Small Group Tutorials

Here to help students catch up, keep up, and move ahead. Book a consultation here.

How to Perform in the new G3 SEC Examinations | Learner’s Guide Vol 1574 | EMS Orchard Magnetic-Response Decision Audit — Physics Limits, Review Capacity and Honest Public Claims

A strong G3 SEC learner can calculate an accurate percentage and still reject the sentence that uses it incorrectly. This integrated English, Mathematics and Science guide completes a fictional orchard magnetic-response project. It shows how to combine Physics observations, Mathematics capacity calculations and English reports without changing an observation into an unsupported public guarantee.

The 2027 Singapore-Cambridge SEC examines English Language K300, Mathematics K310 and Physics K323 as separate G3 subjects. EMS is this series’ teaching approach, not an official combined SEC examination. All committee choices, clips, counts and staffing models below are imaginary classroom practice and not a real metal-screening, procurement, food-safety or technical certification exercise.

  • Audit every quantity’s source, denominator, unit and scope.
  • Separate a recorded magnetic movement from detection, recovery and safety claims.
  • Reconcile 34/40 and 34/100 with a 220-object capacity model and a 230-object reserve requirement.
  • Approve a carefully qualified educational display while rejecting unsupported guarantees.
  • Practise twenty-five worked cross-subject decisions with corrected reasoning.

Vol 1573 English Paper 1 magnetic-trial report · Vol 1572 Mathematics magnetic-response audit · Vol 1571 Physics magnetic sorting · Vol 1570 EMS solar evidence decision · Mathematics Hub · Science Hub · Secondary English Hub · PSLE to Full SBB transition

What an integrated evidence decision is

A strong integrated response asks whether several correct statements actually describe the same claim. The school has a fictional magnetic-response classroom demonstration. Physics explains the interaction and the limits of the test. Mathematics calculates sample proportions and the capacity of an invented manual review station. English prepares a responsible notice for visitors. Each field does a different job. They become dangerous only when a writer substitutes a proportion from one sample for a safety rate in another. This guide teaches the discipline of checking relationships among evidence before making a public statement.

G3 SEC subjects remain separate

The official 2027 SEC G3 syllabuses identify English Language K300, Mathematics K310 and Physics K323 as separate examined subjects. English includes writing, comprehension, listening and oral components. Mathematics has two written papers assessing core content and problem solving. Physics has theory and a practical assessment. There is no examination called EMS that merges those papers. In this series, EMS means the deliberate practice of transferable English, Mathematics and Science reasoning. A learner must always consult the specific official document for actual subject assessment requirements. The integrated case is an original teaching device.

The four-volume route and its division of labour

Vol 1571 owns magnetic materials, field ideas, controlled variables and observation limits. Vol 1572 owns the two-way table, denominator choices, graph arithmetic and hypothetical capacity inequalities. Vol 1573 owns the tone and evidence selection of a situational email to a school coordinator. This volume owns a final combined decision. No chapter needs to repeat every explanation from the other three. The learner’s new responsibility is to reconcile the assumptions: a thirty-four count cannot become a universal efficiency; a station throughput cannot become an accuracy rate; a courteous email cannot turn either into a certification.

Define the fictional committee’s decision

Imagine a school committee considering a supervised science-room exhibition. It must choose between three outcomes: display the classroom data with careful labels, postpone the display until the charts are corrected, or reject a proposed headline claiming universal hazard removal. The committee is not approving a commercial screen or inspecting food. The issue is educational communication. The decision criteria are whether the recorded figures are represented accurately, whether visitors can understand the observation rule and whether the presentation avoids unsupported safety claims. Making the decision narrow is essential: vague objectives invite broad assertions that no supplied evidence can support.

Build an evidence register with three sources

The Physics record says that forty labelled steel clips were observed at each of three separations using one bar magnet, with movement recorded only within ten seconds. The Mathematics record calculates fractions and introduces a separate manual-review capacity model. The English record is a proposal email that recommends a table and a limitation notice. Write each source’s purpose beside its figures. If an eventual headline makes an assertion about safety, ask which source measured safety. In this exercise the answer is none. That missing source cannot be manufactured by adding together figures from the other registers.

