Mathematics makes electrical data useful
A row of measurements can look precise without answering an important question. Three voltages and three currents do not automatically show how much electrical energy a small solar panel will supply tomorrow, or how much electricity a school could save over a year. To make a defensible decision, you must know which quantities belong together, which calculation is requested, and which assumptions connect the mathematics to the situation. This chapter uses a small school solar-panel demonstration to train the step from accurate arithmetic to informed mathematical judgement.
The immediate method is to define each variable with units, organise the paired measurements, convert current into amperes, calculate power from voltage and current, choose the correct comparison baseline, read a graph with its scale, and test a proposed operating plan against any limits. When the question supplies only three observed conditions, do not invent a continuous rule that the data do not establish. That habit makes the difference between a smooth-looking answer and one that a critical examiner or reader can actually trust.
This is a learner guide for Singapore-Cambridge Secondary Education Certificate Mathematics at G3. The official 2027 Mathematics K310 syllabus specifies Paper 1 and Paper 2, each worth 90 marks and 50% of the total, each lasting 2 hours 15 minutes. Paper 2 includes a real-world-context question towards the end, intended to assess ability to interpret information, select methods and communicate an appropriate result. Here the solar-panel observations are an original teaching case, not an official specimen paper, expected future examination question or energy-generation prediction.
A continuity map for the three-subject learning route
Start with Volume 0946: Physics solar-panel shading, which introduces the safe low-voltage experiment and the physical limits of the readings. If current and voltage graph vocabulary is unstable, read Volume 0913 on circuit tables, graphs and gradients. For an earlier example of how a numerical plan changes when reserves and operating conditions change, read Volume 0915 on emergency-light runtime constraints. These links lead to distinct subject-owner lessons rather than another copy of this chapter.
The earlier Volume 0943 on leaf-disc timed observations is useful for a different reason. It shows that mathematically correct calculations can still mislead when a dataset excludes observations that did not reach an endpoint. The solar-panel lesson teaches the parallel problem: a correct power value can mislead when it is described as the panel’s maximum output, its energy for a whole day or its conversion efficiency. In both examples, state what has actually been measured.
A fictional dataset, stated before the working
The school uses a small low-voltage solar module, with a fixed nominal 20-ohm load and suitable meters, under a supervised demonstration. An opaque card covers approximately zero, one-quarter or one-half of the module surface. In the fictional no-cover condition, the load has voltage 3.6 volts and current 0.18 ampere. At one-quarter cover, it has 3.0 volts and 0.15 ampere. At one-half cover, it has 2.4 volts and 0.12 ampere. Assume each pair was recorded at the same instant under the indicated condition.
The three pairs have an intentionally simple relationship: dividing each voltage by its current gives 20 ohms. These numbers were written for teaching, not obtained from a commercial product, real rooftop installation or SEAB test. The area covered is an approximate fraction of the panel surface, not a measurement of the solar energy reaching the panel. Resist the temptation to deduce input irradiance, maximum-power operation or future sunlight from these numbers.
1. Translate the prose into a clean data table
Write four columns. Column one is the covered fraction: 0, 0.25 and 0.50. Column two is potential difference across the load in volts: 3.6, 3.0 and 2.4. Column three is current in amperes: 0.18, 0.15 and 0.12. Column four, added by calculation, is power delivered to the load in watts. The heading of each column should identify units. This turns a paragraph into something a reader can inspect without repeatedly searching for numbers.
The order of operations matters. Do not put voltage from one condition beside current from another. Do not invent a fourth condition because an evenly spaced table looks attractive. And do not confuse covered fractions with light intensity percentages; a quarter of the area covered is not necessarily a quarter less light energy entering the entire electrical arrangement. The table is a record of a specific test, not a general model of solar energy.
2. Calculate three powers accurately
Electrical power at the measured load is P = VI. For no cover, P = 3.6 × 0.18 = 0.648 W. For one-quarter cover, P = 3.0 × 0.15 = 0.450 W. For one-half cover, P = 2.4 × 0.12 = 0.288 W. These results let us compare one operating point in each condition. Keep more calculation digits than you eventually display if the question asks for a particular level of precision.
