Skip to content

Accessibility in Surveys & Forms

Accessibility in REDCap Surveys & Forms

Article ID RC-SURV-11 — Accessibility in REDCap Surveys & Forms
Domain Surveys
Applies To Survey and instrument designers; anyone answering an institutional accessibility question about REDCap
Requires Varies by item — see §3. Most improvements land between 16.0.6 and 17.3.0
Verified Against REDCap v17.4.1 (Standard) / v17.3.7 (LTS) — changelog review. Assistive-technology behaviour has not been independently tested
Prerequisite RC-SURV-01 — Surveys: Basics
Version 1.1
Last Updated 2026-08
Author See KB-SOURCE-ATTESTATION.md
Related Topics RC-SURV-02 — Survey Settings: Basic Options and Design; RC-SURV-03 — Survey Settings: Behavior, Access and Termination; RC-FD-06 — Online Designer: Instrument & Field Management; RC-MLM-01 — Multi-Language Management; RC-INFRA-03 — REDCap Versions, Release Lines & Patching
Synonyms is redcap accessible; redcap screen reader support; WCAG compliance redcap; section 508 redcap surveys; keyboard navigation surveys; accessible surveys for participants with disabilities; VoiceOver redcap; low vision survey design; accessibility of redcap forms; datepicker keyboard access; what is a VPAT; accessibility conformance report; WCAG AA meaning; section 508 vs WCAG

Scope and a caveat about what this article is. This consolidates the accessibility work REDCap has shipped and the settings that affect accessibility, drawn from release notes through 17.4.1 Standard / 17.3.7 LTS. It has not been validated by testing with assistive technology, and it is not a conformance statement.

REDCap does not publish a WCAG or Section 508 conformance level, and neither does this article. If your institution needs a formal accessibility statement — a VPAT or ACR, or an answer to a procurement questionnaire — that must come from your own evaluation or from the REDCap Consortium, not from a changelog summary. Those terms are defined in §1.1. What follows tells you what has been improved and what you can control; it does not tell you that any given survey is conformant.

1. Overview

Accessibility in a REDCap survey has two independent parts, and confusing them is the most common mistake:

Layer Who controls it Examples
The platform REDCap's developers; your patch level Screen-reader announcements, keyboard focus indicators, ARIA state on widgets
The instrument You, the designer Field labels, choice wording, colour and contrast choices, text size, whether the survey is navigable at all

Upgrading REDCap improves the first. It does nothing for the second. A survey with unlabelled fields, colour-only meaning, or instructions embedded in an image remains inaccessible on the newest release — see §4.

Between 16.0.6 and 17.3.0 REDCap shipped a sustained run of accessibility improvements, most of them concentrated in the 17.1.x releases. That is the single strongest argument for keeping a public-facing survey instance current: participants with disabilities are the population most affected by an old patch level, and they are also the ones least able to work around it.

1.1 The terminology

These terms turn up in institutional accessibility requests, usually without explanation.

Term What it means
WCAG Web Content Accessibility Guidelines — the international standard for web accessibility, published by the W3C. Organised into three conformance levels: A (minimum), AA (the level most institutions and laws require) and AAA (strictest, rarely required in full). "WCAG 2.1 AA" is the common procurement bar
Section 508 A provision of the US Rehabilitation Act requiring federal agencies to make electronic and information technology accessible. Since 2018 its technical requirements are WCAG 2.0 AA. US institutions receiving federal funding often apply it to their own systems
EN 301 549 The European equivalent, referenced by the EU Web Accessibility Directive. Also built on WCAG
VPAT Voluntary Product Accessibility Template — a standard form, maintained by the Information Technology Industry Council, that a vendor fills in to state how their product measures up against an accessibility standard, requirement by requirement. It is a blank template, not a certification
ACR Accessibility Conformance Report — a completed VPAT. Strictly, "VPAT" is the empty form and "ACR" is the filled-in document, though in practice people say "VPAT" for both. This is the artefact a procurement office is usually asking for
Assistive technology (AT) Software or hardware a person uses to access a computer — screen readers (JAWS, NVDA, VoiceOver), screen magnifiers, switch devices, voice control
Screen reader Software that reads screen content aloud and lets a user navigate by headings, links and form fields. It depends entirely on the underlying markup being correct — which is why field labels matter so much (§4)

What a VPAT/ACR actually tells you. It is a self-assessment by the vendor, not an audit or a certification. Its value is that it is specific: it walks each criterion and states Supports, Partially Supports, Does Not Support or Not Applicable, with explanatory notes. A procurement office asking for one wants that itemised detail, not a yes/no answer.

REDCap does not publish one, which is why §5 exists — you will need to answer such a request from your own evaluation, or escalate it to the Consortium.