Source one: the mixed-material trial

The fictional mixed tray contained forty steel clips, thirty aluminium squares and thirty plastic tags. At two centimetres, thirty-four steel clips visibly moved within ten seconds. No aluminium squares or plastic tags met the rule. The observation count is thirty-four of one hundred across the deliberately chosen mixed tray; among the steel clips it is thirty-four of forty. These percentages measure different groups. Both belong in a fully labelled school chart if they help the audience. Neither is a measured percentage of hazardous material detected or removed from a real facility.

Source two: the same steel clips at different distances

In the separate distance comparison, the forty steel clips were reset for tests at two, four and six centimetres. The invented responses were thirty-four, twenty-one and eight. The test does not supply multiple independent replicate runs or a calibrated measure of field strength. Its defensible trend is that fewer clips met the movement rule at greater separation in this particular arrangement. A straight descending line through the three counts is a representation of the invented dataset, not proof of a universal field law. Those limits matter when the committee prepares a caption.

Source three: a hypothetical review station

The Mathematics extension stipulated a capacity of fifty-five objects per station per hour under a constant-rate planning model. With two independent stations and two hours, the nominal capacity is 220 objects. A batch of two hundred fits with space for twenty additional objects. If the policy adds a separate thirty-object reserve, the requirement becomes 230, so the two-station plan fails by ten. Three stations would give nominal capacity of 330 under the same simplifying assumptions. These are hypothetical scheduling calculations; no station was built, timed or validated in the magnetic-response trial.

Why the three sources are not interchangeable

The first two records concern visible movement in a magnet experiment. The third concerns the number of objects a separately defined manual review workflow could process. It does not report how accurately objects are inspected. These quantities have different units, methods and scopes. One cannot combine a nominal capacity of 220 with an observed 85 per cent steel response and claim that 187 dangerous objects will be safely removed. That calculation would silently assume the object mix, trial probabilities, workflow relationships and detection properties. None are established.

Begin every audit with the nouns

Before recomputing a percentage, identify the nouns in the proposed statement: steel clips, all objects, reviewed objects, hazardous metals, food items or people. Those categories are not synonyms. Thirty-four labelled steel clips moved in the original setup. The experiment did not classify dangerous metals or measure food safety. Replacing ‘steel clips’ with ‘hazardous fragments’ changes the claim even if every number stays identical. This is why integrated reasoning needs careful reading as much as arithmetic. In the language chapter, a precise noun is not a cosmetic preference; it is part of the evidence.

The denominator must travel with the number

Eighty-five per cent arose from thirty-four divided by forty labelled steel clips. Thirty-four per cent arose from thirty-four divided by the deliberately mixed hundred objects. The committee should never print ’85 per cent of all items’ or ’85 per cent safe’. The first confuses samples; the second changes the measured outcome. The correct public caption can avoid percentages altogether and state the counts. When a percentage genuinely improves understanding, accompany it with the group and test condition. Even a mathematically valid fraction becomes misleading when the caption discards what was counted.

Response is not the same as recovery

The trial defined response as visible movement toward a magnet within ten seconds. It did not measure whether the object was collected, separated from anything else, inspected independently or disposed of. Those are separate operational outcomes. A school poster saying ‘thirty-four clips responded’ is supported by the fictional record. Saying ‘thirty-four clips were reliably removed’ would require a measurement that the source did not provide. This distinction is especially important because familiar verbs—find, catch, recover, protect—sound interchangeable in casual conversation. In technical claims, each verb describes a different event.

The value of one clearly stated limit

An effective visitor notice could say: ‘The results show visible movement of labelled classroom objects under specified conditions. They do not establish whether a real screening system would reliably find or remove hazards.’ The notice does not undermine the lesson. It tells visitors what to learn. The stated limitation is more informative than a page of vague legal-sounding cautions. A student can apply the same approach to other contexts: a thermometer reading is not proof of food quality; a photograph is not proof of causation; a convenient sample is not an entire population.

When a zero cannot support a universal claim

The mixed tray showed zero movement among thirty aluminium squares and thirty plastic tags. The committee may state that observation with the sample count. It should not publish that aluminium has no possible magnetic interaction under any conditions or that plastic items could never move in any apparatus. Physical details, hidden components and the observation rule vary. The stronger scientific insight is that an ordinary bar magnet interacts differently with materials, not that the classroom made a universal safety classification. A zero in a table is a valid finding within the table’s domain, not a promise about the wider world.

