Quick answer. A strong integrated G3 SEC response uses the correct subject skill for each question: explain the current-produced magnetic effect in Physics, compare the right denominators in Mathematics, and write only evidence-supported conclusions in English. A small classroom model can demonstrate an idea; it cannot prove universal reliability or certify any installed door system.
This concluding EMS guide gives a fictional evidence board, worked calculations, careful language revisions and a timed decision exercise. It follows Volumes 1117, 1118 and 1119 in that order, bringing together the distinct skills without treating EMS as a separate examination subject.
- Science: K323 Physics, current, solenoid, core and mechanical release.
- Mathematics: K310, pass counts, unequal samples, pooled proportions and constraints.
- English: K300, purpose, qualifiers, inference, summary and public wording.
- Decision: publish only what the recorded evidence can support.
The final task is to decide what the record can support
The first three guides in this sequence teach different skills. Volume 1117 explains a low-voltage electromagnet classroom model and the conditions that affect release. Volume 1118 compares unequal sample sizes without losing denominators. Volume 1119 reads two fictional notices for omission, inference and purpose. This final volume puts those skills together in a decision that demands restraint. The question is not ‘Can I say something impressive?’ It is ‘What is the strongest conclusion this evidence can actually support?’
Imagine a student committee planning a teaching display about magnetism. The committee has a model, a small set of trials, a set of inconsistent records and draft wording for visitors. The committee is not evaluating a real installed safety device, and students must not modify building equipment. The work is entirely educational. To succeed in the exercise, the candidate must audit the physics claim, compute numerical proportions, identify language that overstates the case and recommend a responsible next step.
A simple evidence hierarchy
Start with raw observations: current readings, switch position, armature retained or released, gap and timing. Next come calculations: pass proportions and comparisons across groups. Next come interpretations: which hypothesis is consistent with the observations, and what else could explain the result? Last come public claims and recommendations. Each level depends on the earlier levels. If a summary statement contradicts the raw record, correct the summary. If a proposed inference needs a variable that was not measured, label that inference uncertain.
A student can organize the material in four columns headed Observation, Calculation, Interpretation and Decision. An observation is ‘the plate remained attached in twelve of fifteen trials’. A calculation is ’12/15 = 80%’. An interpretation is ‘the model retained the plate more often in this set than in the other set’. A cautious decision is ‘repeat matched trials before making a broad public reliability claim’. Keeping those columns distinct helps prevent a persuasive sentence from being mistaken for a measurement.
Know which official assessments this practice supports
The 2027 Singapore-Cambridge SEC replaces the separate GCE N(T), N(A) and O-Level qualifications in line with Full Subject-Based Banding. Each G3 subject is examined at its own subject level. The current school-candidate codes are K300 for English Language, K310 for Mathematics and K323 for Physics. The older reference codes, used up to 2026, are 1184, 4052 and 6091 respectively. Codes matter when finding the correct official syllabus and past examination material.
The invented EMS case is not a separate SEC examination paper. It is cross-subject practice. English Paper 2 assesses comprehension and careful inference; Mathematics Paper 2 includes a final real-world application question; Physics Paper 2 requires explanations and data-based reasoning, while Physics Paper 3 assesses practical skills. The 2027 English paper weightings are 35%, 35%, 10% and 20% for Papers 1 to 4. Mathematics has two equally weighted papers. Physics has 30%, 50% and 20% for Papers 1 to 3. Use the official syllabus for precise descriptions, not a composite case invented for practice.
The fictional record you must audit
Version A uses 120 coil turns and records twelve holds from fifteen trials. Version B uses 150 turns and records fourteen holds from twenty trials. Current in A was measured at 0.35 A at the start and finish of the set; for B, the operator records only a supply voltage and not the actual coil current. The gap was held constant by a spacer for A but manually estimated for B. A spring was fitted for release in both arrangements, but two observations about release were left blank.
Four pieces of information matter immediately. First, the pass proportions are 80% and 70%, so B’s higher raw number does not make it the better observed rate. Second, the number of turns differs but the current was not controlled or recorded in the second set, so there is no clean isolation of the turn-count effect. Third, the gap control differs, which introduces another possible influence. Fourth, missing release observations prevent a claim that the model released correctly in every attempt. Do not add those unrecorded events to the pass column by assumption.
Physics audit: what does switching the current do?