A student who writes 3.6 × 180 = 648 W after reading 180 mA has lost the factor of one thousand. Another who writes 3.6 + 0.18 = 3.78 has used a meaningless operation on two differently dimensioned physical quantities. Check the units before touching the calculator. Then interpret the answer: 0.648 W is a rate of electrical energy transfer to this load under this condition, not an accumulated energy amount.
3. Use dimensional checks as a mathematical safety net
Watts and joules answer different questions. One watt is one joule per second. If power is multiplied by seconds, the resulting unit is joules; if power is divided by seconds, the resulting unit would not represent the energy transferred over that interval. A quick unit inspection can catch an error even when the arithmetic appears tidy.
Suppose the no-cover output were constant for 600 seconds. E = Pt = 0.648 × 600 = 388.8 J. A candidate may calculate the right number yet still misinterpret the result by writing 388.8 W. The multiplication by time has changed the quantity. Conversely, a ten-minute energy result cannot be used as an instantaneous output rating. Examinations that place several quantities in one paragraph reward precise reading as much as quick computation.
4. Explain the three different comparisons
From no cover to one-quarter cover, power decreases by 0.648 − 0.450 = 0.198 W. From no cover to one-half cover, it decreases by 0.648 − 0.288 = 0.360 W. From one-quarter to one-half cover, it decreases by 0.450 − 0.288 = 0.162 W. Each statement is true, but each starts and ends at different conditions. Include those conditions when discussing the result.
Do not make a comparison sound more dramatic by swapping its baseline unnoticed. If the task asks how much power is lost when the covered fraction increases from one-quarter to one-half, the 0.162 W answer is relevant. If it asks how much the half-covered output differs from no cover, 0.360 W is relevant. Knowing all three figures does not compensate for answering the wrong comparison.
5. Percentage change begins with a denominator
The reduction from the no-cover state to quarter cover is 0.198 divided by 0.648, multiplied by 100%, giving about 30.6%. The reduction from the no-cover state to half cover is 0.360 divided by 0.648, multiplied by 100%, giving about 55.6%. Between quarter and half cover, the reduction is 0.162 divided by 0.450, multiplied by 100%, giving exactly 36% for the stated numbers.
These are percentage reductions in measured output power relative to specified starting conditions. They are not percentage reductions in sunlight, panel efficiency or electricity bills. A percentage is not fully explained until the reader knows what makes up one hundred percent. A strong answer can sometimes avoid ambiguity by giving both the absolute decrease in watts and the clearly based percentage.
6. Distinguish a percentage from a percentage point
The covered fraction changes from 0% to 25% and from 25% to 50%. Those are differences of 25 percentage points in the stated fraction of area covered. The measured output reduction from 0.648 W to 0.450 W is approximately 30.6 percent relative to baseline power. These two percentages refer to different quantities and cannot be subtracted as though they describe a single outcome.
If someone says ‘a 25 percent rise in shading produced 30.6 percent less electricity’, ask whether the first figure means a 25-percentage-point change in covered area or a relative percentage increase in a previously shaded fraction. A rise from zero has no ordinary percentage-increase denominator. Clear mathematics begins by naming the variable, not by choosing an eye-catching number.
7. Ratios reveal useful relationships, not necessarily causes
At quarter cover, the ratio of measured power to no-cover power is 0.450/0.648 = 25/36, about 0.6944. At half cover, it is 0.288/0.648 = 4/9, about 0.4444. Those retained-output fractions are useful if a question asks what portion of the baseline remains. They do not prove that the same fractions apply under different sunlight or with a different load.
The load’s measured voltage-current ratios happen to be 3.6/0.18 = 20, 3.0/0.15 = 20, and 2.4/0.12 = 20 ohms. The numerical consistency supports the fictional load description. It does not prove a physical law linking shaded area to voltage. Ratios answer the question you construct; choose those with a relevant meaning.
8. Plot output against the correct horizontal variable
Use covered fraction on the horizontal axis, from zero to at least one-half, and power in watts on the vertical axis. Plot (0, 0.648), (0.25, 0.450) and (0.50, 0.288). If instead you label the horizontal axis ‘percentage of light blocked’, you claim a measurement that the experiment did not make. The opaque card obscures area, not a calibrated fraction of the total solar energy at the cells.
