English exam answers succeed when the evidence, explanation and wording make the student’s reasoning recoverable. A correct idea can remain invisible if the response is too vague; a long quotation can occupy space without proving the claim; and a polished paragraph can still lose direction when it follows a memorised answer formula instead of the question actually asked.
How to write better English comprehension answers therefore requires more than PEEL, PEE, PETAL or any other classroom acronym. Frameworks can remind students to make a point, use evidence and explain it, but the examination skill is deeper: decode the task, select the smallest sufficient evidence, state a defensible claim, build the evidence-to-meaning bridge, calibrate certainty and stop when the required relationship is complete.
This world-facing handbook continues the How English Examination Works series. It connects directly to the unseen-passage handbook and the question-decoding handbook. Its distinct ownership is answer construction: turning understanding into concise, visible, assessable English without letting a paragraph formula replace judgement.
The 50-second quick read
QUESTION → CLAIM → EVIDENCE → BRIDGE → SCOPE → STOP. Decode the task. State the answer. Select evidence that carries the answer. Explain why that evidence supports the claim. Match certainty and scope to the source. Stop when the job is complete. If a framework helps you remember those operations, use it; if it forces irrelevant stages, leave it behind.
Part I. An answer is a visible reasoning object
1. The examiner cannot mark the thought you did not write
Students often say, I knew what I meant. The examination response must make enough of that meaning visible to another reader.
The task is transmission, not private understanding.
2. Correct evidence is not automatically a complete answer
A quotation can be relevant yet unexplained. If the question asks what a detail suggests, copying the detail stops before inference.
Evidence is material; explanation gives it a job.
3. A claim can be correct but too broad
The character is upset may fit the passage but fail to distinguish embarrassment, disappointment or irritation when the evidence supports a more precise interpretation.
Precision should increase only as far as evidence permits.
4. A long answer can still be incomplete
Length cannot compensate for a missing causal or analytical hinge. Five sentences of paraphrase may never explain why the evidence matters.
Count relationships, not sentences.
5. A short answer can be sufficient
If a retrieval question asks for one stated reason, one accurate sentence may fulfil it. Expanding automatically can create new errors.
Proportion is part of control.
6. Evidence selection is a reasoning act
Choosing one phrase rather than the whole paragraph shows which detail carries the claim.
The smallest sufficient evidence often makes explanation easier.
7. Explanation is the bridge
The bridge answers: why does this evidence support this claim? Without it, point and quotation can remain adjacent but disconnected.
Practise the because even when the final answer does not literally contain that word.
8. Scope belongs in the answer
At this point, mainly, one reason and according to paragraph four constrain what can be claimed.
A strong answer respects boundaries after decoding, not only before writing.
9. Certainty belongs in the answer
Shows, suggests, may imply and proves carry different strengths. Choose language that matches evidence.
Overclaiming can turn a plausible inference into an inaccurate one.
10. Own words require meaning control
Paraphrase should preserve agency, cause, condition, time and degree.
Changing vocabulary while changing meaning is not successful reformulation.
11. Paragraph frameworks are scaffolds
PEE, PEEL, PETAL and similar structures can remind learners of operations, but no acronym can determine what every question needs.
The question remains the governing architecture.
12. Stopping is a skill
Once the answer has fulfilled the task, additional speculation can weaken it.
A stopping rule protects precision and time.
Part II. Original answer-construction laboratory
13. Original passage: The Returned Umbrella
Aisha noticed the umbrella only after the bus had left. It leaned against the bench, black, dry and neatly folded despite the rain. She looked down the road, but the bus was already turning beyond the traffic lights. On the handle, a strip of masking tape carried one word in blue ink: KUMAR.
The shelter was empty. Aisha could have left the umbrella where it was, but the rain had begun to blow sideways. She moved it behind the timetable panel, where the roof covered the bench more completely, then photographed the name and sent the picture to the neighbourhood group chat.
Ten minutes later, a reply appeared: That is my father’s. He is already halfway home. Please leave it there; I will collect it.
Aisha read the message twice. Then she moved the umbrella back to the bench, this time where it could be seen from the pavement.
14. Retrieval question
Question: What name was written on the umbrella? Answer: Kumar.
Nothing is gained by explaining the rain, bus or group chat. The task is explicit retrieval.
15. Cause question
Question: Why did Aisha move the umbrella behind the timetable panel? Answer: She wanted to protect it from the sideways rain because that part of the shelter covered it more fully.
