Quick Read
A total mark can tell us that performance changed.
It may not tell us whether the change came from concept knowledge, route selection, language load, checking, speed or one unusually difficult question cluster.
The Resolution Limit asks how fine a distinction our current evidence is actually capable of supporting.
The One-Sentence Answer
A learning measure should support only diagnoses at the level of detail it can genuinely distinguish; coarse evidence can justify coarse conclusions, while fine conclusions need finer evidence.
Where the Resolution Limit Sits in Curie
The Blind Spot Test asks what the evidence cannot see at all.
The Resolution Limit asks what the evidence can see, but only approximately.
A thermometer may detect temperature without identifying the source of heat. In the same way, a mark may detect weaker performance without locating the exact educational mechanism.
Coarse Measures and Fine Measures Answer Different Questions
| Evidence source | Useful resolution | What it should not claim alone |
|---|---|---|
| Total examination mark | Broad performance under that paper | Exact weak cognitive operation |
| Topic score | Performance inside one content area | Whether failure is concept, selection or execution |
| Independent worked solution | Route, execution and checking can become visible | Long-term retention unless returned later |
| High-resolution Tutorial observation | Prompting, hesitation, explanation, recovery | What survives when the Tutorial is gone |
| Delayed mixed task | Retention + selection + independence | Every reason for failure without further narrowing |
Diagnosis: Do Not Ask a Coarse Instrument a Fine Question
“The score fell by eight marks” may be accurate.
“The learner’s conceptual representation system has weakened” may be far more specific than the mark alone can justify.
The correct next step is not to reject the score. It is to use a finer observation if the distinction matters.
Resolution Depends on the Educational Decision
Sometimes coarse resolution is enough.
If a parent only needs to know whether a topic deserves extra review, a broad performance signal may suffice. If the tutor must decide whether to reteach the concept or repair execution, finer evidence is required.
Use the lowest resolution that safely answers the decision, and increase resolution only when the decision genuinely needs it.
English
A weak comprehension score may reveal that something is wrong, but it may not distinguish passage meaning, inference, evidence selection, response structure or language control. A Tutorial can increase resolution by separating those operations.
For developmental English stages, continue through the English Tutor route.
Mathematics
A low Mathematics mark can be a coarse alarm. Finer evidence may reveal that concepts are secure but representation fails, or that route selection is accurate while arithmetic execution is unstable.
For developmental Mathematics progression, continue through the Mathematics Tutor route; detailed Mathematics diagnosis remains with the subject-specific Mathematics pages.
Science
A Science answer can be wrong because terminology is missing, the mechanism is reversed, evidence is misread or the learner cannot transfer the model into a changed setup. The answer identifies a problem; finer questioning locates it.
For developmental Science stages, continue through the Science Tutor route.
Additional Mathematics
One failed A-Math question may contain representation, method choice, algebra, domain and verification layers. The final answer has low resolution about which layer failed. The working process has much higher resolution.
The Discriminating Test Raises Resolution Efficiently
When a coarse measure reveals a problem but not its location, The Discriminating Test can increase resolution without reopening everything.
Supply one representation, remove one prompt, simplify one wording layer or hold one method constant. Observe what changes. The purpose is to distinguish nearby explanations with the smallest additional intervention.
Resolution Can Be Too Fine
More detail is not always better.
If the next action would be the same whether the problem is one of two very similar subtypes, resolving the difference may add burden without educational value. This connects directly to The Stopping Rule.
Repair: Match Diagnostic Detail to Evidence Detail
- State what the current evidence directly shows.
- Identify the finest distinction it can reasonably support.
- Ask whether the next decision requires more detail.
- If yes, choose one higher-resolution observation.
- Use Observed, Inferred, Unknown to separate direct evidence from interpretation.
- Stop increasing resolution when the next safe repair is already clear.
Tutorial and Tutor
A Tutorial can raise resolution because the tutor can see intermediate steps, prompts, timing, explanation and self-correction rather than only the final answer.
The Tutor dashboard should preserve that distinction:
“The paper shows weaker Mathematics performance. Tutorial evidence narrows the current problem to representation before route selection. We do not need to call the whole topic conceptually weak.”
What Parents Can Ask
- What can this score genuinely distinguish?
- Are we making a finer diagnosis than the evidence allows?
- What one observation would increase resolution?
- Would more detail change the next teaching action?
- Can we begin a narrow repair without diagnosing everything?
Voyage: Mature Learners Need Finer Self-Observation, Not Endless Self-Measurement
Across the Voyage, learners become better at distinguishing “I do not understand this” from “I understand it but cannot retrieve it,” or “I know the method but misread the representation.” That is useful higher-resolution self-knowledge.
Darwin: New Environments Sometimes Demand Higher Resolution
The Darwin Series asks when an old method stops being enough. A transition to harder work may require a finer diagnosis because broad labels that once guided practice no longer locate the new bottleneck.
Boundaries
- Do not infer a fine mechanism from a coarse score alone.
- Do not collect fine-grained data that will not change the teaching decision.
- Do not confuse more detail with more truth.
- Do not reduce a human learner to a set of microscopic measurements.
- Keep every diagnosis within the resolving power of its evidence.
The Core Rule of the Resolution Limit
Match the precision of every learning claim to the resolving power of the evidence, increasing observational detail only when a finer distinction would genuinely change the next educational action.
Frequently Asked Questions
Is a total mark too coarse to be useful?
No. It is useful for broad performance questions. It simply should not be asked to identify mechanisms it cannot distinguish.
What is the simplest way to raise resolution?
Observe an intermediate operation: ask the learner to represent, select, explain, execute or check separately rather than looking only at the final result.
Marie Curie Series · Blind Spot Test · Discriminating Test · Observed, Inferred, Unknown · Stopping Rule · Tutor
eduKate Sengkang | Small groups of up to 3 | 1.5-hour tutorials | Properly Taught Kids Shine a Bright Light Into the Future.
eduKate Sengkang 