A greenhouse class has learned that an indicator colour does not identify every substance in water. It now examines a different fictional instrument: a classroom sensor whose numerical signal changes with the concentration of a known harmless model solute. The student must use a straight-line graph, interpret a gradient and intercept, estimate an unknown within the calibrated range and recognise the limits of the model. This guide turns one practical story into precise G3 Mathematics skills instead of pretending that a neat equation is a guarantee about real samples.
The first Singapore–Cambridge Secondary Education Certificate (SEC) examinations take place in 2027. The official G3 Mathematics syllabus is K310, corresponding to the earlier 4052. There are two compulsory papers, each 90 marks and 50% of the subject, each lasting two hours 15 minutes. Paper 2 includes a final extended problem involving application to a real-world context. The data, device, signal units and classroom decisions below are invented; this is an independent study guide, not a past paper or a certified measuring method.
Quick route: Label quantities and units; recognise a straight-line relationship; calculate a gradient and intercept; construct an equation; interpolate carefully; reject unjustified extrapolation; interpret repeated readings and error intervals; explain answers in context; finally solve a longer integrated problem. Vol 1387 owns acid-base chemistry and indicator tests. This Mathematics volume uses a separate fictional linear signal, not a linear conversion of pH. The previous Vol 1384 remains the owner of cold-chain rate, timetable and feasibility calculations.
1. Understand what the quantity represents
A numerical signal is not automatically a concentration. The fictional sensor produces a reading R in arbitrary signal units when exposed to a reference solution of stated solute concentration C in milligrams per litre. A calibration relates these two distinct quantities. We choose concentration for the horizontal axis and sensor reading for the vertical axis. If the axes are switched, the gradient has a different meaning even though the points lie on the same straight line.
Write the units beside each variable before calculating: C measured in mg/L, R measured in signal units. Then translate any answer into words. A solution might have a sensor reading of 22 units, but it would be false to report “22 mg/L”. The whole exercise is designed to prevent that kind of category error. In an examination, unit discipline often saves marks before any difficult algebra begins.
2. Establish a reference table
The fictional calibration points are (C,R) = (0,5), (2,15), (4,25) and (6,35). Their first coordinates are concentrations in mg/L; the second coordinates are instrument signal units. This is a deliberately exact teaching model. Real measurements often have scatter, and a suitable calibration procedure must be validated rather than assuming four perfect readings. Here the values are supplied only to practise proportional reasoning and straight-line graph methods.
Ask what changes between successive rows. Concentration rises by 2 mg/L, and the signal rises by 10 units each time. The constant ratio of changes suggests a linear relationship over the stated range. The data do not establish the relationship at negative concentration or beyond 6 mg/L. A reliable mathematical answer states the pattern in the given interval instead of quietly extending it to every imaginable value.
3. Calculate gradient from two points
For a straight line, the gradient is change in vertical quantity divided by change in horizontal quantity. Using (2,15) and (6,35), m = (35 − 15) ÷ (6 − 2) = 20 ÷ 4 = 5. The units are signal units per mg/L, because the numerator and denominator represent different measurements. Saying “the gradient is five” is numerically correct but omits its physical interpretation.
The meaning is that the model signal increases by five instrument units for each additional 1 mg/L of the model solute within the calibration range. An exam response should include a suitable two-point calculation and a sentence. If the gradient were negative, its meaning would be a decline in signal as concentration rises. The sign is not a decorative feature; it describes the direction of the relationship.
4. Find the vertical intercept
The table shows that at concentration zero the sensor gives a signal of five units. Therefore the vertical intercept is 5. With gradient 5, the model can be written R = 5C + 5. Students sometimes assume an instrument must read zero when a reference concentration is zero, but that is not justified by the supplied data. The fixed offset is precisely what the intercept records.
An intercept has an operational meaning only if the relevant x-value and units make sense in the model. Here zero concentration is an explicit reference point. In an unrelated graph, extrapolating the intercept to an unobserved condition may not be legitimate. A strong learner interprets an intercept in the given context and asks whether the reference point was actually measured. The algebra is short; the thinking makes it useful.
5. Verify the equation with another point
Do not stop when a proposed line passes through two points. Substitute C = 4 into R = 5C + 5 to obtain R = 25. This matches the third reference pair. C = 6 gives R = 35, matching the fourth. Checking additional supplied points is an easy way to detect arithmetic or transcription mistakes before using the equation on an unknown sample.