The answer preserves action plus cause. Saying because it was raining is less precise because the umbrella was initially dry and the passage specifically explains the more protected position.
16. Inference question
Question: What does Aisha’s decision to photograph the name suggest about her intention? Answer: It suggests that she wanted to identify or contact the owner rather than simply keep the umbrella safe at the shelter.
The inference grows from an action that would be unnecessary if protection alone were her aim.
17. Structure question
Question: Why is it significant that Aisha moves the umbrella twice? Answer: The first move protects it from the weather, while the second makes it easier for the owner’s family to find. The repeated action therefore shows her purpose changing as new information arrives.
A structural answer needs both occurrences and the changed relationship between them.
18. Evidence-selection question
Question: Which detail most clearly shows that Aisha checks the reply carefully? Answer: She “read the message twice.”
The phrase carries the point directly. Quoting the whole paragraph would hide the useful evidence inside unnecessary material.
19. Evaluation question
Question: A reader says Aisha acts only because she is worried the umbrella will get wet. How far is this supported? Answer: The claim is only partly supported. Her first action protects the umbrella from rain, but photographing the owner’s name and later moving the umbrella where it can be seen show that she also wants to help return it.
The word only makes the later evidence decisive.
20. The answer-construction lesson
The six answers use different architectures because the questions require different jobs. A universal paragraph template would overbuild the retrieval answer and could underbuild the evaluation answer.
The stable machinery is not paragraph shape. It is task → claim → evidence → relationship → scope.
Part III. Evidence selection
21. Start from the claim
Do not collect quotations first and hope one becomes useful. Decide what the answer needs to establish, then select evidence that tests or supports it.
Claim-first selection reduces quotation dumping.
22. Use the smallest sufficient quotation
If one phrase carries the relationship, quoting four lines can make the important language harder to see.
Small evidence creates room for explanation.
23. Sometimes one word is enough
A modal such as might or an adverb such as reluctantly can carry a precise analytical point when context is already clear.
Single-word evidence is useful only when the word genuinely bears the claim.
24. Sometimes one word is not enough
A structural contrast, changing attitude or causal sequence may require two moments or a larger sentence.
Match evidence scale to claim scale.
25. Evidence can be paraphrased
Some tasks permit or require own-word reference rather than direct quotation. Preserve meaning and attribution.
Follow the current examination instructions.
26. Evidence should be discriminating
Choose the detail that makes your interpretation stronger than plausible alternatives.
A generic detail may support several incompatible claims weakly.
27. Repeated evidence can show movement
Two appearances of the same object can reveal changed meaning when compared.
Do not treat recurrence as mere duplication.
28. Negative evidence needs caution
What a passage does not say can matter, but absence is harder to establish than presence.
Use omission only when the text creates a clear expectation.
29. Counterevidence belongs in evaluation
After selecting support for an evaluative claim, search deliberately for material that limits it.
A judgement becomes stronger when it survives its best challenge.
30. Source boundaries still govern
Excellent evidence from the wrong paragraph, text or speaker can remain invalid for the question.
Selection begins inside the permitted evidence window.
Part IV. Quotations without quotation dumping
31. Embed only when it helps
A quotation can sit inside a grammatical sentence, but elegant embedding is not more important than accurate analysis. Do not distort source wording merely to make it fit your syntax.
32. Quote exactly when quoting
If quotation marks are used, preserve the selected wording accurately. If you need to change the wording substantially, paraphrase instead, subject to the task rules.
33. Do not quote the question back
Repeating the wording of the task does not add textual evidence. Use response space for the claim, source and relationship.
34. Avoid orphan quotations
A quotation placed after a claim with no explanation can leave the reader to infer the connection. Build the bridge explicitly when the task requires analysis.
35. Avoid quotation sandwiches that become rituals
Point, quote, explain can be a useful teaching sequence, but not every answer needs identical layers. Retrieval and summary can require different architectures.
36. Quote for the distinction
If your point depends on uncertainty, quote the modal or qualifier. If it depends on repetition, show the repeated pattern. Evidence should expose the feature you discuss.
37. Context can be paraphrased around a short quotation
You can explain the situation in your own words and quote only the decisive phrase. This often keeps the answer concise.
38. Do not stack quotations as substitutes for reasoning
Three quotations do not automatically create a stronger paragraph than one well-explained quotation.
39. Use two quotations when relationship is the point
Before-and-after structure, comparison and changing tone often need evidence from both states.
