NIGEL BEVAN is Head of Usability Services at the National Physical Laboratory in the UK. He has first degrees in physics and psychology, and a PhD in man-machine interaction. He was technical co-ordinator of the European-funded MUSiC project which developed methods for measuring usability. These have been adopted by a wide range of companies, particularly in UK and Sweden. Nigel is now manager of the European INUSE and RESPECT projects which have set up a network of Usability Support Centres around Europe. The centres offer a comprehensive set of state-of-the-art methods and services for supporting usabilty throughout the design process. Nigel participates in several international standards groups, and is editor of ISO 9241-11 (Guidance on usability) and ISO 14598-1 (Evaluation of software quality - General guide), and contributed to the development of ISO 13407 (Human-centred design process for interactive systems).
The European Usability Support Centres are in the process of
extending
their range of services to include web design and evaluation, building
on
their existing expertise and methods. In line with our general experience
with usability, the top priority is to provide guidance, tools and
methods
to help organisations which do not already have expertise with usability
to
produce systems (web pages) which have minimum levels of usability.
Our
approach to usability is a broad one, where usability is synonymous
with
"quality in use", and is (should be) the major objective of systems (and
web page) design. Following ISO 9241-11 and ISO 14598-1, the
definition of
usability (also quality in use) is: The extent to which a product can be
used by specified users to achieve specified goals with effectiveness,
efficiency and satisfaction in a specified context of use.
For web pages, this means that the usability goals which provide the
basis
for evaluation need to be specified under the following headings:
CONTEXT OF USE
USERS: What are the skills, motivations and previous experience of
each
anticipated user group?
TASKS: For what identifyable purposes will each user group be visting
the
web site? What will be their tasks? How motivated will they be to
persist
in achieving their task objectives?
ENVIRONMENTS: For each user group, which web browsers will
be used, what
type of machines, with what type of screens and what speed of web
access?
What access will users have to assistance if they encounter problems?
Which combinations of users and tasks are most important for the
organisation producing the web site? (For example we carried out this
analysis for an organisatiation which had produced a web site
containing
comprehensive community and tourist information for their city, with
the
enthusiastic support of many local community groups. It turned out that
the business case for investing in this web site was to attract more
visitors to the city, yet much of the information was provided only in the
local language (Dutch), and the site had never been evaluated from the
perspective of a tourist!)
USABILITY CRITERIA
What is the relative importance of criteria for effectiveness, efficiency
and satisfaction?
EFFECTIVENESS: How many of the goals of the intended users can
the web site
support (eg how much of the information required by a potential tourist
is
available on the site?) Will a typical user accessing the site easily
locate all the information relevant to their goal?
EFFICIENCY: How much time and effort will be required to locate
the
required information?
SATISFACTION: How satisfied will the user be, and how much will
they enjoy
using the site?
It is important to prioritise these goals for each web site (potentially
for each user group).
We refer to the analysis described above as "Usability Context
Analysis",
and we have structured tools and methods to support the
comprehensive
analysis and documentation of these issues. We recommend that this
process
should take place prior to design, to elaborate the requirements from a
user centred perspective. It is equally important to obtain this
information prior to any evaluation, to ensure that the evaluation criteria
accurately map on to the intended context of use of the web site.
Given this background information, standard usability methods can be
used
for evaluation, depending on the time and resources available. Heuristic
evaluation can be used, but we strongly recommend evaluation by
representative users carrying out representative tasks (matching the
identified context of use). As with any development, this should start
early in design (see the recommendations below).
EVALUATION OF ESTABLISHED SITES
A product is being developed (by HFRG, Cork and Nomos
Management Ltd,
Sweden) to collect usability data on the use of existing websites, using
a
method called WAMMI (see the workshop application by Richard
Whitehand).
This will extend the experience gained with the psychometric SUMI
questionnaire to use in a web environment. Users will answer multiple
choice questions to assess the perceived usability of the web site. They
will also answer questions which classify them into one of the potential
user/task groups who are expected to use the site. It will then be
possible to compare the perceived usability of the site for different user
groups. This can be used to make the case for providing resources to
improve the site, and to compare the perceived usability before and
after
redesign.
DRAFT GUIDANCE FOR WEB SITE DESIGN
This is prposed minimumal guidance for web page developers. The
intention
is to alert web page developers to the essential usability activities that
will help ensure that they produce a usable site.