It is possible for two points to determine a line that fails to represent the remaining data well. In this ideal example, all four are consistent, but later we will examine what a small discrepancy means. Exam questions reward the habit of verifying a derived relation rather than assuming that a tidy formula must be right. This habit also prepares students for more complex real-world contexts in Paper 2.
6. Solve for an unknown using substitution
Suppose an unknown model sample gives R = 22 signal units. Substitute into the calibrated equation: 22 = 5C + 5. Subtract 5 from both sides to obtain 17 = 5C, then divide by 5 to find C = 3.4 mg/L. The concentration estimate is within the measured range of zero to six, so this is interpolation rather than extrapolation.
Write the answer as a conditional estimate: “According to the stipulated linear calibration, the sample corresponds to 3.4 mg/L.” This is more accurate than calling it a laboratory-certified concentration. The maths derives a result from a model; science must still determine whether the instrument and sample support the model. A correct equation can be used on unsuitable data and lead to an unwarranted practical conclusion.
7. Recognise interpolation
Interpolation is estimation between values covered by the observed calibration interval. Our unknown signal of 22 sits between reference readings of 15 and 25. Its estimated concentration of 3.4 mg/L lies between 2 and 4 mg/L. This result can be checked without redoing the entire algebra: 22 is seven signal units above 15, which is seven-tenths of the ten-unit interval. Seven-tenths of the corresponding 2 mg/L interval is 1.4 mg/L, so the estimate is 3.4.
Using both an equation and proportional reasoning provides a reasonableness check. A student who obtains 34 mg/L should notice that the answer lies far outside the interval bracketed by the reference readings. The exam skill is not just calculation; it is noticing whether a result fits the graphical and physical constraints of the stated model.
8. Reject unsupported extrapolation
Suppose the sensor now reports R = 41. The equation would give C = (41 − 5) ÷ 5 = 7.2 mg/L. The arithmetic is valid, but the reference measurements stopped at 6 mg/L. The true device might not remain linear beyond this interval. Therefore the result is an extrapolated model output, not a calibrated reading supported by the data.
In a real-world style mathematics question, distinguish “what the model predicts” from “what can be trusted”. If asked to compute, show 7.2 mg/L and label it conditional. If asked whether the result is reliable, say that it exceeds the tested range and needs additional reference measurements or other validation. The strongest answer responds to both the arithmetic and the meaning of the question.
9. Explain why pH is not this sensor
The prior Chemistry guide discussed Universal Indicator and pH. Here the instrument measures a deliberately chosen numerical response to a reference solute concentration, not pH. The model R = 5C + 5 is stipulated for this educational example; it must never be used to transform pH readings into concentration. The pH scale does not behave as an ordinary linear measure of quantity.
Cross-subject learning becomes dangerous when students transfer the shape of one calculation onto an unrelated scientific property. Mathematics supplies modelling tools; it does not prove that a particular model fits every substance. In exam work, pay attention to the meaning of the variable given by the problem. If a question describes temperature, voltage or concentration instead, check the relationship it actually specifies before reaching for a familiar formula.
10. Choose scales and label axes
A good graph uses horizontal and vertical scales that represent the provided range clearly. For these reference values, the horizontal axis could run from 0 to 6 mg/L, while the vertical axis could run from 0 to 40 signal units. Mark axes with quantities and units. Plot (0,5), (2,15), (4,25) and (6,35) accurately, then draw the line supported by the data.
If a graph has a vertical axis labelled only “reading”, the reader cannot distinguish sensor signal from concentration. If the origin is omitted, a change in scale must be made transparent; otherwise the diagram could exaggerate an apparent trend. Graph construction is a communication skill. Make it easy for someone else to reproduce the interpolation from your diagram without relying on the accompanying prose.
11. Avoid misleading graphical presentations
Imagine a presentation graph that begins its vertical axis at 20 instead of zero and shows readings of 21, 22 and 23 as very tall separate bars. A reader may wrongly think the differences are enormous. A truncated axis can be legitimate for some purposes, but its use should be explicit and proportionate. In this study context, compare the absolute values and the scale before accepting a dramatic visual impression.
G3 Mathematics includes identifying why a statistical diagram may mislead. It is not enough to say “the graph looks wrong”. Name the feature: omitted baseline, unequal intervals, inconsistent scaling, missing units or an unsuitable display type. Explain how that feature changes the reader’s interpretation. The problem may contain no algebra at all; it can still demand mathematical judgement.
