A recommendation is only as strong as its evidence
Two student teams have made simple periscopes for a school display. Team A’s device is compact and easy to carry. Team B’s is longer and produces a clearer view in a controlled classroom test. The student writer is asked to recommend one device to a teacher organising a demonstration for younger pupils. An answer such as ‘B is better because its image is clearer’ might contain a relevant point, but it has not yet explained what matters to this reader, how the comparison was made or whether the observed advantage survives the actual demonstration conditions.
This is an English problem before it is a technical one. A writer must establish purpose and audience, select details from given material, distinguish observation from interpretation, make a fair comparison and reach a proportionate recommendation. The writing should remain readable even when its evidence comes from a table, a report, a diagram or a short video. The Physics and Mathematics guides immediately before this article explain how the devices might work. This volume teaches the different skill of turning evidence into a coherent judgement without inventing certainty.
The same thinking helps with comprehension. A writer may tell readers that a device is ‘the most reliable design’, while a table reveals it was tested only once in bright conditions. A critical reader should notice that the claim is broader than the observation. A school examination does not require the reader to be an engineer. It requires the reader to understand language and meaning: what is stated, what is implied, how a comparison is framed, what support is provided and which conclusions remain uncertain.
What the official G3 English assessment actually requires
For 2027, SEAB lists G3 English Language as SEC syllabus K300, replacing the earlier O-Level reference code 1184. Paper 1 Writing lasts 1 hour 50 minutes, carries 70 marks and contributes 35%. It includes Editing, a 250–350-word Situational Writing task involving a visual text, and a 350–500-word Continuous Writing task chosen from four topics. The set genres are not restricted to the classroom ‘comparative review’ label used in this guide. The task’s actual genre, points, audience and purpose always govern the response.
Paper 2 Comprehension lasts 1 hour 50 minutes, contributes 35%, and includes a visual text among its materials, narrative or recount reading, non-narrative reading and a summary task of about 80 words. Paper 3 Listening contributes 10%, and Paper 4 Oral Communication contributes 20%. Comparison, evaluation and accurate use of details support multiple parts of the subject. They are skills, not an extra paper. The official SEAB G3 school-candidate listing, updated 26 October 2026, should be consulted for the correct examination year.
The K300 assessment objectives include understanding written and multimodal texts at literal, inferential and evaluative levels; using written English for purpose, audience and context; synthesising and summarising information; and presenting spoken ideas effectively. A good comparative review is therefore a training context in which several official skills meet. It is not a promise that a periscope comparison or any exact review prompt will appear in a 2027 paper. The examples here are original practice texts, not specimen or past-year questions.
Define the decision before collecting adjectives
Before writing, ask four questions. Who will read this? What decision does the reader have to make? Which criteria matter to that decision? What evidence is available for those criteria? If a teacher needs a safe demonstration for a busy corridor, portability and reliable viewing from a marked position may matter more than decorative finish. If the same teacher needs pupils to learn reflection geometry, visible mirrors and easy access for adjustment may matter more than the sleekness of the tube.
An adjective such as ‘excellent’, ‘superior’, ‘innovative’, ‘clear’ or ‘practical’ does not complete a comparison. It creates a question: in what respect, compared with what, under which conditions, and supported by what observation? The strongest revision often replaces a large adjective with a small fact. ‘Model B had a clearer view’ becomes ‘In the three classroom trials reported, observers could read the six-centimetre card through B on all three occasions, while A succeeded once.’ That statement is narrower but far more useful.
We can then connect the evidence to the reader’s purpose. If pupils must read small labels from a fixed distance, those trial results support B. If the devices will be carried by young children between stations, size, weight and handling evidence may change the recommendation. A comparison is not simply a competition to list more positives for the preferred choice. It is a reasoned connection between tested differences and the real decision.
