The point of the evidence board
A reported iron-recovery figure is not a description of everything that happened in a furnace. A chemical equation is not a full description of an industrial plant. A good summary may be accurate about what a writer said but not establish that the writer was right. This chapter brings the three questions into the same room while keeping their answers separate. The result is a disciplined way to judge a claim without pretending that a calculation proves more than it does.
Suppose a fictional student exhibition announces that a process recovered 780 kilograms of iron from 2,000 kilograms of ore. The number alone does not tell us the grade of the ore, the iron-bearing mineral, the theoretical output, the operational losses, the fuel use or the source of the measurement. A learner who notices these gaps has already improved the quality of the answer. A learner who can supply the right calculations and write a qualified conclusion has turned that noticing into examination performance.
Read the quick answer first. Identify what material entered the process; identify what reaction is proposed; identify what quantity has actually been measured; calculate using the correct denominator; distinguish reported fact from inference; then state the narrowest conclusion that the evidence permits. These six moves apply to many problems beyond iron ore. They are not six formulae to memorise. They are six checks that keep chemistry, mathematics and English from being confused.
All companies, operating records and conversations used here are fictional teaching examples, not observations from a real mine or school event. No exercise is an official SEAB question. Whenever a figure appears, it describes the stated example only. The purpose is to learn a transferable method for G3 examination tasks and for responsible reading outside examinations.
A route through earlier volumes
Start with Volume 0950: reduction chemistry if the roles of carbon, carbon monoxide, iron oxide and hydrogen are uncertain. Continue with Volume 0951: ore grade and recovery mathematics when the difficulty is choosing a mass basis or percentage denominator. Use Volume 0952: attribution, inference and summary if the difficulty is recognising who made a claim and how a writer qualifies it. This volume does not replace their subject explanations. It asks the learner to decide when each explanation is relevant, and when one kind of evidence cannot answer another kind of question.
The wider Sengkang G3 SEC English, Mathematics and Science guide gives the level framework. Learners still repairing basic reading and numerical confidence may use the PSLE scoring and transition route as a bridge, without assuming that PSLE-style exercises are SEC examination papers. For deeper topic ownership use the Science Hub, Mathematics Index and English Index. These are reading routes rather than a demand to read everything at once.
What the 2027 examinations actually assess
The Singapore–Cambridge Secondary Education Certificate begins in 2027. The official SEAB SEC overview explains that students take subjects at G1, G2 or G3 and that the overall examination standards remain aligned with the corresponding earlier qualifications. For 2027 school candidates, SEAB’s G3 syllabus index identifies G3 English Language as K300, Mathematics as K310 and Chemistry as K324. These codes are identifiers, not claims that all three subjects share a paper format.
Under G3 Chemistry K324, scientific reasoning involves using models, interpreting experimental evidence and explaining how observations support conclusions. Under G3 Mathematics K310, candidates need mathematical methods, reasons and contextual interpretation. Paper 1 and Paper 2 each carry 90 marks and 50% weight; both last 2 hours and 15 minutes. Under G3 English Language K300, students read non-narrative texts with care, distinguish ideas and provide passage-based answers. English Paper 2 carries 50 marks and 35% weight, with a duration of 1 hour and 50 minutes. This cross-subject workshop combines learning; it is not itself a separately assessed EMS paper.
A recurring mistake is to use one subject’s assessment requirement as if it were another’s. A balanced equation earns no English comprehension credit if the question asks what a speaker implied. A polished paragraph cannot repair a wrong conversion of tonnes into kilograms. And a correct mathematical percentage does not prove that the factory’s original measurement was dependable. Your first examination task is therefore not to compute. It is to read the command word and identify the discipline being assessed.
From oxide to iron: the chemical boundary
A common school model is reduction of iron(III) oxide by carbon monoxide: iron(III) oxide + carbon monoxide produces iron + carbon dioxide, balanced as Fe₂O₃ + 3CO → 2Fe + 3CO₂. In that reaction iron(III) oxide loses oxygen and is reduced. Carbon monoxide gains oxygen and is oxidised. In an alternative simplified hydrogen route, Fe₂O₃ + 3H₂ → 2Fe + 3H₂O. These equations allow a comparison of reaction products, not an unsupported claim about the life-cycle emissions, cost or feasibility of whole processes.
