How to answer comment questions is to make a relevant observation or judgement about the material and support it with evidence. Strong comments go beyond repeating what is visible but remain anchored to the data, source, text or result provided.

The word comment can feel vague because different subjects use it differently. In practice, it usually asks the learner to say something meaningful about the evidence: a pattern, implication, limitation, significance or noteworthy feature.

This guide explains how to read the stimulus, identify what is worth commenting on, support the observation, qualify the language and avoid drifting into unsupported opinion.

This guide is written for students, parents and educators who want a practical answer to how to answer comment questions in exams. The purpose is to turn the command word into a repeatable operating system: identify the task, select the correct depth and evidence, construct the response, check it against the question and then practise again on a changed prompt.

The 60-Second Answer

Read the exact question and stimulus. Identify the most relevant pattern, feature, result or implication. State the comment directly. Support it with evidence from the material. Explain briefly why the observation matters if the mark value requires it. Qualify the wording where the evidence is limited. Avoid personal opinion that is disconnected from the source.

Working definition: A comment question asks the learner to make a relevant evidence-based observation, interpretation or brief judgement about material presented in the question.

Why Comment Questions Matters

Comment requires selection because not every feature of the stimulus deserves equal attention.

The response usually needs to move one step beyond description by saying what the evidence implies or why it is noteworthy.

Evidence anchors the answer and prevents comment from becoming unsupported opinion.

The useful standard is whether the learner can recognise the command word in an unfamiliar assessment and produce the right type of reasoning without relying on a tutor, template or memorised answer.

The Hidden Problem: Comment Does Not Mean Say Anything You Notice

Students sometimes treat comment as permission to offer a general reaction.

An assessment comment must still serve the question and the stimulus.

The strongest response identifies the most relevant observation and shows exactly which evidence supports it.

The Operating Model

Decode → Inspect Evidence → Select Noteworthy Feature → State Comment → Support → Interpret → Qualify → Check Relevance.

Each stage has a different function. Interpretation identifies what the command requires. Selection chooses the relevant material. Construction turns that material into a markable response. Checking verifies scope, completeness and precision. Delayed retesting shows whether the method transfers beyond one familiar question.

How to answer comment questions: Step by Step

1. Read the exact wording

Clarify what aspect of the result, source, graph, method or statement deserves comment.

Do not respond to the entire topic.

2. Inspect the evidence

Look for patterns, anomalies, contrasts, limitations or implications.

Use what is actually present.

3. Select the strongest observation

Choose the feature most relevant to the task.

Do not list every possible comment.

4. State the comment directly

Write the observation or judgement in clear language.

Avoid vague phrases such as this is interesting.

5. Support with evidence

Use values, quotations, features or facts from the stimulus.

Make the support explicit.

6. Explain significance where needed

State why the pattern or limitation matters to the question.

Match depth to marks.

7. Qualify appropriately

Use language such as suggests, appears, limited by or consistent with where certainty is constrained.

Do not overclaim.

8. Check relevance

Ask whether the comment answers the specific request rather than offering a general opinion.

Then stop.

What Strong Performance Looks Like

• The question target is clear.
• A relevant feature is selected.
• The comment is direct.
• Evidence is explicit.
• Interpretation adds value.
• Certainty matches evidence.
• Personal opinion is avoided.
• The answer remains concise.

These indicators are behavioural rather than personal. They can therefore be taught, rehearsed, checked and improved.

Common Failure Modes

1. Personal reaction

The answer says I think this is surprising.

Replace reaction with evidence-based observation.

2. Description only

The response repeats the pattern without saying what it implies.

Add a meaningful interpretation.

3. Evidence-free claim

A judgement is stated without support.

Quote the relevant data or source feature.

4. Every-feature list

The answer comments on everything visible.

Select the strongest relevant point.

5. Overclaim

Limited evidence is treated as proof.

Use qualified language.

