Integrating Manual and Automatic Evaluations to Measure Accessibility Barriers

Explicit syntax and implicit semantics of Web coding are typically addressed as distinct dominions in providing metrics for content accessibility. A more down-to-earth portrait about barriers and their impact on users with disabilities could be obtained whether any quantitative synthesis about number and size of barriers integrated measurements from automatic checks and human assessments. In this work, we present a metric to evaluate accessibility as a unique meas-ure of both syntax correctness and semantic consistence, according to some general assumptions about relationship and dependencies between them. WCAG 2.0 guidelines are used to define boundaries for any single barrier eval-uation, either from a syntactic point of view, or a subjective/human one. In or-der to assess our metric, gathered data form a large scale accessibility monitor has been utilized.