40. Stop quoting when the evidence is already visible
Once the necessary textual support is present, move into explanation rather than continuing to copy.
Part V. Explanation: the missing middle
41. Explanation answers why
Why does this action suggest reluctance? Why does this word make the claim cautious? Why does this structural delay matter?
The answer should expose the reasoning step.
42. Paraphrase is not always explanation
Restating a quotation in simpler words can establish meaning but may still stop before significance.
Ask what the paraphrased meaning does in relation to the question.
43. Reader effect needs specificity
This shocks the reader is weak unless the text and task support that exact response.
Prefer explaining what expectation is disrupted or what information becomes newly uncertain.
44. Technique labels are not explanation
Metaphor, repetition and short sentence identify forms. Analysis explains their local contribution.
Name the method only when it helps.
45. Cause must not be invented
A character looks away; the answer claims guilt. The behaviour may support discomfort, avoidance or another interpretation, but guilt needs more evidence.
Calibrate the bridge.
46. Context belongs in the explanation
The same phrase can mean different things depending on who says it, when and after what event.
Local context constrains interpretation.
47. Structure needs before and after
A delayed revelation matters because it changes what earlier information means or when the reader can know it.
Position plus consequence creates structural explanation.
48. Comparison needs a shared axis
Explaining A and B separately is not enough. State the consequential similarity or difference.
The bridge can run across texts.
49. Evaluation needs weighting
After support and counterevidence are identified, explain which carries more weight for the exact claim.
Evidence counting is not evaluation.
50. Summary explanation is usually selection logic
A summary does not need analytical commentary on every point. Its reasoning appears in choosing, separating and compressing the right information.
Do not force analytical paragraphs into a compression task.
Part VI. PEEL, PEE, PETAL and other frameworks
51. Why frameworks help
They can reduce cognitive load for beginners by reminding them that a claim needs evidence and that evidence may need explanation.
52. Why frameworks fail
They fail when the student writes a link sentence no question needs, hunts a technique because the acronym contains T, or expands a one-mark retrieval item into a paragraph.
53. Use frameworks as checklists after thinking
First decode the question and decide the required relationship. Then use the framework only if it helps you make that relationship visible.
54. Do not grade acronym compliance
A strong answer can use a different sentence order and still fulfil the task. The assessment criteria, not classroom acronym purity, determine formal credit.
55. Frameworks should fade
As control grows, the learner should be able to choose answer architecture independently. Permanent dependence can become a new bottleneck.
56. Retrieval may need R: response
A direct factual answer can be complete without quotation-analysis-link stages.
57. Inference may need claim + evidence + bridge
The core relationship is observation → interpretation. A concise answer can make that visible in one or two sentences.
58. Language analysis may need claim + evidence + local meaning + contribution
Here the bridge is richer because the task asks how a writer’s choice creates meaning.
59. Evaluation may need claim + support + counterevidence + weighting
A simple PEEL paragraph may underrepresent the need to test the supplied statement.
60. Writing tasks need entirely different architecture
Creative writing and directed writing should not be forced into comprehension-answer frameworks. Their units of work are scenes, claims, source transformations and reader effects.
Part VII. Answer length and stopping rules
61. There is no universal sentence count
The required length depends on task, marks, evidence and examination system. Current official guidance outranks internet formulas.
62. One requested point means one distinct point
Do not manufacture multiple alternatives when the task asks for one reason unless explanation requires development.
63. Two reasons must be semantically different
Rephrasing one reason twice does not create two points.
64. A paragraph ends when its job is done
If claim, evidence and required relationship are clear, another generic effect sentence may weaken rather than strengthen it.
65. High-mark does not mean maximum length
More complex tasks often require more development, but irrelevant length remains irrelevant.
66. Use a completeness question
Can another reader recover exactly why my evidence answers this question? If yes, check scope and move on.
67. Use a speculation stop
If each new sentence becomes less directly supported than the previous one, you may have passed the strongest answer.
68. Protect time across the paper
An answer can be locally excellent but globally costly if it consumes time needed for required later work.
69. Revision should shorten some answers
Improvement is not always expansion. Removing duplication can make reasoning more visible.
70. Concision is not minimalism
Do not cut the bridge simply to be brief. Remove redundancy while preserving the reasoning the task needs.
