Page 274 -

CHAPTER
10
Usability testing

10.1 INTRODUCTION
Usability testing is often known as “user research.” Often, in usability testing,
we’re not researching the user, we’re researching the interface. We’re trying to
figure out how to make a specific interface better. However, user research is also
a broader term that may include elements of design and development, such as
personas, user profiles, card sorting, and competitive research that generally
might not be considered “research” by those who consider themselves researchers
(Kuniavsky, 2003). Furthermore, usability testing as a research method (utilizing
representative users and representative tasks) can be used to learn more about how
people interact with interfaces, even when the goal is not fixing the interface, but
instead learning more about users and interactions. So, in usability testing, maybe
we are researching the user?

10.2 WHAT IS USABILITY TESTING?
Usability testing, in general, involves representative users attempting representa-
tive tasks in representative environments, on early prototypes or working versions
of computer interfaces (Lewis, 2006). If that sounds like a very broad definition, it is
meant to be that way. The world of usability testing includes:
• testing prototypes that have only been built on paper (known as paper
prototypes);
• testing screen mock-ups or wireframes which have no functionality
• testing screen layouts which have partial functionality
• testing prototypes that look complete but have a human behind the scenes
responding (known as the “Wizard of Oz” technique);
• testing working versions of software before it is officially released;
• testing software that has already been implemented in existing systems.
The interfaces being usability tested are typically screen layouts for desktop, lap-
top, or tablet computers, as well as smart phones and other mobile devices. Usability
testing can also be done to evaluate physical interaction with devices. Mobile devices
frequently need usability testing, since the interaction approaches (such as multi-
touch screens) are newer, more content is stuffed into a smaller screen size, and it can

Research Methods in Human-Computer Interaction. http://dx.doi.org/10.1016/B978-0-12-805390-4.00010-8 263
© 2017 Elsevier Inc. All rights reserved.