Monday, November 19, 2007

Demystifying Data Analysis


If you're intrigued and sometimes frustrated by the analysis of qualitative data for a design project you may enjoy Rachel Hinman's short essay on the Adaptive Path site.

When you've collected your mountains of data:
So what do you do? What process do you follow? I’ve asked user experience industry folks these very questions. They invariably described their process as “deep thinking” and “marinating in the data” or “cranking it out.” Descriptive? Yes. Helpful? Not really.
She goes on to more helpfully explain a little about her process through:
  1. What we saw/heard
  2. What it means
  3. Why it matters
She recommends the use of simple techniques like Participant boards and a board of Emerging Insights. As with all data analysis don't be scared to get your thinking out big. Several walls and a bucket of Post-its come in handy. Making the data and emerging insights visible is especially important when you're working in teams (which hopefully you are).

Have a read for yourself: Demystifying data analysis

Needle in a haystack from ralphlosey

3 comments:

omar said...

marinating in the data! i love that. jono, your desk always seems covered with stuff. marinating might be a euphemism for being crushed!

on the teams comment: doesn't it get sort of strange that in this age of interdisciplinary work and more and more complicated projects that the output for a phd student, in the end, must be their own novel work. why can't two people, on a team, have a wonderful accomplishment together?

ir's funny actually. i know these identical twins who are doing a phd at the same institution, on the same topic. so far, all their papers are jointly authored by the twins (and some other people). they work better together!

anyway getting up off my observation soapbox.

Caoimhe Mc Mahon said...

Great post! We are about to jump into the initial analysis phase of our Masters project and it's a huge help to be able to find articles written on this topic which can sometimes seem like a 'black art'.Thanks!

Jono said...

Hi Caoimhe, thanks for taking the time to comment. Really it's Rachel who you should thank =)

If you haven't already you could do worse than check out the regular clustering advice in the textbooks:
Beyer, Hugh, and Karen Holtzblatt. 1998. Contextual Design: Defining Customer-Centered Systems. San Francisco, California, United States: Morgan Kaufmann.

and

Glesne, Corrine, and Alan Peshkin. 1992. Becoming Qualitative Researchers: An Introduction. White Plains, New York, United States: Longman.

and

Patton, Michael Quinn. 2002. Qualitative Research and Evaluation Methods. 3rd ed. Thousand Oaks, California, United States: Sage Publications Inc.
(from my bibliography)

Also, I took a look at your site and thought you may be interested in a project called Bounce by Irene Chien and Greg Niemeyer (among others). I found 2 links:
Bounce on Greg Niemeyer's homepage and a note about Bounce on the San Jose expo page
It's about connecting the old and the young.

And btw Omar - guess I never did reply, but I totally agree with your twin contribution thought. It's a little ridiculous. Fix it would you?