Separate observation, inference, claim and recommendation
An observation reports what was seen or recorded: ‘The observer read four of five labels through Device A.’ An inference interprets that result: ‘A may be adequate for large classroom symbols, but less reliable for small labels under these conditions.’ A claim asserts a property: ‘A is suitable for every age group.’ A recommendation connects the available evidence to a purpose: ‘Use A for the short supervised demonstration if portability matters more than label detail.’ These jobs are related but not interchangeable.
Writers often move too quickly from one successful trial to a universal statement. ‘The picture was bright yesterday’ does not mean ‘the image is always bright’. A phrase such as ‘in our indoor test’, ‘for the tested viewing position’, ‘based on the recorded results’ or ‘for this specific demonstration’ can restore proportion. This is not timid writing. It is the language of a person who knows exactly what a conclusion rests on.
A weakly supported claim should not be made stronger by adding more confident adverbs. ‘Undoubtedly’, ‘clearly’, ‘without question’ and ‘certainly’ do not create evidence. An effective comparison may end in a firm decision while still admitting a relevant limit: ‘Choose B for this event, because it produced a readable image in all three recorded trials, although the small sample means we should test it again in the final venue.’
The anatomy of a balanced paragraph
A useful paragraph begins with a comparative point, then gives evidence for both sides, then explains its significance for the audience. For example: ‘B provides a more dependable view in the reported test. Three pupils successfully read the target card through B, compared with one through A. Because the event involves quick demonstrations to unfamiliar users, the reduced need to reposition the device is valuable. The observations were made indoors, so a final check under the event lighting is still needed.’
This paragraph is stronger than a two-column list because it has a direction. It is not stronger simply because it is longer. Notice the movement: comparison, evidence, interpretation, condition. If the question asks for a short answer, the writer may not need all four sentences. If it asks for a developed recommendation, leaving out interpretation would make the evidence feel unconnected. A skilled student adjusts the length to the command, not to a template remembered from practice.
Compare like with like
A fair comparison requires the same or clearly described conditions. If Team A was tested at a distance of one metre and B at two metres, the result cannot automatically be treated as a controlled comparison of image clarity. If one model was used by an experienced operator and the other by a first-time pupil, differences in familiarity may affect speed. If the test card changed, the difficulty may have changed too. A good reader notices these hidden variables; a good writer does not conceal them.
The English task is not to perform a full statistical analysis where none is asked. It is to read what the source gives and not exceed that evidence. Words such as ‘more’, ‘fewer’, ‘at least’, ‘up to’, ‘about’, ‘on average’ and ‘in the sample’ alter the reach of a statement. ‘Most testers preferred B’ is not equivalent to ‘B is preferred by everyone’. ‘B worked twice as often as A’ may be mathematically true in a tiny test while still being too weak to support ‘B will always be twice as reliable’.
A complete model of a school recommendation
Original practice situation. A teacher asks a student to recommend which of two periscope prototypes should be used at a Primary 6 science open house. A supplied note reports that Model A is 28 centimetres long and weighs 180 grams; it gave readable target-card views in two of four trials. Model B is 42 centimetres long and weighs 260 grams; it gave readable views in four of four trials. Both were tested indoors with the same card from a marked position. The student must recommend a model, explain the relevant trade-off and suggest one practical follow-up. The following is a practice response, not an official SEAB task.
Model response: a recommendation with an identifiable audience
Dear Ms Lim,
I recommend using Model B for the Primary 6 science open house. Although it is longer and heavier than Model A, the available test results suggest that it will give visitors a more dependable view of the target card.
The two devices were tested indoors with the same card and viewing position. All four trials with Model B produced a readable image, while only two of the four trials with Model A did so. That difference matters because the open house will involve visitors who may have only a short time at each station. We should choose the device that is less likely to require repeated adjustments during the demonstration.
Model A does have an advantage. At 28 centimetres and 180 grams, it is smaller and lighter than Model B, which measures 42 centimetres and weighs 260 grams. It may therefore be easier to carry and store. However, the main purpose of this station is to let visitors observe how reflected light changes direction. For that purpose, reliable viewing is more important than saving 14 centimetres of length.
