Abstract
This paper describes several approaches for the evaluation of software-user interface. First, a software-user interface test (SUIT) was developed as an aid in the discrimination between acceptable and unacceptable interface. This test also provided a usability profile of software through the examination of component scales. The point-biserial correlation between the test items and the criterion was .8633 (p ≤ .005) indicating a high degree of predictive validity. Another methodology discussed involved the evaluation of software through the development of user interface models. A model of an interface was developed and ways of improving the interface efficiency were analyzed. A method was also presented for finding the optimal number of questions per panel given certain display and user characteristics.
Get full access to this article
View all access options for this article.