Key points are not available for this paper at this time.
Usability is an important aspect of any interactive system, but only one aspect leading to its success. Thus, when evaluating a system, it is useful to have short questionnaires that can lead to reliable, sensitive and valid measures of usability. In this commentary, I discuss two papers that aim to develop two such measures: one is for a four-item usability metric and the other a single-item usability. Whilst care is taken in both cases to produce suitable metrics, the processes have flaws that influence the value of these metrics. The underlying problem seems to lie with the application of psychometric methods when there are better methods available in this context. Even allowing for this, in the end there remains the problem of construct validity that undermines any such efforts.
Paul Cairns (Wed,) studied this question.