Accessibility Testing Requirements – Perceivable – Adaptable
In the series of reviews I’m looking at WCAG level A / AA (Web Content Accessibility Guidelines) requirements from testing perspective.
Common Definitions
Human Testing
- Brief – “You know as you see it” – quick scanning is sufficient to check and make a judgment. This does not include effort of logging defects.
- Detailed – Either interaction (execution of tests) with the functionality is necessary or detailed review/analysis must be done in order to verify a requirement.
Tool-Assisted Checking
- Not Available – Tools cannot help in checking for the specified requirement or they help very insignificantly.
- Partial – Tools provide significant help in checking for the specified requirement, either by saving time in parsing content source or through visualization. Human does the verification and judgment.
- Full – Tools capable of locating and verifying the specified requirement. Brief review of checking results is sufficient to make a judgment.
Perceivable – Adaptable
Create content that can be presented in different ways (for example simpler layout) without losing information or structure.
Tools play major role in parsing content and identifying missing or invalid tags.
Application-specific GUI design requirements must be set to create compliant and consistent adaptable content.
Specific knowledge of accessibility supporting mark-up used in the project is required; with that knowledge any generally skilled tester can perform the assessment.
Semantic Markup
Semantic markup is used to designate headings (<h1>), lists (<ul>, <ol>, and <dl>), emphasized or special text (<strong>, <code>, <abbr>, <blockquote>, for example), etc. Semantic markup is used appropriately.
Human Testing – Very Detailed
Tool-Assisted Checking – Partial
Testers can rely on tools to identify obvious problems with semantic markup, i.e. performing tool-assisted “Smoke Test”.
However, this requirement has hundreds of possible valid and invalid implementations, where right/wrong heavily depends on the context. Human skill, experience, and judgment required to recognize and analyze those situations.
For significant amount of pages to test systematic approach must be in place to ensure proper test coverage.
For the newly developed application systematic implementation approach should be used, to standardize implementation patterns and create reusable UI elements.
This type of application-specific end-to-end checking can be well covered within GUI Automation.
Tabular data
Tables are used for tabular data. Where necessary, data cells are associated with their headers. Data table captions and summaries are used where appropriate.
Human Testing – Detailed
Tool-Assisted Checking – Full
Testers can rely on tools to identify problems with markup of the tables.
Captions, summaries, and descriptions must be assessed for relevance and completeness.
Tables, especially with input controls, are also subject for a thorough Keyboard Operability testing.
Related form elements
Related form elements are grouped with fieldset/legend.
Human Testing – Brief
Tool-Assisted Checking – Full
Tools may identify grouping problems in some cases.
Testers to perform visual inspection of all forms to verify the grouping. Legend must be assessed for relevance and completeness.
Meaningful Sequence
The reading and navigation order (determined by code order) is logical and intuitive.
Human Testing – Detailed
Tool-Assisted Checking – Not available
Tools can help little here.
Testers need to perform detailed review. This testing can be combined with functional testing.
“Logical and intuitive” is a subjective definition. Certain form of requirements (UI Mockup, checklist, etc) would help in establishing common interpretation in the project and reduce misunderstandings.
Instructions – visual
Instructions do not rely upon shape, size, or visual location (e.g., “Click the square icon to continue” or “Instructions are in the right-hand column”).
Human Testing – Brief or Detailed
Tool-Assisted Checking – Partial
This is a very context-specific requirement. Tools may help identifying certain problems. Brief or detailed review might be required.
Instructions – auditory
Instructions do not rely upon sound (e.g., “A beeping sound indicates you may continue.”).
Human Testing – Brief
Tool-Assisted Checking – Not available
Tools don’t verify that, brief review by a human is required.