6. Irrelevant criticism

A limitation is mentioned that does not affect the question.

Explain why the limitation matters.

7. Vague significance

The answer says this is important without explaining why.

State the consequence.

8. Long essay

A brief comment question becomes overdeveloped.

Use marks to control depth.

How the Strategy Changes Across Subjects

Science

Comment on data quality, anomalies, trends, method limitations or whether evidence supports a conclusion.

The subject standard remains decisive. General exam technique can organise the response, but discipline-specific knowledge determines whether the content is actually correct and sufficient.

Mathematics

Comment on a graph, model, estimate, distribution or result using quantitative evidence.

The subject standard remains decisive. General exam technique can organise the response, but discipline-specific knowledge determines whether the content is actually correct and sufficient.

English

Comment on language, tone, argument or textual evidence with a brief supported interpretation.

The subject standard remains decisive. General exam technique can organise the response, but discipline-specific knowledge determines whether the content is actually correct and sufficient.

Humanities

Comment on sources, trends, claims or outcomes using contextual and source evidence.

The subject standard remains decisive. General exam technique can organise the response, but discipline-specific knowledge determines whether the content is actually correct and sufficient.

Primary School, Secondary School and Examination Years

Primary school

Use the frame I notice X; this suggests Y because of Z. Keep the evidence visible and the inference modest.

Use short prompts and concrete examples. Adults can model the response structure, then fade the scaffold as the learner becomes able to select the correct response type independently.

Secondary school

Students should practise moving one step beyond description while avoiding full evaluation unless asked.

Secondary learners should increasingly decide scope, evidence, depth and checking without waiting for a teacher to signal what belongs in the answer.

Before major examinations

Use data, source and text stimuli from authentic papers and calibrate response length to marks.

Near major exams, mix command words in the same practice set so the learner must identify the required thinking before starting to write.

A Focused 60-Minute Practice Session

• 10 minutes — identify the most comment-worthy feature in five stimuli.
• 10 minutes — write direct one-sentence comments.
• 15 minutes — add evidence and significance.
• 10 minutes — practise qualification language.
• 10 minutes — distinguish comment from describe and evaluate.
• 5 minutes — trim overlong answers.

The exact timings can change. What matters is the sequence: command identification, independent answer, feedback, targeted repair and fresh retest.

Diagnostic Checklist

• Is the target clear?
• Is the selected feature relevant?
• Is the comment direct?
• Is evidence used?
• Does interpretation add meaning?
• Is certainty calibrated?
• Is personal opinion avoided?
• Does the answer match the mark value?

Use the checklist to identify the earliest decision that failed. A weak response may begin with a command-word error long before the final sentence appears.

Practice Laboratory

Practice 1: Noteworthy-feature sort

Rank possible observations by relevance to the question.

Choose the strongest.

Practice 2: Description-to-comment

Take a descriptive sentence and add what it suggests or why it matters.

Keep the evidence.

Practice 3: Evidence tag

Attach one precise value or quotation to each comment.

Check support.

Practice 4: Qualification ladder

Rewrite proof-like claims as suggests or consistent with where appropriate.

Match certainty to evidence.

Practice 5: Limitation relevance

State one limitation and explain exactly what conclusion it weakens.

Remove generic criticisms.

Practice 6: Comment-versus-evaluate

Write a brief comment and a fuller evaluation on the same evidence.

Compare depth.

Practice 7: One-mark comment

Practise concise evidence-based observations under tight limits.

Stop early.

Practice 8: Fresh-stimulus transfer

Apply the same routine to a graph, source and paragraph.

Check transfer.

Three Student Cases

Case 1: Maren gives opinions

Maren writes that a result is good or bad without evidence.

Her tutor requires every comment to point to a feature or value.

Her responses become assessable.

Case 2: Iona only describes

Iona accurately reports a trend but never says what it suggests.

She adds one implication sentence.

Her comments move beyond observation without becoming essays.

