com/processing-qualitative-research-data-with-tinderbox/ I think the potential of something like DevonThink could really be exploited further for such work but I'm still learning how to use DT so I can't really say.

Oh my! Peter's post reminds me I hijacked this thread with a topic discussed elsewhere in this forum, but you cannot be more off-topic here: sorry! But it was spontaneous. So this subject being put into parentheses here...

(And forgive me upon my insisting: Perhaps the notion of "establishing an umbilical cord to the imaginary good object" or something could help with the cloud phenomenon.)

("Umbilical cord" conveying the notion of NOURISHING good object... You could even go so far as saying that couch potatoes ain't nourished enough by their tv alone, so are reduced to stuff that potato chips into themselves, on top, whilst people on their iPad umbilical cord - but must be online - are nourished some way better than that on their deepest psychological level, so can do without chips - it's perhaps a matter of better "appropriation" of the stuff delivered to them? Whilst all that "identification" with tv content leaves you "hungry", not satiated, on your deeper levels?)

this might interest you (cords or no cords): http://savageminds.org/2010/07/26/qda-or-not-qda/

Fredy wrote:

Oh my! Peter's post reminds me I hijacked this thread with a topic discussed elsewhere

The thread is about "Progam with QDA Qualitative Data Analysis features? Coding/tagging blocks of text" and I appreciate all comments related to this.
Going off topic is easy, but pages and pages of off-topic is tedious indeed.

Peter wrote:

this might interest you (cords or no cords):
http://savageminds.org/2010/07/26/qda-or-not-qda/

Excellent discussion; an anthropological case study in itself :-)

"The problem with Evernote is that you are limited to tagging at the note level. With Atlas.ti or NVIVO you can tag individual words if you are so inclined. They are an order of magnitude more flexible than Evernote." I think this describes quite well the importance of "entities" (to use Bill/MadAboutDana's term) in information management. I'd say that most of the software we talk about here works at a much higher level --whole texts for 2/3 pane info managers, paragraphs for outliners.

So it seems that if you want to do QDA, you should probably resort to QDA software.

By the way, the eval.org GDA S/W resource mentioned somewhere in the discussion by Jeremy Trombley does not work, but this one does http://www.eval.org/Resources/QDA.asp

Alexander Deliyannis wrote:

I'd say
that most of the software we talk about here works at a much higher level --whole texts
for 2/3 pane info managers, paragraphs for outliners.

This is a very important distinction! The question of outlining levels often gets blurred or forgotten when discussing "outliners" more generally. This forum could be improved perhaps if we kept that distinction in mind. For instance, Daniel Wessel on his blog makes an important distinction between "content" and "structure" outlining, using the comparison between Circus Ponies Notebook and Scrivener: http://www.organizingcreativity.com/2012/02/outliner-in-scrivener-vs-outliner-in-cpn-structure-scrivener-vs-content-cpn-outlines/

Both of these "levels" are important in my (qualitative) analysis and writing and one measure of a good app (for me) is one that allows working across these levels. QDA software, in my opinion, has traditionally done the "content" better but we're seeing some interesting convergence with apps like Tinderbox and DevonThink (I think).

As a lawyer, QDA is appealing to analyze documents and evidence in preparation for witness examinations and legal argument. Transcripts, statements, and audio and video recordings, and images can be excerpted by source, and individual fact, and issue; and then filtered on these fields. Casemap is the standard for this analysis. This thread caused me to look at QDA software as an alternative. I am evaluating, for free, Dedoose, a Java QDA app. Aspects are superior to Casemap. Notably, it is much more visual. Color is used well, and excerpts instantly and can be viewed in clear, coded context. Multi-media support is valuable. Java is less than ideal, but for me, so is Casemap's native Windows platform, which I run virtualized on OS X, my platform of choice. The deal breaker presently for Dedoose is poor PDF support. Imported PDFs apparently are barely supported. I could not excerpt from standard OCRed scanned PDFs. Instead of importing PDFs, Dedoose users are advised by Dedoose to convert PDFs to RTF or .doc format. Some of the other QDA applications indicate full PDF support. I have not tried them yet.

