textual user data: workflows? tools?

4 Nov 2013 - 4:41pm
2 years ago
3 replies
3987 reads

I'm fairly new to UX and very new to UX research.  I'm working on a research project involving hundreds of user interviews, so we will have several hundred textual transcripts to sort through and derive meaning from.  I haven't seen the data or interview questions yet, so I don't have a lot of information to share here.

Does anybody out there have suggestions for how to approach this?  Tools or methodologies you work with?  off the top of my head, I imagine tagging responses with keywords, then doing something with a spreadsheet, recording interview question numbers and corresponding keywords to start.

thanks for your input!


4 Nov 2013 - 6:28pm
Eva Miller
Frankly, I can't imagine too many situations where hundreds of people would need to be interviewed, especially if the interviews were qualitative in nature. Why did this happen?

It may be that you won't actually need to review all of that data. I'd try to identify a small set of "fruitful" interviews and analyze those first. Forget "demographics" or market-driven definitions and make a hypothesis based on common behaviors or information needs to identify the types of interviewees you want to focus on. Grab useful quotes from their transcripts, create an "elemental" meaning for each quote (these will start repeating pretty fast), then categorize more broadly by user issue or need (larger themes that drive your design work). Indi Young's Mental Models is a pretty good guide for a way to do this.

Even 6-12 people is often enough to get the insights you need. If you keep seeing the same things over and over in the smaller sample you analyze, you are probably done.

11 Nov 2013 - 4:26am
David More

If you are stuck with making sense of hundreds of interviews, there is software for the job - look for Nvivo or NUD.IST from QSR.

11 Nov 2013 - 1:41pm

You can write your own script to do some basic Nature Language Processing stuff.

Syndicate content Get the feed