A current flowing in the coil produces a magnetic field. A soft-iron core may strengthen the magnetic effect. In a suitable classroom model the iron plate can be attracted and retained, depending on the coil and geometry. When the circuit opens, the energized magnetic effect decreases. The spring can then separate the armature if the arrangement and forces permit it. Power removal does not, by itself, prove that a separate repelling force acted.
For the written answer, identify which part is directly observed. If the table says the switch was opened, that is a control action. If an ammeter says the current fell to zero within its precision, that is electrical evidence. If the plate moves away, that is mechanical evidence. A complete explanation connects these facts without pretending that any single observation establishes all others. This is why a well-designed table includes both the electrical state and the release outcome.
Physics audit: the materials and their limits
An iron armature becomes magnetized in the presence of a suitable field and is commonly attracted. The behaviour of a permanently magnetized piece needs different analysis. Do not conclude that reversing current must make an ordinary soft-iron plate be repelled. Reversing current reverses the polarity of the field produced by a solenoid with a fixed winding. The induced magnetization and actual force on the plate depend on material and arrangement.
A useful examination response distinguishes direction of field from magnitude of attraction. Another useful distinction is between the temporary magnetic effect in soft iron and the greater tendency of some steels to retain magnetism. Neither material description automatically supplies a release time. To assess release, the student needs observations of the actual mechanical result. Qualified explanations are not incomplete; they are accurate when the problem has genuinely omitted necessary information.
Mathematics audit: rebuild the counts before the chart
The table contains A: 12 passes and 3 failures from 15. B: 14 passes and 6 failures from 20. Combined: 26 passes, 9 failures, 35 attempts. Report the percentages as 80%, 70% and 26/35 ≈ 74.3%. The combined rate is not 75%, because A and B were tested different numbers of times. Calculations must be based on the correct denominators, not on a visually attractive graph or a rounded claim in a draft poster.
If one result is later discovered to have been copied twice, the correct response is not to guess which result to remove. Look for a traceable trial identifier. If the duplicate truly represents one extra copy of a success in B, the corrected B denominator and success count would both decrease by one, giving 13/19 rather than 14/20. If the duplication involves another outcome, the correction differs. The identity of the duplicate must be known before calculating a replacement rate.
Mathematics audit: choose the exact comparison being asked
‘Which group had more successful trials?’ is answered by B, with fourteen compared with twelve. ‘Which group had the higher observed proportion?’ is answered by A, with 80% compared with 70%. ‘Which group is proven more dependable?’ cannot be answered from these figures alone, because of the uncontrolled gap, unmatched current and limited samples. These are three different questions. Good examination performance comes from resisting the temptation to provide the same answer to all three.
If a task asks for the relative increase of A’s pass rate over B’s, the calculation is (80−70)/70 × 100% ≈ 14.3%. If it asks for the percentage-point gap, the answer is 10 percentage points. If it asks what share of all documented failures came from B, calculate 6/9 = 2/3. Read the denominator implied by the wording before choosing an operation. A wrong population produces an elegant but irrelevant answer.
English audit: the most dangerous words are often small
Draft A says, ‘The model held its plate in fourteen trials.’ Draft B says, ‘The holder has never failed.’ The first can be a true but incomplete count; the second contradicts the six failures recorded for Version B. It is not merely optimistic; it is factually unsupported. The word ‘never’ removes conditions that are essential to the meaning. When a student revises the text, the aim is to restore measured boundaries without burying the reader in unnecessary caution.
A better sentence is: ‘The classroom model retained its plate in fourteen of twenty recorded trials under the tested conditions.’ This is not a complete approval statement, but it accurately conveys the data. For a visitor-facing explanation, add one clear sentence: ‘This is a demonstration of electromagnetism, not a test of any installed building equipment.’ The result is accessible, honest and useful. It also illustrates why English reading and writing matter to scientific communication.
What to do with conflicting testimony
A club member says, ‘I heard the plate release every time.’ The record has two blank release fields and no timing data. Neither dismiss the witness automatically nor convert the recollection into an official measurement. Classify the statement as an account that may warrant checking. Ask what was observed directly, whether each trial had a written release entry and whether the witness saw all attempts. Preserve the distinction between recorded data and retrospective memory.
In an examination-style decision, a strong statement is: ‘The witness reports consistent release, but the written log contains missing entries, so the record does not verify that every trial released as expected.’ It is possible for the witness to be sincere and for the available evidence to remain insufficient. This dual recognition is an advanced communication skill. It avoids the false choice between blind belief and unwarranted accusation.
