Uncovering the Role of Expectations on Perceived Web Accessibility
Amaia Aizpurua, Myriam Arrue, Markel Vigo · 2013 · Proceedings of the 15th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '13) · doi:10.1145/2513383.2513411
Summary
This short paper investigates how user expectations shape the perception of web accessibility for blind users, arguing that guideline compliance alone does not guarantee a satisfying user experience. The authors note a key gap: prior research found that only 50.4% of problems encountered by screen reader users were covered by WCAG 2.0 success criteria, with many problems relating to subjective factors like unmet expectations about content and functionality. The study recruited 11 legally blind screen reader users (4 female, average age 45, range 21-64) who navigated four restaurant websites with varying accessibility compliance levels (two higher, two lower) using JAWS on Internet Explorer or VoiceOver on Safari. Participants completed three tasks on each site — free navigation, finding gastronomic information, and finding menus/making reservations — followed by semi-structured interviews. They rated each site on a 7-point Likert scale for perceived accessibility. Thematic analysis of interview transcripts focused on how expectations influenced accessibility perceptions. Participants' expectations were shaped by both previous web browsing experiences and real-life experiences with similar businesses (restaurants).
Key findings
The study identified seven types of expectation-related factors affecting accessibility perceptions: content found where not expected, content not found where expected, content not found as expected, expectations on unaccessed content, expectations on content labeled as accessible, functionality not working as expected, and expected functionality not present. The most striking finding was that website W4, which had lower guideline compliance (containing Flash content, frames, and layout tables), received a median accessibility rating of 6 out of 7 — comparable to the more compliant sites W1 and W2. This appeared to be because W4's navigation menu consisted of 8 clear, concise text links that perfectly matched participants' expectations for restaurant content (history, location, menu, wines, specialities, bookings). Most participants were unaware of the Flash-based dynamic content and therefore did not perceive the associated accessibility issues. Conversely, W1 had higher compliance but received polarized ratings because it lacked a traditional menu — offering only pictures of dishes with rare names — which violated expectations. One participant noted: "nor menu not even anything, it's all messed up." Another participant created higher expectations for W1 specifically because she recognized it as an accessible version of the site, then judged it more harshly when content was missing.
Relevance
This paper challenges a foundational assumption in accessibility practice: that standards compliance equates to user satisfaction. The finding that a technically non-compliant website can be perceived as more accessible than a compliant one — simply because it meets user expectations about content and structure — has important implications for how we evaluate and prioritize accessibility work. For practitioners conducting user testing, the results suggest that capturing user expectations before testing (through pre-interaction interviews) is essential for interpreting results accurately. The study also highlights that perceived accessibility is shaped by factors beyond the web itself — users' real-life experiences with similar services influence what they expect to find online. This connects accessibility to broader UX principles: a website that is technically navigable by a screen reader but fails to provide expected content is not truly accessible in any meaningful sense. The work supports the argument that accessibility evaluation should combine automated compliance testing with qualitative user research.
Tags: web accessibility · blind users · screen readers · user experience · perceived accessibility · expectations · usability testing · WCAG · qualitative research
Standards referenced: WCAG 2.0