[Sigia-l] Usability Test Questions
Dan Linsky
danl at davidandgoliath.com
Thu Aug 12 16:45:27 EDT 2004
Hey Eric. I read your book - and all of the other books on the subject.
I am familiar with many of the techniques that have been suggested.
In this case, the participants will not know what they are testing until
they get here. All they know is its worth $75.
This is ad agency usability, not development firm usability. Basically
there is a disagreement between the client and the agency as to the
usability of a certain navigation model. We would like to prove that it
is useable without leading the user while they are thinking the tests
will prove otherwise.
Kind of a bastardization of the discipline, but necessary nonetheless.
I am trying to develop some navigation-based tasks but getting into the
site to find content is good for maybe 2 out of 10. This is why I am
looking for more "creative" suggestions.
- Dan
Dan Linsky
Information Architect
208 N. Market St., Suite 500
Dallas, TX 75202
(469) 227-3158
-----Original Message-----
From: Eric Reiss [mailto:elr at e-reiss.com]
Sent: Thursday, August 12, 2004 3:29 PM
To: Thomas.Donehower at eurorscg.com; Dan Linsky
Cc: sigia-l at asis.org; sigia-l-admin at asis.org
Subject: Re: [Sigia-l] Usability Test Questions
Thomas Donehower wrote:
Ask questions that will [snip] give a clear indication as to what
content they represent. Some examples (Looking at homepage): What do
you think this site is about? What do you think it is for?
And I write:
I've asked test subjects the same open-ended questions...but...
Isn't it fair to assume that most real-life visitors already have
some kind of expectation of the site's subject and purpose before
they arrive?
Your questions are valid, but I wonder if they shouldn't be phrased
differently. After all, what would be gained by plunking someone in
front of a house and asking "Who do you think lives here?"
An alternative question might be, "Do you think the person you want
to visit lives here?"
Hence, open-ended questions are fine, but the test subject must have
some predetermined goal if the questions are to have any real value.
Cheers,
Eric
e-reiss.com
copenhagen, denmark
More information about the Sigia-l
mailing list