Case 3: Leonie overclaims

Leonie treats one small dataset as definitive proof.

She practises calibrated language and limitation statements.

Her conclusions become more defensible.

How Parents Can Help Without Taking Over

Parents can support comment questions by asking the learner to explain the command and response plan rather than supplying content. Useful prompts include: “What does the command word ask you to do?”, “Which evidence belongs?”, “How many parts are needed?”, “What would make this incomplete?”, and “How will you know when to stop?”

Once the learner can perform the decision reliably, remove the prompt. The goal is independent exam control.

How Tutors Can Use a Three-Student Small Group

In a three-student tutorial, comment questions becomes visible because one learner can answer, another can challenge whether the command has been satisfied, and the third can check against a rubric or mark scheme. Roles then rotate.

After discussion, each learner completes a fresh individual question. Shared explanation becomes useful only when it transfers into solo performance.

How to Measure Progress

• Personal-opinion language decreases.
• Comments become more evidence-based.
• Description-only responses decline.
• Qualification improves.
• Limitation statements become more relevant.
• Answer length becomes more controlled.
• Cross-subject transfer improves.
• Low-mark comment questions become faster.

Marks are the final indicator, but process improvements often appear first: more direct openings, fewer irrelevant sentences, better evidence selection, stronger scope control and faster command-word recognition.

A Four-Week Implementation Plan

Week 1 — Select and state

Practise identifying one relevant feature and commenting directly.

Use simple stimuli.

Week 2 — Add evidence

Attach values, quotations or visual features.

Check support.

Week 3 — Add interpretation and qualification

Practise what the evidence suggests and where limits apply.

Compare with evaluate.

Week 4 — Simulate

Complete mixed comment questions under time.

Review whether every comment is evidence-based and relevant.

Evidence and Responsible Use

Cornell’s exam guidance emphasizes decoding what a question is asking and tailoring the response to that directive. For comment-type prompts, this supports a focused observation grounded in the provided material rather than a generic personal reaction. Princeton’s preparation guidance similarly stresses using specific evidence in practice responses.

Command-word meanings can vary slightly by subject, school and examination board. Use official syllabus documents, teacher guidance and mark schemes as the final standard, while using this framework to build transferable reasoning habits.

• Cornell: How to Tackle Exam Questions
• Cornell: Words to Watch for in Essay Questions
• Princeton: Preparing for Exams

Frequently Asked Questions

What does comment mean in an exam?

Usually make a relevant observation or brief judgement about the material and support it.

Is comment the same as describe?

Comment usually goes one step further by interpreting significance or implication.

Is comment the same as evaluate?

Not usually; evaluate generally requires more systematic criteria and weighing.

Should I give my personal opinion?

Only if the task explicitly asks for it; otherwise ground the comment in evidence.

How long should the answer be?

Use mark allocation and subject convention.

Can I comment on an anomaly?

Yes if it is relevant to the question and supported by the evidence.

Should I use cautious language?

Yes when the evidence supports only a tentative conclusion.

How do I practise?

Use varied stimuli and require every comment to cite its evidence.

Helpful Reading on eduKateSingapore

• How to Read Exam Questions
• How to Answer Describe Questions
• How to Answer Explain Questions
• How to Check Your Work
• How to Answer Data Interpretation Questions

A Comment Is an Evidence-Based Observation With Meaning

The learner should say something worth noticing and show why the evidence supports it.

Find the feature. State the comment. Support it. Explain its significance if required. Qualify where necessary. Then stop.

Properly taught kids shine a bright light into the future.

Extended Diagnostic Workshops

Workshop 1: Read the exact wording × Description only

Set up one fresh exam-style question where the learner must practise read the exact wording while watching specifically for description only. Clarify what aspect of the result, source, graph, method or statement deserves comment. Require an independent response before any model is shown so the actual decision becomes visible.