Keep the model honest. A real iron ore is not a pure flask of iron(III) oxide, and the mass fraction of a named mineral must be known before it is used in a calculation. Industrial ore, gangue, moisture, transport, heat requirements and downstream processing complicate any real comparison. If a question provides a simplified composition, use it and label it as an assumption. Never introduce extra facts that the problem did not give and then present them as measured data.
A practical student’s evidence may concern a colour change, a measured mass, a gas produced, or a change after heating. Such evidence requires a control and a clear measurement description. The question ‘what evidence indicates that reduction occurred?’ is different from ‘how many tonnes could theoretically be obtained?’ The first asks about a chemical change and how it was observed. The second calls for relative atomic masses and a mass calculation.
Build the mass chain before calculating
An invented ore batch weighs 2,000 kg and is stated to contain 65% Fe₂O₃ by mass. It therefore contains 1,300 kg of Fe₂O₃. For classroom arithmetic use Fe = 56 and O = 16. A formula unit of Fe₂O₃ has mass 2(56) + 3(16) = 160 relative units; the iron within it accounts for 112 units. The theoretical iron mass is therefore 1,300 × 112/160 = 910 kg. If 780 kg of iron is recovered, the stated recovery against this theoretical maximum is 780/910 × 100%, about 85.7%.
Now compare three tempting percentages. The ore contains 65% Fe₂O₃. The chemical iron content of Fe₂O₃ in this model is 70%. The actual recovered iron as a fraction of the total incoming ore is 780/2,000 × 100% = 39%. None of those is the recovery relative to theoretical iron, which is about 85.7%. Each number answers a different question. A student who writes only ‘efficiency was 39%’ has silently changed the denominator and probably the intended meaning.
If an answer asks for the highest possible iron output, do not calculate it using the measured recovery. If it asks for actual iron output, do not substitute the theoretical value without a supplied recovery assumption. If it asks whether the process is environmentally preferable, there is not enough information: an iron mass alone does not account for the source of hydrogen, source of energy, material handling, emissions or other impacts. A precise ‘cannot be determined’ can be a better scientific answer than a confident guess.
Read public statements as evidence, not as instructions
Consider a fictional newsletter statement: ‘The new method delivers more than eighty-five per cent of the iron in the ore.’ That may sound consistent with an 85.7% recovery calculation, but the denominator is misleading. In the classroom model, the process recovered 85.7% of the theoretical iron, not 85.7% of the total ore mass. Correct the sentence rather than merely replacing one percentage with another. The correction must name the basis on which the percentage was calculated.
Attribution matters too. ‘The operator said the method uses less fuel’ identifies a speaker; it does not independently verify fuel use. ‘A school calculation estimated a theoretical maximum’ describes a model; it is not an industrial audit. ‘A journalist observed the conveyor operating’ is an observation of movement; it does not establish the rate of iron recovery. Good writing keeps these voices and methods distinct. Accurate numbers deserve accurate nouns and verbs.
An informed reader can accept a limited statement and reject an inflated version of it. For example: ‘In this fictional exercise, 780 kg was recovered from a batch with 910 kg of theoretical elemental iron, giving about 85.7% recovery on that basis.’ This leaves room for more evidence rather than demanding it be treated as the final word on the process. The habit of specifying what a figure means is useful in experiments, reports, consumer claims and other forms of scientific communication.
A practical six-column evidence record
Make a small record with six columns: exact claim, source, measurement or model, quantity and units, limitation, and defensible revision. A claim about yield may have a verified mass calculation but an unverified sampling method. A claim about emissions may have a chemical equation but no energy inventory. A claim about cost may have a quoted price without including transport. Do not collapse such differences into a single verdict of true or false when the record supports only a more nuanced conclusion.
