Page 276 -

10.3 How does usability testing relate to “traditional” research? 265

on the right, which would seem to continue to the next screen. In reality, clicking the
yellow arrow causes the user to upgrade their seat. To continue without an upgrade,
the user needs to click on the textual link on the left, which is small (in comparison to
the arrow) and not obvious. This is a confusing and potentially misleading interface
flaw (whether it was intentionally misleading is a question that we will not address).
This is a very minor flaw to change. However, it will have a major improvement on
user interaction and performance.
The range of usability testing is quite broad. Usability testing can involve hundreds
of users, have a number of controls, and use a true experimental design. Usability
testing can also involve a researcher sitting down next to three users, watching them
go through the interface, and then taking basic notes on where the problems are.
While both of these exercises can be called usability testing, it is more likely that the
former would be considered research and would be published. Usability testing can
involve hypothesis testing, tight controls, control groups, and a large enough number
of participants to determine statistically significant differences. However, that’s not
the way that most usability testing happens (Rubin and Chisnell, 2008). Why? In
industry, the extra time needed to plan controls and do random assignments, and the
high number of participants needed, are often a barrier to entry (Rubin and Chisnell,
2008). If the choice is that you must do all of those or nothing at all, businesses often
choose to do nothing. Therefore, more flexible, easier, and quicker methods are often
used. Where does usability testing end and research begin? It’s an unclear, fuzzy line
and the distinction is not all that important.

10.3 HOW DOES USABILITY TESTING RELATE
TO “TRADITIONAL” RESEARCH?
Usability testing can be considered a close cousin of traditional research methods,
and is often known as “user research.” In reality, the approaches utilized in usability
testing are often the same as those used in classic research. Metrics utilized in usabil-
ity testing include measurement of task performance and time performance, similar
to experimental design. Methods utilized as part of usability testing include surveys
to measure user satisfaction. Observation techniques, from ethnography, are often
utilized in usability testing. Key logging and clickstream analysis (see Chapters 12
and 13) can be utilized in usability testing. As discussed in other chapters of this book
(primarily Chapters 12 and 13), there are many automated data collection methods
that could be used for usability testing. In usability testing, the rights that participants
have are the same as in any other type of research. The names of the participants must
remain anonymous, participants must be informed of their rights and sign some type
of informed consent form, and participants have the right to leave the research at any
time, just as in traditional research.
However, usability testing often has different end goals. Usability testing is pri-
marily an industrial approach to improving user interfaces. As an industrial approach,
there is little concern for using only one research method or having strict controls.
In fact, Wixon goes as far as to say that usability testing has more in common with