The response repeats the pattern without saying what it implies. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 2: Inspect the evidence × Overclaim

Set up one fresh exam-style question where the learner must practise inspect the evidence while watching specifically for overclaim. Look for patterns, anomalies, contrasts, limitations or implications. Require an independent response before any model is shown so the actual decision becomes visible.

Limited evidence is treated as proof. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 3: Select the strongest observation × Long essay

Set up one fresh exam-style question where the learner must practise select the strongest observation while watching specifically for long essay. Choose the feature most relevant to the task. Require an independent response before any model is shown so the actual decision becomes visible.

A brief comment question becomes overdeveloped. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 4: State the comment directly × Evidence-free claim

Set up one fresh exam-style question where the learner must practise state the comment directly while watching specifically for evidence-free claim. Write the observation or judgement in clear language. Require an independent response before any model is shown so the actual decision becomes visible.

A judgement is stated without support. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 5: Support with evidence × Irrelevant criticism

Set up one fresh exam-style question where the learner must practise support with evidence while watching specifically for irrelevant criticism. Use values, quotations, features or facts from the stimulus. Require an independent response before any model is shown so the actual decision becomes visible.

A limitation is mentioned that does not affect the question. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 6: Explain significance where needed × Personal reaction

Set up one fresh exam-style question where the learner must practise explain significance where needed while watching specifically for personal reaction. State why the pattern or limitation matters to the question. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says I think this is surprising. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 7: Qualify appropriately × Every-feature list

Set up one fresh exam-style question where the learner must practise qualify appropriately while watching specifically for every-feature list. Use language such as suggests, appears, limited by or consistent with where certainty is constrained. Require an independent response before any model is shown so the actual decision becomes visible.

The answer comments on everything visible. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 8: Check relevance × Vague significance

Set up one fresh exam-style question where the learner must practise check relevance while watching specifically for vague significance. Ask whether the comment answers the specific request rather than offering a general opinion. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says this is important without explaining why. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 9: Read the exact wording × Description only

Set up one fresh exam-style question where the learner must practise read the exact wording while watching specifically for description only. Clarify what aspect of the result, source, graph, method or statement deserves comment. Require an independent response before any model is shown so the actual decision becomes visible.

The response repeats the pattern without saying what it implies. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 10: Inspect the evidence × Overclaim

Set up one fresh exam-style question where the learner must practise inspect the evidence while watching specifically for overclaim. Look for patterns, anomalies, contrasts, limitations or implications. Require an independent response before any model is shown so the actual decision becomes visible.

Limited evidence is treated as proof. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 11: Select the strongest observation × Long essay

Set up one fresh exam-style question where the learner must practise select the strongest observation while watching specifically for long essay. Choose the feature most relevant to the task. Require an independent response before any model is shown so the actual decision becomes visible.

A brief comment question becomes overdeveloped. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 12: State the comment directly × Evidence-free claim

Set up one fresh exam-style question where the learner must practise state the comment directly while watching specifically for evidence-free claim. Write the observation or judgement in clear language. Require an independent response before any model is shown so the actual decision becomes visible.

A judgement is stated without support. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 13: Support with evidence × Irrelevant criticism

Set up one fresh exam-style question where the learner must practise support with evidence while watching specifically for irrelevant criticism. Use values, quotations, features or facts from the stimulus. Require an independent response before any model is shown so the actual decision becomes visible.

A limitation is mentioned that does not affect the question. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 14: Explain significance where needed × Personal reaction

Set up one fresh exam-style question where the learner must practise explain significance where needed while watching specifically for personal reaction. State why the pattern or limitation matters to the question. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says I think this is surprising. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 15: Qualify appropriately × Every-feature list

Set up one fresh exam-style question where the learner must practise qualify appropriately while watching specifically for every-feature list. Use language such as suggests, appears, limited by or consistent with where certainty is constrained. Require an independent response before any model is shown so the actual decision becomes visible.

