[Sigia-l] Straw poll re: card sorting

[Todd Warfel]

Billie,

They are better for qualitative than quantitative, typically.  
However, they can be used to gather quantitative data. We worked on a  
large financial intranet several years ago and one of the internal  
exercises they were running was closed card sorting (to add to an  
existing structure) with around 50 participants. They did use cluster  
analysis for doing the final crunching (each card had a number on the  
back, which was logged into the excel file that was in turn run  
through cluster analysis).

One of the great things about card sorting, however, is the actual  
conversations that are had during the exercise. The dialogue between  
participants (we typically run five groups with three participants  
each - 15 participants total), provides a great deal of insight into  
their mental models and expectations - this is something that you  
typically won't see in the raw "data".

On Aug 18, 2005, at 2:37 PM, Billie Mandel wrote:

> Our esteemed colleagues Lou R and Peter Mo suggest that card sorts  
> are better for
> qualitative than quantitative analysis,


Cheers!

Todd R. Warfel
Design & Usability Specialist
--------------------------------------
Contact Info
Email:    twarfel at mac.com
AIM:       twarfel at mac.com
Blog:      http://toddwarfel.com
--------------------------------------