[Sigia-l] card sorting by survey?

[Todd R.Warfel]

Well, I think Whitney hit most of the highlights on Card Sorting.

There are several different methods for performing Card Sorts, but they 
all have a common goal - you're looking for patterns. I've tried 
several different methods and typically prefer open card sorts with 5 
groups of 3 people each (15 total participants). I've found that this 
does two things:

1) Yields a large enough sample size to establish patterns for 
structure.

2) Groups of users tend to be able to help each other out on items that 
are ambiguous, whereas individuals may get stuck. Additionally, they 
tend to help each other figure out where things should go when 
individuals may get stuck. This is why I'm not much of a fan of things 
like EZSort, or simply giving an individual an Excel file to sort by 
themselves.

Another good point was brought up by Donna - by giving the user an 
application to do the sorting in, they can tend to get hung up on the 
application/environment instead of the actual task. It can be 
distracting.

Again, as Whitney pointed out, there's no silver bullet. There are 
several things you can do to cut down on the time, need of a conference 
room, and still yield valuable results. Here's what I would do:

Print the items to be sorted on Avery labels - 5160 works well. Place 
each item on an index card. Shuffle the cards and place a number on the 
back (can help reduce input time after the first sort). Pull a group 
together at someone's desk, or in the cafeteria at a table. Give them 
30 minutes, provide them some simple instructions, and let them go at 
it. Record the results and repeat until finished. If you can be there, 
either in person or remotely, you can pick up on behavior patterns 
better.

In my experience, this can take less time than setting things up in 
EZSort or an application. But it is still time consuming. I recently 
had to work on getting a client to buy-into doing this very thing. It 
was a bit of a challenge, as we're under tight time constraints. 
However, when explaining the benefit and that we'd spend less time in 
the end figuring out the navigation than passing it back and forth 
among the key stakeholders and letting them "suggest changes," they 
bought in w/o too much extra effort.


On Oct 24, 2003, at 2:21 PM, Whitney Quesenbery wrote:

> Use remote usability methods (you watch them work, and talk on the 
> phone during the session)
>
> Debrief afterwards in a follow-up call with a random subset of your 
> participants (or all of them) so you can get at the reasons.
>
> Or,  go for large numbers of participants and look for quantitative 
> significance.

Cheers!

Todd R. Warfel
User Experience Architect
MessageFirst | making products easier to use
--------------------------------------
Contact Info
voice: 	(607) 339-9640
email: 	twarfel at messagefirst.com
web: 	www.messagefirst.com
aim: 	twarfel at mac.com
--------------------------------------
In theory, theory and practice are the same.
In practice, they are not.