Abstract. The majority of research into web accessibility has focused on identifying and eliminating the problems that people with disabilities encounter when interacting with the Web. In this paper we argue that we need to move away from studying user problems to studying how people with disabilities apply interaction strategies while browsing the Web. In this paper we present a study of 19 print disabled users, including blind, partially sighted and dyslexic people, interacting with a variety of interactive Web 2.0 web applications. The participants undertook tasks using concurrent and retrospective protocols to elicit information about how they interact with web content. The result of this study was a collection of 586 strategic action sequences that were classified into seven different types of strategy. Differences in the application of strategies between the user groups are presented, as well as the most frequent strategies used by each user group. We close the paper by discussing some implications for the design of websites and assistive technologies as well as the future directions for empirical research in accessibility.Keywords: Web accessibility, user study, user strategy, print disabled Web users, blind Web users, partially sighted Web users, dyslexic Web users.
IntroductionThe Web provides many opportunities for users to take part in work and leisure activities online. The evolution of Web 2.0, in which websites allow users to participate in creating, contribute and share content [9], makes our connections to both other individuals and organizations stronger in the virtual world. It is important that all users be able to participate equally in these activities, including people with disabilities who use different assistive technologies and interact with the Web in very different ways 668 C. Power et al.from mainstream users. However, recent studies show that people with disabilities still encounter large numbers of problems on the Web [1,7,12,14]. While existing web accessibility good practice covers some of those problems encountered by disabled users, there are still a substantial number of reported problems that have no clear solution. What are the design principles that can help close that gap and better support users with disabilities? When trying to address problems encountered by mainstream users on the Web, researchers and practitioners fall back on empirical research and design principles about user interactions with the Web and other interactive systems. However, when trying to address problems encountered by disabled users, relevant empirical studies about user interactions are rare. As a result, we are unable to go back to "first principles" when trying to either remove or, at least, reduce the impact that problems have on users with disabilities. We need more empirical work eliciting and understanding interaction strategies that disabled users apply when interacting with the Web. With that information, we can construct a framework for analysing why disabled users encounter the problems ...