2. Accessibility-Relevant Settings You Control

These are existing REDCap features that bear directly on accessibility. They are documented in full elsewhere; collected here because they are what a designer can actually act on.

Setting Where Notes
Text-to-Speech Survey settings Reads survey text aloud. Can be off, on-by-default, or available-but-off. Sends survey text to a third-party service — check with your IRB before using it on sensitive surveys. See RC-SURV-03 §3.9
Survey Text Size Survey design options Larger base text benefits low-vision participants without requiring browser zoom. See RC-SURV-02 §4.4
Survey Text Font Survey design options Font choice affects legibility, particularly for dyslexic readers. See RC-SURV-02 §4.5
Survey Theme Survey design options Custom themes are where contrast problems get introduced. A theme matching institutional branding is not automatically legible. See RC-SURV-02 §4.6
Enhanced Signature field (17.1.0+) Online Designer Accepts a hand-typed signature as well as a scribbled one. This matters: a mouse-drawn signature is impossible for many keyboard-only and motor-impaired users, so on any consented study the enhanced field is the accessible choice. See RC-FD-06
Multi-Language Management Project setup Beyond translation, MLM sets the page's lang attribute correctly (§3.1), which is what tells a screen reader how to pronounce the content

Note: Text-to-speech is a participant convenience, not a substitute for screen-reader compatibility. A participant using their own screen reader is not helped by REDCap reading text aloud in parallel — the two compete.

3. Platform Improvements by Version

All versions below are Standard. An LTS instance receives these only if its line was cut from a Standard release at or above the version shown — see RC-INFRA-03.

3.1 Screen reader support

Version Improvement
16.0.6 With MLM enabled on a survey, the lang attribute on the page's html tag is set to the MLM language code where it maps to an ISO 639-1 code. This is what lets a screen reader select the right pronunciation rules
16.0.7 Descriptive Popups made more accessible to screen readers on surveys and data entry forms
16.0.9 Screen reader users can navigate inside the date/time picker
16.0.9 Fixed handling of HTML BR tags in field labels. VoiceOver on Chrome announced a BR in a label as "Empty Group", breaking the reading flow. Chrome-specific
17.1.0 Survey pages announce page changes — e.g. "You are currently on page X of Y of the survey title Z." Without this, a screen reader user could not tell whether submitting a page had done anything
17.1.0 The language selector widget communicates its expanded/collapsed state, so assistive technology reports when the menu opens or closes
17.1.1 Dismissing the "required fields" error popup now scrolls to and focuses the first incomplete required field, instead of leaving the user to re-traverse the page hunting for it
17.1.4 After a failed login, the error message sits closer to the login fields and is exposed to the screen reader on redirect back to the page

3.2 Keyboard navigation and visible focus

Version Improvement
16.0.9 Keyboard users can reach the date/time picker at all. Previously tabbing skipped straight past it — a hard block, not an inconvenience
16.1.1 Record Status Dashboard rows highlight on hover and on keyboard focus
17.1.0 Survey navigation buttons ("Next Page", "Save & Continue", "Save & Return Later") show a glowing highlight on hover or keyboard focus
17.1.3 Survey form fields and buttons show a clear visible highlight when reached by keyboard — the single most useful change for anyone navigating without a mouse or with low vision
17.1.3 The slider control highlights its surrounding question in green on focus, consistent with every other field type, and the track and knob have improved contrast when active

3.3 Visual presentation

Version Improvement
17.3.0 Button styling adjusted on surveys and project pages to improve accessibility

3.4 A regression worth knowing about

Version caveat (16.0.9–16.1.1 Standard): The 16.0.9 datepicker accessibility work introduced a regression — erratic flashing of the datepicker in the context of ASIs and Alerts & Notifications. Closing the picker returned focus to the input field, whose onfocus handler immediately reopened it. Fixed in 16.1.2, which kept the accessibility improvements while restoring normal behaviour.

This is a useful illustration of why accessibility fixes deserve the same regression testing as any other change: the fix was correct in intent and broke ordinary use for everyone.

4. What REDCap Cannot Fix For You

Every improvement in §3 addresses the platform. The most common accessibility failures in REDCap surveys are in the instrument, and no upgrade touches them:

  • Field labels that don't describe the field. A screen reader reads the label. "Please enter below:" tells the user nothing.
  • Meaning carried by colour alone. Red text marking required items is invisible to a colour-blind participant and to a screen reader alike. Say "required".
  • Instructions embedded in images. Text in an uploaded image is unreadable to assistive technology. Use the field label or a descriptive field.
  • Descriptive text used as a heading without being one. Section Headers give structure that screen readers can navigate by; bold descriptive text does not.
  • Custom themes with insufficient contrast. Branding colours frequently fail contrast ratios. Check before deploying, not after a complaint.
  • Matrix fields that are very wide. They are difficult to navigate with a screen reader and awkward on small screens, where many participants with disabilities are.
  • Time limits and forced-response patterns that assume a participant can complete quickly.

