The difference between a claim and what a source actually says
A public notice can be grammatically clear and still be wrong. A caption may report that three samples were investigated when the original record shows that only two were measured. It may use the word “all” where the source said “some”, or present a prediction as an observation. In English Language Paper 2, reading well means recovering not only what an author wants readers to believe, but what the supplied words and evidence actually permit. The ability is useful across school subjects and public life.
This learner’s guide is an original, examination-style non-narrative reading workshop. It gives you two fictional versions of a science-display notice, a small supporting table, a correction memo, questions at increasing depth and worked model answers. You will learn to distinguish a broad headline from a qualified statement, track dates and corrections, infer purpose without inventing motive, analyse language effects and write a concise summary whose ideas are genuinely supported by the designated passage.
Quick route: identify who issued the text and for whom; mark the source, scope and condition of each claim; recognise when a revision replaces an earlier figure; distinguish fact, inference and prediction; explain words in context; collect only eligible summary points; paraphrase precisely; review what the question actually asks before answering.
Read the preceding Volume 0834, Chemistry chloride-condition decisions, and Volume 0835, Mathematics sample concentration records if you need the science and numeracy foundations. This chapter is the English reading owner: it asks how evidence, caveats and editorial choices shape a non-narrative text. For earlier practice with qualified notices, follow Volume 0815; for earlier source attribution, see PSLE English Learner’s Guide 0077.
Official exam anchor, checked 5 November 2026: The SEAB 2027 G3 SEC list includes English Language K300. Paper 2 Comprehension is 1 hour 50 minutes, 50 marks and 35% of the total assessment. Section A addresses a visual text, Section B a narrative or recount, and Section C non-narrative texts and an approximately 80-word summary. The passages and questions here are original practice, not a past or predicted examination paper. Chemistry K324 supplies the science cross-check, not an invitation to conduct experiments with hazardous gases or molten chemicals.
Chapter 1. Read what the heading can and cannot guarantee
The heading of a notice is the reader’s first frame. It may invite people to visit, reassure them about a procedure, warn them of risk or announce a correction. A strong candidate identifies those purposes by pointing to wording rather than inventing an author’s private intention. “Important Correction to the Exhibition Guide” explicitly signals a change. “Everything You Need to Know” promises broad coverage; a small poster may not be able to keep that promise. “Proven for Every Sample” makes an absolute claim requiring much stronger support than “predicted under specified conditions”.
In comprehension, some questions ask for literal retrieval: what did the memo say? Others demand an inference: why might the author state the scope of the correction so carefully? Both depend on the supplied words, but the second requires a small, defensible step beyond them. That step should never become a leap into speculation. You do not know that the school tried to deceive people merely because it issued a correction. You can say it wants readers to use the updated version and not repeat an earlier mistake.
Chapter 2. Meet the two versions of the fictional exhibition notice
The scenario is entirely fictional. No chemical experiment is proposed for children. The display exists only on paper. Three labelled, theoretical conditions are discussed: M, molten sodium chloride; D, dilute aqueous sodium chloride; and C, concentrated aqueous sodium chloride. All predictions assume inert electrodes and the simplified syllabus model. A separate spreadsheet records fictional salt masses and solution volumes; these numerical records do not prove that particular gases were observed. When reading the passage, keep these two types of statement apart.
Text A — the exhibition notice before review
SCIENCE WEEK: CHLORIDE MAKES TWO GASES
Visit the Student Evidence Corner to discover what our chloride samples demonstrate. The exhibition combines a model of electrolysis with a table of solution measurements. The first display card shows molten sodium chloride. The second describes dilute sodium chloride in water. The third describes concentrated sodium chloride in water. These three arrangements are different, but all three demonstrate the same products at their electrodes. Every chloride sample gives sodium metal at the negative electrode and chlorine gas at the positive electrode.
The collection table contains Batch A, with 12 g of salt in 300 mL; Batch B, with 9 g in 450 mL; and Batch C, with 15 g in 150 mL. Because Batch C has the greatest salt mass, its concentration is only slightly greater than A’s. The table proves that the quantity of dissolved salt by itself is enough to predict the gases that will be formed. Our science team therefore recommends that every reader select the batch with the largest mass when comparing possible outcomes.
