Why accessibility — the legal and ethical case
Accessible design is both a civil-rights obligation and better teaching. In the United States, several laws establish that educational programs and their digital materials must be usable by people with disabilities.
Americans with Disabilities Act
Prohibits disability discrimination in public accommodations and services. Courts and the DOJ apply it to websites, learning platforms, and digital course content, not just physical spaces.
Rehabilitation Act of 1973
Bars programs receiving federal funds — including most schools and universities — from excluding people with disabilities, and grounds the right to reasonable accommodation.
Rehabilitation Act, as amended
Requires federal agencies' electronic and information technology to be accessible; its 2017 refresh formally adopts WCAG 2.0 Level AA as the technical standard, widely used as a procurement benchmark.
Accessibility vs. accommodation
Accessibility is designing the environment so it works for the widest range of people from the outset — proactive and for everyone. An accommodation is an individual, after-the-fact adjustment (extended time, a captioner, an alternate format) for a specific student, usually arranged through a disability services office. Robust accessibility reduces — but never fully eliminates — the need for accommodations.
The curb-cut effect
Curb cuts were built for wheelchair users, yet they help people with strollers, delivery carts, luggage, and bicycles. Designing for disability routinely benefits everyone: captions serve viewers in noisy or sound-off settings and language learners; clear headings help every reader skim; keyboard operability helps power users. Access is a rising tide.
WCAG and the POUR principles
The Web Content Accessibility Guidelines (WCAG), published by the W3C's Web Accessibility Initiative, organize all success criteria under four principles. Content must be Perceivable, Operable, Understandable, and Robust — POUR.
Available to the senses
Information can't be invisible to all of a user's senses.
- Text alternatives for images
- Captions & transcripts for media
- Sufficient color contrast
- Don't rely on color alone
Usable by anyone
Interface and navigation must be operable by all input methods.
- Full keyboard operability
- No keyboard traps
- Enough time; pausable motion
- Descriptive page titles & link text
Clear and predictable
Content and operation must make sense.
- Readable, plain language
- Predictable, consistent navigation
- Labels & instructions on inputs
- Helpful error identification
Works with assistive tech
Content must be interpretable by a wide range of user agents.
- Valid, well-formed markup
- Correct name, role, value
- Screen-reader compatible
- Status messages announced
Minimum
The most basic barriers removed. Necessary but rarely sufficient on its own.
The common target
The standard most laws, institutions, and procurement policies require (e.g. 4.5:1 text contrast, captions, resize/reflow). Aim here.
Enhanced
The highest bar (e.g. 7:1 contrast, sign language). Not required for whole sites, but valuable where feasible.
WCAG success criteria by principle
Under each POUR principle sit guidelines, and under those the testable success criteria — each numbered and assigned a conformance level (A, AA, AAA). A representative selection is shown below in plain language; WCAG 2.2 defines 87 in total.
