Accessibility

Website Calculator Accessibility Checklist: 18 Release Tests

A reusable WCAG 2.2-oriented QA sheet for calculator inputs, validation, dynamic results, keyboard operation, touch targets and embeds.

Direct answer

A calculator passes the basic release gate when every control has a persistent programmatic label, instructions and errors are associated with the right field, the whole task works by keyboard, dynamic results are announced without disrupting focus, color is not the only signal, targets meet WCAG 2.2 sizing rules, and the layout reflows without two-dimensional scrolling.

Definition

An accessible website calculator lets people understand inputs, enter values, correct errors and receive the result using different input methods and assistive technologies. Accessibility applies to the complete task, not just the form's first screen.

Key findings

Verified 24 August 2026

  • The result state is part of the form and needs an accessible name, clear units and an appropriate announcement strategy.
  • Placeholder text is not a substitute for a persistent label or instruction.
  • WCAG 2.2 Level AA sets a 24 by 24 CSS pixel minimum target rule with stated exceptions; larger controls remain easier to use.
  • Automated scans cannot verify formula meaning, focus order through conditional steps or whether a result announcement is useful.

What should be tested before calculation?

Run these checks on the real published calculator, including the host page when it is embedded. Save the URL, viewport, browser, keyboard sequence, assistive technology if used, result and screenshot or recording for every failure.

Tests 1 to 6: structure and input
#TestPass evidencePrimary mapping
1Page and calculator have descriptive headingsHeading outline explains the taskWCAG 2.4.6
2Every control has a persistent labelAccessible name matches visible purposeWCAG 1.3.1, 3.3.2, 4.1.2
3Units and formats are explicitCurrency, period and valid range remain visibleWCAG 3.3.2
4Related choices are groupedFieldset and legend or equivalent relationship existsWCAG 1.3.1
5Required status is conveyedNot dependent on color or a symbol aloneWCAG 1.3.1, 3.3.2
6Autocomplete and input purpose are appropriateBrowser and assistive technology receive the intended purposeWCAG 1.3.5

How should keyboard and pointer operation be tested?

Start with the pointer disconnected. Tab from the page address bar through the calculator, change every control, activate Calculate, reach the result and continue to the next action. Reverse with Shift+Tab. Focus must remain visible, logical and unobscured.

Tests 7 to 10: operation
#TestPass evidencePrimary mapping
7Complete task by keyboardNo mouse-only slider, tooltip or submit actionWCAG 2.1.1
8Focus order follows the taskConditional fields enter the sequence where expectedWCAG 2.4.3
9Focus is visible and not obscuredIndicator remains visible under headers and dialogsWCAG 2.4.7, 2.4.11
10Targets meet size or spacing ruleAt least 24 by 24 CSS pixels or a documented exceptionWCAG 2.5.8

How should errors and results behave?

Test a missing required value, a non-number, the lower and upper boundary, and a conflicting combination. The error needs plain-language text, programmatic association and a correction route. Do not erase valid entries after another field fails.

Then calculate a valid result and change one input. The result should identify its value, unit, time period and relevant assumption. Use a polite live region for a concise update when appropriate; avoid announcing every keystroke in a long formula.

Tests 11 to 15: validation and result
#TestPass evidencePrimary mapping
11Invalid fields are identifiedError text names the field and problemWCAG 3.3.1
12Correction is suggestedAccepted range or format is statedWCAG 3.3.3
13Errors are programmatically connectedaria-describedby or equivalent relationship worksWCAG 1.3.1, 4.1.2
14Dynamic result is communicatedA concise update is announced without moving focus unexpectedlyWCAG 4.1.3
15Meaning does not rely on colorText and values explain positive, neutral or negative statesWCAG 1.4.1

What layout and embed tests remain?

Zoom text to 200%, test a 320 CSS pixel viewport and inspect both the host page and calculator document. Tables may scroll within a labeled region, but the primary calculation flow should not require simultaneous horizontal and vertical scrolling.

Tests 16 to 18: reflow and embedding
#TestPass evidencePrimary mapping
16Text resizes and content reflowsNo lost controls or clipped result at zoom and narrow widthWCAG 1.4.4, 1.4.10
17Motion is avoidableReduced-motion preference removes nonessential transitionsWCAG 2.3.3
18Embedded context is named and escapableiframe title is useful and focus does not become trappedWCAG 2.1.2, 4.1.2

What does this checklist not prove?

Passing these checks is not a certification or legal opinion. It does not replace testing with disabled users, a broader WCAG conformance evaluation or review of the surrounding product journey. Treat the sheet as a repeatable release gate and retain failures as work items rather than averaging them into a score.

Method and evidence

Evidence type: 18-test WCAG mapping and manual QA decision sheet

  1. Mapped calculator states to W3C forms guidance and relevant WCAG 2.2 criteria.
  2. Separated automatically detectable checks from keyboard, screen-reader and cognitive checks that require human judgment.
  3. Covered the empty state, invalid state, calculated result, conditional branch and embedded context.
  4. Defined pass evidence so a reviewer can reproduce the decision instead of recording only a score.

Topic score: 4.68 / 5. Business fit 4.5, verified demand 4, distinct intent 5, original evidence 5, citation usefulness 5, feasibility 5.

Primary sources

Limitations

  • The checklist is a release aid, not a claim of WCAG conformance or legal compliance.
  • Assistive-technology behavior varies by browser and platform, so representative combinations still need direct testing.

Verification and corrections

Checklist mapping to current W3C guidance verified 24 August 2026.

Recommended retest: Retest before every calculator release and after changes to fields, logic, results, charts or embeds.

Found an error or a changed standard? Use the correction process and include the page URL and primary evidence.

Next step

Apply the evidence to your next release

Use the published method, keep a dated test record and revisit the result after the calculator or its operating rules change.

Open the testing protocol