Software tools for qualitative data organization
November 12, 2014 6:32 AM Subscribe
I'm about to start a qualitative research project that will require organizing a number of ideas from different sources. During my Ph.D. I used a variation on the index card method, but I'm curious what the state of the art is for software tools.
The key elements that I need are the ability to manage hundreds of data points (usually either quotations or short summarized notes on a single idea), to manipulate the grouping of data points so I can put like things together and vary the order within those groupings, and to add metadata to the data points (e.g., source, a ranking of how important the claim is).
The ability to visually/spatially manipulate the data points is helpful - I have strong spatial skills and one of the things I liked about the index card method during my dissertation was that I could see the structure of the argument coming together as everything fell into its place, and I could literally move pieces around in order to figure out how to present the logic.
I don't need the software to do any analysis on its own. I'm not looking for a statistics package, and I don't need frequency analysis or any other qualitative statistical tool. I just need a good search capability.
I also don't need the software to do bibliographic management at any level of sophistication. If I can put a note on a data point saying "from interview X", that's good enough for what I am doing.
I have heard some people use Evernote for this kind of work, but I'm not sure if there are drawbacks I should be aware of or if there are better/easier to use tools out there. I tried Devonthink during the dissertation and found it too overwhelming (but it's possible I just didn't give it enough time).
I'm using a Mac. The ability to work on my iPad as well as my desktop would be a plus but not necessary. Free is strongly preferred.
The key elements that I need are the ability to manage hundreds of data points (usually either quotations or short summarized notes on a single idea), to manipulate the grouping of data points so I can put like things together and vary the order within those groupings, and to add metadata to the data points (e.g., source, a ranking of how important the claim is).
The ability to visually/spatially manipulate the data points is helpful - I have strong spatial skills and one of the things I liked about the index card method during my dissertation was that I could see the structure of the argument coming together as everything fell into its place, and I could literally move pieces around in order to figure out how to present the logic.
I don't need the software to do any analysis on its own. I'm not looking for a statistics package, and I don't need frequency analysis or any other qualitative statistical tool. I just need a good search capability.
I also don't need the software to do bibliographic management at any level of sophistication. If I can put a note on a data point saying "from interview X", that's good enough for what I am doing.
I have heard some people use Evernote for this kind of work, but I'm not sure if there are drawbacks I should be aware of or if there are better/easier to use tools out there. I tried Devonthink during the dissertation and found it too overwhelming (but it's possible I just didn't give it enough time).
I'm using a Mac. The ability to work on my iPad as well as my desktop would be a plus but not necessary. Free is strongly preferred.
Evernote is good for notes and organization and saving information in one place and maybe less good for regrouping thoughts. You might try something like Trello. Which is basically your index cards in a browser.
posted by bluefly at 7:37 AM on November 12, 2014
posted by bluefly at 7:37 AM on November 12, 2014
Atlas.ti is made for organizing qualitative research. I didn't find it very complex to learn, though it's not as simple as index cards. You can get a copy with a student discount for $100 and I think a trial version is also available. I'd encourage you to check it out.
posted by reren at 7:40 AM on November 12, 2014 [1 favorite]
posted by reren at 7:40 AM on November 12, 2014 [1 favorite]
Wellll... Tinderbox feels very clunky and buggy and crashy and overwhelming at first. And I sometimes fear for my data. But, once you get used to all its quirks, it's pretty great, and it seems to be exactly what you're describing. It handles custom metadata in a structured and consistent way.
http://www.eastgate.com/Tinderbox/
posted by zeek321 at 7:41 AM on November 12, 2014 [1 favorite]
http://www.eastgate.com/Tinderbox/
posted by zeek321 at 7:41 AM on November 12, 2014 [1 favorite]
(Never heard of Atlas.ti. Watching the video, I think it might be better for you than Tinderbox.)
posted by zeek321 at 7:55 AM on November 12, 2014
posted by zeek321 at 7:55 AM on November 12, 2014
I played around with dedoose for qualitative data a few years ago. I'm not sure what the current state of that program might be, but that might be worth checking out.
posted by anthropophagous at 8:07 AM on November 12, 2014
posted by anthropophagous at 8:07 AM on November 12, 2014
Social science researcher here: Not a lot has changed in the past three years since this related AskMe question.
