Reference Chapter 147 · Judge professional teaching fairly, improve what happens in classrooms and protect educators from arbitrary decisions.
A teacher receives an excellent annual appraisal because students obtained high examination grades. Yet many of those students entered the year already achieving at a high level. In another school, a teacher takes a class with weak algebra foundations and helps pupils make substantial progress, but the final examination average remains lower. Elsewhere, an observer visits one unusually difficult lesson and turns a single impression into a career-changing judgement. An appraisal system that cannot distinguish starting points, instructional competence, contextual constraints and genuine improvement may be administratively tidy but educationally unfair.
The gold standard is a national teacher appraisal and professional-review system that sets legitimate standards in advance, draws on multiple appropriate sources of evidence, differentiates coaching from consequential decisions, gives teachers meaningful feedback and establishes fair review procedures. It should protect children from consistently inadequate instruction while also protecting professional educators from unsupported accusations and crude numerical targets.
This chapter covers employed teachers after appointment: formative and summative appraisal, classroom observation, student work, professional judgement, evaluator training, calibration, employment consequences, improvement plans and appeal. It complements Chapter 111 on Early-Career Induction and Probation, Chapter 120 on Continuing Professional Development, and Chapter 123 on Workload and Protected Preparation. Those address professional support and job design; this chapter addresses the consequential judgement of classroom work and how that judgement can be corrected.
1. State the two purposes of appraisal without confusing them
The OECD TALIS 2024 chapter on the demands of teaching distinguishes formative appraisal, which helps educators improve practice, from summative appraisal, which evaluates performance for accountability or other consequential decisions.
A National Teacher Appraisal Purpose Register should identify which conversation is developmental, which judgement may affect employment or compensation, and which evidence or rights apply to each.
A mentor may help a teacher discover why pupils misunderstand fractional equations. That exchange is not necessarily a formal disciplinary hearing. A formal finding about persistent inability to teach required curriculum needs the relevant law, standard and review, not merely a coaching note.
Where the same evidence may serve both purposes, teachers must know the boundary beforehand. Trust requires accurate expectations rather than an assurance that nothing will count followed by a surprise personnel decision.
2. Use recent international evidence as context, not a copied rule
The OECD Education at a Glance 2026, indicator on teacher and school-head appraisal, describes how appraisal frameworks vary and reports that 24 of 30 responding countries or economies have formal policy frameworks covering at least one type of teacher appraisal.
Its discussion of TALIS finds widespread appraisal, but frequency, responsible assessors, evidence and consequences differ. A country’s decision to appraise annually, less often or through specified career stages is an institutional and legal choice, not an OECD requirement.
A National Appraisal Evidence Note should distinguish survey responses, regulatory requirements, actual school practice and independently validated teaching outcomes.
These sources can help compare designs. They cannot establish that a particular teacher is effective or ineffective without appropriate evidence of that person’s work.
3. Establish legitimate teaching standards before observing staff
A teacher needs to understand what competent practice means in the relevant subject and career stage.
A Professional Teaching Standards Crosswalk should connect lawful registration and employment expectations with subject knowledge, coherent explanations, assessment, accessible participation, safeguarding, ethical conduct and appropriate communication.
Avoid generic adjectives such as dynamic, charismatic or inspiring unless the framework explains what observable educational practice they represent. A quiet teacher may offer excellent mathematics explanations and valid feedback.
Teaching standards must also distinguish responsibilities held by a classroom teacher, specialist, middle leader or school principal. A teacher should not be penalised for failing to perform powers the institution never assigned.
4. Make the teaching assignment part of the evaluation
The difficulty of the timetable, subject, learner starting point, facilities and legitimate support influences what can be observed.
A Teaching Context Record should identify the actual class, curriculum stage, available instructional time, subject prerequisites, learner access needs and material disruptions.
Context is not an excuse for poor instruction. It is evidence that helps determine whether an outcome reflects teacher practice, missing support or another part of the education service.