12. Use an inverse relationship correctly
From R = 5C + 5, solve for C to obtain C = (R − 5) ÷ 5. This form makes it easier to convert many sensor readings into concentration estimates. Check the units of both sides: the subtraction takes place between signal readings, then division by five signal units per mg/L returns mg/L. A bare numerical rearrangement is less explanatory than one that tracks the quantities.
For R = 30, C = 25 ÷ 5 = 5 mg/L. For R = 10, C = 5 ÷ 5 = 1 mg/L. Both fall within the calibrated interval. For R = 2, the equation gives a negative concentration, which is not meaningful for this model. That is a warning to investigate a reading, an offset or the model’s validity, not a licence to claim that the sample has a physical concentration of −0.6 mg/L.
13. Distinguish mathematical error and measurement uncertainty
A student may calculate C = 3.4 mg/L correctly while the sensor reading itself remains uncertain. Algebraic correctness means the equation was solved correctly under the supplied assumptions. Measurement uncertainty means the input may represent a range rather than one exact value. These are different questions. A rounded display that says 22 does not necessarily mean the underlying signal was exactly 22.000000.
If the problem explicitly gives an uncertainty interval, keep it with the estimate. If no interval is supplied, do not invent a scientific uncertainty percentage and call it measured. You may still state that unmodelled measurement error could affect reliability. A mathematics exam often distinguishes exact values, rounded values and stated tolerances. Learning to articulate those distinctions improves both calculation and interpretation.
14. Propagate a stipulated signal interval
Assume the fictional instrument reports R = 22 ± 1 signal unit under the stated method. Then R may lie from 21 to 23. Using C = (R − 5) ÷ 5, the lower estimate is (21 − 5) ÷ 5 = 3.2 mg/L and the upper estimate is (23 − 5) ÷ 5 = 3.6 mg/L. Thus the model yields an interval from 3.2 to 3.6 mg/L.
We have used the smallest and largest allowed inputs because the calibration relationship is increasing. This reasoning is more reliable than reporting an unsupported exact answer. The interval also reveals the influence of the gradient: a one-unit signal change corresponds to 0.2 mg/L in this model. A flat or steep line would produce a different conversion. This is a useful example of connecting graph interpretation to numerical error bounds.
15. Do not confuse an interval with a probability
The interval 3.2 to 3.6 mg/L follows from an assumed input bound. It does not automatically mean there is a 95% probability that the true concentration lies inside it. A probability statement requires a probability model or statistical basis that has not been provided. A bound and a confidence interval are different objects, even if both are written with two endpoints.
In G3 Mathematics, the discipline is to use the information the question gives. If a display uncertainty is explicitly ±1, applying an interval calculation is justified. If the question supplies a sampling distribution or probability information, those facts may support a different task. Do not add scientific-sounding confidence claims to an otherwise simple algebra answer. Clear reasoning often means refusing to claim more than the assumptions allow.
16. Take repeated readings properly
Consider three independent readings from the same fictional sample: R = 21, 22 and 23. Their arithmetic mean is (21 + 22 + 23) ÷ 3 = 22, with a range of 2 signal units. Convert each value through the calibration: 3.2, 3.4 and 3.6 mg/L. The mean of those estimates is 3.4 mg/L. Because our calibration relation is linear, the converted mean equals the conversion of the mean signal.
That equality depends on the stipulated linear model. Do not assume it holds for every nonlinear transformation or every chemical quantity. The three readings tell us about observed variation, not whether the sensor is correctly calibrated against an independent reference. When answering an exam question, report the mean, an appropriate spread statistic and the qualification that the result relies on the supplied calibration.
17. Know when the median helps
Suppose the recorded signals were 21, 22, 40. The mean becomes 27.67 to two decimal places, while the median is 22. The large third reading pulls the mean upwards; the median remains a better description of the centre of these three observations if the extreme value does not represent normal behaviour. The result should trigger checking, not silent deletion.
An exam question may ask which measure better represents the typical reading. Explain why, referring to the extreme value and the purpose of the summary. Do not automatically prefer median in every situation. If the readings come from a genuinely skewed population, the mean and median answer related but different questions. If the high value was a transcription mistake, the correct response is to verify the record. Context determines the sensible statistic.
18. Interpret spread rather than just central tendency
Two instruments could have the same mean response of 22 but very different repeatability. Device A might give 21, 22, 23; Device B might give 12, 22, 32. Both means are 22, but Device B varies much more widely. Its range is 20 rather than 2. An answer that reports only the equal means conceals an important difference in reliability of repeat readings.
