Forum Posts Feed
Progam with QDA Qualitative Data Analysis features? Coding/tagging blocks of text?
< Next Topic | Back to topic list | Previous Topic >
Pages: < 1 2 3 4 5 6 7 > Last ›
Posted by Dr Andus
Aug 3, 2012 at 09:02 PM
Carrot wrote:
>I’m
>wondering if you might possibly make a short video outlining the process, or perhaps
>illustrate the instructions you provided with screenshots?
>One
>problems- I really do not like the idea of using markup language - It just doesn’t seem
>to make sense to me to use that (I simply have too many things to try to learn and
>remember- why not just click on toolbar icons?)
>I’m very curious- do you think I
>should begin using ConnectedText rather than AtlasTI? I have about 18 documents each
>about 100 pages, each one my fieldnotes for different research trips.
I’d love to but unfortunately I’m travelling until end of Aug. w/o proper PC or internet connection.
Re CT, there is a bit of a learning curve but so is w/ Atlas or NVivo. As for markup, you literally only need 3 or 4 codes to do what I described. I believe it’s faster to type == than to select text and click on an icon to code something.
Whether Atlas or CT that’s depends on your personal preferences and whether you’re willing to put in a couple of days to learn CT. I couldn’t quite warm to Atlas and preferred NVivo. However CT is just far more flexible and I can import masses of data and organise it the way my mind works. It is essential to sign up to the CT forum and read around and ask for help.
Posted by Fredy
Aug 3, 2012 at 09:37 PM
Carrot, you say you will probably use such a program for years. Let’s suppose after your PhD you’ll be well paid. You think 1-2 Benjamins now and 6 others then will be too much money, for such “overpriced” sw, and you say many, many others think as you do. Some weeks ago, in the UltraRecall forum I (as schferk) explained in depth WHY sw developers have lost all interest in refining their babies. People who think like you are the culprits, I fear, for such a development to extreme poverty in functionality of available sw, and, as I explained there also, the cloud crowd will not be better off neither, all to the contrary.
Daly, did it ever occur to you what the original idea behind a citation function in a forum sw was? Just a hint: SELECT BITS to comment on those bits.
ConnectedText, it’s a pleasure to see you’re considered here, lately, as the Swiss pocket knife to any text processing problem whatsoever; it’s heart-warming to see one developer flourish at least, even if it’s for the wrong reasons. At soon as you’ll be able to prepare my coffee and do my laundry, let me know, I’ll happily buy.
Carrot, if you really want to become independent of your proprietary “format”, may it for bucks or for other factors: Consider using a programmable editor. 1 line 1 paragraph. Each line starting with at least one code “super-paragraph” (this can be automatted), more codes at libitum, here or at the end of lines or anywhere within, just use dollar, pound, whatever leading signs.
Many such editors can sort lines by block content, i.e. chars x to y of these lines (hence the utility to carefully design your codes(‘lengths, incl. blank codes)), not just by their start chars. Many such editors can filter lines by multiple criteria (= codes = “fields” with specific content); regex goes without saying. They’re all many years old - so what, except for Unicode, perhaps?
There are some special editors like KEdit that seem to facilitate such filtering at first sight - I’m not advocating these traps, though.
You can buy for cheap such editors, or even “steal” them. All that’s needed for everything you’ll ever need here is perhaps one day of learning the corresponding syntax, incl. some statistical functionality that might be needed.
Then, with your PhD, you’ll probably work within a corporation that buys some sw, fortunately. Corporations buy sw, individuals by iPads. (800$ every years’s nothing for such lifestyle hardware, but beware of applics costing 5$ - they must be 1,50 or they will be unacceptable). That’s why in many years here, nothing really new here, and nothing will come.
Posted by Carrot
Aug 4, 2012 at 02:28 AM
Dr Andus wrote:
>
>Re CT, there is a bit of a learning curve but so is w/ Atlas or
>NVivo. As for markup, you literally only need 3 or 4 codes to do what I described. I
>believe it’s faster to type == than to select text and click on an icon to code
>something.
>
>Whether Atlas or CT that’s depends on your personal preferences and
>whether you’re willing to put in a couple of days to learn CT. I couldn’t quite warm to
>Atlas and preferred NVivo. However CT is just far more flexible and I can import masses
>of data and organise it the way my mind works. It is essential to sign up to the CT forum
>and read around and ask for help.