I suggest that we use Model B for the display and retain Model A as a comparison model. Before the event, we should repeat the reading test with a few first-time users under the lighting conditions of the actual venue. That would help us check whether the result remains consistent outside the initial classroom test.
If you agree, our group can prepare simple labels showing where the two mirrors are positioned and why the light follows a different path from direct sight.
Yours sincerely, A Student
The response makes an explicit recommendation, uses the supplied facts, recognises the opposing advantage and proposes a bounded follow-up. It does not claim that four successful trials prove performance in every setting. In an actual K300 situation, the candidate must follow the stated role, format and every required content point, which may differ from this invented example.
Thirty-four cases: editing evidence into fair and useful English
English evidence case 01: The superlative without a comparison
A poster announces, ‘Our periscope is the best ever made.’ The source describes only two school prototypes in one lesson.
Revised reasoning. Revise the claim to specify the comparison: ‘Of the two school prototypes tested in this lesson, B produced the clearest recorded view.’
What the reader learns. The word best requires a comparison class and criteria. Without them it is marketing language, not a supported finding. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 02: One successful result
A report states that an observer read a sign through Device A on a single attempt. The draft says, ‘A always gives a clear image.’
Revised reasoning. A faithful version is: ‘A produced a readable image in the one attempt reported.’ It may be too early to decide whether the performance is consistent.
What the reader learns. Universal terms such as always and never demand much stronger evidence than a single observation. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 03: A hidden change in distance
A was tested from one metre and B from two metres. The writer declares A easier to use because its target looked larger.
Revised reasoning. The apparent difference could involve the changed distance rather than only the device. State the two conditions and avoid a simple ranking of optical performance.
What the reader learns. A fair comparison identifies the condition that changed. Precise language can make an unfair trial visible. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 04: The unsupported causal link
Several visitors preferred B, and B had a blue casing. The draft says its colour caused the preference.
Revised reasoning. Unless visitors explained that the colour influenced them, the result only shows a preference alongside a colour difference. The causal explanation has not been established.
What the reader learns. Sequence and association are not proof of cause. Distinguish what the source records from what a writer is tempted to infer. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 05: The ambiguous pronoun
The review says: ‘A was heavier than B, and it was easier to handle.’ A reader cannot tell what ‘it’ refers to.
Revised reasoning. Repeat the intended noun: ‘Although A was heavier, testers found A easier to handle’ if that is what the evidence says.
What the reader learns. A shorter sentence is not clearer when its reference is ambiguous. Repetition can be the best editing choice. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 06: Mixing up two users
A teacher operates one prototype, while a six-year-old uses the other. The review attributes the shorter trial time entirely to the design.
Revised reasoning. State who operated each model and why the unequal experience matters. The available result is not a controlled test of the equipment alone.
What the reader learns. Readers need relevant conditions, not only attractive numbers. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 07: Percentages with no sample size
A report states, ‘100% of users succeeded’, without mentioning that only one pupil tried the model.
Revised reasoning. Write ‘The single pupil who tested it succeeded.’ If a percentage is retained, include the denominator.
What the reader learns. A true percentage can still mislead when the reader is invited to imagine a much larger group. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 08: The word ‘safe’ without a test
A design note describes a cardboard casing with taped edges. The review claims that it is completely safe for young children.
Revised reasoning. A more defensible statement is: ‘The casing has taped edges, but the device still requires a handling check and supervised use.’
What the reader learns. Safety is not established solely by a favourable description. Keep conclusions within documented checks. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 09: A glowing adjective
The sentence reads: ‘The exciting, incredible and fantastic view impressed everyone.’ The only source records three successful card readings.
Revised reasoning. Replace the stacked adjectives with the test observation and, if relevant, a genuine quotation from a participant.
What the reader learns. Emotional words cannot substitute for evidence. They may have a place in a review, but should not carry its logic. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 10: The changing criterion
The introduction promises to compare image readability. The conclusion recommends the lighter device without explaining why weight has become decisive.
Revised reasoning. If portability is the main criterion, establish that before the evidence. If readability is decisive, return to it in the conclusion.