Devonthink is appealing for case analysis, too. I've used it for storage and recall for years. One requirement for expanding its use to replace Casemap is that I have so far been unable to create a script to excerpt from PDFs with a link-back reference to the original PDF and its page number. There is a standard contextual (right click) option for copying a page link, but it's burdensome to use it for each clipping. There is also a script for annotating a PDF, but it works only for whole PDFs, not excerpts of them. Another awkward limitation is obtaining intersection of tags, or of groups. Smart searches or smart groups permit intersection searches; and the tag view does. Neither is ideal, however. Smart searches are not meant to be and cannot be created rapidly ad hoc. The smart search smart folders create clutter until deleted. And the tag view is awkward to use for intersection searches because it contains tags from the whole database. Finding the group-specific tags which are needed in the sea of database-wide tags, especially when one is under stress such as in a trial, is problematic. Nesting tags does not help because the database view seems to be only flat, even if the tags are nested in Devonthink's tag/group browser pane. Finally, printing is inadequate. To print multiple files, one must combine the files into one file. I have been unable to select multiple files and then simply issue one print command to print them.

Thank you for this sobering review yooj. Very helpful!

I am not a lawyer and was unaware of both Casemap and Dedoose. They both sound interesting but I note the issues you raise. I just had a quick look Dedoose. Another drawback is that it is cloud-based which is a problem (for me) when working with sensitive information. This goes against the ethical review guidelines that I must follow for some of my work.

I recognize your frustration with PDF incorporation. Atlas.ti finally managed to get this working about a year ago. However I find that it generally slows with bigger projects. I also find A7 a disappointing upgrade, nothing really new over A6 and clumsy on the whole, now that I've switched to a Mac. I'm curious of anyone has comparative experience with other QDA software.

Do you know Mindjet's program? I think it's called Mind Manager. I don't know it yet but it looks like a mind mapper on steroids that might have something to offer above and beyond basic mind mapping.

Your review of DT is also helpful, especially since I am just starting to learn it. I've tried to find some real-world screencasts of how people use it but so far I've only found one (from 2009): http://www.youtube.com/watch?v=UH2L6WSA0UY If you felt so inclined to upload a screencast of your use of DT I would certainly be interested! An easy free app to create screencasts is Jing. (I've proposed this now to a few people but so far no one has bitten!)

I am also interesting in finding better ways to annotate/tag PDFs. I have yet to explore Sente - DT or other such combinations. BibDesk might also have something to offer:
http://www.organognosi.com/category/bibdesk/ Here is a screencast of someone using Bibdesk and Skim together (both free apps) for annotation: http://jenniferclaro.wordpress.com/2011/08/12/a-simple-academic-workflow/ It looks pretty slick but BibDesk is designed to work with LaTeX apparently, which I know nothing about. I'm not sure how much of an issue this posses when working with Word, Scrivener, for instance.

Peter, I'm using AtlasTi 6 for some time and I was thinking in upgrading to A7. Do you really feel it's not worth the 75eur upgrade price?
Do some of the users know maxqda?
Thanks

Well, I guess it's all relative. My advice is to download a trial and give it a whirl. Perhaps it depends on how central it is to your workflow, if you plan to stick with Altas.ti no matter what (or considering alternatives), if you are able to get done what you need to do without upgrading... all the usual considerations. Oh and whether or not you can afford it of course! I do think it's time they revamp the interface (at least) and make a Mac version. As for me, I'm starting to look at other options, although I've used Atlas.ti for more than 10 years.

Peter wrote:

For instance, Daniel Wessel on his blog makes an important distinction between “content” and “structure” outlining,
using the comparison between Circus Ponies Notebook and Scrivener

Very useful distinction. I would add that the 'content' outline maintains the integrity of the full text, whereas the 'structure' approach breaks it down to non-continuous blocks of text.

If I understand correctly what QDA software does, it also maintains the integrity of the text ('content' approach) while offering the ability to annotate it at a very detailed level (the 'entities' are very fine). Last but not least, unlike much of the 'content' software often discussed here, QDA software does not provide a writing/editing environment, but is limited to referencing. Is this correct?

Carrot wrote:

Dr. Andus,
Thanks very much for your reply. I looked at your posting about
ConnectedText http://www.outlinersoftware.com/topics/viewt/3799/40
I'm
wondering if you might possibly make a short video outlining the process, or perhaps
illustrate the instructions you provided with screenshots?
I'm sure many of us
would greatly benefit from that.

