Information Technology Reference
In-Depth Information
usability evaluations; and the
context of use
, which takes into account parameters such
as type of users, user age, work environment, etc.
3. Select the Web Artifacts to Be Evaluated.
The artifacts selected may depend on
either the Web development method or the technological platform. The artifacts to be
considered might be:
Platform-Independent models
(e.g., content/domain models,
navigational models, abstract user interface models) which are obtained as output
from the analysis and design stages of an MDWD process;
Platform-Specific models
(e.g., specific user interface models, database schemas) which are obtained as output
from the model transformation stage of an MDWD process; and
Code models
(e.g.,
source code, final user interfaces) which are obtained as output from the code genera-
tion stage of an MDWD process.
4. Select Usability Attributes.
The Web Usability Model is used as a catalog in order
to select which usability attributes will be evaluated. This Web Usability Model is
based on the decomposition of the usability characteristic proposed in the ISO 25000
(SQuaRE) [16] quality model. The first version of this model was presented in [11] and
has been improved by considering other usability guidelines (e.g., [18]), in order to
discover 15 new usability attributes that are relevant to the Web domain. The Web
Usability Model currently considers two different views: usability of the software
product, and usability in use. In this paper, we focus solely on the usability of the soft-
ware product, since it can be assessed during the Web development process by inspect-
ing Web artifacts. Usability from a software product perspective is decomposed into
seven sub-characteristics:
Appropriateness recognisability
,
Learnability
,
Ease of use
,
Helpfulness
,
Technical accessibility
,
Attractiveness
, and
Compliance
. These are
also decomposed into other sub-characteristics and measurable attributes. The Web
Usability Model, including all the sub-characteristics attributes and their associated
metrics, is available at
http://www.dsic.upv.es/~afernandez/CAiSE11/
WebUsabilityModel.
The outcomes of the above activities represent the
Evaluation Requirements
that
will be used as input by the next stage.
3.2 Specification of the Evaluation
The aim of this stage is to specify the evaluation in terms of which metrics are in-
tended to be applied and how the values obtained by these metrics allow usability
problems to be detected. The activities involved in this stage are described below.
1. Select the Metrics to Be Applied.
The Web Usability Model is used to discover
which of the metrics are associated with the usability attributes selected. Metrics al-
low us to interpret whether or not these attributes contribute to achieving a certain
degree of usability in the Web application. The Web metrics that are included in the
Web Usability Model were taken from different sources: surveys that contain metrics
that had been theoretically and empirically validated (e.g., Calero
et al.
[6]); quality
standards (e.g., SQuaRE [16]) and Web guidelines (e.g., W3C [26]). Each metric was
studied by considering its parameters (i.e., purpose, interpretation, artifacts to which it
can be applied, etc.) in order to provide a generic definition of the metric. The aim of
providing a generic definition is to allow metrics to be applied to artifacts of different