What the reader learns. A coherent evaluation keeps its criteria visible. New information can change the judgement, but the reasoning must be explained. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 11: A result reported as a prediction
An indoor trial produced good visibility. The article says, ‘Visitors will see clearly at the outdoor booth.’
Revised reasoning. Use ‘The device worked in the indoor trial; outdoor visibility remains to be tested.’
What the reader learns. Future conditions are not identical to recorded conditions. A careful writer separates observed performance from expected performance. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 12: A difference without units
The writer says B is fourteen longer than A. The source table uses centimetres.
Revised reasoning. Supply the unit: ‘B is 14 centimetres longer than A.’ If relevant, connect this difference to storage or handling.
What the reader learns. Numbers without units can make a comparison meaningless. English clarity depends on preserving the quantitative noun. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 13: A negative built from silence
The report does not mention any complaints about A. The writer concludes that no one found it uncomfortable.
Revised reasoning. The report contains no complaint data. Say that comfort was not measured, rather than turning missing evidence into evidence of comfort.
What the reader learns. Absence of a record is not automatically a negative result. This distinction is useful in comprehension and evaluation. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 14: Overclaiming from a diagram
A drawing shows two mirrors at approximately 45 degrees. The caption claims the exit ray must reach every visitor’s eye.
Revised reasoning. A more limited caption explains the intended ray path under the diagram’s assumed alignment and viewing position.
What the reader learns. An idealised diagram represents a model, not every physical outcome. Use conditional wording where the assumptions matter. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 15: Two sources disagree
One source says B was easier to carry; another reports B is heavier. A student deletes the contradiction.
Revised reasoning. Preserve both facts and ask whether ‘easier to carry’ refers to grip, balance or a different tester group. Weight alone does not settle handling preference.
What the reader learns. Contradictions invite closer reading. Do not erase inconvenient evidence to make a paragraph smooth. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 16: A conclusion that never answers
After listing dimensions and test results, the writer ends: ‘Both periscopes have advantages.’ The task asks which to choose.
Revised reasoning. Name the preferred model and tie it to the audience’s principal need. You can still acknowledge the other model’s strength.
What the reader learns. Balance does not mean indecision. A recommendation must make the requested decision. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 17: An unfair absolute
Two of four observers preferred A; two preferred B. The student writes, ‘B was much more popular.’
Revised reasoning. State the equal preference split. If other criteria favour B, explain those separately rather than editing the survey to fit the conclusion.
What the reader learns. A writer can recommend B without pretending it won every measure. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 18: The very long sentence
One sentence contains the venue, user ages, trial scores, dimensions, limitations and recommendation, separated by repeated commas.
Revised reasoning. Divide the sentence by its jobs: set the context, compare results, explain relevance and state the decision. Use transitions to show how they connect.
What the reader learns. A clear review makes relationships visible. Sentence length should follow thought, not the wish to sound sophisticated. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 19: A missing source attribution
A promotional leaflet says B is durable. The review presents durability as an established fact although no endurance test is supplied.
Revised reasoning. Identify it as the manufacturer’s or team’s claim. Do not promote an unverified statement into an independent finding.
What the reader learns. Attribution changes the strength of a sentence. ‘The team claims’ is different from ‘testing showed’. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 20: The misleading average
Average setup time is 40 seconds for A and 35 for B, but B has one trial taking 90 seconds. The writer calls B reliably faster.
Revised reasoning. The lower mean may coexist with considerable variation. State that the average was lower but the individual times varied, and check whether the outlier matters.
What the reader learns. A comparison of averages need not imply consistency. Select words that describe the data actually shown. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 21: Confusing ease with speed
An observer completes the setup quickly but reports that the instructions were difficult to understand. The writer calls the setup easy.
Revised reasoning. Time and perceived difficulty measure different things. The fair account reports both rather than treating one as proof of the other.
What the reader learns. Define the criterion. A word like easy may refer to learning, effort, speed or physical handling. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 22: A changed audience
A review written for science teachers emphasises optical terms. The same text is copied unchanged onto a Primary 3 visitor card.
