Emerging Issues, Solutions & Challenges from the Top 20 Issues Affecting Web Application Accessibility
David Hoffman, Lisa Battle · 2005 · Proceedings of the 7th International ACM SIGACCESS Conference on Computers and Accessibility (Assets '05) · doi:10.1145/1090785.1090833
Summary
This poster paper presents the results of a second in-depth analysis of hundreds of accessibility issues documented across real web application projects at the U.S. Social Security Administration (SSA), compared with findings from a prior study of over 1,000 accessibility issues at the same organisation. Working as consultants to the SSA, the authors collected accessibility test results from more than 80 applications produced over a three-year period by multiple groups of accessibility specialists. Using an open card sort methodology, they categorised the issues, named user experience problem categories, and ranked them by frequency of occurrence and severity. The combined frequency and severity scores produced a ranked "Top 20 Accessibility Issues" list, with each issue mapped to the specific user groups affected and one or more potential solutions.
Key findings
The Top 20 accessibility issues as of May 2005, ranked by combined frequency and severity, were: (1) keyboard inaccessibility — users cannot access objects by keyboard; (2) non-HTML file and application challenges that can be insurmountable; (3) missing or broken field label associations; (4) lack of comparable text access to displayed information; (5) unclear link purpose without surrounding context; (6) missing field-level help and examples; (7) graphics not readable due to inability to change font or colour; (8) confusing, missing, or conflicting hot keys and browser commands; (9) users unaware of errors or unable to find them on the page; (10) focus not moving to the correct location; (11) poor or missing skip-to-content implementation; (12) graphics without text equivalents; (13) radio buttons and checkboxes not readable as grouped questions; (14) unclear table and grouped control context; (15) no visible focus indicator; (16) illogical tab order; (17) images as action poses problems for voice input users; (18) dynamic content changes; (19) unclear push button purpose; and (20) application timeouts during multi-tasking. The authors note that solutions for one user group sometimes compromise the experience for another, leading them to advocate for alternate views optimised for different user groups rather than a single universal design.
Relevance
This practitioner-oriented study is valuable because it draws from real-world accessibility testing across dozens of government web applications, providing an empirically grounded ranking of the most common and impactful accessibility problems rather than a theoretical checklist. Many of the Top 20 issues identified in 2005 — keyboard accessibility, missing labels, unclear link purpose, absent focus indicators, dynamic content challenges — remain among the most prevalent accessibility barriers found in web audits today. The finding that keyboard inaccessibility ranked as the single most frequent and severe issue reinforces its status as the foundational accessibility requirement. The authors' observation that some accessibility solutions for one user group can negatively affect another group foreshadows the ongoing tension between universal design ideals and the practical reality that personalised or alternate views may sometimes better serve diverse needs.
Tags: web accessibility · accessibility evaluation · accessibility barriers · design patterns · web applications · keyboard accessibility · screen readers