Manual testing to automatic testing A formal training programme for testers at Bands 3 to 7

Accessibility check

Every resource in the reading list is checked before the cohort starts, following Universal Design for Instruction (spec, "Inclusion and reasonable adjustments"). A resource that fails a check needs an alternative format, or a replacement, before it is used.

Checklist for each resource

# Check How
1 Screen reader Navigate the resource with a screen reader (for example VoiceOver or NVDA). Headings, links, tables, and code blocks are announced sensibly.
2 Keyboard only Every link, control, and video player works without a mouse. Focus is visible.
3 Contrast Text, including code blocks, meets WCAG 2.2 AA contrast (4.5:1 for normal text). Check light and dark themes.
4 Zoom and reflow Readable at 200% zoom with no loss of content, and at 400% with no horizontal scrolling for prose.
5 Captions and transcripts Every video has accurate captions. A transcript exists or can be provided.
6 Alternative formats The content is available in at least one other format: text for video, a downloadable or printable version for web pages, or a summary from the mentor.
7 Plain language Jargon is explained, or the session explains it.
8 Images and diagrams Meaningful images have text alternatives. Diagrams are explained in text.
9 Code samples Code is real text, not an image, so it can be read aloud, copied, and enlarged.
10 Tools Tools used in the resource (editor, browser, CI) work with the assistive technology people use.

Status values: not yet checked, passed, passed with alternative (say which), failed (say what replaces it).

Automated scan

accessibility-scan-results.md holds the latest automated axe-core scan of every web resource in the reading list. To re-run it before each cohort:

cd scripts/accessibility-scan
npm install
npx playwright install chromium
npm run scan

The scan is an organiser's tool, run by the training lead, not by participants. Re-run it before each cohort, and whenever the reading list changes.

The scan supports checks 3 and 8 only, so every status below stays not yet checked until a person completes the manual checks. Findings from the 2026-10-07 scan of the JavaScript and Selenium reading list that need action before the cohort starts:

  • testingexamples.github.io fixture page: form fields without labels (label, select-name). The fixture markup is a fixed contract, so do not change it for the course. Instead, tell screen reader users which fields are which, and give them the exercise sheet in modules/module-10-browser-automation-fundamentals/exercise-sheet.md, which names every fixture.
  • testingexamples Given-When-Then page and the fixture spec on GitHub: scrollable code blocks cannot be reached by keyboard (scrollable-region-focusable). Provide the code as text files, or report the issue to the site's maintainer.
  • roles-skills.github.io: colour contrast failures on 14 elements (color-contrast). Offer the self-assessment TSV files in instruments/ as the alternative format, and report the issue to the site's maintainer.
  • loinc.org and iso.org: blocked the automated browser (HTTP 403). Check them by hand.

Resources

Module Resource Status Checked by Date Notes
Module 6 roles-skills reference website not yet checked
Module 6 roles-skills self-assessment guide not yet checked
Module 6 UK GDaD PCF website not yet checked
Module 6 Programme specification not yet checked
Module 6 Capability self-assessment instrument (instruments/) not yet checked Must also work when completed in conversation
Module 7 What is automatic testing? not yet checked
Module 7 What is the purpose of automatic testing? not yet checked
Module 7 What is the automatic testing pyramid? not yet checked Check any pyramid diagram has a text description
Module 7 What is browser automatic testing? not yet checked
Module 7 What is continuous integration automatic testing? not yet checked
Module 7 How does Six Sigma lead manual testing into automatic testing? not yet checked
Module 7 Automation candidate analysis template (TSV) not yet checked Check in the spreadsheet tools people use
Module 8 Structured JavaScript course (Decision 3) not yet checked Videos need captions
Module 8 MDN JavaScript Guide, and using Promises not yet checked
Module 8 Mocha and Node.js assert documentation not yet checked
Module 8 What are related concepts for automatic testing? not yet checked
Module 8 How to start learning automatic testing? not yet checked
Module 8 Visual Studio Code and its documentation not yet checked Check screen reader mode and high-contrast themes
Module 8 Kata guide and kata files not yet checked
Module 9 Pro Git book not yet checked
Module 9 The team's git hosting service (pull requests, review) not yet checked
Module 10 Fixture site not yet checked
Module 10 Fixture contract not yet checked
Module 10 Selenium JavaScript skill not yet checked
Module 10 Selenium JavaScript demo not yet checked
Module 10 Selenium JavaScript demo, and the existing suite exercise not yet checked
Module 10 Selenium documentation: WebDriver, locators, waits, select lists not yet checked
Module 10 Selenium IDE browser extension not yet checked Recording happens in a browser extension panel: check it with a screen reader and keyboard only, and plan an alternative, such as writing the steps by hand with the mentor
Module 10 Exercise sheet not yet checked
Module 11 Given-When-Then Examples not yet checked
Module 11 NHS Wales worked example and spec not yet checked
Module 11 Selenium documentation: page object models not yet checked
Module 11 Failure evidence in test-results/ (screenshots, page source, console logs) not yet checked Screenshots are images: the page source and Mocha messages are the text alternative
Module 11 Given-When-Then template and worked example not yet checked
Module 12 HL7 FHIR R4 specification not yet checked
Module 12 HAPI FHIR (background only) not yet checked
Module 12 MDN: using the Fetch API not yet checked
Module 12 LOINC not yet checked
Module 12 FHIR sandbox guide not yet checked
Module 12 Read-an-existing-suite exercise not yet checked
Module 13 What is DevOps for automatic testing? not yet checked
Module 13 Selenium documentation: Chrome options; Mocha reporters; GitHub Actions job matrix not yet checked
Module 13 The organisation's CI service not yet checked Check pipeline logs and reports with a screen reader
Module 13 CI failure triage template and triage exercise not yet checked
Module 14 Clinical risk management process and hazard log not yet checked
Module 14 Information governance policy not yet checked
Module 14 Synthea not yet checked
Module 14 @axe-core/webdriverjs documentation not yet checked
Module 14 Traceability matrix template (TSV) not yet checked
Module 14 Synthetic data and secrets checklist not yet checked
Module 15 What metrics help automatic testing? not yet checked Check any embedded video has captions
Module 15 How does artificial intelligence help automatic testing? not yet checked
Module 15 Selenium documentation: waits (flaky tests) not yet checked
Module 15 Flaky-test investigation and suite-health report templates not yet checked
Module 16 Capstone briefs, scope agreement, presentation outline not yet checked
Module 17 How does Six Sigma lead manual testing into automatic testing? not yet checked
Module 17 Certification body's Green Belt course materials and exam (Decision 8) not yet checked Ask the body for its accessible formats and exam adjustments
Module 17 Green Belt project charter template not yet checked

When a check fails

  1. Record what failed in the Notes column.
  2. Find an alternative: another format, another resource with the same content, or a mentor-led session.
  3. If a person's reasonable adjustment depends on it, tell the training lead before the module starts.
  4. Recheck at the 6-monthly maintenance review.

View this document on GitHub materials/accessibility-check.md