OK, I managed to get started with the CT tutorial, here is the first instalment. Sorry, no time for videos, though hopefully I can add a screenshot here and there:
http://drandus.wordpress.com/2012/09/21/why-use-connectedtext-for-qualitative-data-analysis/

Dr. Andus,
Thanks very much for creating the tutorial.
I'm now very much intrigued by CT thanks to your posting of your experiences.

I've been using MyInfo, and I find it extraordinary useful for data collection, but I've not found it so useful for discovering links in my data. Nor have I found it particularly useful for writing papers. I began using Scrivener recently, and rely on it for writing, and MyInfo as my data repository. I was not able to figure out how to get CT to function in a way similar to Scrivener, which seems far more geared towards writing.

I'll look forward to further instalments. I hope you will be able to provide a detailed description with screenshots showing your process and the results. (PS. good screenshot capture program is the open-source "Greenshot")

Carrot wrote:

I began using Scrivener recently, and rely on it for writing, and MyInfo as my
data repository. I was not able to figure out how to get CT to function in a way similar to
Scrivener, which seems far more geared towards writing.

There is a way to emulate (sort of) the main view of Scrivener in CT, by using the Outliner pane and linking items in the outline to a given topic, so that by double-clicking on it CT opens the given topic. See Manfred's description:

http://takingnotenow.blogspot.co.uk/2008/06/connectedtext-scrivener-for-windows.html

I hope you will be able to provide a detailed description with
screenshots showing your process and the results. (PS. good screenshot capture
program is the open-source "Greenshot")

Thanks for the Greenshot suggestion, I like it a lot! It should make things easier...

Carrot wrote:

I hope you will be able to provide a detailed description with
screenshots showing your process and the results. (PS. good screenshot capture
program is the open-source "Greenshot")

Speaking of tutorials and screenshots, does anyone know of a good free screenrecording software for making tutorials? I see there are some free ones here, but in case anyone's got a specific recommendation...

http://www.videohelp.com/tools/sections/desktop-screen-capture?orderby=Type

I've only toyed with it and not done anything serious, but CamStudio (free, open source) seems pretty decent.

shatteredmindofbob wrote:

I've only toyed with it and not done anything serious, but CamStudio (free, open
source) seems pretty decent.

Thanks for that, shatteredmindofbob

Carrot wrote:

I'll look forward to
further instalments. I hope you will be able to provide a detailed description with
screenshots showing your process and the results.

Here is the next instalment (some tips on how to get started with CT - for novices):

http://drandus.wordpress.com/2012/09/27/getting-started-with-connectedtext/

Carrot wrote:

I hope you will be able to provide a detailed description with
screenshots showing your process and the results.

Well, it turned out to be a mammoth task, but I managed to complete my tutorial series on how I use CT for qualitative data analysis. If you already use CT, then reading the last two posts might suffice. If you're new, then I suggest starting at the beginning. Here is the main page and the table of contents:

http://drandus.wordpress.com/connectedtext-tutorials/

1. Why use ConnectedText for qualitative data analysis?
1.1. CT as an alternative to CAQDAS such as Atlas.ti and NVivo.

2. Getting ConnectedText
2.1. The difficulty of understanding CT by first-time users.

3. Getting started with ConnectedText
3.1. Installing CT the first time.
3.2. Sources of information and help.
3.3. Tips for desktop layout configurations.

4. Setting up ConnectedText
4.1. The issue of having to use markup in CT.
4.2. How to avoid having to type markups.
4.3. Customising markup colours to aid reading and writing.

5. Designing your QDA project for ConnectedText
5.1. Developing a research project design prior to importing data.

6. Importing your data into ConnectedText
6.1. Organising the import process.
6.2. Importing particular file types.
6.3. Linking to files, websites and programmes.

7. Preparing for coding in ConnectedText
7.1. a bit of theory and final preparations before launching into coding.

8. Coding process flow in ConnectedText
8.1. visual representation of coding process flow and project structure.
8.2. description of coding process with relevant markups.

9. Summary and example of coding in ConnectedText
9.1. step-by-step description of coding process.
9.2. uses example and screenshots.

Questions and comments welcome.

Note that I don't know anything about QDA.

The set of articles are a good underrated topic but many of the things felt imbalanced.

The installation portion felt too long and tailored towards an extreme subset of a a general audience but then Atlas.ti and NVivo were too quickly name dropped as if the audience is supposed to be an expert of QDA and there was a lack of true detailed comparison in the transition from one software to another especially as no true specific subject appeared to be needed to be researched or requiring of an enlightenment. We're never really sure if you were making a personal wiki on a subject you know of and then just adding supplementary notes or whether you were brainstorming new perspectives as you add data into the software. It was like a general project template that had a bit of an obscure software comparison but none of the desperate qualifier that comes with a user needing to tackle an issue that they seriously need to research.