Students can join the exhibition discussion after class. All results are confirmed, and no uncertainty remains. The display describes chemical knowledge that should not be copied as a home experiment.
Text A contains a useful final safety sentence and several factual problems. It has also blended a theoretical model, a mass-per-volume table and an assertion of experimental confirmation. When you are asked what Text A claims, report its claim accurately even if the claim is scientifically wrong. When asked whether it is justified, cross-check the supporting material. “What the writer says” and “what the writer has shown” are not interchangeable answers.
Text B — corrected notice issued after review
SCIENCE WEEK: READ THE CONDITIONS BEFORE THE CONCLUSION
Revised display text, replacing the first notice. We have corrected several statements in the earlier exhibition summary. The original headline suggested that all sodium chloride conditions produce the same electrode products. That is not the standard prediction. The state of the electrolyte matters, and the dilute-versus-concentrated distinction changes the expected anode product in the simplified inert-electrode model.
For molten sodium chloride, sodium is expected at the cathode and chlorine at the anode. For dilute aqueous sodium chloride, hydrogen and oxygen are expected at the cathode and anode respectively under the syllabus contrast. For concentrated aqueous sodium chloride, hydrogen is expected at the cathode and chlorine at the anode. These are theoretical predictions tied to stated conditions. The exhibit does not present gas collection or gas-test observations as proof that a physical experiment took place.
The solution-measurement table also needs a correction in interpretation. Batch A has 12 g in 300 mL, equivalent to 40 g/L. Batch B has 9 g in 450 mL, equivalent to 20 g/L. Batch C has 15 g in 150 mL, equivalent to 100 g/L. The greatest dissolved mass and greatest concentration happen to belong to C in this table, but the conclusion about concentration comes from mass and volume, not mass alone. These values must not be treated as universal numerical thresholds for which chemical product forms at an electrode.
The paper display is intended to teach careful reading of scientific conditions and measured quantities. No learner should imitate electrolysis of salts, handle molten materials or produce chlorine outside a properly authorised professional setting. Readers should use the revised notice when quoting this exhibit and should indicate clearly whether they are describing a model, a calculation or a verified observation.
The corrected notice is less sweeping but more reliable. It does not claim the samples were actually electrolysed, and it separates a description of conditions from a calculation of concentration. The phrase “under the syllabus contrast” narrows the scope of the product prediction; “must not be treated as universal numerical thresholds” warns against transferring one type of information to another. A good comprehension answer notices those boundaries, not merely the three product pairs.
Chapter 3. A short editorial memo provides a third perspective
Text C — note to student guides
The revised exhibit should replace the first notice on all printed handouts. Staff should not describe the earlier errors as a “small typo”. They change the meaning of the main claims. Please retain the version date and refer questions about the chemical model to the approved syllabus material. The concentration spreadsheet can be used for a Mathematics discussion, but it must not be offered as an experimental record of gases collected. If a visitor asks what was observed in a laboratory, explain that this is a classroom paper case, not an account of a real electrolysis experiment. Students may discuss the science; they should not handle hazardous materials.
Text C gives a different kind of source: an instruction to the people presenting the material. It is not intended as public promotional copy, and its voice is firmer. It tells staff what to replace, what not to say and how to answer a predictable question. When a question asks you to compare Text B and Text C, first identify each intended audience. A public notice should explain what readers need to know; a staff memo can directly command an operational correction. Differences of tone may be explained by purpose without accusing either author of dishonesty.
Chapter 4. The first reading pass: identify the architecture
On your first read, do not try to memorise every chemical name. Mark the purpose of each paragraph. In Text A, paragraph one introduces the exhibit and makes a general claim. Paragraph two gives figures and makes an unsupported conclusion. Paragraph three invites readers and asserts certainty. In Text B, paragraph one announces a correction, paragraph two gives the science model, paragraph three explains measurements, and paragraph four sets boundaries and safety limits. Text C tells guides how to use the revision. This map makes retrieval faster because you know where each kind of answer belongs.