One school may have a functioning laboratory; another may have cancelled most practical work because a facility was unsafe. Their teachers should not be judged by a misleadingly identical practical-lesson count.
5. Use observation to describe evidence rather than personality
The OECD TALIS 2024 evidence finds classroom observation a common appraisal method. But a visit is only as useful as the observer’s understanding of the intended lesson.
A Classroom Observation Evidence Sheet should record the concept being taught, prior knowledge, actual explanation, tasks, student responses, formative checks and relevant constraints.
“Teacher lacked energy” is weak evidence. “Several pupils continued to invert only the numerator when dividing algebraic fractions, and the lesson did not test that misconception” is a more concrete instructional observation.
The observer should distinguish what occurred from interpretations and recommendations. Professional judgement still matters, but it should be anchored in defensible examples.
6. Never treat a single observed lesson as the entire career
A class may be unusually distracted, a laboratory temporarily unavailable or a lesson deliberately focused on independent consolidation.
A Repeated Observation and Evidence Rule should define when multiple observations, student work, relevant planning and professional dialogue are necessary for consequential judgements.
The actual sampling model should reflect the job and national requirements. Do not invent a fixed global number of observations sufficient for every appraisal.
A lesson that raises a serious safeguarding or professional concern may require immediate competent action; that is a separate issue from using one ordinary lesson as a complete appraisal of teaching quality.
7. Give subject expertise a genuine role in evaluation
A general evaluator may understand classroom climate but lack the expertise to judge an unusual mathematics method or scientific explanation.
A Subject-Competent Appraisal Panel should identify when a qualified subject reviewer is necessary, especially where the finding concerns accuracy of mathematical reasoning, laboratory instruction or assessment interpretation.
For example, the equivalent transformations \(2(x+3)=10\) and \(x+3=5\) are mathematically consistent, though lesson suitability depends on the problem and pupils. A reviewer unfamiliar with the concept should not penalise a teacher for using a valid alternative method.
The school should distinguish pedagogical preference from academic error. Different legitimate teaching approaches can meet the same subject standard.
8. Treat student examination results as one piece of evidence
A high class average may reflect earlier attainment, selective admissions, tutoring, pupil motivation, teaching and other factors. A low average can also conceal substantial progress from a difficult starting point.
A Student Outcome Interpretation Gate should identify the assessment, pupils included, starting levels, curricular coverage, measurement reliability, relevant supports and changes over time.
Do not use raw examination averages as the sole basis for teacher promotion or punishment. Nor should a ministry discard learning evidence entirely. Appropriate student work and valid outcomes are important, but their limitations must be understood.
See Examination Score Comparability for the underlying measurement concerns.
9. Judge explanation and feedback through learner thinking
A polished slide presentation is not proof that students understand the lesson. An engaged discussion is not sufficient if key concepts remain incorrectly taught.
A Learning Evidence Observation may sample students’ initial responses, guided practice, independent attempts and later transfer to new problems.
When a teacher introduces algebraic fractions, the evaluator might examine whether pupils understand equivalent forms, not merely whether they completed a worksheet.
The evaluator should not claim to diagnose every student’s learning from a brief visit. Select appropriate evidence and identify what further checking is necessary.
10. Separate professional performance from worker popularity
Students and families may offer useful feedback. A teacher’s popularity is not identical to competence, and criticism after a difficult grade or discipline decision may not establish misconduct.
A Stakeholder Feedback Reliability Review should distinguish the question asked, respondent population, relevant experience, possible conflicts and whether an allegation requires separate investigation.
A child may accurately report that they cannot hear the teacher’s explanation; the school should investigate the issue. A parent demanding a higher grade is not automatically a reliable judge of marking quality.
Student voice matters when it informs a fair examination of educational services. It should not become a popularity contest.
11. Design fair use of professional portfolios
Teachers may submit lesson plans, assessment examples, feedback, reflection and development records. Such evidence can show sustained practice but may also generate unnecessary paperwork.