Distinguish mechanical outcomes before computing a rate
A school model can have more than one outcome: the plate may be held when powered, release when unpowered, fail to hold, or remain attached unexpectedly. If the task asks about successful holding, only the correct ‘hold under power’ observations belong in that numerator. If the task asks about release after switching off, those require a separate record. Combining both outcomes into one poorly defined ‘success’ column hides the very failure mode the investigator needs to understand.
Define the success criterion before observing data. For an integrated exercise, it might be ‘held under the stated powered condition and subsequently released after power removal when the spring acted’. That compound outcome is more demanding than ‘held’. The current table does not provide enough release information to compute the combined criterion. A capable student should refuse to manufacture a new percentage from the wrong measure. Naming the missing data is part of solving the problem.
Design a fair follow-up trial
A defensible follow-up compares two winding counts at a matched gap, core material, plate geometry, spring arrangement and measured current. It records each attempted trial with a unique identifier and a clearly stated result. The number of turns is the independent variable; a measured holding or release outcome is the dependent measure. The set-up should be teacher supervised, low voltage and appropriate to school apparatus. It is not a procedure for inspecting, bypassing or altering any real safety device.
The plan should say how the results will be used to reach a conclusion, not merely list apparatus. If students want to test whether changing turns altered the likelihood of successful retention in a particular model, they must first ensure that conditions and the definition of success remain comparable. Multiple trials may show variability. They will not prove a universal probability across every design. The scientific principle is to isolate the claimed cause and report the remaining uncertainty.
Use a decision table rather than a slogan
Organize possible actions into three categories: supported now, possible after more evidence, and not justified from this record. Supported now: explain that the model demonstrates a magnetic effect of current; report the exact trial counts; invite students to examine how a switch changes the energized state. Possible after controlled retesting: compare two versions more confidently. Not justified: assert that a model always retains or always releases; claim any built-in safety certification; infer an exact release time without timing measurements.
This decision table is also good preparation for English response questions. Instead of writing a paragraph full of repeated warnings, write one decisive sentence followed by its reason. For example: ‘Approve the educational demonstration as a model of a current-produced magnetic effect only if supervised and clearly described; do not publish a universal reliability claim because the trial records are incomplete and the setups were not fairly matched.’ The conditional wording is purposeful and grounded.
How to handle an unfamiliar question under examination time limits
Begin by identifying which subject skill the subquestion requires. If it asks for a fraction, solve the fraction before debating the physics. If it asks why an armature moved, trace current, field and force. If it asks how a poster influences readers, identify wording and effect rather than explaining the circuit again. Cross-subject integration is useful training precisely because it forces the reader to choose the right lens for each instruction.
Next, mark the evidence source: table, diagram, quotation or instruction. Use the table for calculations, the diagram for directions and connections, and the prose for claims and conditions. Then provide an answer with the minimum necessary steps and a justified conclusion. Leave time to check that your result matches the original question. Accuracy is more important than an elaborate introduction that consumes the minute needed for a complete calculation.
Make errors visible enough to repair
After an integrated attempt, create an error ledger with columns for the wrong step, the reason it failed, the correct method and the next test. ‘Averaged 80% and 70%’ becomes ‘ignored unequal denominators’ and is repaired by pooling counts. ‘Said power-off repels’ becomes ‘confused reduced attraction with repulsion’ and is repaired by identifying the spring’s role. ‘Said always’ becomes ‘removed qualifier’ and is repaired by retaining the sample boundary.
The next test should change a variable so the learner must apply the repaired rule again. Use a different sample size, a reversed coil direction or an altered headline. A student who can reproduce the answer to the old question may not yet have corrected the underlying mental model. A short, deliberately varied retest often reveals whether the correction is secure.
The final public sentence
A sample of suitable student-facing wording is: ‘Our low-voltage classroom model shows how current in a coil can produce magnetic attraction. In the recorded trial sets, the plate was held twelve times out of fifteen and fourteen times out of twenty, respectively. The groups used different controls, and some release observations are missing, so the results do not establish reliable performance in other settings.’ This is relatively concise and respects the evidence.
There is still room for an engaging invitation: ‘Visitors can explore the science of electromagnets and practise reading experimental evidence.’ The educational appeal is genuine without promising technical safety. Good writing does not need to conceal uncertainty to hold a reader’s attention. It needs to state what has been learned clearly, explain the unresolved question and describe the next responsible step.