G3 Mathematics includes range, interquartile range and standard deviation as measures of spread. Choose the measure requested by the exam, and interpret it in the stated units. If the question asks about a fair instrument comparison, consider whether both devices were tested under the same conditions and with the same number of observations. Variability is meaningful only when the comparisons themselves make sense.
19. Compare data on a common basis
If one group takes five readings and another takes twenty, a raw total of values is not a useful instrument comparison. A mean may help, but unequal samples and different procedures still matter. Comparing instruments requires matching reference solutions, temperature conditions and measurement protocol as far as the question states. Mathematical fairness begins with understanding how the numbers were produced.
In a Paper 2 context, the task might mix graphs, means and costs in one narrative. Before calculating, write a short note naming the quantity that must be compared. Is the aim lower typical reading, less spread, less cost per reading or fewer invalid measurements? Different questions produce different choices. A sophisticated answer may say that two options cannot be ranked uniquely from the given information.
20. Compute percentage changes with a defined base
Suppose a model signal rises from 20 to 25. The increase is 5, and the percentage increase relative to 20 is 5 ÷ 20 × 100% = 25%. If it then returns from 25 to 20, the decrease is 5 ÷ 25 × 100% = 20%. Equal numerical steps do not generally give equal percentage changes because the reference bases differ.
Never apply “25% increase” directly to the concentration unless the sensor relationship allows that inference. With a nonzero intercept, the percentage change of the signal does not equal the percentage change of concentration. For example, signal 20 corresponds to C = 3 mg/L, and signal 25 corresponds to C = 4 mg/L. Concentration rises by one-third, not 25%. The intercept matters even in a simple percentage question.
21. Practice unit conversions
A concentration expressed in mg/L means milligrams per litre. One litre equals 1000 millilitres. If a hypothetical solution contains 4 mg of solute per litre, a 250 mL portion contains 1 mg under the stipulated uniform-mixing assumption. Show the chain: 250 mL = 0.25 L; amount = 4 mg/L × 0.25 L = 1 mg. This is a mathematical model, not a recommendation for preparing or treating any real water.
Converting between quantities is often more useful than memorising a formula. If a final answer has units of mg/L when the question asks for mass in mg, something is missing. Look at what the multiplication or division does to the units. In context problems, every row in a table may use a different unit, so write a common unit before making comparisons or totals.
22. Read a dilution model critically
Suppose 10 mL of a fictional uniform solution at 8 mg/L is diluted with solvent to a final volume of 40 mL, with no solute lost or added. The concentration in this simplified model is 8 × 10 ÷ 40 = 2 mg/L. The key phrase is “final volume”. Adding 40 mL of solvent to the initial 10 mL would instead give a 50 mL final volume and a different answer.
In a mathematics answer, list the assumption that the amount of solute remains constant. The manipulation is not automatically valid when chemical reactions occur or when volumes do not behave additively. The context here is a fictional soluble model substance. A student should not generalise this calculation to changing pH or neutralising an acid. The previous Chemistry volume separates dilution from chemical reaction for exactly this reason.
23. Calculate total measurement costs
A classroom project budgets $18 for calibration references and $1.20 for each additional test. If n further tests are completed, a simple linear cost model is T = 18 + 1.20n dollars. For n = 25, T = 48. This has a gradient of $1.20 per test and an intercept of $18, just as the sensor graph has a variable rate and a fixed offset.
The mathematical structure is similar, but the units and meaning differ. In a cost graph the horizontal variable is the number of tests, which must be a non-negative whole number. A fitted intercept might reflect a fixed charge. Comparing these two models helps students understand that y = mx + c is a structure, not a context-specific formula. Always interpret slope and intercept in the actual story.
24. Solve a budget inequality
If the available budget is $60, then 18 + 1.20n ≤ 60. Subtract 18: 1.20n ≤ 42. Divide by 1.20: n ≤ 35. Since the number of tests is a whole number, the maximum is 35 under the supplied cost model. A result of 35.0 is not the final story; it means the class can afford at most 35 tests after paying the initial fee.
If each test requires an additional sample container not included in the per-test price, the model would be incomplete. In examination problems, use the conditions specified and do not invent extra charges, but recognise when a model is conditional. Showing the inequality and final integer constraint is more defensible than writing a bare number. Many long Paper 2 questions end with this type of feasibility decision.