Revised reasoning. Keep the scientific fact but adjust the explanation: ‘The first mirror sends light down the tube; the second sends it towards your eye.’
What the reader learns. Audience awareness changes wording and selection, not the truth of the mechanism. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 23: Unclear comparative structure
Paragraph one describes every feature of A, and paragraph two describes every feature of B. The reader must hunt across both paragraphs to compare.
Revised reasoning. Organise by the criteria that decide the task: readability, handling and suitability for the event. Discuss A and B together under each point.
What the reader learns. Point-by-point comparison helps the reader make a decision. Block organisation can work, but it needs deliberate signposting. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 24: Evidence placed after the conclusion
The first line orders the teacher to choose B before the writer mentions the task criteria or trial results.
Revised reasoning. A direct recommendation can appear early in a formal message, but it should be followed promptly by clear grounds; it must not stand as unsupported command.
What the reader learns. Structure follows purpose. A recommendation email may open with the decision, while an analysis may build towards it. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 25: A modal verb changes the meaning
The source says a device may work outdoors after adjustment. The candidate paraphrases it as ‘It will work outdoors’.
Revised reasoning. Restore the modal: ‘It may work outdoors if it is adjusted.’ The condition is part of the information.
What the reader learns. Can, may, should, must and will are not interchangeable. They encode degrees of certainty and obligation. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 26: The missing limitation in praise
B has four successful trials. A critical reader is asked whether ‘B is guaranteed reliable’ is justified.
Revised reasoning. No. Four successful trials support performance in those trials, not a universal guarantee. A reader should explain the gap between sample and claim.
What the reader learns. Evaluative comprehension asks for a reasoned interpretation, not a generic ‘yes’ or ‘no’. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 27: The vague comparison word
The sentence says ‘A performed better’. The report contains measures for weight, view clarity and setup time.
Revised reasoning. Name the dimension: ‘A was 80 grams lighter’, or ‘A had a shorter mean setup time’ if supported. ‘Better’ is incomplete without a criterion.
What the reader learns. A good comparative adjective is attached to an observable or meaningful measure. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 28: A misleading heading
The heading says ‘Proven for all conditions’. The body reports only four indoor tests with one viewing card.
Revised reasoning. Rewrite the heading to match scope: ‘Indoor reading trial: results from four attempts’.
What the reader learns. Headings frame readers’ expectations. An accurate body does not neutralise an exaggerated title. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 29: The quotation used as universal evidence
One visitor says, ‘I could see the sign easily.’ The writer declares every pupil will find the device easy.
Revised reasoning. Attribute the remark and keep its scope to that visitor. A quotation records a perspective, not the distribution of opinions.
What the reader learns. The source is relevant but small. A precise writer preserves its usefulness without inflating it. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 30: An attractive but irrelevant statistic
A table reports the colour of the cardboard for each model. The task asks which design allows the clearest demonstration.
Revised reasoning. Mention colour only if the audience’s decision requires it. Give priority to recorded visibility and handling evidence.
What the reader learns. Not every supplied detail belongs in the answer. Relevance is an active selection skill. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 31: A conditional recommendation
B is clearer indoors, but the organiser has not confirmed whether the event will be outdoors.
Revised reasoning. Recommend B provisionally for the currently tested conditions and request a test at the final venue before committing.
What the reader learns. A conditional conclusion is not a weak conclusion when the missing information genuinely affects the decision. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 32: The summary with too much detail
A comprehension question asks for the main comparison, but the answer reproduces every figure, tester name and sequence of assembly.
Revised reasoning. State the key contrast and the criterion it affects. Add a figure only when it is necessary to demonstrate the contrast.
What the reader learns. A concise answer can show deeper selection than a long recital. Match detail to the question command. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 33: A title that promises a winner
A review is titled ‘The Winner’ before the writer has checked whether one design dominates on every criterion.
Revised reasoning. Choose a descriptive title such as ‘Periscope Designs for a School Demonstration’ until the decision is justified.