The temptation to call a trend a law

At distances of two, four and six centimetres, the sample count fell by thirteen clips for each two-centimetre step. A Mathematics student can calculate the equal segment gradients correctly. A Physics student should still refuse to claim that magnetic force falls by a constant number of clips per centimetre. Clip count is an observation subject to a movement threshold, not a direct force measurement. A responsible committee might connect the three points visually, but its caption should say the graph shows observations at the three marked positions, not a predictive relationship valid everywhere.

Two-pass arithmetic without an independence assumption

One fictional extension retests the six previously nonresponsive steel clips after changing the setup. If three of these six now respond, thirty-seven unique clips have moved at least once: 34 + 3 = 37, or 92.5 per cent of the original forty. This is a cumulative count in a new two-stage procedure. It is not evidence that two identical passes independently succeed with some universal probability. Because the second group is selected after failing the first, its behaviour may be systematically different. The committee must not convert this optional teaching example into an operational guarantee.

Where an independent model belongs

A second optional mathematical exercise stipulates independent miss events with probability 0.15 on each pass. Multiplying 0.15 by 0.15 gives a theoretical double-miss probability of 0.0225, or 2.25 per cent. That is a correct calculation under the stipulated assumptions. The magnet trial did not establish independence, representative sampling or a true 0.15 miss probability in a real process. The committee should leave this theoretical figure out of its exhibition claims unless the display is explicitly about how modelling assumptions change numerical predictions. Mathematics can explore hypothetical models without pretending that they have been experimentally validated.

Capacity is not accuracy

A station processing fifty-five objects per hour is described by throughput. The station could be fast and unreliable, or slow and careful; the model does not tell us. The magnet experiment measured a threshold response. Neither variable establishes classification accuracy of the manual review station. An integrated answer should therefore keep two questions apart: ‘Can the scheduled stations process the specified number of objects within two hours?’ and ‘What performance has been validated for the review process?’ The first can be answered conditionally from given arithmetic. The second cannot be answered from the supplied fictional data.

Compare two hundred with two hundred and thirty

The ordinary requirement is two hundred objects. At two stations, fifty-five per hour each, for two hours, capacity is 220. In that scenario the arithmetic passes with a twenty-object margin. Under a separately stipulated requirement of two hundred and thirty including reserve, the same plan fails by ten. There is no contradiction because the requirements differ. A careless report might cite the favourable two-hundred case while claiming it satisfies the two-hundred-and-thirty policy. The correct response names the scenario alongside its verdict. Context should not vanish as soon as the inequality is solved.

What changing to three stations actually proves

If three identical stations can each handle fifty-five distinct objects per hour, two hours gives 330 nominal slots. Under the hypothetical model, this meets the 230-object requirement. It does not prove that staffing is available, that the equipment performs as assumed, that every item receives a meaningful review or that the wider project is safe. In an examination, state the mathematical answer and its conditions. The most common integrated overclaim is to turn ‘feasible under stated throughput assumptions’ into ‘validated in operation’. Real-world decision-making demands a second kind of evidence that the question did not supply.

Check the time window

The magnet movement rule is ten seconds per clip observation in one classroom procedure. The manual review throughput is fifty-five objects per hour in a separate hypothetical workflow. These figures cannot automatically be combined to calculate a total inspection time. We do not know whether the clip tests are sequential, whether loading and resetting take extra time, whether stations perform magnetic observations or a different review, or how throughput was derived. Unit conversion is not the missing step. The missing step is a defined process relationship. A good solver asks for that information rather than inventing a workflow.

Build the decision matrix

Make a text-only matrix with three proposed public statements. A: ‘In the classroom trial, thirty-four of forty steel clips moved at two centimetres within ten seconds’—supported by the invented observational source. B: ‘The two-station model can review two hundred items in two hours’—conditionally supported by stipulated capacity, with assumptions stated. C: ‘The magnet removes all dangerous metal and guarantees visitor safety’—unsupported and must be rejected. The matrix separates observational support, conditional modelling and unjustified promises. An examination answer earns clarity by identifying why a statement belongs in its category.

A communication release gate

Before a caption goes on display, test it against five questions: Is every noun faithful to the original sample? Does every percentage have the right denominator? Are distance, duration and test conditions named where needed? Are model assumptions made explicit? Does the conclusion avoid certification, safety and general reliability claims not tested? The committee can approve a clearly scoped educational caption and reject a sweeping guarantee in the same meeting. This is a better outcome than either uncritical enthusiasm or refusing to communicate any experimental result.