Use a small set of annotations rather than highlighting the page until every word looks equally important. Underline changes in condition: “molten”, “dilute aqueous”, “concentrated aqueous”. Box source phrases: “the earlier notice”, “the revised notice”, “the memo”. Circle sweeping terms such as “all”, “every”, “confirmed” and “no uncertainty”. Put a discreet question mark beside a strong assertion that the text does not visibly support. Your marks should save time, not become a second essay.
Chapter 5. Qualifiers are not decorative words
A qualifier defines how far a claim extends. “All” covers the entire stated group; “some” does not. “Expected” indicates a prediction; “observed” refers to reported evidence. “Under specified conditions” prevents a rule from pretending to apply everywhere. In a high-stakes reading question, replacing one of these words can reverse the meaning of a sentence. A candidate who ignores qualifiers may identify the general topic correctly yet miss the precise point under test.
Text A says “all results are confirmed”. Text B says the products are “theoretical predictions tied to stated conditions”. Their difference is not simply that one notice is longer than the other. The second distinguishes the status of the evidence. You can support that interpretation directly by referring to the mention of no gas-collection or gas-test observations. The correct inference is narrow: the revised notice does not claim actual laboratory confirmation. It does not prove that experiments never take place anywhere in the school or that the writers have no scientific knowledge.
Chapter 6. Measure the size of an inference
Inference is the movement from stated evidence to a reasoned conclusion that is not given in exactly those words. Text C instructs guides not to call the error a “small typo”. We can infer that the correction affects substantive meaning rather than merely spelling or punctuation. We should not infer that the first notice was intentionally misleading, because the memo supplies no evidence about intention. A safe answer uses the source to justify the smallest conclusion that fully answers the question.
Suppose the question asks why the memo recommends keeping the revision date. One defensible explanation is that readers need to distinguish the corrected version from copies of the earlier notice. The purpose follows from the instruction to replace old handouts. Saying that the date protects the school from litigation would be an unsupported addition. Examination readers are rewarded for inference from details, not for imaginative speculation about motives or consequences.
Chapter 7. Language effect requires a reader and a purpose
If asked about the effect of “EVERY CHLORIDE SAMPLE” in Text A, do not simply reply “it emphasises the sample”. Explain that the absolute wording makes the claim sound comprehensive and certain, encouraging readers to accept it without considering different conditions. In the revised heading, “READ THE CONDITIONS BEFORE THE CONCLUSION” is an imperative. It directs the reader to slow down and examine the premises before accepting the outcome. One sounds sweeping; the other teaches caution.
Language-effect answers often work best with three parts: the chosen word or feature, the meaning created by it and its likely effect on the intended reader. A command may prompt action. Repetition may stress a central contrast. Technical vocabulary may confer precision if it is explained, but may confuse an unfamiliar audience if left unexplained. Avoid generic statements such as “it captures attention” when the wording has a more specific effect.
Chapter 8. Text A is not useless because it is wrong
A comprehension task sometimes gives an inaccurate claim on purpose. The candidate may still need to identify its intended audience, locate a phrase or explain how the argument develops. Do not answer every question as a scientific fact-check. If asked “Which batch does the writer recommend?”, the literal answer is the batch with the largest mass, because that is what Text A says. If asked “Is the recommendation justified by the figures?”, the answer requires evaluating concentration rather than merely repeating the recommendation.
Separating reporting from evaluating is especially important when two texts disagree. The first part of an answer may say “Text A claims X”; the second may state “Text B corrects this because Y”. This structure prevents a contradiction from becoming your own. It also makes the source of each idea visible. It is the English equivalent of checking which denominator belongs to which mathematics calculation.