Perceivable
| Criterion | Level | What it means in practice |
|---|---|---|
| 1.1.1 Non-text Content | A | Every image, icon, chart, and control that conveys meaning has a text alternative; purely decorative images are marked so assistive tech ignores them. |
| 1.2.2 Captions (Prerecorded) | A | Prerecorded video with audio carries synchronized captions of speech and meaningful sound. |
| 1.3.1 Info and Relationships | A | Structure conveyed visually — headings, lists, table headers, labels — is also encoded in the markup so it survives for screen readers. |
| 1.4.1 Use of Color | A | Color is never the only way information is conveyed; a text label, icon, or pattern carries the same meaning. |
| 1.4.3 Contrast (Minimum) | AA | Normal text has at least a 4.5:1 contrast ratio against its background; large text (≈24px, or 18.66px bold) at least 3:1. |
| 1.4.11 Non-text Contrast | AA | UI components and meaningful graphics (button borders, form outlines, icon parts) have at least 3:1 contrast against adjacent colors. |
Operable
| Criterion | Level | What it means in practice |
|---|---|---|
| 2.1.1 Keyboard | A | All functionality is available from a keyboard alone — no action requires a mouse. |
| 2.1.2 No Keyboard Trap | A | Keyboard focus can always move away from any component using standard keys; users never get stuck. |
| 2.4.4 Link Purpose (In Context) | A | Each link's purpose is clear from its text or immediate context — no bare "click here" or naked URLs. |
| 2.4.7 Focus Visible | AA | The keyboard focus indicator is always visible, so users can see where they are on the page. |
| 2.4.11 Focus Not Obscured | AA | New in 2.2 — a focused component is not entirely hidden behind sticky headers, footers, or overlays. |
| 2.5.8 Target Size (Minimum) | AA | New in 2.2 — interactive targets are at least 24×24 CSS pixels, or have adequate spacing, so they are easy to activate. |
Understandable
| Criterion | Level | What it means in practice |
|---|---|---|
| 3.1.1 Language of Page | A | The page's primary language is set in the markup (e.g. lang="en") so screen readers pronounce it correctly. |
| 3.2.3 Consistent Navigation | AA | Navigation that repeats across pages appears in the same relative order each time. |
| 3.3.1 Error Identification | A | When an input error is detected, the item is identified and the problem described in text. |
| 3.3.2 Labels or Instructions | A | Form fields have visible labels or instructions so users know what to enter. |
| 3.3.7 Redundant Entry | A | New in 2.2 — information a user already entered in a process is auto-populated or available, not re-requested. |
Robust
| Criterion | Level | What it means in practice |
|---|---|---|
| 4.1.2 Name, Role, Value | A | Every UI component exposes its name, role, and current state to assistive technology — native HTML does this for free; custom widgets need correct ARIA. |
| 4.1.3 Status Messages | AA | Messages that report status without moving focus (a saved confirmation, a search count) are announced via live regions. |
Accessibility self-audit
Check every practice your course materials already meet. This produces a live coverage estimate across the POUR principles — a quick heuristic to prioritize fixes, not a substitute for testing with real assistive technology and users.
Color contrast checker
Enter a foreground (text) and background color as hex values. This computes the exact WCAG contrast ratio from each color's relative luminance and reports pass or fail against the AA and AAA thresholds — the same math a formal audit uses.
- AA · normal text—
- AA · large text—
- AAA · normal text—
- AAA · large text—
- UI / non-text (3:1)—
Thresholds: AA needs 4.5:1 for normal text and 3:1 for large text (≈24px, or 18.66px bold); AAA needs 7:1 and 4.5:1. UI components and graphics need 3:1.
The CAST UDL framework
Developed by CAST, Universal Design for Learning is a framework for designing flexible learning experiences that reduce barriers for the widest range of learners. It rests on three principles — provide multiple means of Engagement, Representation, and Action & Expression. (The UDL 3.0 Guidelines, released in 2024, refresh this framework and foreground learner agency and the dismantling of barriers.)
Multiple means of Engagement
Motivation and interest vary. Offer choice and relevance, sustain effort and persistence, and support self-regulation.
- Give students options in topics or tools
- Make tasks authentic and relevant
- Vary challenge and offer mastery feedback
- Build in goal-setting and reflection
Multiple means of Representation
Learners perceive and comprehend information differently. Present content in more than one format.
- Offer text, audio, and video versions
- Caption media; provide transcripts
- Define vocabulary; clarify syntax
- Use graphics, models, and examples
Multiple means of Action & Expression
Learners differ in how they navigate and demonstrate what they know. Offer varied ways to act and express.
