Evaluating accessibility by simulating the experiences of users with vision or motor impairments

J. Mankoff H. Fait R. Juang User studies involving users with disabilities often incur greater financial cost and complexity than those involving general populations. Consequently, accessibility issues may not be identified during the earlier phases of software design, when designs are still malleable. Additionally, it can be difficult to create controlled studies with multiple groups of very similar subjects due to the extremely heterogeneous nature of the impact of many motor and visual disabilities. This paper examines the feasibility of simulating the interaction experiences of users with low vision or motor impairments. Based on empirical studies in the literature of the impact of these impairments on the experience of computer use, we have developed EASE (Evaluating Accessibility through Simulation of User Experience), a tool that can help developers identify disability-related usability problems early in the design process. EASE can also be used to allow fine-grained exploration of user capabilities that are difficult to control, such as achievable typing speed. We present a study of the use of word prediction software that illustrates the value of fine-grained control over typing speed and that also shows word prediction is most useful at typing speeds between 5 and 8 words per minute.