A graph can reveal a trend in the measured range. It cannot guarantee what happens beyond the points. A smooth line through three readings may help a reader see direction, but it should not be used to prove that power at 75% cover equals a predicted value. Extrapolation introduces an assumption about the shape of the relationship. In a mathematics problem, state that assumption if the question asks you to use it.
9. Calculate average gradients between the points
Between zero and quarter cover, the change in power is −0.198 W over 0.25 of the covered fraction, so the interval gradient is −0.792 W per unit covered fraction. Between quarter and half cover, the change is −0.162 W over 0.25, so the interval gradient is −0.648 W per unit fraction. Both are negative because output power decreases in the supplied data as covered fraction rises.
The two gradients are different. You should not describe the entire observed relation as exactly linear. If the horizontal axis were expressed in percentage points rather than unit fractions, the first gradient would be −0.00792 W per percentage point. The number changes because the unit of horizontal distance changes. This is why a gradient without its axis context can be misleading.
10. Calculate a whole-interval gradient with caution
Across zero to half cover, the average gradient is (0.288 − 0.648)/(0.50 − 0) = −0.720 W per unit covered fraction. This number lies between the two adjacent interval gradients, but it is not the gradient at every possible covered fraction. It summarises a change between endpoints. An answer that calls it the exact local gradient at 40% cover would assume a model that the three observations do not establish.
A useful checking habit is to ask whether the sign and magnitude match the picture. The output falls as cover increases, so the average gradient should be negative. If you obtained +0.720, you may have reversed one subtraction without reversing the other. Consistency of order in the numerator and denominator is more important than memorising a diagram of a downward line.
11. When is interpolation justified?
Interpolation means estimating between known points using a chosen rule. Suppose a question explicitly asks for a straight-line estimate between the quarter-cover and half-cover observations. At 0.375 covered fraction, halfway between 0.25 and 0.50, linear interpolation would give halfway between 0.450 and 0.288 W, or 0.369 W. The calculation is internally correct under the stated straight-line assumption.
But it is an estimate, not an observed power at 37.5% cover. Real photovoltaic modules need not behave linearly with shading. State the assumption and mark the value as estimated. If a question does not supply a model, do not silently insert one and present the answer as physically guaranteed. Mathematical sophistication includes understanding the circumstances under which a calculation is meaningful.
12. A reverse question is not the same as a forward one
A forward problem gives the covered fraction and requests power from the observed table. A reverse problem gives a required power and asks which of the three tested conditions meets it. If a demonstration load needs at least 0.40 W, the no-cover condition at 0.648 W and the quarter-cover condition at 0.450 W meet the threshold; the half-cover condition at 0.288 W does not.
That conclusion concerns only the three tested conditions and the defined threshold. You cannot deduce a precise maximum permitted covered fraction without an approved function or further measurements. If the question states you may assume linear interpolation between a pair of observations, you can solve an approximate threshold. Without it, the defensible answer is the set of conditions that the evidence actually covers.
13. Solve one conditional threshold model
Assume, for this worked question only, that power varies linearly between the quarter-cover and half-cover measured points. With covered fraction x between 0.25 and 0.50, the model is P = 0.450 − 0.648(x − 0.25). For a required output of at least 0.400 W, solve 0.450 − 0.648(x − 0.25) ≥ 0.400. Rearranging gives x − 0.25 ≤ 0.050/0.648, so x ≤ about 0.3272.
This bound is a mathematical consequence of the assumed linear segment. It is not a universal safe shaded fraction for actual photovoltaic panels, since the model has not been experimentally validated between the two points. If a student reports 32.7% as a guaranteed engineering design threshold, the arithmetic may be correct but the conclusion is not. The qualifying words protect the meaning.
14. Distinguish design planning from real output prediction
Planning a science demonstration may require deciding whether a lamp will receive enough electrical power under one of the measured test conditions. If the load needs at least 0.300 W, the quarter-cover condition passes and the half-cover condition fails, according to the fictional table. If the threshold is reduced to 0.280 W, all three recorded conditions pass. These are simple comparisons rather than a forecast.
A real plan must also consider variation between observations, changes in sunlight, any additional electrical components and the load’s actual requirements. Those facts are outside the three-row table. In an examination, solve the stated simplified mathematical problem first, then explain the boundary where the real-world decision requires more evidence. Avoid adding complexities that obscure an answer while also avoiding unjustified certainty.
