ASSESSING THE PROGRESS OF IMPLEMENTING WEB ACCESSIBILITY - An Irish Case Study

Abstract

Abstract: In this paper we attempt to gauge the implementation of web accessibility guidelines in a range of Irish websites by undertaking a follow-up study in 2005 to one conducted by McMullin three years earlier (McMullin, 2002). Automatic testing against version 1.0 of the Web Content Accessibility Guidelines (WCAG 1.0) using WebXact online revealed that accessibility levels had increased among the 152 sites sampled over the three-year period. Compliancy levels of A, AA and AAA had risen from the 2002 levels of 6.3%, 0 % and 0 % respectively to 36.2%, 8.6 % and 3.3 % in 2005. However, manual checks on the same sites indicated that the actual compliance levels for 2005 were 1.3%, 0 % and 0 % for A, AA and AAA. Of the sites claiming accessibility, either by displaying a W3C or ‘Bobby ’ compliance logo, or in text on their accessibility statement page, 60 % claimed a higher level than the automatic testing results indicated. When these sites were further manually checked it was found that all of them claimed a higher level of accessibility compliance than was actually the case. As most sites in the sample were not compliant with the WCAG 1.0 for the entire set of disabilities, the concept of ‘partial accessibility ’ was examined by identifying those websites that complied with subsets of the guidelines particular to different disabilities. Some disability types fared worse than others. In particular blindness, mobility impairment and cognitive impairment each had full support from at most 1 % of the websites in the study. Other disabilities were better supported, including partially-sighted, deaf and hearing impaired, and colour blind, where compliance was found in 11%, 23 % and 32 % of the websites, respectively. 1

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 2 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.