On Oct 8, 2009, at 9:26 AM, James Page wrote:
Have a hypothesis before you conduct your test, and don't fish. But
surely this
applies to traditional lab testing as well. And is the reason why
Web Stats
are great as Hypothesis generating tool, but not a tool to test a
hypothesis.
James,
I believe it can work the other way as well. A lab test of a deployed
product may expose a source of friction that does not completely
prevent the observed users from completing their tasks. That
observation can lead to a hypothesis that the rate of abandons from a
particular page in a particular situation could be high. Analysis of
log data, or an A-B test against a revised design, can verify or
refute the hypothesis.
I like to say, somewhat simplistically, "Web stats tell you what,
usability studies tell you why." No matter which you discover first,
the quantitative "what" or the qualitative "why", you will sometimes
benefit from seeking to discover the other.
Larry Tesler
________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... [email protected]
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help
