Heuristic evaluation is a discount usability engineering method
for quick, cheap, and easy evaluation of a user interface
design.
Heuristic evaluation is the most popular of the usability
inspection methods. Heuristic evaluation is done as a systematic
inspection of a user interface design for usability. The goal
of heuristic evaluation is to find the usability problems
in the design so that they can be attended to as part of an
iterative design process. Heuristic evaluation involves having
a small set of evaluators examine the interface and judge
its compliance with recognized usability principles (the "heuristics").
What is Heuristic Evaluation?
Heuristic Evaluation (originally proposed by Nielsen and Molich,
1990) is a discount method for quick, cheap, and easy evaluation
of the user interface.
The process requires that a small set of testers (or "evaluators")
examine the interface, and judge its compliance with recognised
usability principles (the "heuristics"). The goal
is the identification of any usability issues so that they
can be addressed as part of an iterative design process.
Heuristic Evaluation is popular in Web development circles
because it requires few resources in terms of money, time
or expertise. So any developer can enjoy the benefits of usability
testing - not just those with thousands to spend on a professional
assessment. Heuristic Evaluation is characterised by:
- Small test scenarios that use paper mock-ups or screen
shots, which can easily be changed from one test situation
to the next
- An informal basis for assessment that doesn't require
psychologists
- A high success rate - so only a handful of testers are
needed
- A few key guidelines
How can I Use Heuristic Evaluation on my Site?
- Plan Your Evaluation
How will you test your interface? Heuristic Evaluation typically
employs one of the three main approaches:
- Develop a set of tasks and ask your evaluators
to carry them out.
Identify and test the tasks that are critical to your site's
success - you'll want all visitors to be able to perform
these - and any elements expected to cause difficulty for
your site visitors.
- Provide evaluators with the goals of the system,
and allow them to develop their own tasks.
An example goal might be "users should be able to find
out how much product x costs." Evaluators can then
break this goal down into appropriate tasks, and test each
in turn.
- Ask evaluators to assess your dialogue elements.
Ask evaluators to go through the interface a number of times
and examine and assess the efficacy of those elements of
your Website that contribute to a dialogue with your site
visitors.
Choosing which method to use will depend on you, the time
that you have available, and on your evaluators. For example,
if you were evaluating with young children, the most appropriate
method would be to develop a set of tasks and ask them to
carry them out. Children will find this much more achievable
than trying to develop their own tasks, or assessing your
Website elements without any obvious aims.
Choose your Evaluators
The more evaluators you use, the more usability problems you'll
reveal. However, studies on the subject have shown that the
benefit/cost ratio decreases at about five evaluators. So
who should these evaluators be?
- Those with experience - if you can find 5 evaluators
who are experts in software ergonomics, and in the field
in which the software is applied, a well-planned evaluation
program will typically find 81%-90% of usability problems
with your interface
- Those without experience - if you don't have 5 free experts
at your fingertips, don't worry. A student with no knowledge
of software ergonomics will find 22% to 29% of usability
problems.
Heuristic Evaluation is known to find more than 90% of usability
problems if it's performed by 3 to 5 experienced people...
but remember, one evaluator is better than none!
Heuristic Evaluation - a Step By Step Guide
- Review the Heuristics
Once you've decided which approach you'll take, and you've
selected your evaluators, you'll need to brief these people
on the ten heuristics you want them to assess your site
against.
Note: Examples of what they should look out for can be found
in the Useful Websites section at the end of this article.
Visibility of System Status
Probably the two most important things that site visitors
need to know are:
- Where am I?" and
- Where can I go next?".
So it's essential that your interface keeps users informed
about what's going on. To test this, your evaluators should
look for appropriate feedback within a reasonable time
following each user interaction.
For example, once a user clicks the 'Submit' button on
your order form, within a few seconds they'll require
feedback that tells them their order has been received.
This feedback might appear in the form of a separate page,
or popup, which also
contains a 'back to site' link indicating where the user
can go next.
2. Match Between the System and the Real World
The system should speak the users' language,
using words, phrases and concepts that are familiar to
the user, rather than system-oriented terms. Even though
you might use what is considered standard jargon for the
topic on which your site focuses, consider including a
further simplification or explanation of the words you've
used.
Your evaluators should make sure you've followed real-world
conventions, and that your information appears in a natural
and logical order. A real world concept applied on the
Web is the shopping cart. On many sites, you click once
to select an item (the equivalent of picking it up off
the shelf in a real store), click again to "add to
basket" (or place it in your trolley) and then a
third time to confirm your intention to buy (or move to
the checkout).
3. User Control and Freedom
Site visitors often choose system functions by mistake,
and will need a clearly marked "emergency exit"
to leave the unwanted page without having to go through
an extended dialogue. While there's a definite need for
order to exist in your site, a greater degree of user
control may be required to cater to the needs of more
experienced users.