Chapter 9. Reading numerical evidence in a language passage
Paper 2 is not a Mathematics paper, but numbers are still meaning-bearing language. Text A’s table gives both grams and millilitres; the flawed sentence pays attention only to mass. The reader does not need a lengthy algebraic derivation to recognise why that is incomplete. Concentration requires mass divided by volume. In Text B the same data are recalculated as 40, 20 and 100 g/L. The reading skill is to connect the correction to the evidence, not to write a new chemical law.
If an English question asks what is misleading about “the quantity of dissolved salt by itself is enough”, quote or paraphrase the relevant condition: different solution volumes. Do not turn the response into a speculative account of electrolysis equipment or gas production. Good cross-curricular reading respects the limits of the source. A numerical figure can establish its own unit relationship without proving an unrelated scientific event.
Chapter 10. Prepare the summary by selecting, not copying
In the G3 SEC K300 Paper 2 Section C summary task, the question identifies what aspect of the passage you must summarise. The task is not to compress the entire passage indiscriminately. For this workshop, suppose the prompt says: Using Text B only, summarise the corrections made to the original exhibition notice and the advice given to readers. Write approximately 80 words. The relevant material lies across Text B’s paragraphs; you should gather the changes in scientific prediction, the concentration reasoning, evidence status and reading/safety advice.
First, make a working list: earlier notice wrongly treated the electrode products as identical; molten NaCl gives sodium and chlorine in the model; dilute aqueous gives hydrogen and oxygen; concentrated aqueous gives hydrogen and chlorine; concentrations depend on both mass and volume; table values cannot set universal chemical thresholds; predictions are not proof of laboratory observations; use the revised notice; distinguish models, calculations and observed evidence; avoid unsupervised hazardous chemical work. These are candidate points, not yet a final summary.
Next merge connected ideas. You need not repeat “under the inert-electrode model” in every sentence if it is clear once. A good summary preserves the essential contrasts without reproducing the display’s full numerical table if it is not necessary. Do not write “the exhibition proved that…” because the passage expressly denies providing observed gas results. Avoid adding an example from another volume. The source boundary matters as much as the word limit.
Model summary, approximately 80 words: “The revised notice corrects the claim that all sodium chloride conditions produce identical electrode products. Under the stated model, molten salt yields sodium and chlorine, dilute aqueous solution yields hydrogen and oxygen, and concentrated aqueous solution yields hydrogen and chlorine. It explains that concentration depends on both dissolved mass and solution volume, not mass alone. The table cannot establish universal product thresholds or actual gas observations. Readers should use the correction, distinguish predictions from measurements and avoid attempting hazardous experiments.”
Count words according to the examination’s instructions rather than relying blindly on a word processor that may split hyphenated terms differently. More important, check that every point can be located in Text B and that each verb accurately represents the source. A summary of the wrong question cannot be rescued by excellent grammar. A summary that changes “expected” into “proved” loses a crucial distinction even when its length is perfect.
Twenty-five short reading decisions, with model explanations
Attempt each item before reading its answer. These questions deliberately vary what must be found: a stated fact, a supported inference, a language effect, a cross-text comparison or the precise boundary of a summary. They build exam control without relying on a rigid formula.
Reading decision 1
Task: What is the main purpose of Text A’s opening paragraph?
Model reasoning: It introduces the school exhibition and attempts to persuade the reader that its three chloride conditions lead to a single product pattern. The opening explains the display topic and then makes an overconfident generalisation. Name both functions without endorsing the claim as scientifically correct.
Reading decision 2
Task: Which sentence in Text A makes the strongest universal claim?
Model reasoning: The sentence saying that every chloride sample produces sodium at the negative electrode and chlorine at the positive electrode. The word “every” removes exceptions, even though the text itself lists different states and concentrations. Quote only what is needed to identify the scope.
Reading decision 3
Task: How is Text B’s heading different in purpose from Text A’s?
Model reasoning: Text A promotes a confident conclusion, while Text B directs readers to examine the conditions before drawing conclusions. The imperative in the revised heading encourages careful reasoning. Mention the reader-facing effect rather than merely saying the titles are different.
Reading decision 4
Task: What does “replacing the first notice” establish?