There were also too many points that felt like CT was a supportive software. The flow chart focus obviously wasn't made in CT and most of the CT transitions were described rather than animated thus leaving behind a strong workflow hole where in you saw the finished product but you don't know what the psychological mindset benefits are. It really was less Design and more Getting Started on this software by inserting a vague topic to research with little respect given to the diverse natures behind a person's learning process and the diverse nature of how a topic can be researched. (A biography for a person for example can be easily made by a researcher but the subtle distinction between basic and informed vs in-depth while still being accurate and telling multiple stories from different perspectives is very important for an outliner to be able to distinguish and warn the user of.)

Finally, from a blog design PoV, there were points were it felt you were getting lazy and were being held back by the image insertion.

I'm not saying you're writing style is absolutely horrible for your target audience or even for random visitors because it's certainly not going to get many critics from this forum but when you have a writer start using terms like daisy-chaining and naming sections DRA case study (instead of an actual test subject) it's just a step above from the generic demo examples provided by the software maker themselves. There's some value there as a preview but there's also an apathy that comes with the supposed intention of helping someone design their own research.

It's like the software isn't supposed to carry the brunt of the data. It's just supposed to be a bag. You put something in. You close it. You reopen it. Very un-research-like in a real tough research subject like when dual comparison of multiple data points is needed by the user to view the outline in such a way that they can easily compare multiple facets without falling into biases. I'm not saying you, as an actual researcher, with the IQ or EQ or other types of strengths and associated knowledge that CT will personally bog you down. But the overall text you have written is very common of text that bog people down who are curious about the subject of say QDA but don't possess the knowledge, training, intelligence and critical thinking to just insert things into CT. In short, the text you have written does not present the evidence to the audience of why CT should be used for QDA. A question you yourself imposed at the beginning of the tutorial.

This doesn't mean the internet isn't full of these types of articles. I already said you might not receive much critique here. As the minority though, rare subjects such as the one you have written is sad to witness go down the drain despite tackling a very important subject in the realm of outlining. There really could be more written on how your mind interprets importing as a process of reduction. Or how you pre-shape your final outline considering researchers (general ones not QDA) are supposed to treat findings as neutral objects with no final predetermined form. Or how a specific non-data could be imported incrementally (especially when the data still doesn't exist and is being collected solely to be stored in CT). There's tons of those questions left into the mystery of your mind.

Forgot to add this: My biggest personal critique though is that there is no indication in the articles themselves of now no longer needing to read any of your posts in this forum related to CT or QDA.

For a set of articles of that length, a topic absorbing all the things you've written here and planting a type of study or paper or summary of all that has passed would have been more true to answering the question of why to use CT for QDA than what you have written.

Foolness wrote:

Forgot to add this: My biggest personal critique though is that there is no indication
in the articles themselves of now no longer needing to read any of your posts in this
forum related to CT or QDA.

For a set of articles of that length, a topic absorbing all
the things you've written here and planting a type of study or paper or summary of all
that has passed would have been more true to answering the question of why to use CT for
QDA than what you have written.

Thanks for your comments. You make a lot of valid points and if I was making a living doing software tutorial writing then I would certainly go back and address those issues. However, as this is not my job or business, it is what it is for now, as I don't have more time to keep rewriting them.

But I'm happy to address any specific questions about using the software and the coding and research process itself. I may add more posts in the future which might clarify some things retrospectively as well.

In the end though I'm afraid it is the prospective interested user who will need to make the effort to make sense of it. This is for a very specialised and narrow (but motivated) audience: for PhD students and other researchers who have a genuine problem of being frustrated with the dedicated CAQDAS tools on the market for a host of reasons (cost, technical rigidity, implicit philosophical bias etc.) and are looking for an alternative.

Having said that, I could imagine other types of users making some sense of it as well for totally unrelated uses that I never thought of.

I've added a more generalised model of the qualitative data analysis process I have described in my CT tutorials, with a chart that hopefully expresses things more clearly:

http://drandus.wordpress.com/2012/10/06/caqdas-model-for-connectedtext/

Nice and clear! What did you use to make the chart? VUE again?