A model visitor caption

A suitable caption reads: ‘In our classroom demonstration, thirty-four of forty labelled steel clips moved toward a bar magnet within ten seconds at two centimetres. The number meeting that rule was twenty-one at four centimetres and eight at six centimetres. We used one specified setup. The demonstration shows how observed magnetic response can differ with distance and material; it does not test real-world hazard removal.’ This text is precise without being complicated. It allows visitors to see the pattern and the uncertainty. It does not claim the result has been repeated independently or generalised.

A model committee conclusion

The committee could recommend displaying the fictional school experiment as a supervised educational illustration, provided the results chart names the samples and conditions and the caption explicitly avoids real safety claims. It should not endorse a proposed headline about total hazard removal. The hypothetical manual review timetable may be shown as a separate Mathematics exercise, labelled as a model rather than an actual operational plan. That decision uses all three subjects without stretching any one beyond its evidence. It is an evidence decision, not an equipment validation or food-handling recommendation.

The difference between useful caution and excessive hedging

An answer that says ‘perhaps something might have happened’ is not careful; it is unclear. The observed counts in this invented scenario are definite because the task stipulates them. The appropriate caution belongs to conclusions beyond the data. We can state thirty-four of forty at two centimetres confidently, but should not say that eighty-five per cent of all hazards will be found in the future. This balance is a mark of good scientific English. It keeps the reader informed while showing precisely which limits matter.

Diagnose hidden assumptions in a single sentence

Consider: ‘With two stations and an eighty-five per cent detection rate, the school will be protected.’ The clause ‘two stations’ comes from the manual throughput model. ‘Eighty-five per cent’ comes from the steel-only movement count. ‘Detection rate’ changes the recorded variable. ‘Protected’ adds an unsupported safety outcome. The sentence hides at least three untested links. The best correction is not to adjust the arithmetic. It is to separate the unrelated claims, restore the correct nouns and remove the unsupported conclusion. Many advanced examination errors begin as a single plausible sentence.

How a learner should answer a multi-source question

First read the final decision required. Next label the sources P, M and E for Physics, Mathematics and English. Beside every figure, write its unit, denominator, condition and source. Then identify which statements are observations and which are hypothetical models. Only after that should you calculate or draft the final recommendation. This ordering prevents attractive figures from controlling the argument. In unfamiliar source-based questions, disciplined annotation can save more time than trying to hold every datum in working memory.

Why a graph, table and email are different evidence tools

A graph helps the reader see the ordered relation among distance settings and count. A two-way table keeps categories and denominators from being confused. An email identifies a recipient, purpose and proposed action. The choice of format depends on the communication question. Converting a table into prose requires preserving the group definitions; converting a graph into a claim requires preserving the measured axes and limits. Students often lose accuracy in the translation between formats. This is why an integrated audit checks both numbers and language.

Avoid the false certainty of many decimal places

The invented count-based probabilities can be calculated exactly as fractions of the sample. Writing the proportion of steel among the sixty-six nonresponsive objects as 0.0909090909 does not increase the study’s scientific credibility. It merely displays more digits. The Physics test still has limited repeated evidence and the classroom sample is deliberately chosen. In Mathematics, show enough precision to answer the numerical question; in a public caption, prefer a clear fraction or rounded percentage when appropriate. Precision in calculation is not the same as confidence in inference.

The role of repeated independent observations

The original distance figures give one result per condition after resetting the same forty clips. They do not supply the variation we might expect over separately repeated runs. A better future classroom study could predefine placement and timing, repeat the experiment, record counts across runs and discuss the spread. That would improve evaluation of repeatability. It still would not automatically validate a public safety claim or a commercial process. To change the application, a study needs an appropriate purpose, sample frame, operating conditions and independent validation. The lesson is to match evidence design to the strength of the desired conclusion.

Distinguish cause from coincidence

The classroom pattern is consistent with separation influencing the response under the specified arrangement. It is not enough, on its own, to establish an exact cause for every individual clip that moved or failed to move. Differences in friction or orientation could contribute to observed thresholds. The Physics chapter suggested controlling such factors. The Mathematics chapter described the trend without declaring a universal law. The English chapter used appropriately limited wording. The combined response is stronger because each discipline prevents a tempting inference from running beyond its foundation.