- Let students show learning multiple ways
- Support planning and strategy use
- Vary tools and assistive technologies
- Scaffold goal-setting and progress
UDL vs. accommodation — proactive vs. reactive
| Dimension | UDL | Accommodation |
|---|---|---|
| Timing | Proactive, at design | Reactive, after the fact |
| Who benefits | All learners | One identified student |
| Trigger | Anticipated variability | Documented disability |
| Scope | The whole environment | An individual adjustment |
They work together
UDL and accommodations are complementary, not competing. Strong universal design shrinks the number and intensity of individual accommodations needed, but some students will always require specific, legally protected adjustments. Design universally first; accommodate individually where gaps remain. Accessible base materials also make each accommodation faster and cheaper to fulfill.
The full UDL guidelines
Each of the three principles breaks into three guidelines, and each guideline into concrete considerations (formerly "checkpoints"). CAST arranges them from Access to Support/Build to Executive/Internalize, with the goal of developing expert learners who are purposeful, resourceful, and strategic.
Multiple means of Engagement
Options that recruit interest and sustain the will to learn — the affective networks of the brain.
- Optimize individual choice and autonomy
- Optimize relevance, value, and authenticity
- Nurture joy and play; address bias and threat
- Heighten salience of goals and objectives
- Vary demands and resources to optimize challenge
- Foster collaboration, community, and belonging
- Offer action-oriented, mastery feedback
- Promote expectations and beliefs that motivate
- Develop coping skills and strategies
- Support self-reflection and assessment
Multiple means of Representation
Options for perceiving and comprehending information — the recognition networks of the brain.
- Support opportunities to customize display
- Offer alternatives for auditory information
- Offer alternatives for visual information
- Clarify vocabulary, symbols, and structures
- Support decoding of text and notation
- Cultivate multiple languages and dialects
- Illustrate through multiple media
- Connect prior knowledge to new learning
- Highlight patterns, big ideas, and relationships
- Guide information processing
- Maximize transfer and generalization
Multiple means of Action & Expression
Options for strategic action and expressing what one knows — the strategic networks of the brain.
- Vary methods for response and navigation
- Optimize access to tools and assistive technologies
- Use multiple media for communication
- Use multiple tools for construction and composition
- Build fluencies with graduated support and practice
- Guide appropriate goal-setting
- Support planning and strategy development
- Facilitate managing information and resources
- Enhance capacity for monitoring progress
Assistive technologies
Accessible design is what lets these tools work. Knowing how learners actually reach your materials makes the "why" behind each practice concrete.
Speech & braille output
Software such as JAWS, NVDA, and VoiceOver renders the page as synthesized speech or a refreshable braille display, navigating by headings, landmarks, links, and form fields.
- Depends on real headings and labels
- Reads alt text and table headers
- Announces ARIA name, role, value
Screen & text zoom
Low-vision users enlarge the interface with OS magnifiers or browser zoom, often 200–400%. Content must reflow to a single column without loss of information or horizontal scrolling.
- Needs responsive reflow (WCAG 1.4.10)
- Text resize to 200% (WCAG 1.4.4)
- No meaning lost when enlarged
One or two switches
People with limited mobility operate a computer with one or more switches that scan through controls. This works only when everything is reachable and operable without a mouse.
- Requires full keyboard operability
- Logical focus order and visible focus
- Adequate target size and timing
Voice control
Dictation and voice-control tools (Dragon, Voice Control, Voice Access) let users click controls and enter text by speaking — they rely on controls having visible, accurate names.
- Visible label matches accessible name
- Controls are real, named elements
- Label in Name (WCAG 2.5.3)
Text for media
Captions, transcripts, and audio description let deaf, hard-of-hearing, blind, and low-vision learners — and anyone in a sound-off or hands-busy setting — reach multimedia content.
- Synchronized captions on video
- Transcripts for audio
- Audio description of key visuals
Focus & comprehension aids
Text-to-speech, reading rulers, reflow/reader modes, and word prediction support learners with dyslexia, ADHD, or cognitive differences — and benefit from clean structure and plain language.