For example, the sentence ‘The reduction step produces water’ can be correct for the simplified hydrogen equation. It does not follow that the entire system has no emissions. The first part belongs in the chemical mechanism column; the second would require a life-cycle comparison. The sentence ‘The process recovered 780 kg’ is a reported measurement that needs a source, scale accuracy and product identification. The sentence ‘This is 85.7% of the theoretical maximum’ is a calculation that can be checked from stated inputs.
Use this register under time pressure. Read once for topic, then again to mark the exact quantity and its basis. Put a short annotation beside a question: chemistry, mathematics, English, or combined reasoning. Do the required work in that order. If an earlier stage is unsupported, do not let a later stage quietly inherit a certainty it has not earned.
Twenty-eight original evidence drills
For each fictional case, first write an answer in one or two sentences without looking below it. Then study why the answer is limited. The explanations make the exercises useful for independent revision rather than merely supplying an answer key.
01. Ore mass versus oxide mass
Situation. A crate record gives 1,500 kg of ore, with 60% Fe₂O₃ by mass.
Question. Find the oxide mass and say why it is not the same as iron mass.
Worked response. The oxide mass is 900 kg. The compound includes oxygen, so 900 kg is not 900 kg of elemental iron.
Why it matters. The word ‘grade’ refers here to the specified Fe₂O₃ fraction. It does not automatically state the elemental-iron fraction. A careless answer overstates the output before any recovery losses are considered.
02. Theoretical iron from oxide
Situation. The 900 kg Fe₂O₃ in the previous case is assumed to react fully.
Question. Calculate the theoretical iron mass with Fe = 56 and O = 16.
Worked response. The iron fraction is 112/160 = 0.70, so the maximum is 630 kg.
Why it matters. Show the formula-mass step. Simply multiplying by a remembered percentage hides why the fraction is appropriate and makes it harder to spot a wrongly identified ore mineral.
03. Reported recovery
Situation. A team reports 546 kg iron from the batch whose theoretical limit is 630 kg.
Question. Calculate recovery against theoretical iron and interpret the result.
Worked response. 546/630 × 100% = 86.7% to three significant figures.
Why it matters. The phrase ‘against theoretical iron’ belongs in the conclusion. The result is not the proportion of the original ore mass, and it does not include a judgment about cost or energy.
04. A different denominator
Situation. A poster calls 546 kg ‘36.4% iron recovery’ using the original 1,500 kg ore.
Question. Explain why the arithmetic might be correct but the label misleading.
Worked response. 546/1,500 × 100% = 36.4%, which is recovered iron as a fraction of incoming ore.
Why it matters. The number has a legitimate meaning. It is not, however, the same recovery percentage calculated relative to the theoretical elemental-iron content of the specified oxide.
05. Unknown moisture
Situation. Two batches have identical recorded wet mass, but one contains more water.
Question. Can their dry iron-bearing content be compared reliably from wet mass alone?
Worked response. No. More water leaves less dry material within a fixed wet mass, unless further composition data are supplied.
Why it matters. State what measurement is missing. The issue is not a sophisticated formula; it is a missing definition of the input to which a percentage is applied.
06. Carbon monoxide and oxygen
Situation. A diagram shows Fe₂O₃ reacting with CO and the product CO₂.
Question. Name what is reduced and what is oxidised.
Worked response. Fe₂O₃ is reduced by loss of oxygen; CO is oxidised by gain of oxygen.
Why it matters. Do not decide oxidation and reduction from the word ‘iron’ alone. Track oxygen transfer in the specified reaction rather than making a general statement about the atmosphere.
07. Hydrogen route
Situation. A class compares Fe₂O₃ + 3H₂ → 2Fe + 3H₂O with the CO reaction.
Question. Identify a valid chemical distinction and an invalid broad conclusion.
Worked response. The hydrogen reduction equation produces water instead of carbon dioxide in the stated reaction. Calling the entire industrial route carbon-free from that equation alone is unsupported.
Why it matters. The source and production of hydrogen, energy inputs and other stages matter to a whole-system assessment. Separate chemical products from a life-cycle assertion.
08. The mass of oxygen
Situation. A sample contains 160 g pure Fe₂O₃ and is fully reduced under the simplified model.