A Teacher Appraisal Portfolio Standard should identify necessary artefacts, permitted examples, confidentiality, submission limits and which decisions each item supports.
A portfolio of fifty glossy lesson templates is not more reliable than a small set of accurately analysed student learning examples.
The school should avoid requiring teachers to disclose protected learner records in unsecured files merely to satisfy an evaluator’s request.
12. Train evaluators in accuracy and bias
Appraisal can suffer from recency effects, personal preference, unconscious assumptions, conflicts and inconsistent rubrics.
An Evaluator Calibration Programme should use common observation examples, valid standards, independent scoring, discussion of disagreement and controlled updates.
Where possible, significant decisions should be moderated or reviewed by another competent assessor under national rules.
A panel should distinguish an observable mathematical error from a style difference, and persistent professional concerns from a single difficult interaction.
Evaluator competence is itself a public education quality requirement.
13. Manage relationships and conflicts of interest
A line manager may have a legitimate employment supervisory role, but a personal dispute should not secretly shape a consequential rating.
A Appraisal Conflict and Recusal Record should identify significant interests, prior involvement and how independent review is arranged under the applicable rules.
Not every disagreement requires recusal. The standard is fair procedure with documented reasons and appropriate authority.
Teachers need a credible route to raise an appraisal concern without assuming the original assessor can unilaterally dismiss every challenge.
14. Keep developmental feedback specific and useful
The TALIS 2024 analysis reports that on average across participating OECD systems, about 65% of teachers are engaged in post-appraisal discussions on improving their practice, while around 46% are offered training or development plans. These are survey descriptions, not proof that every follow-up improves teaching.
A Teacher Feedback-to-Practice Record should identify the actual issue, evidence, practical support, classroom trial and later check.
A vague recommendation to become more engaging is less useful than a targeted plan to model ratio relationships and examine unfamiliar independent student solutions.
15. Distinguish a development plan from a formal improvement notice
A capable teacher may want to deepen expertise. Another may require structured support to meet established standards. A third may face a serious formal performance process.
A Professional Support and Consequence Map should distinguish voluntary learning, mandated development, formal competence intervention and disciplinary matters.
Teachers must know which route applies and the relevant legal rights. A supportive discussion should not covertly become evidence for a sanction if the system has promised a different status.
Conversely, serious substantiated safeguarding or teaching concerns cannot be ignored under the banner of informal coaching.
16. Make a competence improvement plan teachable
When an educator is genuinely struggling, an improvement plan must explain what competent practice looks like and what assistance the institution will provide.
A Time-Bound Professional Improvement Plan should name specific issues, valid evidence, qualified support, realistic practice opportunities, review date and employment consequence under applicable law.
A teacher who cannot explain proportional reasoning may benefit from subject mentoring, observed modelling and independent checks of pupil understanding. Telling the teacher simply to improve student scores supplies no pedagogical pathway.
The plan must preserve students’ access to appropriate teaching while allowing a fair, evidence-based decision about whether the required standard has been met.
17. Separate professional ethics investigations from ordinary appraisal
Allegations involving abuse, discrimination, corruption or examination malpractice may require independent professional or statutory procedures.
A Conduct–Competence Jurisdiction Test should identify whether the concern is about teaching quality, employee conduct, child protection or another legal matter.
Do not bury a serious safeguarding allegation in a routine appraisal form. Nor should a teaching-method disagreement be dressed up as a misconduct allegation to avoid ordinary professional review safeguards.
See Teacher Ethics, Registration and Accountability.
18. Explain lawful consequences before the appraisal cycle
National systems may link appraisal to confirmation, salary progression, promotion, ongoing professional registration or improvement obligations.
A Teacher Appraisal Consequence Notice should state the actual legal and employment basis, criteria, decision-maker and review.
The ministry must not promise financial rewards or legal consequences simply because a comparative OECD report describes their use elsewhere.
