5

I do a lot of qualitative coding and assessment as part of my research activities. Usually my colleagues and I simply use generic tools for this process like text editors and spreadsheets, but I know that there's a lot of software out there (NVivo, Atlas.ti, etc.) that is designed for making qualitative data analysis "better".

If these pieces of software were inexpensive, I'd probably buy a package and try it out, but they're very costly (about $600 per seat) and appear rather intimidating.

Can anyone with experience in qualitative data analysis comment on whether these tools are useful (or not useful) and are worth the high price? (And you get a cookie if you can describe a use case that really highlights why the tool is powerful!)

Irwin
  • 8,847
  • 4
  • 33
  • 41
  • 3
    High price or free, the tools are scientifically useless if you don't know what they actually do... If you use it as a dumb tool, then the inferences you glean from its output are probably going to be unreliable as well... –  Apr 25 '13 at 20:59
  • 2
    I don't know any of these software. I looked at their webpages, read it, and still don't understand what they do. I don't know what is “qualitative data analysis” but it looks like I'm not doing it :) – F'x Apr 25 '13 at 21:02
  • @JohnQ.Public I understand that, but it's also valuable to learn about what tools are there that might make your life easier. As F'x mentioned, those web pages are obscure. And I'm doing qualitative data analysis! I figured I'd ask here just in case someone with experience knows and can explain their benefits. – Irwin Apr 25 '13 at 21:10
  • I hope that by making data "better" you mean visually and not by arbitrary transformations that make the data become what you want it to become. – Bitwise Apr 26 '13 at 03:21
  • Absolutely not. I said in my question that I'm wondering if these tools make the analysis better, as in easier to handle, easier to see the patterns, etc. – Irwin Apr 26 '13 at 05:00
  • I think this question might better belong at http://scicomp.stackexchange.com/ – Ben Norris May 01 '13 at 01:11
  • 1
    @BenNorris while it might fit someplace else, I think it is a fine fit here. – StrongBad May 01 '13 at 08:34
  • I posted it here because qualitative data analysis isn't a computational science activity. This might fit with ux.stackexchange.com though but they are more about visual design rather than HCI research methods. – Irwin May 02 '13 at 00:33
  • This seems like a 'tools for reasearch' question. Tools are generally on-topic, as discussed here. – earthling May 31 '13 at 11:23
  • 1
    You should also check out the R Qualitative Data Analysis package (RQDA), which is free and open source and uses R, which has a huge support community... Also, re: moving the question: if it were to be migrated anywhere else, I'd think stats.SE would be the place - it's a data analysis question, after all. – naught101 Sep 08 '15 at 01:17

3 Answers3

2

My answer might not be as useful to you because it only deals with my personal experience. I often deal with mixed methods work and qualitative data analysis is something I do about ~25% of my time.

I have used Atlas.ti because it is provided free on my school server. However, realistically, I used Excel to code themes, categories etc. following a grounded theory approach (Glaser and Strauss, 1967). I don't find much use for it but I know significant numbers of professors and graduate students who swear by it. It seems to be the SPSS of the qualitative world.

One powerful use of Atlas.ti that I can attest to is organization, search and visualization of categories, codes, themes and performing some simple reliability measures (Cronbach, Krippendorf) etc.

However, the same can also be done in Excel with some elbow grease. :)

Shion
  • 9,582
  • 1
  • 42
  • 61
1

The tools can be useful, but note they are designed to help you manage and 'tag' qualitative data: you still have to do the analysis yourself.

A typical use case would be where you have transcripts from dozens of interviews. The software will help you go through and find quotes about particular topics, so you can quickly find them all later. Then you can read all the quotes on a topic, and see if there is consensus or difference of opinion, or if certain respondents (eg. from one area, or over a certain age) differ.

While some of the tools are expensive, other qualitative data analysis software like Quirkos are quite cheap, but they all have free trials so you can see if they will help you. You can find independent reviews from the University of Surrey. There are also open-souce tools like Transana and Taguette.

Daniel Turner
  • 31
  • 3
0

I've used DeDoose in the past and found it helpful as there is a feature that can easily tell you of some codes are more or less associated with others. Although it's probably feasible to do this broadly, using a visual assessment, getting a clear confirmation was nice.

Another upside of DeDoose is that you can import audio/video and code these types of media.

I thought it was greatly helpful to code interviews and fairly straightforward textual data. I'm currently trying to coerce it into coding tables from Excel and it just won't do what I want (which is how I ran into your question), so it definitely is not perfect!

What I find helpful is leveraging the free month trial on some of these applications to see if I can actually import the data that I want to analyze.

curious
  • 1,020
  • 1
  • 12
  • 18