A parent’s route from PSLE into SEC evidence questions

At primary level, a child may describe what changed and what stayed the same in a simple experiment. G3 requires more careful handling of graphs, proportions, variables, scientific models and source-based interpretation. A parent can support progress by asking four recurring questions: What exactly was measured? Which number answers the question? What has the writer assumed? What conclusion follows without exaggeration? These questions work across English, Mathematics and Science. The goal is not to memorise an orchard story but to rebuild an independent method for unfamiliar examination contexts.

A staged revision routine

Begin with five minutes retrieving the key definitions without notes. Rebuild the hundred-object table from scratch. Explain the difference between thirty-four of forty and thirty-four of one hundred. Draw a clearly labelled distance graph. Solve the two-station capacity inequality, then change the reserve requirement and recalculate. Finally draft a three-sentence visitor caption and identify one unsupported headline. Repeat after feedback using a different fictional context. This deliberate sequence moves from recall to calculation, interpretation, communication and evaluation. Each stage becomes a prerequisite for the next.

What mastery looks like in a timed paper

A strong Physics response explains why the material and method affect the observed movement and can evaluate a controlled test. A strong Mathematics response produces the right fraction, graph and feasible integer decision with explicit assumptions. A strong English response selects the evidence needed for a particular audience and writes a truthful recommendation. An integrated practice session checks whether the three outputs agree on the facts. No single subject mark scheme replaces the others. The deeper skill is being able to say what is known, what is assumed and what further evidence would be needed.

The end of this learning sequence

This four-volume magnetic-response sequence began with material properties and a fair test, moved through numerical interpretation, then practised an audience-aware written report. It ends with a release decision that refuses to turn school data into unwarranted safety claims. The next topic in the wider learner series should introduce a genuinely new scientific principle or assessment skill, not merely rename this one. For now, the learner should be able to explain the whole chain in their own words, identify one weak link and repair it using evidence.

Twenty-five worked cross-subject decisions

These original practice questions link the experiment’s scientific meaning to the mathematics of samples and the language of careful public reporting. The goal is to repair assumptions, not merely produce a numerical answer.

Practice 1: The supported quantity

Question: What was directly observed at two centimetres? Thirty-four of forty labelled steel clips visibly moved toward one bar magnet within ten seconds. This describes an operational classroom observation. It does not establish a real-world hazard-detection rate or complete physical removal.

Practice 2: The mixed-tray count

Question: What percentage of all one hundred labelled objects met the rule? Thirty-four per cent. The eighty-five per cent figure refers only to the forty steel clips. A good conclusion names the denominator in words rather than printing a percentage without its group.

Practice 3: Different meanings of forty

Question: Why is the forty-clip test different from the hundred-object classification? The forty-clip distance experiment used one material category at several separations. The mixed tray compared material groups at a fixed separation. The research questions and denominators differ even when one count happens to match.

Practice 4: A false metal claim

Question: Why reject ‘every metal is detected’? The classroom magnet did not elicit visible movement among the thirty aluminium squares, and the experiment was not a validated detection system. The correct educational statement is that materials can respond differently to a magnetic field.

Practice 5: A false safety claim

Question: Does a school science-room demonstration establish food safety? No. No hazards, representative sampling or safety standards were tested. State that this is a classroom illustration and separate any actual professional inspection from the exercise. Do not call a count of moved clips a certificate.

Practice 6: Distinguish test outcomes

Question: Can ‘moved’ be replaced with ‘removed’? No. The response criterion recorded visible movement toward the magnet within ten seconds. Collection, separation and verified removal would require additional measurements. Accurate verbs are part of the evidence.

Practice 7: Count nonresponses

Question: How many of the forty steel clips did not respond at two centimetres? Six. The correct sentence is ‘Six did not meet the movement criterion.’ It is not valid to infer that six must be nonmagnetic because friction and placement can affect visible motion.

Practice 8: Read the distance record

Question: What happened at four and six centimetres? Twenty-one and eight of forty steel clips responded. These counts show the observed decline across the three measured distances. They do not directly measure field strength or justify extrapolating a universal detection law.

Practice 9: Two graph axes

Question: Which axes are honest? Separation in centimetres on the horizontal axis and number meeting the response rule out of forty on the vertical axis. Labelling the y-axis as magnetic force would substitute an unmeasured quantity for the recorded count.