Retest the sequence, not the story
Take a fresh problem in which one group has more raw successes but a smaller success percentage. Make the core steel instead of soft iron. Replace a direct observation with a witness recollection. Change ‘may’ to ‘will’ in a public notice. These variations reveal the same four hazards: confusing conditions, using the wrong denominator, inferring more than was observed and exaggerating certainty. Prepare for examination transfer by recognizing those hazards even when the school-club story disappears.
A well-prepared learner should be able to use the method on a heating experiment, a transport timetable, a statistical survey or an unfamiliar narrative passage. The particular apparatus is a teaching frame, not the destination. The destination is a reliable way of thinking under time pressure: identify the question, trace the evidence, apply the right rule, and communicate a conclusion whose strength matches the available information.
Twenty integrated transfer cases with worked corrections
The next tasks are original educational drills. They resemble the mix of calculations, explanations and interpretation that students must handle across their different subjects; they are not copies of SEAB questions and do not constitute an additional official paper. Attempt each question, check the model correction, and explain why an attractive alternative answer would be wrong.
Transfer case 1: Wrong raw-count comparison
Prompt: Version B has fourteen successes and A has twelve. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: B has more successes, but the observed proportions are B 70% and A 80%. Both statements can be true. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Separate counts from rates. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 2: A combined percentage
Prompt: A report averages 80% and 70%. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Pooling gives 26/35 ≈ 74.3%, not 75%, because the two groups have different numbers of trials. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Add numerators and denominators separately. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 3: A missing current reading
Prompt: Version B used more coil turns but did not record current. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Turn count cannot be isolated cleanly as the cause; changing winding may also change resistance and current. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Ask what was controlled. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 4: An uneven gap
Prompt: A had a fixed spacer and B used an estimated gap. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Different geometries could affect attraction, so the observed rate difference is not evidence of a sole winding effect. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Identify an alternative explanation. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 5: Unlogged release
Prompt: Two records show no release outcome. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Do not mark them successful releases. Request the actual observation or state that the combined outcome cannot be calculated. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Missing is not success. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 6: An absolute headline
Prompt: The public draft says ‘never fails’. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: The record contains failures. Revise it to a qualified report of the observed pass fraction. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Accuracy beats enthusiasm. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 7: A switched-off coil
Prompt: A writer says the coil repels the plate when off. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: The energized magnetic effect falls, but movement depends on the spring or other forces. No automatic repulsion follows. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Name the mechanism. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 8: Reversing current
Prompt: The diagram swaps coil terminals. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: With the same winding, solenoid polarity reverses. That does not mean ordinary unmagnetized iron must be repelled. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Polarity and attraction differ. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 9: A percentage point
Prompt: A is 80% and B 70%. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: A exceeds B by ten percentage points, rather than by ten per cent relative to B. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Choose the requested measure. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 10: A relative comparison
Prompt: A report asks for increase relative to B’s rate. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Compute (80−70)/70 × 100% ≈ 14.3%. State the baseline B. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Denominator defines the comparison. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 11: A witness remembers
Prompt: Someone says every release worked. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Treat this as testimony, not a complete set of recorded results. Verify against the missing fields. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Do not turn memory into data. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 12: A still photograph
Prompt: The poster displays a plate attached while powered. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: It supports that one visible state, not the release sequence or long-run reliability. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: One image proves little about time. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 13: An ambiguous success field
Prompt: The log says ‘pass’ but never defines what passed. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Define whether the criterion is retention, release or both before calculating a rate. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Measure the intended outcome. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 14: An impractical proposal
Prompt: The committee wants to test a real building fire door. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Reject that as a student activity. Use only an approved classroom model and qualified oversight for any real equipment. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: The case is educational. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 15: A data-based Physics answer
Prompt: The examiner asks for two possible reasons for inconsistency. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: Point to documented differences such as gap or unmeasured current, and explain how each may affect the result. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Use evidence-supported possibilities. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 16: An English purpose question
Prompt: Text A records limitations; Text B invites visitors. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: A informs accurately and B promotes. Analyse how each selects and frames the same result. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Do not confuse purpose with topic. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 17: A Mathematics word problem
Prompt: Thirty-five trials include nine failures. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: The pass count is 26. If selecting uniformly from 35 recorded cards, P(pass-labelled card) = 26/35. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Define the random experiment. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 18: An unjustified probability claim
Prompt: A club treats its 70% sample proportion as universal. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: A small, non-matched observational sample cannot by itself justify a population-wide probability. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Keep conditions and scope. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 19: A missing unit
Prompt: A note reports power as 2.4 with no unit. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: At 6 V and 0.4 A the electrical power is 2.4 W, not 2.4 N. Force needs separate evidence. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Units reveal errors. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Transfer case 20: A weak final conclusion
Prompt: The student says ‘probably safe because most trials passed’. Give a clear answer, including any numerical calculation, physical explanation or textual evidence that the task requires.
