IB predicted grades influence university applications and help the IB evaluate whether awarded results appear appropriate. They should be accurate forecasts grounded in student work, not aspirational targets, rewards for effort, or responses to pressure from families and universities.
The most effective IB predicted grades best practices combine a common school policy with subject-specific professional judgment. Coordinators should standardize evidence quality, moderation, documentation, bias checks, and review routes without imposing one numerical formula on subjects that assess different skills.
What the IB officially expects
The IB definition of a predicted grade is the teacher's prediction of the grade a candidate is expected to achieve, based on evidence of the candidate's work and knowledge of IB standards. Predictions are required for subjects, Theory of Knowledge, and the Extended Essay. Schools should avoid both under-prediction and over-prediction.
The IB's Programme standards and practices expects schools to communicate predictions accurately and maintain consistent, fair, and transparent reporting processes. However, public IB documentation does not prescribe a universal weighting for mocks, coursework, class tests, and performance trends. Each school must create a defensible process aligned with current subject guides, assessment criteria, and IB grade descriptors.
How bias enters predicted-grade decisions
Bias is not always deliberate discrimination. It often appears when irrelevant information, inconsistent evidence, or cognitive shortcuts affect how performance is interpreted.
| Risk | Possible distortion | Appropriate control |
|---|---|---|
| Halo effect | Participation or likability is mistaken for attainment | Separate academic evidence from behavior |
| Anchoring | An early result fixes expectations too firmly | Consider recent and representative evidence |
| Confirmation bias | Contradictory work is discounted | Summarize the complete evidence pattern |
| Recency bias | One recent result receives excessive weight | Review several comparable assessments |
| Opportunity bias | Unequal conditions affect available evidence | Check access arrangements and assessment conditions |
| Pressure bias | Admissions requirements encourage inflation | Separate entry thresholds from attainment judgments |
Ofqual's objective-judgment guidance is not an IB regulation, but it provides useful principles. It recommends excluding irrelevant factors such as personality, behavior, family background, sibling performance, and knowledge of university requirements.
Context still matters when it affects evidence validity. A student entitled to approved access arrangements should receive them where applicable. Context may explain missing or unrepresentative evidence, but it cannot justify an unsupported increase or decrease.
Establish a school-wide evidence framework
Standardization does not mean assigning identical percentages to mocks or coursework. Mathematics: Analysis and Approaches and Visual Arts necessarily use different evidence. Coordinators should standardize the principles governing evidence, while departments explain their subject-specific application.
Evidence should be:
- Relevant to the current subject guide
- Representative of content and assessment objectives
- Recent, while accounting for sustained performance
- Independent, with supported work clearly identified
- Comparable in conditions and difficulty
- Authenticated, especially for coursework
- Sufficient, rather than based on one result
Each department should publish a short protocol explaining the evidence normally considered, treatment of incomplete components, and circumstances requiring professional judgment. The RevisionDojo parent guide to predicted grades can help schools communicate why methods differ between subjects.
Use a common predicted-grade workflow
1. Set policy before considering individual outcomes
Publish responsibilities, timelines, evidence requirements, moderation rules, confidentiality expectations, and review procedures before teachers propose grades. Predetermined procedures are less likely to shift in response to a difficult case.
2. Calibrate teachers against IB standards
Subject teams should independently mark representative scripts using current criteria or markschemes, compare decisions, and discuss important grade boundaries. The IB assessment guide explains how standardization and quality controls support consistent marking. Schools cannot reproduce the complete IB awarding process, but they can reduce dependence on which teacher marked the work.
3. Record the evidence behind every prediction
Use a common template recording the proposed grade, assessment results, coursework evidence, trends, missing evidence, access arrangements, and a concise standards-based rationale. Departments need not use identical weightings, but every teacher should explain why the complete evidence supports the decision.
4. Moderate within departments
A second specialist should review borderline cases, unusual profiles, new teachers' judgments, and unexplained differences between classes. Anonymized script review can reduce the influence of names or prior expectations. Moderation by exception is usually more practical than re-marking everything.