15. Time turns output power into a conditional energy amount
For constant no-cover output over ten minutes, the energy transferred to the load is 0.648 W × 600 s = 388.8 J. For quarter cover it would be 0.450 × 600 = 270 J. For half cover it would be 0.288 × 600 = 172.8 J. The phrase ‘constant output’ is essential because the experimental table contains instantaneous readings, not ten minutes of continuous monitoring.
If a school demonstration lasts exactly ten minutes and conditions vary, a better estimate of energy would require readings over time. One measurement at the start cannot establish the total. For examination purposes, perform E = Pt when the question explicitly states power is constant or asks for an estimate under that assumption. Good mathematical reading recognises the conditional phrase that allows the formula to be used.
16. Work through a fictional energy target
Suppose an entirely hypothetical demonstration display needs 200 J delivered to a load during a ten-minute session. Under constant outputs matching the table, no cover supplies 388.8 J and quarter cover 270 J, both above the target. Half cover supplies only 172.8 J and falls short by 27.2 J. This analysis assumes a constant measured output, the stated time and no additional losses between the measured load and the target.
Do not reinterpret this result as a battery capacity or a promise that the sun will supply the necessary energy in real conditions. If a real device needs storage, power electronics and safety components, new information is required. The purpose of this problem is to build a habit: define the target, calculate all comparable outputs with the same assumptions, and make the decision on the correct units.
17. Model operating time using an inequality
Consider a hypothetical condition where the load receives 0.450 W continuously and must receive at least 405 J. From E = Pt, the time needed is t ≥ 405/0.450 = 900 seconds, or fifteen minutes. A candidate who divides time by power or leaves the result in joules has confused the quantities. The inequality sign matters: nine hundred seconds is the minimum in this simplified model, not the maximum.
Now ask whether the experiment supports a guarantee of 0.450 W for fifteen minutes. It does not, because the 0.450 W figure came from one fictional paired reading. The inequality is a correct conditional calculation under a constant-power assumption. This separation of model and evidence is especially important in real-world-context questions, which may ask for a practical recommendation after the mathematics.
18. Divide a task into intervals only when necessary
Some tasks combine different operating conditions. Imagine a deliberately simplified ten-minute session with four minutes at 0.648 W, three minutes at 0.450 W and three minutes at 0.288 W. Treat each interval as constant. The energy contributions are 0.648 × 240 = 155.52 J, 0.450 × 180 = 81 J and 0.288 × 180 = 51.84 J. Total energy delivered to the specified load would be 288.36 J.
Do not average the three power readings equally unless the intervals are of equal duration. When durations differ, a correct time-weighted average power is total energy divided by total time. Here 288.36 J / 600 s = 0.4806 W. A neat mean of three readings would be 0.462 W, a different quantity that ignores the time distribution. The right average depends on what the question wants to represent.
19. Weighted averages can hide a choice of weights
If the intervals last four, three and three minutes, the weights are 4/10, 3/10 and 3/10. The time-weighted mean power is (4/10)(0.648) + (3/10)(0.450) + (3/10)(0.288) = 0.4806 W. The units remain watts because the weights are dimensionless fractions of time. If the durations change, the weighted mean must be recalculated.
Do not confuse time weighting with probability weighting or with the fraction of the panel covered. All three can appear in the same dataset but describe different things. A skilled learner writes beside a weight what it represents. If you cannot explain why the weights add to one and what population or duration they cover, the weighted average may be answering a question that nobody asked.
20. Prepare for data that are not perfectly consistent
Real readings rarely produce exactly V/I = 20 ohms for every condition. A 20-ohm load may have tolerance; meters may be rounded; temperature can change resistance; or current and voltage may be recorded at slightly different times. If one pair implies 18.5 ohms, do not immediately rewrite it to 20 to make the table tidy. First identify measurement resolution and whether the assumption of an unchanged load is justified.
An important error is ‘repairing’ data by replacing inconvenient values with ones calculated from a nominal specification. Mathematics can check for inconsistencies, but it cannot determine the correct measurement without additional evidence. A good answer says what is inconsistent, shows the calculation, and proposes a sensible check. The aim is not to manufacture perfect tables; it is to draw the strongest defensible conclusion from the actual table.
