The final BITV-Test offers website owners an assessment of the state of accessibility of their new or re-launched website.
Important note (March 15, 2019): The test procedure has been updated. This content is currently being revised.
The final BITV-Test is based upon the Barrierefreie Informationstechnik Verordnung (BITV) 2.0, a directive which mandates accessibility for all internet sites of German federal administrations.
The scope of the website to be included in the final test is defined in co-operation with the website owner. Usually, it is the entire website including historical sections and sections that contain documents for download, for example, in PDF format.
The page sampling is carried out independent of the website owner. Evaluators determine the total number of pages in the sample with regard to the complexity of the site and the number of different templates used. If present, pages with data tables, search results or multimedia content as well as integrated functions and services should be included (e.g., user comments or local search).
When testing web applications, all relevant aspects of selected processes must be included in the page sample. Evaluators include dynamically displayed content as additional states during page sampling. They describe the states and document the way they can be called up. Checkpoint ratings and comments pertaining to dynamic states of a page should clearly reference the respective state.
If any aspect of a web site is to be excluded, this has to be clearly stated in any publication of the test results.
The final BITV-Test is a tandem test, in other words, two qualified evaluators test independent of each other and harmonise their results only once they have finished their respective test runs.
Tandem testing increases the reliability of test results. Sometimes, one evaluator will have spotted something that the second evaluator has failed to notice. In addition, there are sometimes differences regarding the assessment of aspects of implementation that can be sorted out in the arbitration phase.
The more comprehensive tandem test may sometimes unearth more accessibility issues than preceding design support tests of the same site, leading to a worse result.
The basis of the assessment is the procedure laid out in the inventory of test steps. Some of the 49 checkpoints refer not to individual pages but to the entire site.
The weighing of individual checkpoints reflects the importance of the respective requirement for the accessibility of the site.
Faced with serious accessibilty problems that would make the site inaccessible for particular types of users, the BITV-Test provides the option to mark down the total test result to "badly accessible".
For the overall assessment of a site, the weighted results of the 49 checkpoints are added. The maximum score is 100 points. Sites that achieve 90 points or more are considered to have good accessibility. A site with a result of 95 points or more is considered very accessible.
The client receives a detailed PDF test report that documents ratings and comments per checkpoint and page. The report may also include recommendations.
Website owners can publish the results of successful BITV final tests to document conformance to BITV 2.0 and the degree of accessibility of their website. A publication must not be selective: the complete result of the test report must be given.
In case of a successful test result, website owners can place the 90plus logo or 95plus logo on their web site to reference the test result. In addition website owners can ask for inclusion of their site in the 90plus listing of accessible websites.
The idea of the BITV-Test is that it can be carried out with reasonable effort. This is why the test focuses on a representative sample of pages. The downside of sampling is that not all deficiencies may have been captured. Moreover, the test result is a snapshot documening the site's accessibility at a given point of time only. As the site and its content change, the test result may quickly become outdated.
Our experience shows that it is usually recommended to carry out a design support test to remedy exisiting deficiencies before embarking on a final test. It is then more likely that the final test will lead to a good or very good result.