Model reasoning: The revised notice is intended to supersede the earlier text as the current guidance. It does not mean the earlier notice never existed or that every copy has already been removed. Distinguish editorial intention from completed distribution.
Reading decision 5
Task: Why does Text B say the electrode products are predictions?
Model reasoning: It makes clear that the display describes a chemical model tied to specific conditions and does not offer gas collection or gas-test observations proving a physical experiment took place. The answer should link the word “predictions” to the stated absence of observational proof.
Reading decision 6
Task: What role does the word “respectively” play in the dilute aqueous example?
Model reasoning: It pairs hydrogen with the cathode and oxygen with the anode in that order. Ignoring “respectively” could reverse which product belongs to which electrode. This is a precise referential function, not a general emphasis.
Reading decision 7
Task: How is mass different from concentration in the two notices?
Model reasoning: Mass names an amount of dissolved material; concentration relates that mass to the volume of solution. Text A wrongly treats mass alone as sufficient, while Text B gives concentrations using both quantities. The correction is about how comparisons are made, not merely larger numbers.
Reading decision 8
Task: Why might the instruction “must not be treated as universal numerical thresholds” matter?
Model reasoning: It prevents readers from turning the fictional concentration data into a chemical rule about which gas forms. The table supports a quantitative comparison, while the electrolysis prediction needs specified chemical conditions. Explain the risk of crossing from one form of evidence to another.
Reading decision 9
Task: What can a reader infer about the reason for Text C’s firm tone?
Model reasoning: The memo is addressed to staff responsible for replacing notices and answering questions, so direct instructions are appropriate. It stresses that the errors change meaning. Do not invent disciplinary consequences or assume individual staff were intentionally dishonest.
Reading decision 10
Task: Why is “small typo” rejected in Text C?
Model reasoning: The mistakes alter scientific predictions and numerical interpretation, rather than merely spelling or presentation. The phrase would minimise a substantive correction. Point to the content changes as the reason.
Reading decision 11
Task: Which text should a guide use when giving the current public account?
Model reasoning: Text B, according to the instruction in Text C to replace the first notice. A good answer identifies the direct source for the decision, not simply the version that appears more technically detailed.
Reading decision 12
Task: Does Text C establish that gas was observed?
Model reasoning: No. It expressly tells guides to say the case is a classroom paper exercise, not a real electrolysis experiment. Do not treat the references to “samples” and “exhibition” as evidence of laboratory testing.
Reading decision 13
Task: Explain the effect of “no uncertainty remains” in Text A.
Model reasoning: It gives readers an impression of complete confidence and invites acceptance of the conclusions. The expression is misleading because the notices mix predictions with unverified claims. Link the strength of the wording to its influence on a visitor.
Reading decision 14
Task: What is one contrast between a public notice and an internal memo?
Model reasoning: The public notice provides an explanation of the corrected science for visitors, while the memo instructs student guides how to replace materials and respond to questions. Both concern the same correction, but their purposes and audiences differ.
Reading decision 15
Task: Why is “all results are confirmed” not interchangeable with “theoretical predictions”?
Model reasoning: Confirmation refers to support from evidence, while a theoretical prediction describes an expected outcome under a model. The revised notice says no gas observations are supplied. Replacing one phrase with the other changes the evidential strength of the claim.
Reading decision 16
Task: Which numerical value in Text B corresponds to Batch B?
Model reasoning: The corrected concentration is 20 g/L, from 9 g in 450 mL. If asked only for the value, answer briefly. A longer unit-conversion calculation is useful when justification is requested, but not needed for direct retrieval.
Reading decision 17
Task: Is it fair to infer that the author of Text A knew the statements were false?
Model reasoning: No. The text and memo document errors and corrections but say nothing conclusive about the author’s knowledge or intention. A defensible answer can describe the claims as inaccurate without attributing motive.
Reading decision 18
Task: What does the final safety instruction contribute to the revised notice?
Model reasoning: It defines what readers should not attempt and maintains a boundary between classroom explanation and hazardous practical work. It also reinforces the exhibit’s stated purpose as a paper learning activity. Do not present it as evidence that any unsafe incident occurred.