21. Appropriate rounding and calculator discipline
The official K310 syllabus requires exact answers where appropriate and states that nonexact numerical answers should generally be given correct to three significant figures unless another accuracy is specified. For this teaching case, 0.648 W is already three significant figures; 0.450 W also displays three significant figures. The percentage reduction from no cover to quarter cover is approximately 30.6% to three significant figures.
Keep full calculator precision during intermediate working and round the reported result at the end. When a fraction is exact in the fictional values, such as 25/36 for retained power at quarter cover, it can be helpful to keep the fraction until converting to the requested form. Avoid unwarranted precision in any real physical claim: the graph and conclusion should reflect the measurement quality, not just the number of digits in the calculator output.
22. The common fallacy of treating three points as a law
Three measured points can support a comparison in the tested range. They do not prove how a system performs across all possible coverage percentages, weather conditions or equipment types. A student may be tempted to use a quadratic curve simply because three points uniquely determine a quadratic polynomial. Such an expression would fit the teaching numbers exactly but might have no physical justification beyond those observations.
In a mathematics examination, fitting an algebraic model may be a valid task if the question defines the method. In a public claim, a fitted model must be tested and presented with its limits. The ability to solve a polynomial is separate from the judgement that a polynomial is appropriate. Use the model the question provides; explain when its assumptions matter; do not confuse an exact fit to a small sample with a law of nature.
23. Explain the answer to someone unfamiliar with the apparatus
A younger student asks: ‘If half the panel is hidden, why isn’t the loss exactly half?’ A good reply starts with the numbers: in this demonstration, the measured electrical power fell from 0.648 W to 0.288 W, a loss of about 55.6% from the initial output. Then explain that shading changes how the electrical system operates; area covered is not the same as a proportion of output power. Avoid overclaiming about the internal cell design, which this experiment did not investigate.
The words you choose matter. Saying ‘we measured less electricity’ can be ambiguous between power and energy. Prefer ‘we measured lower output power at the load’ when discussing these paired readings. Mathematics becomes useful to real readers when units and baselines remain visible after the final calculation, not only in the rough working.
24. Original examination-style challenges and explanations
The following practice cases are original and use the fictional data unless a changed condition is given. They are not SEAB questions or model marking rubrics. Try each problem first. Show the relevant arithmetic with units and a complete concluding sentence. In the hardest cases, the question is not merely how to calculate but whether the proposed calculation can legitimately answer the claim.
A brief correct solution is better than a long paragraph that hides a wrong baseline. But when a real-world decision is required, a terse number can be incomplete: identify the scenario, threshold and major model assumption. This is why Mathematics Paper 2 contextual tasks reward both working and interpretation.
Practice 1: What is the measured output power without deliberate covering?
Use P = VI with 3.6 V and 0.18 A from the same condition. The product is 0.648 W. A correct solution names the quantity as power delivered to the fixed load at that measured operating point. It is not a maximum rating for the module and not an energy quantity. Include the unit and do not infer unseen time or sunlight exposure.
Practice 2: The current is reported as 150 mA. What is the corresponding power at 3.0 V?
Convert 150 mA to 0.150 A before using P = VI. The answer is 3.0 × 0.150 = 0.450 W. A value of 450 W would result from forgetting the conversion factor. The exercise tests unit reading as much as multiplication, which is why careful table headings matter in unfamiliar real-world contexts.
Practice 3: What fraction of baseline output remains at half cover?
Divide the half-cover power 0.288 W by the no-cover power 0.648 W. The ratio simplifies to 4/9, or approximately 44.4%. This describes retained power relative to the baseline. A 44.4% retained portion is not a 44.4% reduction; the reduction is the complement, approximately 55.6%. Distinguish the quantity that remains from the quantity lost.
Practice 4: How many watts are lost between quarter and half cover?
Subtract 0.288 W from 0.450 W to get 0.162 W. Do not substitute the no-cover reading unless the question explicitly asks for it. The two covered fractions establish the endpoints of the comparison. A unitless number 0.162 would be ambiguous because current, voltage and power all appear in the same situation.
Practice 5: What percentage of power is lost from quarter to half cover?
