Unified Web Evaluation Methodology for websites
Harmonization of evaluation
The Web site evaluation methodology 1.2 (latest version) includes:
- Evaluation guidelines (from within EOWG on Evaluating Web Sites Suite);
- Evaluation tests, intended to address what the evaluation-end-user needs to do when performing an evaluation, and what the evaluation tool developer needs to be aware of;
- User testing protocols will be introduced in version 2.0;
- Automated testing (tool independent), using prototypes of cluster projects.
This is not an exclusive list of what is meant by an "evaluation methodology". These are more focused on what the end user of an evaluation process needs, rather than what the developer of an evaluation tool needs (which is where the WCAG WG is most likely to focus in their test suite development). Such methodologies might be configured to evaluate one Web page, an entire Web site (irrespective of size), or multiple Web sites. The support EAM project has provided a document with necessary information for policy makers to ensure conformance of public sites (e.g. for public procurement and for certification).
Tests for evaluation
The tests for the methodology integrate non-technology specific requirements, technology specific sub-requirements, related questions and procedures. The WCAG 1.0 and 2.0 are the basis for the materials developed and results are input into WCAG WG.
The tests take the following into account:
- Specific to Web Accessibility Initiative (WAI) Documentation - All questions are drawn from current and future WAI Documentation.
- Tool and Browser Independent - Questions are published in a 'pure' form, not directly associated with a current tool or browser. However annotation with examples could be useful.
- Uniquely Interpretable - Questions have only one way of being interpreted, e.g., through simple language and if necessary a specific example.
- Repeatable - 2 different web accessibility evaluators can perform the relevant tests on the same site and get the same results.
- Translatable - The context of tests is optimized for possible translation into other official European languages.
Following is a non definitive list that is taken into account for the methodology:
- Statistical methods for sampling;
- Critical path analysis;
- Computer assisted content selection;
- Manual content selection;
- User testing (later version);
- Interpretation and integration/aggregation of results.
The methodology clearly includes references to the WAI (e.g., WCAG) document(s) it is based upon. It covers evaluations of one Web page, an entire site (irrespective of size), or multiple sites.