Reading decision 19
Task: How would you answer a question asking for a common theme across Texts B and C?
Model reasoning: Both insist that readers use the corrected information and distinguish what is predicted or calculated from what has actually been observed. Support the theme with a detail from each text instead of relying on a vague word such as “science”.
Reading decision 20
Task: Which part belongs in an 80-word summary of corrections, and which does not?
Model reasoning: Include changes to claims, product predictions, concentration interpretation and evidence status because Text B presents them as corrections. Do not insert a story about a visitor arriving late or an industrial process; those details are not in the specified text and do not answer the summary task.
Reading decision 21
Task: How could a candidate explain the role of the semicolon in a compressed response?
Model reasoning: It can join closely related independent clauses without pretending that one is merely an adjective of the other. However, a summary should not sacrifice grammatical clarity to save a word. Use punctuation that expresses the relationship accurately.
Reading decision 22
Task: What difference does “can” make compared with “will”?
Model reasoning: “Can” expresses possibility or capacity, whereas “will” suggests a particular future outcome or strong prediction. If a writer changes “can lead to” into “will lead to” without evidence, the claim becomes stronger than the source supports.
Reading decision 23
Task: If a question requests two errors, why should you not list five?
Model reasoning: A precise two-part answer shows you can select and explain the relevant evidence. Unnecessary extra material may introduce contradictions or unsupported statements. Choose two distinct errors and make each correction clear.
Reading decision 24
Task: Why is a quotation not automatically an explanation?
Model reasoning: A quotation supplies words from the source, but the question may require you to interpret their meaning, purpose or effect. Add an explanatory clause linking the quoted phrase to the point you are making.
Chapter 11. Extend the skill to an unfamiliar notice
Consider a second fictional setting. A student council poster announces a schedule change: “All participants should arrive at 9.00 am.” An accompanying official update says: “Students assigned to the first demonstration should arrive at 9.00 am; other groups should follow the time shown on their invitation.” The mistake is not an obscure fact. It is a failure to preserve the scope of the instruction. The word “all” turns a rule for one group into a rule for everyone.
The same reading process applies. Identify the original claim, find the governing condition, determine whom the exception concerns, and decide which updated source has authority. If the paper asks you to explain why the revised notice is clearer, you can say it separates the two groups and tells each how to find the correct time. Do not conclude that the schedule change was caused by illness, poor planning or an accident; no such reason appears in the text.
Now consider a library notice that reads, “Books returned on Friday may be processed on Monday.” The modal “may” expresses possibility, not a guarantee. Rewriting it as “all books returned on Friday will be processed on Monday” produces a stronger commitment. An attentive candidate can explain the difference even without knowing anything about the library’s internal workflow. Words such as “may”, “usually”, “unless”, “except”, “only” and “at least” are small, but their job is structural.
Chapter 12. Tone is not the same as attitude toward a person
An English comprehension question may ask about tone. Possible descriptions include cautious, urgent, reassuring, critical, instructive and celebratory, but the label must be justified. Text B’s tone is corrective and measured because it acknowledges errors, specifies conditions and avoids claiming more than the source supports. Text C is direct and procedural because it tells guides what actions to take. Text A is promotional and overconfident, especially in its headline and universal claims.
Do not describe Text B as “angry” merely because it says the first notice was incorrect. Correction is not inherently anger. A measured public explanation can disagree firmly with an earlier claim while remaining calm. Likewise, a staff memo that gives commands need not be hostile; its form may simply match an operational purpose. Read the surrounding sentences before assigning emotion to a phrase.
Chapter 13. Answer what is asked, in the length it deserves
Paper 2 time is limited. A retrieval question can often be answered in a phrase or sentence. A language-effect explanation may need two well-connected sentences. A comparison needs a point about each text and the relationship between them. A summary must be planned and edited as a whole. You do not earn clarity by using more words than the task requires. You earn it by choosing words that carry evidence.