WEB USABILITY GUIDELINES
To be usable, web pages need to be:
- effective: enable users to find the required information
- efficient: enable information to be found quickly and without
unnecessary effort
- satisfying: motivate users to access the pages again, and give an
appropriate image of the organisation.
Poorly designed web pages frustrate users, mean that they cannot find
all
the information they require, and give a bad impression of your
organisation.
To produce a usable site it is essential to take a user centred approach
to design.
DESIGN PROCESS
To achieve usable pages requires a user centred design process (based
on
ISO DIS 13407):
1. Define the business objectives of the site:
- what is the purpose of the site and budget for producing it?
- how important is it that users find the site effective, efficient
and satisfying?
2. Identify responsibilities for usability within the design team, and
estimate the budget for usability activities.
3. Specify the intended context of use:
- who are the important users?
- what is their purpose for accessing the site?
- what type of information are they looking for?
- will they want to read it on the screen, print it or download it?
- what type of browsers will they use?
- how large a window will they use with how many colours?
- how fast will their communication links be?
4. Define some anticipated scenarios of use:
- describe some specific examples of people accessing the site, and
what they want to achieve. These will guide design, and can form a
basis for
evaluation.
5. Produce a card (or post it note) for each anticipated page for the
site, and use card sorting techniques to design an appropriate structure.
The structure should make sense to the user, and will often differ from
the
structure used internally by the data provider.
- for guidance on structure see:
http://info.med.yale.edu/caim/stylemanual/M_I_3.HTML
- for guidance on card sorting see:
http://www.sun.com/sun-on-net/uidesign/cardsort.html
6. Mock up a home page and a template for subsequent pages.
- for an example of home page design see:
http://www.sun.com/sun-on-net/uidesign/pagedesign.html
- for guidelines on page design see below.
7. Evaluate a partial mock up of the site with representative users
performing representative tasks
- use first drafts of screens either online or as colour prints
- for the evaluation process see:
http://www.sun.com/sun-on-net/uidesign/papertest.html
8. Produce a working version of a representative part of the site, taking
account of the design principles and evaluation feedback.
9. Evaluate a working version with representative users performing
representative tasks.
- for an example of using a usability lab see:
http://www.sun.com/sun-on-net/uidesign/screentest.html
Evaluations can take anything from a few minutes to a few hours,
depending
on the size of the site and the objectives. Users never behave as
expected! Any evaluation is better than none.
PAGE DESIGN
Some of the key issues are:
1. Design an effective home page
This should establish the site identity and major topics available to the
user. It should fit on one screen, as many users will not scroll.
Optimise the layout by user testing.
See: http://www.sun.com/sun-on-net/uidesign/pagedesign.html
2. Help users find their way
Show users where they are. Use a consistent page layout. Minimise the
need to scroll. For large sites, include Contents, Index and Search.
Include navigational buttons at both the top and bottom of the page.
Provide links to key locations such as: home, contents, other sections,
previous page and next page.
See: http://www.cybertech.apple.com/HI/web_design/find.html
3. Tell users what to expect
Provide a site map or overview. Distinguish between a contents list for
a page, and links to other pages. Give sizes of files which can be
downloaded.
See: http://www.cybertech.apple.com/HI/web_design/tell.html
4. Design for efficiency
Try to make sure users can get to useful information in less than three
clicks. Use small images, use interlaced images, repeat images where
possible.
See: http://www.cybertech.apple.com/HI/web_design/efficient.html
5. Support a range of users
Provide alternate text for images (use ALT tag). Use a maximum 640
pixel
width, or ??? pixels for pages which may be printed in portrait mode.
Avoid frames - use tables instead.
REFERENCE
For a more detailed understanding of the issues, it is worth reading
through:
- case studies: http://www.sun.com/sun-on-net/uidesign/
- principles: http://www.cybertech.apple.com/HI/web_design/intro.html
- design guidance: http://www.sun.com/styleguide
_______________________________________________________________
| Nigel Bevan | Email: Nigel.Bevan@npl.co.uk |
| Head of Usability Services | Tel: 0181-943 6993 |
| National Physical Laboratory | +44 181-943 6993 |
| Teddington, Middlesex | Fax: 0181-943 6306 |
| TW11 OLW, UK | +44 181-943 6306 |
| Web: http://www.npl.co.uk/npl/cise/us |