What the reader learns. A title is a claim. It should establish context rather than prejudge an unresolved comparison. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 34: A confident claim after missing data
The recorded table omits B’s outdoor result. The writer fills the blank with ‘passed’ because B did well indoors.
Revised reasoning. Keep the result missing and state its significance for the decision. Inventing data turns a reasonable report into an unreliable one.
What the reader learns. A blank cell is not permission to guess. Resist the desire to produce a complete-looking story. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
English evidence case 35: The final audience check
A polished paragraph describes angles and mirror geometry accurately but never tells the teacher which prototype to use for an event.
Revised reasoning. Add the decision in ordinary language, supported by the stated comparison and the intended audience’s needs.
What the reader learns. Writing succeeds when a real reader can act on it. Technical correctness does not replace task fulfilment. In an examination answer, keep the claim no larger than the supporting text and select language that makes the writer’s certainty and purpose visible.
How to prepare for a Writing task without memorising an unsafe template
Start by circling the role, recipient, purpose, required points and source material. Decide which points belong together and which need development. For a recommendation, choose criteria that reflect the reader’s actual decision. Write a direct opening, develop the comparison from the supplied evidence and finish with an action or justified conclusion. Check that every requested point has been addressed before improving vocabulary. A polished response that misses a compulsory requirement is still incomplete.
In K300 Situational Writing, the syllabus specifies a 250–350-word response based on a situation that involves viewing a visual text. That word range belongs to the official task; a student’s practice casebook can be longer. If the examination asks for an email, speech, report or another form, the writer should follow that form rather than insist on a review heading. If the question asks for a continuous argument, construct a genuinely argued response instead of forcing in a list of product specifications.
How to use the same skill in Comprehension and Oral Communication
In comprehension, the first job is to identify what the text actually says. The second is to examine what it implies or attempts to make the reader believe. Ask which source supports a claim, whether its words are conditional and whether the writer’s purpose might shape the selection of facts. For an 80-word summary, select the information required by the question and compress relationships without changing their meaning. Do not insert your own evaluation if the task asks only for a summary.
In oral communication, a balanced answer can still take a position. A response to a question about school invention projects might begin with a clear opinion, give a classroom example, weigh a practical concern and reach a concise judgement. The oral task is not a memorised essay spoken aloud. Its usefulness comes from reacting to the actual prompt, developing reasoning and communicating naturally. The video and examiner’s question establish the context.
A seven-day transfer plan
Day one: separate facts, opinions and missing information in two short reviews. Day two: replace unsupported superlatives and absolutes with bounded claims. Day three: practise point-by-point comparisons using one consistent set of criteria. Day four: plan a situational task from role, audience, purpose and visual details. Day five: write one complete response under a reasonable time limit and revise for task fulfilment. Day six: complete a comprehension exercise identifying the language used to persuade and the evidence it supplies. Day seven: discuss two designs aloud, changing the audience from a teacher to a younger pupil while keeping the facts stable.
Revision should not become ritual rereading. Read your paragraph once for meaning, once for task points and once for language accuracy. On the meaning pass, underline every factual assertion and locate its support. On the task pass, check whether the person reading can make the required decision. On the language pass, repair unclear references, mismatched comparisons, tense shifts and sentences that carry too many ideas. Those three passes have different jobs and therefore catch different errors.
Keep the earlier owners in view
This English article relies on the physical model of Volume 0472, Physics Reflection and the measured design reasoning in Volume 0473, Mathematics Ray Geometry. It does not replace either owner. For earlier media-comparison skills, see Volume 0466, Wave Exhibit Reading. For corrected details in a spoken source, see Volume 0470, Listening Corrections.
The PSLE Learner’s Guide on passage evidence offers a foundation for younger readers moving towards secondary-level inference and evaluation. The eduKate Sengkang learning routes give the wider English and Mathematics context. Volume 0475 will use these three subject owners in a single design decision. The main principle to carry forward is simple: make a decision where the task requires one, but never make the evidence say more than it says.
