Contact Us

Document Accessibility

Field: Accessibility · Also called: accessible documents, electronic document accessibility

Document accessibility is WCAG applied to digital documents, such as PDFs and Word files, whether they were created on a computer or scanned from paper. An accessible document can be read and used by a person with a disability, including someone who relies on a screen reader or who uses magnification or high contrast to compensate for low vision. An inaccessible document is a barrier that can leave its content effectively unavailable to people with disabilities.

Scope

For state and local governments in the United States, document accessibility is a legal requirement under ADA Title II. 28 CFR § 35.200 requires public entities to ensure that the web content they provide complies with WCAG 2.1 Level A and Level AA, and 28 CFR § 35.104 defines conventional electronic documents as web content “in the following electronic file formats: portable document formats (“PDF”), word processor file formats, presentation file formats, and spreadsheet file formats.” A PDF posted on a public entity’s website is held to the same WCAG 2.1 success criteria as the web page that links to it.

WCAG was written with web pages in mind, so applying it to a document means reading “document” where a success criterion says “page.” For example, 2.4.2 Page Titled, applied to a PDF, concerns the document’s title. The federal Section 508 Standards make this substitution explicit for non-web documents (E205.4.1). The W3C’s WCAG2ICT note gives informative guidance on applying WCAG outside the web, and defines a document as an “assembly of content, such as a file, set of files, or streamed media that functions as a single item rather than a collection, that is not part of software and that does not include its own user agent.”

Digital native and scanned documents

  • Digital native documents are exported from an authoring tool such as a word processor. Whether they are accessible depends largely on the tool and how it was used, for example whether a PDF export included structure tags.
  • Scanned documents begin as an image of paper. Without optical character recognition (OCR), the file contains a picture of text and no actual text, so a screen reader has nothing to read. W3C technique PDF7 covers performing OCR to provide actual text. OCR alone is often not enough, since recognized text still lacks the headings, lists, tables, and reading order that tags provide (Success Criteria 1.3.1 and 1.3.2). ADA Title II does not prohibit scanned documents, but it requires them to conform like any other web content, and scans of paper documents created before the compliance date may qualify for the archived web content exception. A scan of a paper document created after that date has no date-based exception.

Common requirements

WCAG organizes its success criteria under four principles, often abbreviated POUR, which hold that content must be perceivable, operable, understandable, and robust. Most document accessibility work falls under a small number of those success criteria. The W3C publishes techniques for meeting each one, most of them specific to PDF.

Requirement WCAG 2.1 success criterion W3C technique
Actual text, not an image of text 1.4.5 Images of Text PDF7
Alternative text for images 1.1.1 Non-text Content PDF1
Headings, lists, and tables marked as structure 1.3.1 Info and Relationships PDF9, PDF21, PDF6
Correct reading order 1.3.2 Meaningful Sequence PDF3
A descriptive document title 2.4.2 Page Titled PDF18
The document’s language declared 3.1.1 Language of Page PDF16
Link text that describes the destination 2.4.4 Link Purpose (In Context) PDF11, PDF13
Text with enough contrast against its background 1.4.3 Contrast (Minimum) G18
Form fields with labels, names, and roles 1.3.1 Info and Relationships, 3.3.2 Labels or Instructions, 4.1.2 Name, Role, Value PDF10, PDF12

W3C techniques are informative. They are ways to meet a success criterion, not the requirement itself.

Headings, styles, and Office formats

Headings must be marked as headings, not just made to look like them. In a word processor, that means applying the built-in Title, Heading 1, Heading 2, and Heading 3 styles rather than selecting text and making it bold and centered. Text formatted only by its appearance looks like a heading to a sighted reader but carries no structure for assistive technology, which the W3C documents as failure F2 of Success Criterion 1.3.1, “using changes in text presentation to convey information without using the appropriate markup or text.” Styles applied in the source document carry through as tags when it is exported to PDF with tagging enabled. Frank Rietta demonstrates the difference in the short video Bold Text Can Get You Sued (ADA Title II Explained in 60 Seconds).

The Title style marks the visible title on the page. It is separate from the title stored in the document’s properties, which is the metadata that Success Criterion 2.4.2 relies on.

The W3C publishes techniques for PDF but not for Word, Excel, or PowerPoint files. For those formats, Microsoft’s guidance for making Word documents accessible says to “use the built-in heading styles” and not to skip heading levels, and its guidance on creating accessible Office documents covers the other Office formats.

Important Metadata, the Invisible Content

Much of what makes a document accessible is not visible on the page. Metadata stored in the file, along with structure tags and alternative text, is read by assistive technology but is often not shown to a sighted reader, so a document can look complete while failing WCAG.

Two metadata fields are the minimum for compliance:

  • Title. WCAG 2.1 Success Criterion 2.4.2 requires a title that describes the document’s topic or purpose. For a PDF, technique PDF18 sets it in the Title entry of the document information dictionary. See Document Title.
  • Language. Success Criterion 3.1.1 requires that “the default human language of each Web page can be programmatically determined.” For a PDF, technique PDF16 sets it in the /Lang entry of the document catalog. The requirement applies even when a screen reader would fall back to its user’s configured language. See Document Language.

Other metadata, such as the author, a subject or description, and dates, is not required by WCAG but helps people identify and find documents. Dates also bear on the archived web content exception, whose first criterion turns on when the content, or the paper original it reproduces, was created.

For the document title, technique PDF18 also sets the DisplayDocTitle flag so that viewers show the title in the title bar, making the same metadata visible to sighted readers.

At the scale of a document corpus

A public entity may publish thousands of documents. Assessing them as a document corpus, rather than one file at a time, shows which defects recur, and document provenance traces recurring defects back to the workflow that produced them. Correcting individual files is remediation.

Distinctions

  • Versus automated checking. The W3C states that evaluation tools “can not determine accessibility, they can only assist in doing so,” and that human judgment is required. A document can pass an automated checker and still fail WCAG, for example with alternative text that exists but does not describe the image.
  • Versus PDF/UA. PDF/UA (ISO 14289-1) is a file format standard specific to PDF. ADA Title II incorporates WCAG 2.1, which applies across all document formats, not PDF/UA. The federal Section 508 Standards reference PDF/UA-1 for authoring tools, which must be capable of exporting it when they export PDF (504.2.2).
  • Versus Section 508. Federal agencies are covered by Section 508, whose standards incorporate WCAG 2.0 Level A and Level AA. State and local governments under ADA Title II are held to WCAG 2.1.
  • Versus compliance. Meeting WCAG is conformance, a technical claim. Whether a public entity is compliant with ADA Title II also depends on the rule’s exceptions and dates. See Conformance vs. Compliance and the Preexisting Conventional Electronic Documents Exception.

Sources

See also

Go deeper

Articles