Case Studies - Measuring the User Experience: Collecting, Analyzing, and Presenting Usability Metrics - page 272

product that participants found more or less engaging or frustrating, even if the
participants do not articulate their reaction.

Our Bentley DUC team partnered with Pearson Education, a textbook pub-
lisher that had recently moved into the digital textbook space on the iPad. We
focused this new technique on a usability study that would highlight any differ-
ences in arousal between digital and paper textbooks.

10.5.2 Methods
PARTICIPANTS
We recruited 10 undergraduates who owned and used iPads. Each 60-minute
session was one on one with a moderator from the Bentley Design and Usability
Center in the room with the participant.

TECHNOLOGY
To gather affective measurement and user feedback, we used two innovative
tools. Affectiva's Q Sensor was used to identify moments of increased arousal.
The words selected by participants using the Microsoft's Product Cards enabled
us to understand the direction (positive or negative) of their emotions.

The Affectiva Q Sensor is a wearable, wireless biosensor that measures emo-
tional arousal via skin conductance. The unit of measure is electrodermal activ-
ity, which increases when the user is in a state of excitement, attention, or anxiety
and reduces when the user experiences boredom or relaxation. Since EDA cap-
tures both cognitive load and stress ( Setz, Arnrich, Schumm, La Marca, Troster,
& Ehlert, 2009 ), we used this technology to accurately pinpoint moments of
user engagement with digital and printed textbooks. Affectiva's analysis software
provides markers used to indicate areas of interest in the data. Depending on the
study, areas of interest may include task start and end times. These markers can
be set during a study or post-test.

To better understand the emotions of the user, we utilized a toolkit devel-
oped by Microsoft called Microsoft Product Cards. These cards are given to users
to form the basis for discussion about a product ( Benedek & Miner, 2002 ). The
main advantage of this technique is that it does not rely on a questionnaire or
rating scales, and users do not have to generate words themselves. The 118 prod-
uct reaction cards targeted a 60% positive and 40% neutral balance. A study out
of Southern Polytechnic in Georgia found that cards encourage users to tell a
richer and more revealing description of their experiences ( Barnum & Palmer,
2010 ). This user feedback helped the DUC team assign specific emotions to the
Q Sensor's readings. Without these cards, we would have needed to make infer-
ences about the peaks and lulls found in the Q Sensor data.

PROCEDURE
Each session was structured as follows:

1. When participants arrived, we attached the Q Sensor biosensors to each

of their hands. Participants were asked to walk down the hallway and