On Aug 25, 2009, at 6:18 PM, Nick Gould wrote:
Jared, you are selling Caroline's point short... how about this
scenario:
We report that test participants asked to locate the search box
looked in the upper right corner for it. They told us that this is
where they expected it to be and the eyetracking confirms that this
is where they looked for it.
But that's my point! Who cares if they looked in the upper right for
the search box? The real question that will make a difference in the
performance is, "Did the user find the search box?" Our studies show
that, while users expect the search to be in the upper right, they are
not hampered in any meaningful way if it's anywhere else prominent on
the page. (See http://is.gd/2zZM1.)
Now, maybe you reached for the wrong example here. If you want to try
again, I'm all for it.
But there is something we can learn from this example. Because you
were focusing on the granularity of "where does the user look" and not
"what does the user need", you were focusing on the wrong problem to
solve. Moving the search box from whereever it is to the upper right
would not solve any problem. The data you collected forced you to
focus on the wrong problem, which would commandeer resources to
generate a result that probably won't make the design any better.
I'm not sure which point of Caroline's you think I'm selling short.
Was it this one?
Jared seems to be focusing strongly on point 1. I somewhat
sympathise with
his point of view, in that I've not found that the eye-tracking
stuff adds
greatly to what I can find out from an ordinary observational test
without
eye-tracking.
Because I don't think I sold that point short at all!
So, yes, the ET lent further support to a talk aloud finding. For
some clients - rightly or wrongly - this strengthens their confidence
in the results. That's not razzle dazzle, it's just additional,
consistent feedback. They "said" this and they "did" this. Why is
it any different from reporting where they clicked?*
It's only consistent feedback because you decided to make it
consistent. You could just as easily say that those eye movements are
involuntary and not really indicative of how someone succeeds or fails
with the design.
As you (or someone in this thread) stated, eye tracking records
unconscious behaviors. Applying meaning to unconscious behavior is a
difficult road to go down, because the odds of applying the wrong
meanings are very high.
Sure, it's optional - I don't think anyone claims ET replaces talk
aloud or that it's even necessary for a good study. But it can be a
valid, additional tool (in the right hands) for helping clients to
feel comfortable about the research results.
Yah, you keep repeating this. Still waiting to see what that
additional value is.
If ET doesn't replace traditional stuff, why bother with it? The
traditional stuff is easier, cheaper, better understood, proven to
work, and more reliable.
Jared "The Proof is in The Pudding" Spool
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... disc...@ixda.org
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help
