next up previous
Next: Evaluation Up: User Testing Previous: Post-test Analysis

Usability Inspection

An alternative to user testing is usability inspection . A number of experts in the field of user interface design methodically work through an application and evaluate its usability. There are a number of methods for evaluating usability. For example, it can be judged using a set of heuristics: a set of ideals for user interface designs, or against a cognitive model of the way that people interact with computer programs. The experts are experienced in the design of user interfaces, and can reliably find a large proportion of the major and minor problems in a user interface's design.

Unfortunately, these experts do not come cheap. A cost of US$500 to US$1000 per evaluator is typical . While this approach works best with interface evaluation experts, either software engineers or users who are knowledgeable in the application's target field, can be substituted for the experts. The non-experts are introduced to and taught about user interface evaluation, then allowed to evaluate the system.

While not using experts reduces the cost, the benefits gained are also fewer. From the studies done by Desurvire , it is seen that the proportion of problems in a system that experts found was five to ten times more than non-experts found. This finding was consistent across all different types of usability inspection methods. The experience that experts have with usability guidelines, user testing and the design of user interfaces leads to more effective usability testing and reporting of problems with user interfaces.

Formal methods for evaluating user interfaces, while they may be successful in finding user interface flaws, are still restricted by the fact that they do not use real users. The interfaces are evaluated by an expert using a set of ideals or an approximate psychological model of humans. The other problem with experienced evaluators is that, while they may by skilled in evaluating user interfaces, they may not know much about the field an application is targeted for and skills that people in that field have, for example: how much experience they have in using computers. Nielsen and Mack  state that usability inspections, having experts analyse the interface, is not yet a substitute for user testing with real users. There is no completely accurate model of the human mind predicting how people will think and react when using a user interface. It is hard to predict what a real user will actually do when faced with a given situation. Even the reactions across a number of real users, all faced with the same situation, varies. Because of this, a common approach in an application's design cycle is to initially do a usability inspection first, then, after revising the system and addressing the issues found, doing full user testing.

Nielsen and Mack  point out that when choosing to do either no testing at all or some testing - no matter how limited in scope, formality, number of participants, or skill of the testers - the limited testing is still better than none at all. You are still going to get some insights into the usability of an interface and get suggestions and ideas on how it could be improved.

A true user interface analysis was approximated by a self-evaluation against the ten usability guidelines that Nielsen  proposes, and were discussed in Section [*]. The results of this are presented in Section 6.2.


next up previous
Next: Evaluation Up: User Testing Previous: Post-test Analysis
Steve Smithies
1999-11-13