Decisions should be proportionate, well evidenced and communicated in terms a professional employee can understand.
19. Provide a meaningful response and appeal process
A teacher should be able to inspect relevant findings, correct material factual errors and use an authorised review route when a consequential decision is made.
A Professional Appraisal Review Receipt should identify the decision, original evidence, employee response, reviewer, applicable timeframe, reasons and lawful outcome.
An appeal need not guarantee a changed rating, but it should be capable of correcting factual or procedural problems within its mandate.
A school leader should not arbitrarily erase evidence or alter a rating without a traceable decision.
20. Protect professional privacy and student records
Appraisals may include identifiable pupils’ work, staff health issues, family complaints and confidential employment records.
A Teacher Evaluation Data Governance Standard should identify lawful purpose, necessary information, authorised personnel, retention, correction and restrictions on reuse.
Student records used as evidence should be protected; staff appraisal files should not be shared with vendors or unrelated school personnel.
An AI system that generates a teacher rating from classroom recordings must be subject to the same legal and technical scrutiny as any consequential automated decision.
21. Avoid AI-generated teacher rankings without meaningful oversight
Automated dashboards can count activities, completion and results, but they may not measure subject understanding or teaching validity.
An Automated Appraisal Assurance Gate should examine whether a metric measures the intended professional construct, how errors are detected, whether human expertise reviews outputs and how the teacher can challenge consequential errors.
A teacher posting fewer digital worksheets may be delivering more effective direct instruction. A high number of app interactions does not prove good teaching.
Do not permit a proprietary score to decide a career outcome when its input, meaning and limitations cannot be examined.
22. Examine workload and cost of appraisal itself
An elaborate evaluation cycle may use considerable teacher and leader time without improving practice.
A National Appraisal Burden Review should identify time spent collecting documents, observing, providing feedback, meeting and handling disputes, and the benefit of each step.
Different career stages may warrant different levels of observation or support. The OECD EAG 2026 review discusses differentiated arrangements, but no one template is universally optimal.
Preserve sufficient evidence for legitimate decisions; remove rituals that add documentation without improving fairness or teaching.
23. Connect appraisal findings to training and workforce planning
Recurring problems may reveal gaps in initial teacher education, subject support, material availability or school leadership rather than only individual deficits.
A System-Level Appraisal Learning Report should aggregate appropriately protected findings and identify development priorities for teacher institutes, departments and workforce planners.
A cluster of science teachers unable to deliver practical learning because laboratories are closed should trigger facilities action as well as professional support where needed.
Aggregate findings must be interpreted with care: schools serving challenging populations may report different patterns without being inherently less capable.
24. Use an auditable quality dashboard
A Teacher Appraisal Quality Scorecard may examine: – Clear lawful standards and staff understanding. – Competent evaluator training and moderation. – Useful observation and subject evidence. – Appropriate student outcomes interpretation. – Timely formative feedback and real support. – Conflicts and fairness safeguards. – Valid, proportionate formal consequences. – Teacher response and appeal effectiveness. – Professional workload and personal data protection. – Consistency across schools, subjects and career stages. – Authentic improvements in teaching and pupil capability.
A high percentage of positive ratings may reflect excellent teaching or weak scrutiny. A low rating rate may reflect unusually challenging duties, flawed observation or genuine concerns. Examine mechanisms rather than treating the percentages as self-explanatory.
25. Worked Mathematics case: a strong teacher with a low raw class average
Imagine two Secondary 1 Mathematics teachers. The first teaches a class whose students entered with strong Primary 6 arithmetic; the second begins with learners unable to explain equivalent fractions. At the year end, the first class averages 84 while the second averages 69.
A weak appraisal ranks the first teacher higher using raw results alone. A gold-standard review checks each class’s starting point, quality of instruction, actual student work, classroom opportunities and independent transfer tasks.