Your evaluators should ensure that your site meets the
control requirements of both first-time and experienced
users. An example of a control element might be a "home"
button that appears on every page. It's a simple way to
let users feel in control of the system - they know they
can "go home" (or opt-out) at any stage in the
process.
4. Consistency and Standards
Users should not have to wonder whether different words,
situations, or actions mean the same thing. it's best
to follow the uniform and/or platform conventions to which
your users are accustomed.
If the user want to return to the main page then label
your link "Home" or "Homepage", rather
than some obscure reference.
5. Error Prevention
Even better than good error messages is a careful design
that prevents a problem from occurring in the first place.
The best way to avoid errors is to conduct testing, more
testing, and even more testing. However, if errors do
occur, try to provide user friendly messages in natural
language rather than code.
6. Recognition Rather than Recall
Make sure objects, actions, and options are highly visible.
Your site visitors shouldn't have to remember information
between different parts of their dialogue with your site.
Instructions for use of the system should be visible -
or at least easily retrievable - whenever your users need
them. This increases the chance that your visitors will
be able to recognise where they are, so they won't have
to retrace their path from the home page.
For example, if you create a Website with a lot of submenus,
then use a system that will let the users know what section
they are in at all times. You could do this by leaving
a breadcrumb trail, or maybe applying a color scheme that
differentiates the various sections.
7. Flexibility and Ease of Use
Accelerators, which may be unseen by the novice user,
can often speed up the interaction for the expert user,
and allow the system to cater to both types of visitors.
You might, for instance, allow users to tailor frequent
actions.
Take Amazon. They save the personal information that's
provided by customers upon purchase. Then, each time the
customer makes another purchase, they can retrieve their
information with a single click. In this way, Amazon provides
customers with a way to avoid filling in an extensive
form each time they buy a product at the store.
8. Aesthetic and Minimalist Design
Extraneous information on a page is a distraction and
a slow-down. Make rarely needed information accessible
via a link so that the details are available, but don't
interfere with the more relevant content.
If your "Contact Us" page contains a form, as
well as all your physical contact details such as address,
telephone number etc., there's no need to also include
a map with extensive instructions on how to get to your
premises. Instead, this can be provided on a linked but
separate page - not everyone who fills out the contact
form will wish to see the map every time they visit the
page.
9. Help Users Recognise, Diagnose, and Recover
from Errors
Errors will occur despite all your efforts to prevent
them. Your error messages should be expressed in plain
language with no codes or jargon. They should detail the
problem, and constructively suggest a solution.
For example, if a form is completed incorrectly, your
error message should alert your visitor to this, identify
which fields will need to be refilled, and perhaps highlight
those fields when the user returns to complete the form
after they dismiss the error message.
10. Help and Documentation
Ideally, every online system could be used without documentation,
However, it may be necessary to provide help and documentation
to cater to the needs of all users, and be on the safe
side.
Your evaluators should check to make sure that help and
other documentation:
> is easy to search
> focuses on the user's task
> lists the concrete steps users need to carry out
to achieve their goals
> isn't too large
Often, documentation is fully integrated into a Website.
There should be links from the main help sections into
specific subsections, and vice versa. Help could even
be fully integrated into each page so that users never
feel like assistance is too far away.
Conducting the Evaluation
Conduct the evaluation using either of these methods:
Individual Evaluation - each evaluator reviews
the interface individually and reports problems to you. Individual
evaluation is easily conducted over the Internet. It will
pick up more problems than group evaluation, but takes a lot
more time to complete.
Group Evaluation - evaluators review the
interface as a team, while you record the problems. Evaluators
do not have to agree on a problem - but every issue they identify
should be recorded. Group evaluation requires more planning
than does individual evaluation, as all evaluators need to
be assembled, however, the evaluation need only be conducted
once as all the evaluators can complete their tasks at the
same time.
The most basic form of evaluation is to choose a random page
from your site and see if your evaluators can:
> tell where they are
> tell what they can/should do next
Analysing your Results
Once your evaluators have:
> worked their way through the tasks or goals you set,
> evaluated each of these in light of the ten heuristics,
and
> provided their feedback,
you'll need to compile all the information. Remove any duplicates
and combine similar issues. What's left will be a set of problems
or comments that you can address to improve your site's usability.
Remember the Golden Rule!
The golden rule of usability is:
There's no such thing as a "user error"!
Make sure you clarify every problem your evaluators identify
- ask questions so that you understand the specific nature
of the difficulties they encountered. And remember:
> don't argue with your evaluators,
> don't try to "explain away" problems they identified
If an evaluator found an aspect of your site confusing, then
it's more than likely that your Website visitors might have
problems with it too.
The following 0 to 4 rating scale can be used to rate the
severity of usability problems:
0 = I don't agree that this is a usability problem
at all
1 = Cosmetic problem only: need not be fixed unless extra
time is available on project
2 = Minor usability problem: fixing this should be given low
priority
3 = Major usability problem: important to fix, so should be
given high priority
4 = Usability catastrophe: imperative to fix this before product
can be released
|