25. Compare two cost plans with equations
Plan A has a $20 fixed charge and costs $1 per reading. Plan B has a $5 fixed charge and costs $1.75 per reading. Set the totals equal: 20 + n = 5 + 1.75n. Then 15 = 0.75n, giving n = 20. At twenty readings each plan costs $40. At more than twenty readings, Plan A is cheaper; at fewer than twenty, Plan B is cheaper.
Do not decide from the lower fixed fee alone. The break-even point shows how the choice changes with usage. Check the result with an example on each side: ten readings cost $30 on A and $22.50 on B; thirty cost $50 on A and $57.50 on B. This dual numerical check helps catch the common mistake of reversing the inequality after solving the intersection.
26. Distinguish exact and rounded answers
The official K310 guidance generally expects non-exact numerical answers to three significant figures, or angles in degrees to one decimal place, unless a question specifies another accuracy. That convention does not require rounding away exact counts or currency inappropriately. An answer such as 3.4 mg/L has two significant figures because it comes from the ideal calibration; a problem might explicitly request 3.40 mg/L or a different precision.
Show an unrounded intermediate value when later calculations depend on it. Early rounding can create avoidable errors in a long multi-part question. If a final answer is a number of tests, round according to feasibility: a budget maximum may require rounding down, not nearest whole number. The context can override a purely mechanical rounding habit. Explain the choice where necessary.
27. Check inequality direction and feasibility
Suppose a sensor can process no more than eight readings per hour and the class has three hours. The capacity condition is n ≤ 8 × 3 = 24. If the budget permits thirty-five tests, both constraints apply, so the true maximum is twenty-four. A student who reports thirty-five because that was the last calculated limit has ignored the capacity condition.
Many extended real-world tasks combine money, time and equipment. Organise constraints before optimisation: nonnegative integer n, cost limit, time limit, perhaps an available supply limit. The feasible set is the intersection of all requirements. It is not enough for a plan to satisfy one condition. A clear final sentence says which condition binds and why alternative numerical answers are not feasible.
28. Verify a conclusion with substitution
For n = 24, total cost is 18 + 1.20 × 24 = $46.80, within the $60 budget. Hourly capacity over three hours is 24, so the proposed number uses all available test slots. For n = 25, the cost remains within budget but capacity fails. This direct check demonstrates that twenty-four is not just a plausible guess but the largest feasible whole number.
Students should practise testing the boundary and one value beyond it. That approach is especially powerful with inequalities and capacity problems. If the proposed maximum is seventeen, ask whether eighteen breaks a stated rule; if it does not, the model may have been solved incorrectly. In a written response, a short substitution line can be more persuasive than several paragraphs of general explanation.
29. Construct an extended Paper 2 response
An integrated problem may ask for a calibration line, an unknown estimate, an uncertainty interval, a comparison of two test methods and a feasible number of readings. The student must carry results across parts while keeping the units distinct. Write the model first, then label intermediate answers. When a later part introduces an altered cost or capacity condition, do not accidentally keep using an earlier assumption.
The official syllabus describes Paper 2 as having nine to ten questions of varying lengths, with the final question focusing on a real-world application. Our fictional case reflects the *skill* of coordinating several mathematical ideas, not a promise that an examiner will use greenhouse sensors. The transferable approach is to identify the unknown, convert units, build constraints, compute, check and interpret.
30. Practice question: equation and interpretation
Question: calibration readings at 0 and 6 mg/L are 5 and 35 units. Write the linear equation and explain its gradient. Solution: gradient (35 − 5) ÷ (6 − 0) = 5 units per mg/L. The intercept is 5 units, so R = 5C + 5. This means the model signal rises by five units for a one mg/L rise in concentration within the calibrated interval.
A bare equation earns some of the reasoning but can miss the interpretation demanded by the question. Explain which quantity changes when the other increases. If another question switches dependent and independent variables, the gradient must be recalculated or interpreted from the rearranged equation. Reading the axis labels is a small habit that prevents large mistakes in graph problems.
31. Practice question: unknown and uncertainty
Question: an unknown reads 22 ± 1 signal unit. State the model concentration and its allowable interval. Solution: C = (22 − 5) ÷ 5 = 3.4 mg/L. The lower and upper readings of 21 and 23 give 3.2 and 3.6 mg/L. Thus the point estimate is 3.4 mg/L, with a model interval of 3.2–3.6 mg/L under the stipulated input bound.