Make a two-column correction log during practice. On the left, record the weak answer exactly as you wrote it: “The word every attracts attention.” On the right, replace it with a specific explanation: “The word every presents the prediction as universal, encouraging readers to overlook differences between the samples.” This teaches you to make the connection between language and meaning visible. For source evaluation, replace “The author lies” with “The wording is stronger than the supplied evidence supports.”
Practise retrieving facts without importing a claim from another text. If Text A says a thing and Text B corrects it, label each in your working notes. When a question specifies “Using Text B only”, do not use Text C to support a summary point even if Text C seems clearer. Source boundaries are part of the test. The best answer is not the one that knows the most about the topic; it is the one that answers the task from the allowed material.
Chapter 14. A timed practice sequence
Begin with a short untimed reading to establish accuracy. First mark the three conditions and source names. Next answer five direct-retrieval questions without looking at your notes. Then explain two language choices and justify one inference. Finally, write an approximately 80-word summary from Text B alone. Check factual selection and meaning before counting words. Once this sequence is reliable, repeat with the clock running and a fresh invented notice.
On the next day, revisit only the errors you made. If you repeatedly miss exception words, read sentences with “unless”, “except” and “only” and explain the scope of each. If you over-infer motives, compare your answer with the exact source sentence and remove claims it cannot bear. If summary words run over the limit, practise merging related points instead of deleting indispensable qualifiers. Different weaknesses require different repairs.
Chapter 15. A second summary with a different focus
Change the prompt to: “Using Text B only, summarise what the revised notice advises visitors to do when reading the exhibit.” This is not the same as summarising every science correction. Now the main ideas concern reading conditions, distinguishing models from observations, using mass and volume for concentration, avoiding false universal thresholds, following the revised notice and respecting the safety boundary. Technical product pairs may be mentioned selectively, but they are not the entire purpose of this newly framed task.
Model response: “Visitors should read the conditions attached to each electrolysis prediction and avoid treating different sodium chloride forms as identical. When comparing measured solutions, they should consider dissolved mass together with volume. The sample table does not establish a universal concentration threshold for particular products. The notice also distinguishes theoretical outcomes from laboratory observations, which have not been supplied. Visitors should rely on the revised information and must not copy hazardous chemical procedures outside an authorised setting.”
Comparing the two summary prompts is a powerful self-test. Both use the same source, but each asks for a different selection of information. A learner who copies the first summary into the second task has not fully read the new question. Before writing, convert the task wording into a one-line focus statement: “I need corrections and advice” or “I need reader actions.” That statement helps you decide what belongs and what must be left out.
Frequently asked questions
How do I tell the difference between inference and guessing?
An inference is supported by details in the source and does not require an invented backstory. If you cannot point to a phrase or relationship that makes your conclusion reasonable, you may be guessing. State only what the evidence allows.
Should I fact-check the subject matter in every English question?
Use the information supplied and respond to the command word. Some questions ask you to report what a writer claims, even if the claim is later corrected. Others ask you to evaluate the claim against another text. Keep those tasks distinct. Scientific knowledge can help you understand the context, but it cannot replace the exact written evidence.
Can I copy the passage into the summary?
You should express the necessary points economically and, where possible, in your own words. Do not change a qualified statement into an absolute one just to simplify it. The quality of a summary lies in accurate selection, compression and coherence, not in elaborate synonyms.
Why is source attribution important?
It tells readers whether a statement belongs to the first notice, the revision, the staff memo or your own interpretation. Without attribution, you can accidentally repeat a corrected claim as if it were accepted fact. It also helps with cross-text comparison.
What is the most useful final check?
Return to the question. Underline its focus, permitted source and required number of points. Then inspect your answer for claims stronger than the passage. A concise answer with the right evidence is more valuable than a broad answer full of plausible but ungrounded additions.
The principle worth taking into the next paper
Words such as “all”, “may”, “expected”, “observed”, “only” and “unless” carry the boundaries of a claim. An English learner who preserves those boundaries can read news notices, laboratory descriptions, instructions and arguments with greater independence. Read the words, recover the source, test the inference, explain the language and stop where the evidence stops.
