Paper details

Accessibility metrics provide a valuable role in establishing the extent to which a web resource can be effectively used by people with specific access needs. Measuring accessibility should not be restricted to web pages; but rather the extent to which goals can be completed by disabled people relative to other members of the target audience. This can cover alternative equivalent routes, using alternative online resources if necessary or alternative access means (eg. screenreaders or screen magnifiers). This definition of course contains more subjective, experiential issues, that may modify the apparent impact of a specific accessibility barrier – but real life isn’t only black or white.

 

There are typically two approaches how to measure the web accessibility. “Conformance review” approach that rely on some checklist and “user testing” approach – some form of user testing by people who have different disabilities and/or who rely on different assistive technologies.

 

Conformance review is a typical way of evaluation of web sites accessibility. In general, this involves someone with expert knowledge checking whether the site as a whole, or more commonly a selection of pages, comply with a some guidelines (predetermined checklist of criteria) such as WCAG 2.0. This process is sometimes called “accessibility audit”.

 

User testing is usually carry out by a group of users with different disabilities, and different levels of skill in using the internet and their required assistive technology. These users undertake a series of typical website tasks. Their actions are observed (and sometimes recorded) by the evaluator with the aim of identifying the accessibility barriers that maybe encountered.

 

Both approaches have their strength and limitations, and neither can provide a reliable declaration about the accessibility of a site on its own.

 

At the very beginning of Blind Friendly Web project, we had relied only on guidelined approach to the accessibility. After some years, when we often had met limitations of this approach, we started to combine “guidelined” with “user-centered” approach. Now our typical web accessibility test contains results of both approaches. Accessibility expert evaluates web for conformance with some methodology/guidelines and group of 6 users with various visual impairment undertakes some (3 or 4) typical website tasks.

 

This combination brings more realistic view on web accessibility. Users evaluate the possibility to fullfil their needs and accessibility expert can find barriers, that users aren’t able to find – for example blind tester isn’t able to assesses sufficient contrast of text and background color.

 

Our testing model unfortunately has one weekness, that we would like to “fix” – ideal user group should cover all (or let’s say all major) group of users with disabilities – not only user with visual impairment. But on the other hand – it’s better to have one group of testers than none.