You'll see I also recommended Atlas.ti in that previous question, and it's still what I recommend. One thing that has improved tremendously is their training, including both online and in-person courses--it is great.
Atlas.ti was originally developed with grounded theory approaches in mind, but it works equally well with lots of different kinds of approaches to qualitative data analysis. It's absolutely worth the student license fee of $99.
posted by yellowcandy at 8:59 AM on November 12, 2014
You'll see I also recommended Atlas.ti in that previous question, and it's still what I recommend. One thing that has improved tremendously is their training, including both online and in-person courses--it is great.
Atlas.ti was originally developed with grounded theory approaches in mind, but it works equally well with lots of different kinds of approaches to qualitative data analysis. It's absolutely worth the student license fee of $99.
posted by yellowcandy at 8:59 AM on November 12, 2014
Dedoose and Atlas.ti were created for this. Atlas.ti is the classic, but Dedoose is less expensive and (I think) easier to work on for shared projects.
posted by The Elusive Architeuthis at 9:08 AM on November 12, 2014
posted by The Elusive Architeuthis at 9:08 AM on November 12, 2014
I clicked on this question thinking I probably wouldn't answer because everyone else would have already recommended NVIVO, so I'm very surprised to find that I'm the first person making this recommendation. In my experience in academic and for-profit research, NVIVO is the most commonly used software package for qualitative coding. So I'm thinking that all these recommendations might be either field-specific or generation-specific, or both.
posted by OrangeDisk at 11:38 AM on November 12, 2014 [1 favorite]
posted by OrangeDisk at 11:38 AM on November 12, 2014 [1 favorite]
I'm equally amazed that I'm only the second person saying "Nvivo". It's the only qualitative software I've ever seen or heard of anyone using. Never heard of any of the others mentioned above - I work in healthcare research but I've seen courses teaching Nvivo for sociologists and historians and it comes as standard on our university workstations so it's obviously pretty widely used outside of healthcare too.
It's pretty intuitive, there are books and courses out there which will teach you how to do all the gimmicky word-clouds and add-ons, but just for simple tagging you should be able to work it out for yourself.
posted by tinkletown at 12:38 PM on November 12, 2014
It's pretty intuitive, there are books and courses out there which will teach you how to do all the gimmicky word-clouds and add-ons, but just for simple tagging you should be able to work it out for yourself.
posted by tinkletown at 12:38 PM on November 12, 2014
Can you download and play around with both Atlas.ti and NVivo? I've worked with researchers using both of those software packages, and it seemed people clicked with either one or the other. Actually, when people could choose, they tended to go with Atlas.ti. (There are certain fields in which people are pretty much forced to use NVivo, and so that's the only qualitative analysis software they are aware of.)
Free, and works on a Mac: TAMS Analyzer. I've used this myself when I needed a lightweight tool. I found the way you code data in TAMS to be more intuitive than other software packages such as NVivo.
posted by needled at 1:59 PM on November 12, 2014
Free, and works on a Mac: TAMS Analyzer. I've used this myself when I needed a lightweight tool. I found the way you code data in TAMS to be more intuitive than other software packages such as NVivo.
posted by needled at 1:59 PM on November 12, 2014
I just wanted to chime in on Nvivo--it's perfectly fine. But I personally don't find it as intuitive as Atlas. I also think the output from Atlas is a lot better--easier to sort, easier to recategorize, and it's easier to build code categories and supercodes that make sense, rather than having to jury-rig it, as I've found you often have to do with Nvivo.
Plus, Nvivo is more expensive ($120/year vs. $95/2 years) for a student.
posted by yellowcandy at 2:16 PM on November 12, 2014
Plus, Nvivo is more expensive ($120/year vs. $95/2 years) for a student.
posted by yellowcandy at 2:16 PM on November 12, 2014
Response by poster: Just to clarify - I already have the dissertation and am done with school. So, the academic pricing isn't available to me.
posted by philosophygeek at 7:03 AM on November 13, 2014
posted by philosophygeek at 7:03 AM on November 13, 2014
This thread is closed to new comments.
posted by sevenless at 6:54 AM on November 12, 2014