Question. What mass of elemental iron can form?
Worked response. There are 112 g of iron within 160 g Fe₂O₃, so 112 g of iron is the theoretical maximum.
Why it matters. The remaining mass is associated with oxygen transferred out of the oxide in this model. The statement does not reveal the recovered mass from a real incomplete process.
09. Comparing unequal batches
Situation. Trial A recovers 80 kg from 100 kg theoretical iron; Trial B recovers 144 kg from 180 kg theoretical.
Question. Which has higher recovery?
Worked response. Both have 80% recovery relative to their respective theoretical maxima.
Why it matters. Trial B produces more mass because it has a larger basis. Compare like with like: either recoveries or total product, but do not confuse one with the other.
10. Near-one-hundred claim
Situation. A report gives 99.8% recovery but only a single trial.
Question. What should be checked before generalising?
Worked response. Request the batch definition, weighing method, number of repetitions, uncertainty and whether the product measured is pure iron.
Why it matters. A very high percentage is not automatically wrong. Its evidential strength depends on the way it was obtained and whether the reported quantity matches the claim.
11. A percentage over one hundred
Situation. A student divides measured product mass by estimated theoretical iron and gets 106%.
Question. Should they advertise a breakthrough?
Worked response. No. Recheck the feed composition, moisture, product purity, arithmetic and measurements.
Why it matters. A percentage above 100 relative to a true theoretical maximum flags inconsistent inputs or a different product definition. It does not defeat conservation of mass.
12. The words ‘more efficient’
Situation. A supplier calls Process B ‘more efficient’ because it produced a larger batch.
Question. What is missing?
Worked response. The denominator: feed input, energy, time, operating cost or theoretical iron, depending on the intended efficiency.
Why it matters. Efficiency is a relationship, not a compliment. Define the relevant input and useful output, then determine whether the provided records support that relationship.
13. Attribution in a news item
Situation. A headline says, ‘Factory proves greener production’; the article says only that the factory’s manager made the claim.
Question. How should a reader report the information?
Worked response. Write that the manager claimed greener production, and note that no independent emissions comparison is provided.
Why it matters. The article establishes that a statement was made. It does not establish that the scientific or environmental conclusion is true.
14. An isolated graph
Situation. A chart begins its vertical axis at 95% and shows 96% beside 98%.
Question. Why might the visual impression mislead?
Worked response. A truncated axis can make a two-percentage-point difference look dramatic. Read axis labels, baselines and units.
Why it matters. The numbers may be accurate even when their visual framing exaggerates the size of a difference. In English explain effect; in Mathematics quantify it.
15. Percentage points
Situation. Two trials report recoveries of 82% and 89%.
Question. Describe the increase clearly.
Worked response. The difference is 7 percentage points; the relative increase is 7/82 × 100%, about 8.54%.
Why it matters. A percentage-point change and a percentage change are different quantities. The intended phrase in the question determines which calculation is required.
16. Missing product purity
Situation. A furnace product weighs 500 kg but is only 90% iron by mass.
Question. What iron mass can be attributed to the product?
Worked response. 500 × 0.90 = 450 kg elemental iron.
Why it matters. Using 500 kg as pure recovered iron would overstate the result. A product mass and a pure-element mass should never share an unqualified label.
17. Mass-balance language
Situation. A report says ‘all unrecovered ore was wasted’ after giving only recovered iron mass.
Question. Is that inference valid?
Worked response. No. Ore includes gangue, oxygen within compounds and potentially other materials; not all unaccounted mass is equivalent to lost pure iron.
Why it matters. A proper material balance needs identified streams and compositions. In a school exercise, state which outputs the simplified model includes.
18. A conditional comparison
Situation. A question states that both routes use equal external energy and pure reactants.
Question. May the student discuss products under those assumptions?
Worked response. Yes, but explicitly keep the comparison within the assumptions and equations supplied.
Why it matters. A classroom problem often removes real-world complications deliberately. Use those simplifications to answer the set task without presenting them as verified industrial conditions.
