Validation report · SD-01 · v0.1
Frontend Web Development
Validation in progress — not yet validated
Last updated 18/10/2026 · Next review 2027-04-18
Measured validation results
No validation statistics yet.
Reliability, reviewer agreement, false-positive and false-negative rates and predictive evidence will be published only after at least 30 consenting candidates have completed the study. Current recorded evidence: 0 submitted assessment attempts and 0 finalised independent reviews.
Competency framework
| Competency area | Weight | What is measured |
|---|---|---|
| Semantic HTML and accessibility | 25% | Accessible structure, landmarks, labels, keyboard support (WCAG 2.2 AA). |
| CSS layout and responsive design | 25% | Flexbox/grid, responsive breakpoints, reflow at 320px. |
| JavaScript/TypeScript and state | 30% | Correct DOM/state logic, types, async data handling, error states. |
| Performance and testing | 20% | Bundle and render performance, automated tests for key behaviour. |
Proficiency levels
- Foundational
- Completes well-defined tasks with guidance; recognises standard patterns.
- Professional
- Delivers production-quality work independently on typical tasks; explains trade-offs.
- Advanced
- Solves ambiguous problems, improves others' work, anticipates failure modes.
- Expert
- Sets standards, designs systems others build on, resolves novel high-risk problems.
Practical tasks
Build an accessible, responsive form-driven page from a design spec
Measures: Semantic HTML and accessibility, CSS layout and responsive design · 120 min · Tools: Any editor, docs; no AI code generation · Deliverable: Repository link + live URL
Fix and test a buggy component that fetches and displays data
Measures: JavaScript/TypeScript and state, Performance and testing · 90 min · Tools: Provided starter repo · Deliverable: Pull request with tests
Scoring rubric
Each criterion is scored 0–3 by each reviewer independently.
Semantic HTML and accessibility
Accessible structure, landmarks, labels, keyboard support (WCAG 2.2 AA).
- 0 — Not demonstrated or incorrect.
- 1 — Partially demonstrated; significant gaps or errors.
- 2 — Demonstrated to a working standard with minor issues.
- 3 — Fully demonstrated; robust, clear and well-justified.
Partial credit: Scores 1–2 allowed where the approach is sound but incomplete.
Critical errors (automatic fail): Missing form labels or keyboard trap; Plagiarised or AI-generated submission not disclosed
CSS layout and responsive design
Flexbox/grid, responsive breakpoints, reflow at 320px.
- 0 — Not demonstrated or incorrect.
- 1 — Partially demonstrated; significant gaps or errors.
- 2 — Demonstrated to a working standard with minor issues.
- 3 — Fully demonstrated; robust, clear and well-justified.
Partial credit: Scores 1–2 allowed where the approach is sound but incomplete.
Critical errors (automatic fail): Missing form labels or keyboard trap; Plagiarised or AI-generated submission not disclosed
JavaScript/TypeScript and state
Correct DOM/state logic, types, async data handling, error states.
- 0 — Not demonstrated or incorrect.
- 1 — Partially demonstrated; significant gaps or errors.
- 2 — Demonstrated to a working standard with minor issues.
- 3 — Fully demonstrated; robust, clear and well-justified.
Partial credit: Scores 1–2 allowed where the approach is sound but incomplete.
Critical errors (automatic fail): Missing form labels or keyboard trap; Plagiarised or AI-generated submission not disclosed
Performance and testing
Bundle and render performance, automated tests for key behaviour.
- 0 — Not demonstrated or incorrect.
- 1 — Partially demonstrated; significant gaps or errors.
- 2 — Demonstrated to a working standard with minor issues.
- 3 — Fully demonstrated; robust, clear and well-justified.
Partial credit: Scores 1–2 allowed where the approach is sound but incomplete.
Critical errors (automatic fail): Missing form labels or keyboard trap; Plagiarised or AI-generated submission not disclosed
Pass threshold
85%. SkillSignature applies one 85% pass mark across skills by policy. For this skill it will be checked with a modified-Angoff standard-setting panel and against benchmark outcomes from the pilot; any change will be versioned and will not alter credentials already issued.
Validation study plan
Pilot study: each submission scored independently by two approved reviewers before reconciliation, plus an independent benchmark rating by a senior practitioner.
Sample: Minimum 30 consenting candidates per skill before any statistic is published; 100+ before claiming stability.
- Internal consistency (Cronbach's alpha / McDonald's omega) across rubric criteria
- Inter-rater agreement (Cohen's weighted kappa, ICC) before reconciliation
- False-positive and false-negative rates against the independent benchmark
- Pass-decision consistency at the 85% threshold
- Predictive evidence: employer performance ratings at 3 and 6 months, with consent
Fairness: Group comparisons only where self-declared data and sample sizes allow; differences investigated, not hidden.
Consent: Candidates opt in to research use; results reported in aggregate only.
Reviewers
Reviewer qualifications, calibration, double scoring, disagreement handling and appeals for this skill are published on the reviewer standards page.
Limitations
- No candidate has completed this assessment yet; no reliability, agreement, error-rate or predictive statistics exist.
- Competency framework and rubric are drafts written by the SkillSignature Standards Team and await independent subject-matter expert review.
- The 85% pass threshold is a platform-wide policy, not yet confirmed by a standard-setting study for this skill.
- Listing a skill or publishing this report does not mean the assessment is validated.