| dc.description.abstract | In
this
research
we
investigate
the
evaluation
of
usability
evaluations
methods
(UEMs).
In
particular
we
are
concerned
with
evaluating
their
suitability
for
the
evaluation
of
location-‐
aware
systems.
Not
all
approaches
for
the
evaluation
of
UEMs
have
been
extensively
validated
for
such
types
of
dynamic
interaction,
while
their
application
is
not
clearly
documented.
We
overview
the
strengths
of
the
current
approach
and
suggest
how
to
improve
them.
We
examine
navigation
systems
as
examples
for
issues
with
location-‐aware
systems
in
a
contextually
rich
environment.
The
setting
is
very
different
to
a
traditional
desktop-‐based
application.
Take
the
use
of
the
navigation
device
for
example.
It
is
a
secondary
task;
the
primary
task
is
to
safely
drive
the
car.
The
interface
is
continuously
changing
to
adapt
to
the
current
location
of
the
user.
The
user
navigates
in
a
complex
dynamic
environment
encompassing
various
stimuli
and
unpredictable
external
factors.
We
present
in
the
thesis
a
methodological
and
systematic
way
to
approach
the
evaluation
of
UEMs.
A
comparative
study
of
analytical
and
empirical
techniques
was
carried
out,
to
assess
them
in
identifying
usability
problems
within
both
static
and
dynamic
contexts
of
use.
Four
analytical
methods
(CW,
UAN,
EMU,
and
Design
Guidelines)
and
one
empirical
were
compared.
In
this
thesis,
we
validate
the
existing
classification
scheme
of
Blandford
et
al.
(2008)
and
highlight
relevant
issues.
We
present
an
alternative
systematic
approach
building
on
this
scheme
(CoHUM),
and
its
shortcomings
with
dynamic
systems.
We
show
how
a
rigorous
and
systematic
error
analysis
identifies
phenotypes
as
the
outcome
of
empirical
techniques,
whilst
genotypes
are
the
outcome
of
analytical
techniques.
Finally,
we
present
new
dimensions
that
previous
literature
had
not
identified
for
the
evaluation
of
UEMs.
This
research
will
help
future
researchers
by
providing
them
with
a
stronger
methodological
approach
for
comparing
UEMs
and,
in
particular,
categories
of
UEMs. | |