19. Small sample warning
Situation. A trial measures recovery in three carefully selected batches, while the operator processes hundreds each month.
Question. Can the result establish the monthly average?
Worked response. Not without information about how representative the trials are and how variable ordinary batches may be.
Why it matters. Selection matters. A sample designed to demonstrate a method is not automatically a random or representative sample of routine operations.
20. Two definitions of yield
Situation. One table gives recovered product as a percentage of ore and another as a percentage of theoretical iron.
Question. Should the values be averaged?
Worked response. No. First convert them to a common defined basis or keep them in separate columns.
Why it matters. Averaging unlike percentages gives a mathematically neat but conceptually empty number. Matching denominators is a prerequisite, not an optional refinement.
21. A summary with a missing caveat
Situation. A student writes, ‘The method recovered most of the iron and had no impact on the environment.’
Question. Which part is unsupported by extraction-mass data alone?
Worked response. The environmental conclusion is unsupported. Mass data may support recovery, but environmental effects require other measurements.
Why it matters. The correction is not to make every sentence tentative. It is to put certainty exactly where evidence allows it and to remove conclusions that outrun the record.
22. Time and throughput
Situation. Trial A recovers 600 kg in 3 hours; Trial B recovers 650 kg in 5 hours.
Question. Which has the larger average recovered-iron rate?
Worked response. A averages 200 kg/h; B averages 130 kg/h.
Why it matters. A larger total does not necessarily mean a faster process. If the question asks for throughput, divide by the relevant elapsed time and state whether downtime is included.
23. A source conflict
Situation. An internal log reports 540 kg while a press release reports 560 kg for the same batch.
Question. Can a writer silently choose the higher figure?
Worked response. No. Identify the discrepancy and seek whether the records describe different times, scales or purity adjustments.
Why it matters. An unresolved conflict belongs in the final account as a limit. Choosing the more attractive value without explanation creates false precision.
24. Rounding responsibly
Situation. A computed recovery is 546/630 ×100 = 86.666…%.
Question. How should it be presented when three significant figures are required?
Worked response. State 86.7%, retaining adequate precision during calculation.
Why it matters. Do not round every intermediate number before computing the final percentage. Premature rounding can cause a later comparison to cross a threshold incorrectly.
25. Words of causation
Situation. A survey reports higher output after a new process was introduced, but ore grade also increased.
Question. Does the timing prove the process caused all improvement?
Worked response. No. Ore grade is a possible confounding change.
Why it matters. A responsible claim would compare like-for-like inputs or explain the uncertainty. A chronological sequence is not a controlled experiment.
26. Two-statement reconciliation
Situation. A Chemistry page says hydrogen reduction produces water; a Math page gives 85% iron recovery.
Question. Do they together establish the route’s total water use?
Worked response. No. One is a reaction equation and the other is a recovered-iron fraction.
Why it matters. Even two accurate statements cannot be combined unless they measure compatible aspects of the same specified system. Ask what new data are needed before joining them.
27. Visitor question
Situation. A parent asks whether 85.7% recovery means 85.7% of ore was turned into iron.
Question. Give a short, accurate public reply.
Worked response. No. In the example it means 85.7% of the theoretical elemental iron was recovered, based on the assumed oxide content of the ore.
Why it matters. An excellent reply is calm, names the denominator, and avoids technical showmanship. A reader who understands the basis can judge the figure more fairly.
28. Next-question transfer
Situation. A new task concerns copper purification rather than iron ore.
Question. What remains useful from this lesson?
Worked response. Identify material identity, inputs, theoretical output, measured result, units, denominator, source and limits.
Why it matters. Do not carry the Fe₂O₃ mass factor into a copper problem. Transfer the reasoning procedure while rebuilding the subject-specific chemistry from its own equation.
A full exhibition-board rehearsal
Read this invented draft: ‘The school model recovers 780 kg of pure iron from two tonnes of iron-rich ore. This means the process is 85.7% efficient and produces no pollution. Visitors can therefore be confident that the hydrogen route is the best for every factory.’ The editor has been given a composition note stating that 65% of the ore is Fe₂O₃, a school calculation of 910 kg theoretical elemental iron, and a measurement note that the 780 kg was assumed to be pure iron.