- Semantic structure powers reader modes
- Plain language aids comprehension
- Consistent, predictable navigation
Accessible documents
Most course barriers live in ordinary files — syllabi, slides, handouts, PDFs. The same handful of habits fixes the majority of them, in Word, Google Docs, slides, or the web.
| Do | Don't |
|---|---|
| ✓Use built-in heading styles (Heading 1, 2, 3) in order | ✗Fake headings with big, bold, or colored text |
| ✓Write concise alt text for informative images | ✗Leave images unlabeled or stuff keywords into alt |
| ✓Use meaningful link text that describes the target | ✗Paste "click here" or raw, unreadable URLs |
| ✓Build lists with the real list tool | ✗Hand-type dashes or asterisks as pseudo-lists |
| ✓Give data tables header cells and keep them simple | ✗Use tables for visual layout or merge cells needlessly |
| ✓Check reading order matches the visual flow | ✗Rely on floating text boxes that scramble the order |
| ✓Pair color with text, icon, or pattern | ✗Signal meaning with color alone ("items in red") |
| ✓Export tagged PDFs and run the accessibility checker | ✗Scan a paper handout to an untagged image-only PDF |
Accessible math & STEM, data & images
Equations, charts, and complex graphics are where accessibility most often breaks. An equation saved as a picture, or a chart with no described trend, is invisible to a screen-reader user. These carry text and structure, not just pixels.
MathML & LaTeX
Represent equations as real notation, not flat images. MathML is the semantic web standard that assistive tech can read and navigate term by term; authoring tools and LaTeX/MathJax can render to it.
- Avoid equations pasted as pictures
- MathML exposes structure to screen readers
- MathJax renders LaTeX with accessible output
- If an image is unavoidable, add full alt text
Describe the data
A screen reader can't "see" a trend line. Give a short alt text stating the chart's takeaway, and pair every chart with the underlying numbers in an accessible data table or linked dataset.
- Alt text states the main finding, not "chart"
- Provide the source data as a table
- Don't encode series by color alone
- Label axes and units in text
Short + long descriptions
Diagrams, maps, and process figures need a brief alt text plus a longer description nearby — in a caption, adjacent text, or a linked full description — that conveys the same information a sighted reader gets.
- Short alt names the image and its purpose
- Longer prose explains the detail
- Put the description in real body text
- Decorative images get empty alt
Data tables that assistive tech can read
Use real table markup with header cells (<th> and a scope), a caption, and a simple rectangular layout. Screen readers then announce the relevant row and column headers as the user moves between cells, so a number never floats free of its meaning. Avoid merged cells, split tables, and using tables purely for visual layout.
The picture-of-text trap
Formulas, code, tables, and quotes are frequently pasted as screenshots — fast to make, but completely opaque to assistive technology, unsearchable, and blurry when magnified. WCAG 1.4.5 Images of Text asks you to use actual text wherever the presentation can be achieved with markup. Reserve images of text for logos and cases where a specific rendering is essential.
Accessible video & audio
Multimedia needs text and audio alternatives so it reaches deaf, hard-of-hearing, blind, and low-vision learners — and many more besides.
Synchronized text
On-screen text of speech and meaningful sound ("[door slams]"), timed to the video. Essential for deaf and hard-of-hearing viewers. Prefer accurate closed captions.
The full text record
A complete text version of the audio (and key visuals), readable without playing the video — searchable, skimmable, and usable by screen readers and braille displays.
Narrating the visual
A spoken track describing important visual information — actions, text on screen, scene changes — for blind and low-vision viewers, added in natural pauses.
Operable controls
Keyboard-operable play, pause, volume, and caption controls with visible focus; a player that exposes captions and doesn't autoplay with sound.
Auto-captions are a draft, not a deliverable
Automatic speech recognition misses punctuation, technical terms, names, and speaker changes, and struggles with accents and background noise — the errors it produces are sometimes called "craptions." Always review and correct auto-captions before you consider a video accessible.
Captions vs. subtitles
Captions assume the viewer can't hear the audio, so they include non-speech sounds and speaker identification — an accessibility feature. Subtitles assume the viewer can hear but doesn't understand the language, so they translate dialogue only. They are not interchangeable; accessibility needs captions.
