Recomendations for qualitative data management
May 3, 2007 7:00 AM   Subscribe

What are your experiences with qualitative data coding software? Generally, and Apple specific.

I'm preparing for a year-long fieldwork project in which my data will consist primarily of digitally recorded interviews, written interviews, notes, and digital photographs. I'm trying to figure out the best way to keep track of my data on my G3 iBook. Obviously, the volume of data will be fairly large.

I'm looking for specific recommendations about programs, but also general experiences with how well these programs work. Are they worth the time it takes to learn them? What features should I be looking for? What features sound great but aren't that useful? I found Transana in my searching, and it looks like it might be a good way to actually transcribe less, by allowing you to code audio clips. Has anyone tried this? Did it work?

Basically, anything you can tell me about your experiences with these programs other than "this PC-only program is the one and only, best evar!" would be greatly appreciated.
posted by carmen to Education (9 answers total) 2 users marked this as a favorite
 
Transana seems to have something a bit strange going on. They claim to be GPL licensed and have source posted up at http://sourceforge.net/projects/transana. At the same time, their
Download page states:
Because of our current level of funding, we are no longer able to distribute Transana for free. It costs more to develop, support, and distribute Transana than we have been able to bring in through grants and contracts. We had to choose between charging for the program or curtailing active development.

The single-user version of Transana costs $50. The multi-user version of Transana costs $500. Educational institutions located in Wisconsin and state and local governmental bodies in Wisconsin are eligible for a discounted price.
While it's well within their rights to charge people to download a GPL licensed product (as Linux distros do when they sell CDs), it seems rather unethical for the developer to try to sell software that can be downloaded and compiled for free from the very same site if that's what's going on here. I'd hate for you to purchase it not knowing this.

Anyway, about your question. You might want to look into DevonThink. It's an organizer that collects and organizes all sorts of information (text clippings, web pages, audio files, pictures, movies, etc...) and has "AI-like" features that attempt to classify newly input information and find related items, along with powerful search features. It may or may not jive with your way of thinking and organizing, but it's a pretty powerful tool.
posted by zachlipton at 8:56 AM on May 3, 2007


From my experience, Transana is mainly used by people looking to do micro-level transcriptions of video clips (such as used in Conversation Analysis). I don't think it can incorporate data that is not video-based, and I'm not sure how sensitive the coding can get. Unfortunately, as a fellow Mac user, the only other qualitative data coding/management package I've come across is HyperResearch, but other than fiddling with a demo version several years ago, I've never used it. The folks at the Computer Assisted Qualitative Data Analysis page have reviews and resources, you can probably find some people who have used HyperResearch there. For text-only data that I have analyzed I have used simple databases, both in Excel and in FileMaker.
posted by DiscourseMarker at 11:34 AM on May 3, 2007


I have used two different data coding software programs-- NVIVO and Atlas TI. I used NVIVO back when it used to be called NUD*IST. I used both of them on a PC, and I have no idea about their Apple compatibility (although I think Atlas is PC only, can be run using Virtual PC).

Either program will serve your needs very well. NVIVO is a pretty intuitive program, not hard to pick up, and you can get going pretty quickly. It is a little more limited in its approach to organizing data, using hierarchical groupings (what the program calls a "tree"-- think of many large branches (BIG codes) and then other smaller branches coming off it (subcodes, beneath the BIG code). Also, it cannot code audio files (were you planning on transcribing these interviews?)

I currently use Atlas TI. It is REALLY expensive! And not all that intuitive. I get stumped by it sometimes because it doesn't work like a Windows program. In fact, I am taking a training session for the software program next week. However, it is really high-powered. It can code text, audio, and visual files. It does not require you to use hierarchical coding schemes which allows for more creativity and inductive analysis and it can create really cool diagrams for organizing and representing your data. It's a full service program.

Unfortunately, I don't know anything about Transiana.

I hope this helps, and feel free to ask any follow-up questions.
posted by picklebird at 5:28 PM on May 3, 2007 [1 favorite]


you might like to consider joining the very helpful Qual-Software list (http://www.jiscmail.ac.uk/lists/qual-software.html) and posting your question there. Also, check their archive for material relating to Transana -- David Woods, lead developer, posted new information on April 5, and even invited people to contact him directly with their questions.
posted by davemack at 6:01 AM on May 4, 2007


Thanks everyone. This was kind of an obscure question, but you've all given me some places to get started.
posted by carmen at 7:04 AM on May 4, 2007


I decided to give TAMS Analyzer a shot for a small qualitative dataset I had. I had previously used NVIVO on Windows but I wanted to use my own OS X laptop in this case. While getting started was pretty easy, generating reports got pretty hairy. I couldn't quite figure out how to generate the kind of coding reports I was used to from NVIVO and in the end resorted to doing multi-file text searches using BBEdit.

I have used Transana to transcribe and code videotaped interviews. It worked pretty well for that, but it doesn't sound like that's how you intend to use it.
posted by needled at 10:28 AM on May 4, 2007


needled, I'm thinking of having digitally recorded audio files of interviews. The interviews will be open-ended, so it's possible that many of them will be quite long, and quite varied. Many of them will not be in English, or will be in mixed English. What I was hoping with Transana was that I could code the audio, and then only transcribe and translate the most important parts of the interviews, as needed.

In terms of my written stuff, my main goal is to provide what I'm thinking of as a sophisticated index to my notes and transcripts that I will be able to query. So if I want anecdotes of weddings or something, I can use the related keywords to get either a listing of excerpts or a series of links to all the places I've labelled with that keyword. Is that a report (by now, it should be obvious that my experience with this type of software is nil)? If not, does TAMS do that well? And also (if you have the time and inclination) what kind of information to you use reports to generate?
posted by carmen at 10:53 AM on May 4, 2007


Right, when I talked about reports that's what I had in mind. Specifically I wanted a listing of all the excerpts for a particular code / keyword, which according to the TAMS manual TAMS is able to do but I couldn't figure out how to do it on short notice. I was able to get reports on which files had the keyword or how many times it appeared. I should mention I am not a computer novice and can figure out how to use most software packages, even open source ones with minimal documentation, but for some reason something that is very basic and easy to do in NVivo was not so with TAMS.

I just opened up my old Transana project, for which I used Transana 2.12. Transana will support what you have in mind, but the setup will be nonintuitive. The organizing principle is of video clips. It was excellent as a transcription tool, but now that I think back on it I ended up transferring the transcripts to NVivo for further analysis. I think Transana assumes you are doing a particular type of analysis, in the way DiscourseMarker above described.

The upshot, for the research you've described, I think I would go with TAMS over Transana. One concern is that you will be working on a G3 iBook. My G4 iBook kept crashing when I tried to deal with video in Transana, but Transana ran fine on my G5 dual-CPU desktop machine.
posted by needled at 6:55 AM on May 5, 2007


Thanks again needled. This is really valuable information. I think I'm going to gather some random data this summer before I go and do a little crash course in these two products. It's good to know about the non-intuitive stuff, so I can make sure to figure that out before I go.
posted by carmen at 8:46 AM on May 5, 2007


« Older How core training of US soldiers has changed to...   |   Relocating To Pomona, CA : rental housing and... Newer »
This thread is closed to new comments.