First, repair the mathematics: the theoretical maximum is 2,000 × 0.65 × 112/160 = 910 kg. Recovered mass divided by theoretical maximum is approximately 85.7%. Recovered mass divided by incoming ore is 39%. Do not call either number a universal process efficiency. Second, repair the chemistry: a simplified hydrogen-reduction equation yields water in that reaction. It does not establish the environmental footprint of hydrogen production or plant operation. Third, repair the English: claims about the best method for every factory are not justified by one illustrative calculation.
A defensible replacement might say: ‘In this classroom model, two tonnes of ore containing 65% iron(III) oxide could theoretically yield 910 kg of iron. The stated recovered mass of 780 kg represents approximately 85.7% of that theoretical maximum. The model compares reaction products, but it does not assess total energy use, emissions, cost or the feasibility of an industrial process.’ Every sentence earns its place because it answers a likely reader question and can be traced to evidence.
Now change one fact: suppose the recovered 780 kg product is only 95% pure iron. The actual iron is 741 kg. The recovery relative to 910 kg becomes approximately 81.4%. Notice how a product-quality detail changes the numerical conclusion. This is why data tables, assumptions and ordinary nouns can matter as much as a complex equation.
How to practise deliberately rather than randomly
On the first practice day, take six mixed questions and classify each as mechanism, calculation, reading or integrated evidence. Do not time the first attempt. Explain every wrong classification aloud. On the second day, perform six calculations with units, but before each calculation write the quantity sought and the denominator of any percentage. On the third day, read three short claims and write a version that is accurate without becoming unnecessarily timid. On the fourth day, combine the moves into a single case file and present your conclusion in fewer than 120 words.
After marking, keep an error register with the original error, the reason it happened, a corrected example and a fresh test question. ‘Lost a mark for careless work’ is not a useful diagnosis. ‘Used total ore instead of theoretical iron as the denominator’ is. ‘Guessed that a company claim was independently verified’ is. Such entries point towards a specific repair. Work until the repair survives a new question, not merely until the old solution looks familiar.
Students who are secure can practise counterfactual changes: change product purity, sample size, ore composition or the quoted source, then predict which conclusions must change. Students still building foundations should use smaller integers and write the meaning of each mass in words. Both routes pursue the same eventual skill: to keep track of evidence while the surface story changes.
Questions readers frequently ask
Does the SEC have one combined English–Mathematics–Science examination? No. This is an interdisciplinary revision guide. The subject syllabuses and their paper structures remain distinct. Use the current official syllabus for each registered subject.
Is 85.7% a good industrial efficiency? This example does not provide enough real operational information to make that judgment. It is a calculated recovery under stated hypothetical assumptions.
Can I call Fe₂O₃ an ore? An ore is a naturally occurring material containing economically useful minerals; Fe₂O₃ is the chemical formula for iron(III) oxide. A simplified question can state that a particular mass fraction of ore is Fe₂O₃, but the two terms are not interchangeable.
Should I memorise the example numbers? No. Memorise the distinction between input material, theoretical elemental mass, measured output and the definition of a percentage. Recalculate with new numbers.
How should I respond if an exam passage contains a misleading claim? Answer the question asked. Explain the writer’s assertion as a reading task, then evaluate its support only if the question calls for evaluation. Reading accurately is not the same as endorsing the claim.
Next reading
You have now completed the local iron-ore sequence. The next chapter uses a new physical problem: how material is moved after processing and why lift height, travel distance, work and power are not the same quantity. Review Chemistry Volume 0950, Mathematics Volume 0951 and English Volume 0952 when a specific weak link returns. For current assessment documents, always check SEAB’s 2027 G3 school-candidate syllabuses rather than relying on copied examination advice.
The purpose of good preparation is not to appear certain at every moment. It is to notice what kind of problem you have, choose the right method, complete the necessary reasoning, and communicate an answer that remains true when the next detail is revealed.