Accessibility statements & the VPAT
Beyond the content itself, institutions communicate and verify accessibility through two documents: a public accessibility statement that faces users, and a VPAT that faces buyers. Both matter for anyone selecting or maintaining educational technology.
Your public commitment
A page that states an organization's commitment to access, the standard it targets (typically WCAG 2.1/2.2 AA), the site's known limitations, and — crucially — how a user can report a barrier and get help.
- Names the conformance target (e.g. WCAG 2.2 AA)
- Discloses known issues honestly
- Gives a contact route to report problems
- Dated and reviewed regularly
How vendors document conformance
A Voluntary Product Accessibility Template is the industry form a vendor completes to describe how a product meets accessibility standards; the filled-in result is an Accessibility Conformance Report (ACR). Editions map to WCAG 2.x, Section 508, and the EU EN 301 549.
- Each criterion marked Supports / Partially / Does Not
- "Remarks and Explanations" hold the real detail
- Request the ACR before you buy, not after
- Treat vague or empty VPATs as a red flag
Conformance levels in a VPAT
Supports — meets the criterion. Partially Supports — some functionality does not. Does Not Support — the majority is not met. Not Applicable — the criterion does not apply to the product.
Procurement basics
Write accessibility into RFPs and contracts, ask for a current ACR, verify a sample of claims with your own testing, and keep leverage for remediation timelines. Accessibility is cheapest to secure before purchase.
Why it's "shift-left"
Fixing accessibility during procurement and design costs a fraction of retrofitting a deployed system. A statement plus a verified ACR turns access from an after-the-fact complaint into a purchasing criterion.
A roadmap for making an existing course accessible
You don't have to fix everything at once. Work in priority order, remediate highest-impact materials first, and build accessibility into every new item so the backlog shrinks over time.
Inventory your materials
List every document, slide deck, video, and page in the course, and note the format and where it lives.
Triage by impact
Prioritize what every student must use — the syllabus, required readings, graded assessments — and the highest-traffic pages first.
Fix the fundamentals
Apply real headings, alt text, meaningful link text, and lists. These structural basics unblock the most users for the least effort.
Caption and transcribe media
Add and correct captions on videos and provide transcripts for audio, starting with content students are required to watch.
Check contrast and color
Test text and interface contrast against AA thresholds and make sure no meaning depends on color alone.
Run automated + manual checks
Use built-in accessibility checkers and a browser tool, then test keyboard navigation and a screen reader yourself — automation catches only part.
Adopt UDL for new content
Design new materials with multiple means of representation, engagement, and expression so accessibility is built in, not bolted on.
Document and maintain
Keep an accessibility statement, a way for users to report barriers, and a routine to re-check materials each term.
Glossary & further reading
- Assistive technology (AT) — tools such as screen readers, magnifiers, and switch devices that people use to access content.
- Alt text — a concise text alternative for an image, read by screen readers and shown when images fail to load.
- ARIA — Accessible Rich Internet Applications; markup that supplies name, role, and value where native HTML falls short.
- Conformance level — WCAG grades A, AA, and AAA marking how fully criteria are met.
- Contrast ratio — the luminance difference between text and background; AA requires 4.5:1 for normal text.
- POUR — WCAG's four principles: Perceivable, Operable, Understandable, Robust.
- Remediation — retrofitting existing content to remove accessibility barriers.
- Screen reader — software (JAWS, NVDA, VoiceOver) that renders on-screen content as speech or braille.
- Semantic markup — using HTML elements for their meaning (headings, lists, tables) so structure is machine-readable.
- UDL — Universal Design for Learning; CAST's framework of Engagement, Representation, and Action & Expression.
Frameworks & references
This is an independent open educational resource. Framework and standard names are cited for reference only; no affiliation with or endorsement by the W3C, CAST, or any agency is implied.