Thank you for your reply!
I’ll have to experiment more before I can decide which application might be better for my work.
I’ve found ConnectedText quite difficult to figure out.
Your approach sounds quite intriguing and one day, if you have time to make a video or illustrated
set of instructions, I’m sure many users would find great use in it.
You’d mentioned you code your document by adding Headings (up to 5 levels)
I think this will be the major problem- I need to use about 80 different codes to follow different themes in my work.
Perhaps CT will not be suitable for this job, but I’ll purchase a copy and learn how to use it for my writing and other projects.
The other similar open-source solutions (OutWiker, Wikipad, Book-On-a-Stick) are progressing wonderfully, but do not yet seem capable of providing the functions I will probably need.
For now, I will continue to use AtlasTI and experiment with HyperResearch QDA.
-C.
Posted by Alexander Deliyannis
Aug 4, 2012 at 08:49 AM
Fredy wrote:
>Carrot, you say you will probably use such a program for years. Let’s suppose after
>your PhD you’ll be well paid. You think 1-2 Benjamins now and 6 others then will be too
>much money, for such “overpriced” sw, and you say many, many others think as you do.
>Some weeks ago, in the UltraRecall forum I (as schferk) explained in depth WHY sw
>developers have lost all interest in refining their babies. People who think like you
>are the culprits, I fear, for such a development to extreme poverty in functionality
>of available sw, and, as I explained there also, the cloud crowd will not be better off
>neither, all to the contrary.
Though I agree in principle that we should be willing to pay for the tools of our trade, and as a chronic CRIMPer I’ve paid quite a bit more, I have to note that the pricing of software overall seems to me extremely irrational and in many cases unjustifiable. Moreover, I have seen no positive correlation between the price of software, the support provided and its long term viability. In fact, I have seen several instances that could support the negative correlation of the above, i.e. more affordable software is often better supported and available in the longer run.
Examples:
- Notemap, by far the most expensive single pane outliner, whose development has stopped ages ago, and even basic bugs have not been resolved (as Cassius has pointed out on several occasions)
+ MyInfo, a very moderately priced two-pane outliner, continuously updated and excellently supported. There are actually several other positive examples in this category, and I believe that it is the development and support which keep users faithful to any one such program, rather than the (relatively few) differences in features.
- MindManager, the first and arguably most successful mind mapping program, significantly overpriced for what it does (especially given the many alternatives now in existence), now in its 10th incarnation, has maintained its outline mode crippled as if it were an afterthought, and has even lost useful features such as OPML support along the way. You may see here http://www.mindjet.com/pdf/us/MM2012_WinFamilyComparison.pdf and judge whether the development from one version to the other would justify the upgrade price.
+ Noteliner, a modestly promoted but very powerful and completely free program, created and maintained as a labour of love, gaining strong features along the way, several of which have been proposed by it users in this here forum. Again, there are other such examples, like the very special TreeSheets.
Posted by Dr Andus
Aug 4, 2012 at 11:57 AM
Carrot wrote:
>You’d
>mentioned you code your document by adding Headings (up to 5 levels)
>I think this will
>be the major problem- I need to use about 80 different codes to follow different themes
>in my work.
>Perhaps CT will not be suitable for this job, but I’ll purchase a copy and
>learn how to use it for my writing and other projects.
There are several other features in CT that can be used for coding, the Headings are just one of them. I also use the Categories, Attributes, and Properties. The last two allow for some complex inline and search queries for building reports. It probably also depends on your particular qualitative research philosophy and methodology as to how you would want to use these for your purposes to arrive at a desired outcome.
>For now, I
>will continue to use AtlasTI and experiment with HyperResearch QDA.
I think it is a good idea to reallly learn and use one QDA extensively, and then once you’ve learnt the most useful features and got frustrated with the limitations, then trying to remodel your process in CT. I used NVivo for 6 months daily and processed half of my data before I got fed up. Ironically what got to me that I ended up with 100 codes and it became difficult to say which were the important to focus on.
Sorry I can’t do the detailed instructions for you right now for CT. I’ll try to do them when I get back home in Sept. Good luck!