The practical test. Before releasing a public survey, complete it end to end using only the keyboard — no mouse — at the largest text size your participants might use. Most instrument-level accessibility problems surface within a minute, and it requires no specialist tooling.

5. Answering an Institutional Accessibility Question

REDCap managers are routinely asked whether REDCap is "508 compliant" or "WCAG AA". Some honest framing:

  • There is no published conformance level to cite. REDCap ships accessibility improvements; it does not, in its release notes, claim a conformance level.
  • Conformance is per-survey, not per-platform. Even a fully conformant platform will produce a non-conformant survey if the instrument is badly designed. The question "is REDCap accessible?" is not answerable in the abstract.
  • Patch level is part of the answer. An instance below 17.1.3 lacks visible keyboard focus indicators on survey fields; below 16.0.9 the date picker is unreachable by keyboard. Those are concrete, checkable facts about your instance.
  • Escalate formal requests. VPATs and Accessibility Conformance Reports (see §1.1) and procurement questionnaires should go to the REDCap Consortium via the Community Portal, or be answered by your institution's own evaluation.

6. Common Questions

Q: Is REDCap accessible?

A: REDCap has shipped sustained accessibility improvements from 16.0.6 through 17.3.0, particularly around screen reader announcements and keyboard focus. But accessibility is determined jointly by the platform and by how the instrument is designed — a poorly designed survey on the newest release is still inaccessible. There is no published conformance level to point at.

Q: What's the single most valuable version to be on?

A: 17.1.3 or later. That release added clear visible highlights for survey form fields and buttons under keyboard navigation, which affects every keyboard-only and low-vision participant on every survey. 16.0.9 is the other threshold — below it, keyboard users cannot reach the date/time picker at all.

Q: Does turning on text-to-speech make our survey accessible?

A: No. It is a useful convenience for some participants, but it is not screen-reader compatibility and it does not address keyboard navigation, contrast or label quality. It also sends survey text to a third-party service, which needs IRB consideration for sensitive content.

Q: Our participants use screen readers. Anything specific to check?

A: Be on 17.1.0+ so page changes are announced, and 17.1.1+ so required-field errors move focus to the offending field. Enable MLM if the survey is not in the instance's default language, so the lang attribute is set. Then check your own field labels — that is where most remaining problems will be.

Q: We have a signature field and a participant can't use a mouse. What do we do?

A: Use the Enhanced Signature field type (17.1.0+), which accepts a hand-typed signature as well as a drawn one.

Q: A screen reader reads odd "Empty Group" text in our field labels.

A: That is the BR tag issue fixed in 16.0.9, and it only occurs in Chrome with VoiceOver. Upgrade, or restructure the labels to avoid BR tags.

Q: Our date picker flashes repeatedly on alert-related pages.

A: A regression in 16.0.9 through 16.1.1, fixed in 16.1.2. Upgrade.

Q: What is a VPAT, and can we get one for REDCap?

A: A Voluntary Product Accessibility Template is a standard form a vendor completes to state, criterion by criterion, how their product measures against an accessibility standard; a completed one is properly called an Accessibility Conformance Report (ACR). It is a vendor self-assessment, not a certification or an audit — see §1.1.

REDCap does not publish one. Formal conformance documentation should be requested through the REDCap Consortium, or produced by your own institutional evaluation of your instance and your instruments.

7. Common Mistakes & Gotchas

Treating accessibility as a platform property. It is jointly determined. Upgrading fixes focus indicators and screen-reader announcements; it does not fix your field labels, your contrast, or your reliance on colour.

Claiming a conformance level REDCap has not claimed. "REDCap is 508 compliant" is not a statement the release notes support. Describe what your instance does and does not do instead.

Running a public participant-facing instance on an old patch level. Accessibility improvements are exactly the kind of change that only affects a minority of users — who then simply cannot complete the survey and are rarely in a position to report why.

Assuming text-to-speech covers screen reader users. It does not, and the two can conflict. It is a separate feature for a different need.

Using a mouse-drawn signature field on a consent instrument. Pre-17.1.0 signature capture excludes participants who cannot use a pointing device. On an e-consent instrument that is a consent-validity problem, not just a usability one.

Deploying a custom survey theme without checking contrast. Institutional brand palettes routinely fail contrast requirements.

Never testing with the keyboard. A single keyboard-only pass through a survey catches most instrument-level problems and takes a minute.