Usability-based

Usabilty 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. It is also described as a set of attributes that bear on the effort needed for use, and on the individual assessment of such use, by a stated or implied set of users.

From the various definitions, a number of aspects emerge that play a role in usability:

• Effectiveness
  • Are users able to complete their task and achieve their goal with the system?
• Efficiency
  • How much trouble and time does it cost users to do this?
• Satisfaction
  • What do the users think of the ease of operation of the system?
• Ease of understanding
  • How easily does the user understand what the system expects him to input, and how understandable is the output to him?
• Ease of learning
  • How quick and easy is it to learn and remember how to operate the system?
• Attractiveness
  • How attractive does the user find the system, as regards e.g. layout, use of colour, graphics, film clips and interaction?
• Robustness
  • How easily can the users make mistakes in the system; how serious are these, and how easily can they be rectified?

Ways to test usability:

• Questionnaires
  • A means to request the users’ opinion of the system is using questionnaires. While they are also applicable to prototypes or even screen designs, questionnaires are mainly used when the system is ready, or even already in production. When the participants have completed enough questionnaires, an evaluation of the results follows. While it is a relatively cheap method of testing usability, the disadvantage is that the result will not deliver a particularly detailed impression of what is right and wrong in a system. SUMI (Software Usability Measurement Inventory, http://sumi.ucc.ie) and WAMMI (Website Analysis and Measurement Inventory, www.wammi.com) are methods that are based on the use of questionnaires.
• Checklist, interviews
  • A cheap way is the use of usability checklists during other (usually functional) tests, or interviewing the testers and users after working with the system concerning their experience of it.
• Tools
  • Tools are available, especially for web applications, that can carry out all kinds of checks. Examples of these checks are:
    • Are the graphics and animations provided with an alternative (a text box) for supplying the same information in the event that the graphics, animations, etc. are not working? This can be the case if you use a different browser, don’t have a video card or are visually handicapped.
  • Is the size of the graphics too big, making the site slow?
  • Does every page contain a link for returning to the previous page and/or a link for continuing on to the next page?
  • Are the text boxes perhaps too long in a scrolling field?
  • Are all the hyperlinks (still) valid?