The answer comments on everything visible. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 16: Check relevance × Vague significance

Set up one fresh exam-style question where the learner must practise check relevance while watching specifically for vague significance. Ask whether the comment answers the specific request rather than offering a general opinion. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says this is important without explaining why. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 17: Read the exact wording × Description only

Set up one fresh exam-style question where the learner must practise read the exact wording while watching specifically for description only. Clarify what aspect of the result, source, graph, method or statement deserves comment. Require an independent response before any model is shown so the actual decision becomes visible.

The response repeats the pattern without saying what it implies. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 18: Inspect the evidence × Overclaim

Set up one fresh exam-style question where the learner must practise inspect the evidence while watching specifically for overclaim. Look for patterns, anomalies, contrasts, limitations or implications. Require an independent response before any model is shown so the actual decision becomes visible.

Limited evidence is treated as proof. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 19: Select the strongest observation × Long essay

Set up one fresh exam-style question where the learner must practise select the strongest observation while watching specifically for long essay. Choose the feature most relevant to the task. Require an independent response before any model is shown so the actual decision becomes visible.

A brief comment question becomes overdeveloped. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 20: State the comment directly × Evidence-free claim

Set up one fresh exam-style question where the learner must practise state the comment directly while watching specifically for evidence-free claim. Write the observation or judgement in clear language. Require an independent response before any model is shown so the actual decision becomes visible.

A judgement is stated without support. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 21: Support with evidence × Irrelevant criticism

Set up one fresh exam-style question where the learner must practise support with evidence while watching specifically for irrelevant criticism. Use values, quotations, features or facts from the stimulus. Require an independent response before any model is shown so the actual decision becomes visible.

A limitation is mentioned that does not affect the question. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 22: Explain significance where needed × Personal reaction

Set up one fresh exam-style question where the learner must practise explain significance where needed while watching specifically for personal reaction. State why the pattern or limitation matters to the question. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says I think this is surprising. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 23: Qualify appropriately × Every-feature list

Set up one fresh exam-style question where the learner must practise qualify appropriately while watching specifically for every-feature list. Use language such as suggests, appears, limited by or consistent with where certainty is constrained. Require an independent response before any model is shown so the actual decision becomes visible.

The answer comments on everything visible. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 24: Check relevance × Vague significance

Set up one fresh exam-style question where the learner must practise check relevance while watching specifically for vague significance. Ask whether the comment answers the specific request rather than offering a general opinion. Require an independent response before any model is shown so the actual decision becomes visible.

The answer says this is important without explaining why. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Rank possible observations by relevance to the question. Choose the strongest. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 25: Read the exact wording × Description only

Set up one fresh exam-style question where the learner must practise read the exact wording while watching specifically for description only. Clarify what aspect of the result, source, graph, method or statement deserves comment. Require an independent response before any model is shown so the actual decision becomes visible.

The response repeats the pattern without saying what it implies. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Attach one precise value or quotation to each comment. Check support. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 26: Inspect the evidence × Overclaim

Set up one fresh exam-style question where the learner must practise inspect the evidence while watching specifically for overclaim. Look for patterns, anomalies, contrasts, limitations or implications. Require an independent response before any model is shown so the actual decision becomes visible.

Limited evidence is treated as proof. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: State one limitation and explain exactly what conclusion it weakens. Remove generic criticisms. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.

Workshop 27: Select the strongest observation × Long essay

Set up one fresh exam-style question where the learner must practise select the strongest observation while watching specifically for long essay. Choose the feature most relevant to the task. Require an independent response before any model is shown so the actual decision becomes visible.

A brief comment question becomes overdeveloped. After correction, change one meaningful feature—topic, wording, evidence, mark allocation, data, source or context. Then use this related exercise: Practise concise evidence-based observations under tight limits. Stop early. Finish with a delayed retest so success cannot be explained only by immediate memory of the correction.