Do not call the interval a probability distribution, and do not append a food-safety judgement. A full-mark explanation would note that the signal lies within the calibrated range. The problem tests simultaneous mastery of algebra, interpretation and bounded inputs. If the input were outside the range, the correct numerical method could still produce a result, but the reliability statement would need to change.
32. Practice question: two constraints
Question: a class has $60, a fixed calibration charge of $18, and $1.20 per reading. It can complete at most eight readings hourly over three hours. Find the maximum feasible number. The cost constraint gives n ≤ 35. The capacity constraint gives n ≤ 24. Therefore the maximum is twenty-four readings. Verify that the total cost $46.80 remains below the budget and that twenty-five readings exceed capacity.
The solution is deliberately simple enough to check yet complex enough to reveal planning habits. Some students solve each inequality correctly but then select the greater bound. They need a picture of feasible values: zero through twenty-four satisfy both constraints; twenty-five through thirty-five satisfy only the cost bound. A correct answer is both mathematical and contextual.
33. Practice question: a misleading headline
Question: a graph shows readings 21, 22 and 23 with a vertical scale beginning at 20 and a caption saying “threefold increase”. Explain why the headline is misleading. The readings differ by only two units from lowest to highest, and 23 is not three times 21. Cropping the axis may exaggerate the visual difference. The caption cannot be justified by the recorded values.
An excellent response names both the faulty quantitative claim and the graphical technique that amplifies it. Do not merely write “the graph is biased”. Mathematical communication asks you to show which numbers the claim ignores. A careful reader considers numerator, denominator, baseline and units. This skill matters in newspapers, scientific reports and examination data-response questions.
34. Error correction as a deliberate revision routine
When a solution is wrong, identify whether the first unstable step was interpretation, formula, algebra, unit conversion, rounding or final context. For example, C = 22 ÷ 5 = 4.4 mg/L ignores the intercept. A result of 3400 mg/L may come from a thousand-fold unit mistake. A claim that 7.2 mg/L is validated may come from overlooking the tested range. Fix the cause rather than doing twenty random new questions.
Keep an error journal with one original question, the wrong step, the correction and a new self-written variant. In the next session, solve the variant without notes. The ability to catch the same mistake independently is a better sign of progress than simply copying a model solution. A calm, repeatable feedback system is useful in every examination subject.
35. A short study timetable for the skill
Day one: work on coordinates, axes and gradient until the meaning of rise over run is automatic. Day two: form equations and rearrange to find unknown values. Day three: estimate from graph and equation, then mark cases as interpolation or extrapolation. Day four: practise stated intervals, repeated readings and misleading charts. Day five: solve an integrated cost-and-capacity problem under a reasonable time limit, checking work before looking at an answer.
Parents can support this routine by asking the student to explain why an estimate is believable rather than supplying another worksheet immediately. A learner who can state “my value lies within the reference interval and my units match” has developed a protective habit. If the explanation fails, return to the weak step. More volume of practice does not repair a misunderstanding automatically.
36. Link the numerical skill back to the learner pathway
Review the Mathematics Hub for topic-level work and the Science Hub for the scientific meaning of measurements. The Secondary English Learning Hub supports precise written conclusions. For Primary 6 families moving towards Secondary 1, the PSLE Learning Guide offers the habit of showing working before tackling multi-step G3 problems.
The root-trial data guide (Vol 1368) explores more sophisticated spread comparisons; this volume owns the specific act of deriving and using a linear calibration with bounds. The Chemistry guide (Vol 1387) explains why the model must not be mistaken for pH.
37. Final judgement: correct working and modest claims
A successful solution derives R = 5C + 5 from the reference data, finds 3.4 mg/L for a reading of 22, recognises the stated interval of 3.2–3.6 mg/L when the signal is 22 ± 1, and rejects an unvalidated extrapolation. It can also compare sampling costs, apply an integer capacity limit and explain why a graph or statistic may mislead. These are transferable skills grounded in the G3 Mathematics syllabus.
The fictional instrument is a learning device, not a quality certificate. Good mathematical modelling includes setting boundaries on the claim. When the assumptions are stated, the calculations are shown and the conclusion respects the range of evidence, the learner is doing something more valuable than obtaining an answer: they are producing reasoning that another person can check.
Official exam reference
See the SEAB 2027 G3 syllabus directory and K310 G3 Mathematics syllabus. G3 English Language K300 and Chemistry K324 are separate subjects; this fictional case is not a new official interdisciplinary examination paper. Always verify the subject code and current scheme for the candidate’s actual year of examination.