The second teacher’s pupils can now solve fraction equations and explain why both sides of the equation must remain equivalent. The first class has also progressed, perhaps to more challenging algebra. Both accomplishments deserve accurate evaluation against their contexts and responsibilities.
The goal is not to guarantee equal ratings. It is to make the evidence sufficient, fair and educationally meaningful.
26. Common failure modes and repairs
One difficult lesson defines a career
Failure: a single observation substitutes for dependable evidence. Repair: use proportionate repeated evidence and competent review.
Raw examination averages decide all ratings
Failure: pupil intake and opportunities to learn are ignored. Repair: use contextual, valid learning evidence.
Coaching is secretly used for punishment
Failure: teachers cannot seek help honestly. Repair: publish developmental and formal appraisal boundaries.
Evaluators disagree without moderation
Failure: stylistic preferences become inconsistent ratings. Repair: calibrate standards and review consequential cases.
Appraisal produces no instructional support
Failure: a rating replaces professional learning. Repair: convert credible findings into targeted improvement opportunities.
Automated metrics become employment verdicts
Failure: misleading proxies substitute for teaching expertise. Repair: require justified constructs, human oversight and correction rights.
27. The first 180 days
Days 1–30: inventory applicable teacher employment, appraisal and review law, current standards, evaluator training and urgent fairness concerns.
Days 31–60: audit representative appraisals for evidence quality, student context, curriculum expertise, follow-up and workloads.
Days 61–90: publish a National Teacher Appraisal and Fair Review Charter, evaluator calibration framework and clear support–consequence boundaries.
Days 91–120: pilot proportionate observation and feedback cycles, subject moderation and appeal of material factual errors.
Days 121–180: verify procedural fairness, staff understanding, teaching improvements, employment consequences and quality of student learning; repair defective criteria.
28. Gold-standard acceptance tests
- Are professional teaching standards and lawful consequences known before appraisal?
- Can staff distinguish developmental coaching from summative review?
- Is classroom observation linked to subject expertise and authentic evidence?
- Are learner starting points and real opportunities to teach interpreted appropriately?
- Can a consequential finding rely on more than a single unrepresentative lesson?
- Are appraisers trained, calibrated and free from improper conflicts?
- Is useful feedback followed by realistic professional support?
- Can a teacher correct factual errors and use a competent formal review route?
- Are records and automated tools governed proportionately and privately?
- Can the ministry demonstrate fair professional judgement and better instruction rather than merely completed rating forms?
29. Teacher and parent FAQ
Can one classroom observation show whether a teacher is effective?
It can provide valuable evidence, but one visit seldom covers a teacher’s full subject work and classroom conditions. Consequential decisions need the evidence required by applicable rules.
Does a high examination average prove superior teaching?
Not by itself. Pupil starting points, other supports, assessment quality and what was actually taught also matter.
Should every performance concern become a disciplinary case?
No. The school must distinguish ordinary professional development, substantiated competence concerns, formal employment processes and serious misconduct under relevant law.
Can the teacher appeal an unfair appraisal?
Review and appeal rights depend on the actual employment framework. A national system should state competent decision-makers, available routes and deadlines accurately.
30. Reference sources and reading route
Authoritative comparative sources are OECD Education at a Glance 2026: Teacher and School-Head Appraisal Systems, OECD Results from TALIS 2024: The Demands of Teaching and TALIS 2024: Developing Teacher Expertise. OECD describes survey patterns and comparative frameworks, not employment law in every country.
This is Chapter 147 of The Gold Standard of a Ministry of Education Reference Manual. Its practical coaching companion is Ministry V3.0 Reference 108 on Instructional Coaching and Peer Observation. It also connects to Teacher Careers and Labour Relations and Subject Departments and Middle Leadership.
Final principle
The gold standard of teacher appraisal is not the highest rating, the longest observation checklist or the fastest dismissal process. It is a fair and competent professional judgement that protects standards, understands real teaching, helps educators improve and keeps a lawful route open to correct mistakes.