Worked correction: This model does not certify an installed safety system. Conclude only about the reported educational trials and propose further controlled checking. The point is not to write more than needed, but to justify the conclusion through an identifiable source rather than an assumption.
Retest cue: Do not cross the model boundary. Now imagine the number of attempts, the circuit state or the audience changed. Which part of the answer would have to change? That question tests whether the underlying rule has been understood.
Timed mini-mock: decide whether the evidence is ready for display
Scenario: A supervised science club has prepared a low-voltage desk-model display. The log shows Version A with 120 turns and twelve retentions in fifteen trials. Version B has 150 turns and fourteen retentions in twenty trials. A’s current was recorded at 0.35 A, but B’s current was not measured. The same plate material was used; the gap was controlled only for A. Two release records are missing. A draft poster says: ‘The improved holder never fails when powered, so visitors can trust its release.’
Question 1, Mathematics: Calculate the observed pass rates. Worked answer: A = 12/15 = 80%. B = 14/20 = 70%. Combined = 26/35 ≈ 74.3%. B has more passes but a lower rate. Question 2, Physics: Identify two variables that prevent attribution of the difference solely to the turn count. Worked answer: the actual coil current was not matched or measured for B, and the gap was not controlled the same way. Question 3, Physics: What role could the spring play after switching off? Worked answer: as the powered magnetic effect decreases, the spring can supply a restoring force to separate the plate if the mechanical arrangement permits.
Question 4, English: Identify two problems in the poster wording. Worked answer: ‘never fails’ contradicts documented holding failures, and ‘trust its release’ is not supported by a complete release log. Question 5, English: Rewrite a suitable public sentence. Worked answer: ‘Our classroom electromagnet model retained a plate in twelve of fifteen trials in one setup and fourteen of twenty in another. The results illustrate magnetic attraction, but the trial sets differed and some release observations are missing.’ This version preserves relevant facts without inventing safety guarantees.
Question 6, integrated judgement: Give a next step. Worked answer: continue supervised trials with matched, measured current, consistent gap, clearly defined holding and release criteria, and complete trial identifiers; publish only the validated educational observations. A real installed system requires qualified safety processes outside the scope of this classroom lesson. Question 7, challenge: Why would a 75% combined pass claim be wrong? Worked answer: it incorrectly averages unequal group percentages; the pooled observed rate is 26/35, approximately 74.3%.
Question 8, another challenge: Could a carefully worded notice still be engaging? Worked answer: yes. Explain what visitors will learn about current, coils and magnets, rather than promising flawless performance. Scientific restraint is compatible with clear, lively prose. A high-quality response takes a defensible position and supports it with the record, rather than repeating the same warning in several forms.
Frequently asked questions
Is EMS an examination paper?
No. EMS means English, Mathematics and Science as a study combination. The SEC assesses its listed subjects and paper structures separately. This integrated exercise helps learners transfer reasoning between them.
Why do incomplete logs matter?
Because a missing observation cannot be assumed to be success or failure. The definition of the outcome and the denominator must both be clear before a valid rate can be calculated.
What makes an explanation complete?
It identifies the condition, the relevant mechanism, the observation and the reasonable limit of the conclusion. A correct formula or attractive word is not enough without the logical link.
When should a student state uncertainty?
When the question lacks information necessary to support a stronger claim. Specific uncertainty—such as an unmeasured current—is more helpful than vague doubt.
How can I practise under time pressure?
Use short mixed prompts. Identify the subject skill, mark the evidence source, solve, check assumptions, and then try a changed version of the same problem.
Previous volumes, subject owners and official SEC sources
Read the sequence in order: 1117 Physics model, 1118 Mathematics inspection data, and 1119 English comprehension. For the prior electrical-graph bridge see Volume 1116. Subject owners: Secondary English, Secondary Mathematics, Secondary Science, and PSLE science evidence foundations.
Check the official SEC overview, 2027 G3 subject codes and syllabuses, English K300, Mathematics K310, and Physics K323. The subject pages were checked against the 2027 school-candidate list as updated by SEAB on 26 October 2026. Schools remain the source of individual candidate instructions and timetables.