Part VIII. Original weak-to-strong answer workshop
71. Weak inference
Question: Why does Aisha read the message twice? Weak: She is shocked. Repair: She appears to check the reply carefully because it changes what she should do with the umbrella.
The repair avoids an unsupported emotion and connects the repeated reading to the new instruction.
72. Weak evidence dump
Weak: “The shelter was empty. Aisha could have left the umbrella where it was, but the rain had begun to blow sideways…” This shows she is helpful. Repair: Moving the umbrella where “the roof covered the bench more completely” shows that she deliberately protects another person’s property from the rain.
The shorter quotation exposes the purposeful action.
73. Weak language analysis
Weak: The writer uses the adjective dry to make the reader interested. Repair: Describing the umbrella as “dry” despite the rain makes its exposed position more precarious: it has not yet been damaged, but the worsening weather could change that.
74. Weak structure analysis
Weak: The writer repeats the umbrella. Repair: Aisha moves the same umbrella twice for different reasons: first to protect it, then to make collection easier. The repeated action tracks how new information changes her purpose.
75. Weak evaluation
Weak: I disagree because Aisha is kind. Repair: The claim that she acts only to keep the umbrella dry is too narrow. Photographing the name and repositioning the umbrella after the owner’s family replies show a second aim: helping return it.
76. Weak summary
Weak: Aisha saw an umbrella, looked for the bus, moved it, took a photograph, sent it to a group chat, received a message, read it and moved the umbrella again. Repair for a question asking how she tries to return it: She shares the owner’s name with the neighbourhood group and, after receiving a reply, places the umbrella where the collector can see it.
The repair selects actions relevant to the requested category rather than retelling the passage.
Part IX. Diagnostics and practice
77. Ben quotes too much
Adrian asks Ben to highlight the exact three or four words his explanation depends on. Everything else must justify its presence or be removed.
78. Aisha explains orally but not in writing
Jo asks her to say the because aloud, then write that relationship in one sentence. The problem is formulation, not interpretation.
79. Ryan uses PEEL everywhere
Ryan writes Link sentences even in short retrieval questions. His next mixed set bans acronyms until after he has classified the question family.
80. Mira overclaims
Mira’s evidence is good but her verbs say proves and definitely. Her practice uses a certainty ladder and requires the strongest wording the evidence can actually support.
81. Clara repeats evidence
Clara quotes, paraphrases the quote and then quotes it again. Her repair identifies whether the second sentence adds meaning or merely repeats source material.
82. Ethan never stops
Ethan’s first explanation is precise; his fourth becomes speculative. His stopping rule is claim fulfilled + evidence explained + scope checked = move.
83. Evidence-only drill
Give five claims and one passage. Select the best evidence for each without writing explanations. Then justify why nearby alternatives are weaker.
84. Bridge-only drill
Supply the claim and quotation. Write only the sentence that connects them.
85. Claim-only drill
Supply the question and evidence. Write the most precise defensible claim without explanation.
86. Compression drill
Take a six-sentence answer and reduce it until removing one more sentence would damage task fulfilment.
87. Expansion drill
Take an underdeveloped answer and add only the missing relationship, not extra background.
88. Mixed architecture drill
Combine retrieval, inference, language, structure, summary and evaluation questions. The learner must choose answer shape rather than repeat one template.
Part X. Connected routes and final check
89. Question-decoding route
Use the question-decoding handbook when the answer fails before evidence selection because the task itself was misread.
90. Unseen-passage route
Use the unseen-passage handbook when the learner cannot build the source model needed to select evidence.
91. Reading-comprehension route
Use the reading-comprehension handbook for the broader system connecting retrieval, inference, context and meaning.
92. Language-analysis route
Use the language and structure handbook when the missing bridge concerns a writer’s choice and its contribution.
93. Summary route
Use the summary and paraphrase handbook when the task requires selection and compression rather than analytical expansion.
94. Comparison route
Use the comparison and evaluation handbook when evidence must be aligned or weighed across sources.
95. Marking route
Use the marks and feedback handbook to connect visible evidence in answers with the applicable assessment criteria.
96. Final answer check
What exactly is my claim? Which evidence carries it? Have I explained why the evidence supports the claim? Is my certainty proportionate? Have I respected the question boundary? Is there anything left that does not perform a job?
A strong English examination answer is not a formula filled with quotations. It is a compact argument between task, evidence and meaning. The learner’s understanding becomes assessable when the response makes that relationship visible—and becomes efficient when nothing unnecessary is allowed to hide it.