Practice 10: A straight segment

Question: Does an equal observed count decrease per two centimetres prove a linear physical response? No. It provides a segment gradient of minus 6.5 clips per centimetre in the invented data. Thresholds, materials, geometry and unmeasured distances prevent a general law from being concluded.

Practice 11: Theoretical probability

Question: Under a newly stipulated independent two-pass model, why is 0.15 squared only a hypothetical result? The original repeated-clip study did not demonstrate independence or a true miss probability of 0.15. Mathematics can calculate consequences of assumptions without treating them as empirical validation.

Practice 12: A revised cumulative count

Question: If three of the six first-pass nonresponsive clips move in a modified second pass, how many unique original clips have moved? Thirty-seven of forty. This is cumulative within the two-stage invented example, not a general expected recovery rate for real devices.

Practice 13: Review throughput

Question: Two stations at fifty-five objects per hour for two hours yield what nominal capacity? 220 objects, under the explicit constant-rate, independent-work assumption. The result is a scheduling calculation, not evidence of detection accuracy.

Practice 14: Review with reserve

Question: Why do two stations fail a 230-object requirement? Their nominal capacity is 220, ten short. The result differs from the original 200-object case because the required count changes. Both calculations can be correct under their respective assumptions.

Practice 15: Three station model

Question: How much do three idealised stations cover in two hours? Three times fifty-five times two equals 330 objects. This satisfies 230 in the model. It does not verify actual staffing, workflow, accuracy or safety.

Practice 16: Check time units

Question: Can the magnet’s ten-second observation criterion be substituted directly for the manual station’s per-hour throughput? No. The former is an observation interval, while the latter is a separate workflow model. A relationship between them was not established.

Practice 17: Choose the true caption

Question: Which caption is defensible? ‘In our classroom test, thirty-four of forty labelled steel clips moved at two centimetres within ten seconds.’ It states sample, condition and outcome. The claim that every hazard was detected is unsupported.

Practice 18: Choose the true recommendation

Question: What should the committee approve? A supervised educational display with accurate tables and an explicit scope notice, if the materials are appropriate. It should reject sweeping safety guarantees that no source measured. Approval is for a lesson, not an inspection service.

Practice 19: Identify the hidden noun

Question: Why is ‘steel clips’ not interchangeable with ‘dangerous metal fragments’? The samples were labelled classroom objects, not measured hazards. Changing the noun invents a population and purpose. Keep the source’s categories when communicating results.

Practice 20: Define a conditional result

Question: In which scenario does 220 pass? When the requirement is 200 objects and the two stations work as stipulated for two hours. It fails when the requirement rises to 230. Always name the requirement rather than giving an unsupported yes-or-no.

Practice 21: Find a weak inference

Question: ‘The poster uses correct percentages, so its safety guarantee is true.’ Why is this false? Correct fractions establish numerical relationships in a defined sample. They cannot supply missing measurements or validation for a different safety outcome.

Practice 22: Repair the report

Question: Rewrite ‘Our magnet catches 85 per cent of metals.’ A defensible revision is ‘Thirty-four of forty labelled steel clips met the visible movement rule at two centimetres in the classroom setup.’ The revision repairs the numerator’s meaning, denominator and scope.

Practice 23: Separate fact and proposal

Question: Which sentence is an observation, and which is a recommendation? ‘Thirty-four clips moved’ is a stipulated observation. ‘We should post the result with a limitation note’ is a proposed action. A strong English answer keeps the two functions distinct.

Practice 24: Explain what is missing

Question: What would justify a reliable real-world screening claim? Evidence designed and validated specifically for that claim, including representative test materials, operating conditions, accuracy measures and appropriate professional standards. Nothing in the fictional classroom dataset provides that validation.

Practice 25: An end-to-end conclusion

Question: State the integrated verdict in two sentences. The classroom magnet demonstration can illustrate material differences and a distance-related change in visible response, with the counts and conditions accurately displayed. It cannot support a general hazard-removal or safety guarantee, and the separate throughput model must be presented only as a conditional Mathematics exercise.

Official syllabuses and study route

The official 2027 Singapore-Cambridge SEC G3 subject list provides the current codes. Refer to Physics K323, Mathematics K310 and English Language K300 for their own examination formats and objectives. For pupils and parents, follow the linked Sengkang learning hubs and PSLE-to-secondary pathway. The magnetic-response project is not a substitute for examination syllabuses or professional safety verification.