RevisionDojo's resources on standardizing internal assessment and simplifying internal moderation provide adaptable approaches using anchor samples and recorded rationales.
5. Conduct coordinator-level checks
The coordinator should confirm that departments followed the agreed process, not replace subject expertise. Investigate unexplained differences between classes, predictions inconsistent with representative evidence, missing rationales, and changes following family or admissions discussions.
Historical results are a sense check, not a quota. A stronger cohort may justifiably receive higher predictions, provided current evidence demonstrates that improvement.
6. Audit equality and possible bias
Where lawful and compatible with school policy, compare prediction patterns across relevant groups. A statistical difference prompts investigation but does not prove discrimination. Examine attainment, assessment opportunities, teacher patterns, access arrangements, and sample size rather than mechanically changing grades to equalize averages.
7. Approve and retain the record
Require teacher and reviewer sign-off. Securely retain the evidence, rationale, moderation decision, and subsequent amendments according to school data policies. Explain that a prediction is an evidence-based forecast, not a guaranteed result.
The RevisionDojo IB Grade Calculator may support a sense check, but no calculator can replace teacher judgment because future boundaries and examination performance remain uncertain.
Standardize processes without erasing subject differences
| Standardize school-wide | Keep subject-specific |
|---|---|
| Definition and purpose | Appropriate assessment types |
| Evidence-quality principles | Weight given to papers, oral work, practicals, or coursework |
| Moderation and sign-off | Interpretation of subject criteria |
| Documentation and review routes | Selection of anchor samples |
| Rules excluding irrelevant factors | Final judgment within the framework |
Complete departmental autonomy creates inconsistency, while a rigid whole-school formula reduces validity. The appropriate balance is common decision controls combined with specialist interpretation.
Build a fair review process
Students should be able to request a procedural review when evidence was omitted, recorded incorrectly, or treated inconsistently. Reviews need a published deadline, defined grounds, and an impartial reviewer. They should test whether policy was followed and the prediction remains defensible, not negotiate grades according to effort or university requirements.
Record the original decision, evidence, reviewer, outcome, and reason for any change. The RevisionDojo predicted-grade appeals guide offers an evidence-focused structure.
Evaluate accuracy after final results
Compare predictions with awarded grades by subject, teacher, and cohort. Examine exact agreement, results within one grade, and patterns of over-prediction or under-prediction. Because final grades also reflect examination-day performance and moderation, discrepancies should prompt professional inquiry rather than automatic blame.
Small cohorts require multi-year analysis. Use findings to improve assessment design, calibration, and evidence rules before the next prediction cycle.
Common mistakes to prevent
- Treating previous grade boundaries as guaranteed thresholds
- Substituting conduct, effort, or participation for attainment
- Equating target grades with predicted grades
- Changing predictions to satisfy university requirements
- Giving one teacher sole authority over borderline cases
- Using historical distributions as fixed quotas
- Treating calculator or AI estimates as final decisions
- Offering informal reviews without deadlines or records
Conclusion
Reducing grading bias requires a shared evidence framework, calibration against IB standards, documented review, cross-department checks, equality analysis, and a transparent challenge process. Effective systems standardize decision quality without imposing false uniformity across subjects.
RevisionDojo's moderation guidance and Grade Calculator can support consistent practice, but final predictions must remain documented professional judgments made by the school. Coordinators should begin with a common evidence template and scheduled departmental calibration.
Sources and referenced URLs
- IB student assessment and predicted-grade definition
- IB Programme standards and practices
- IB Diploma Programme grade descriptors
- IB assessment guide for teachers and coordinators
- Ofqual guidance on objective judgments
- RevisionDojo parent guide to IB predicted grades
- RevisionDojo guide to standardizing internal assessments
- RevisionDojo guide to simplifying internal moderation
- RevisionDojo IB Grade Calculator
- RevisionDojo predicted-grade appeals guide