Page 177 -

164 CHAPTER 7 Case studies

7.4.4 DEMONSTRATION
Usually shorter and less in-depth than descriptive case studies, demonstrations are
often found in papers describing new designs. Short anecdotes describing how one
or more individuals successfully used a new tool to complete one or more appropri-
ate tasks often complement usability studies, controlled experiments, and other data
documenting the success of the design.
Demonstration case studies can play an important role in describing the success
of a new design or tool, particularly when a controlled user study is inappropriate or
impractical. This is often the case with a complete tool, which may have many ele-
ments and multiple metrics for evaluation. Demonstration case studies can also be
appropriate in cases where the broad scope of the interface may preclude the use of
a controlled study.
Demonstration case studies tend to follow a common pattern. The report gener-
ally starts with an introduction of the participants and their context of use. Other
elements often found in the report include descriptions of how the participants used
the system, problems they faced, strengths of the system design, and discussions of
subjective responses. See the Interfaces for People with Quadriplegia sidebar for a
discussion of such a case study.

INTERFACES FOR PEOPLE WITH QUADRIPLEGIA
Building interfaces for quadriplegic people is a significant challenge: without
the use of their hands, fingers, or feet, these individuals may be restricted to
using input devices that consist of a single action, such as blowing on a straw
or pressing a single switch. Interfaces for such users are generally based on
some form of scanning: a graphical window on the computer screen contains a
grid of buttons that are scanned—highlighted in some predictable order—with
each button being active for a given amount of time. When the desired button is
highlighted, the user activates the switch to make a selection.
Steriadis and Constantinou (2003) include a demonstration case study as a
partial validation of the proposed design of a new interface architecture. This
paper presented widgets for single-switch input devices (“wifsids”) that support
a model of button selection appropriate for both discrete text input from the
keyboard and continuous mouse movement. These widgets were used to build
a set of applications that would run in Windows, supporting cursor movement,
keyboard entry, selection of applications, macros for common functionality, and
other features.
The case study described how the system was used by a 35-year-old
man with amyotrophic lateral sclerosis (ALS), which left him bedridden
and unable to speak. The paper describes how the participant decided to use
a button between his knees to make selections, after having rejected other
inputs as being difficult to use, cumbersome, or unattractive. The description
of the participant's success in learning how to use the system, and in using