Use the starting quarter-cover output as the denominator: 0.162/0.450 × 100% = 36%. A student using 0.648 W as the denominator would answer a different relative-change question. State that 36% describes the decrease relative to the one-quarter-cover reading under the given fictional load conditions.
Practice 6: Find the mean of the three measured power readings and explain its weakness.
The unweighted arithmetic mean is (0.648 + 0.450 + 0.288)/3 = 0.462 W. It is the mean of three chosen conditions, not the average operating power during a day, because the problem gives no durations or frequencies for those conditions. The calculation is valid as a descriptive statistic but cannot support a prediction about daily energy generation.
Practice 7: Find the range of the three measured powers.
The range is maximum minus minimum: 0.648 − 0.288 = 0.360 W. The result summarises how far apart the three values are; it is not an estimate of measurement uncertainty. A range of tested outcomes and an uncertainty interval describe different ideas. Do not claim the module’s power is always inside this range, since untested conditions may produce other values.
Practice 8: Find the first interval gradient and its sign.
The average change in power between zero and quarter cover is −0.198 W over a change in covered fraction of 0.25. The gradient is −0.792 W per unit fraction. The negative sign expresses decreasing output over that interval. If the horizontal axis uses percentage points, the numerical gradient and unit change accordingly.
Practice 9: Find the second interval gradient and compare it with the first.
From quarter to half cover, (0.288 − 0.450)/(0.50 − 0.25) = −0.648 W per unit fraction. Its magnitude is smaller than the first interval’s 0.792, showing that the two observed average rates of decrease are not identical. This does not establish the exact local shape of the relationship between or beyond the data points.
Practice 10: Estimate power at 37.5% cover under a stated straight-line assumption.
The fraction 0.375 is halfway between 0.25 and 0.50. A linear estimate halfway between 0.450 and 0.288 W is (0.450+0.288)/2 = 0.369 W. Label the result as an interpolation under a modelling assumption. The dataset contains no direct observation at 37.5% cover, and a real photovoltaic module is not guaranteed to obey a linear model.
Practice 11: Which observed settings meet a 0.40 W requirement?
The no-cover output is 0.648 W and the quarter-cover output is 0.450 W, both above the requirement. Half cover produces 0.288 W, below it. These are the tested conditions that pass or fail using the stated fictional readings. There is not enough information to certify every intermediate covered fraction without a model or more observations.
Practice 12: Solve the assumed linear threshold P at least 0.400 W between 25% and 50% cover.
Using P = 0.450 − 0.648(x−0.25) over this interval, the inequality gives x ≤ 0.25 + 0.050/0.648 ≈ 0.3272. Under the stated line model, at most about 32.7% cover meets the requirement within the interval. It is a conditional mathematical threshold, not a safe physical design limit for all panels.
Practice 13: Calculate energy from constant half-cover output for ten minutes.
Ten minutes is 600 seconds. E = 0.288 W × 600 s = 172.8 J. The word constant must remain visible: a single power measurement does not prove that actual output stayed constant throughout the period. The correct unit of the product is joules, not watts.
Practice 14: A display requires 200 J over ten minutes. Which test conditions meet this target under constant output?
The no-cover, quarter-cover and half-cover conditions deliver hypothetical energy amounts of 388.8 J, 270 J and 172.8 J. The first two meet 200 J; the half-cover condition does not. This is a mathematical comparison under a simplifying assumption, not an output guarantee for a changing day or a system with unmodelled losses.
Practice 15: Calculate total energy from four minutes unshaded, three quarter-covered and three half-covered.
Convert the intervals to 240, 180 and 180 seconds. The contributions are 155.52, 81 and 51.84 J. Total is 288.36 J under constant output in each interval. Do not average the three powers equally: the durations differ. The calculation is an example of piecewise energy accounting.
Practice 16: Find the time-weighted average power for the ten-minute mixed session.
Divide total energy 288.36 J by total duration 600 seconds to obtain 0.4806 W. Equivalently use weights 4/10, 3/10 and 3/10 on the three powers. An unweighted mean would be 0.462 W, which describes equal weighting rather than this session. Explain why the weights match time.
Practice 17: The panel is measured at 3.0 V and 0.15 A. Can you calculate conversion efficiency?
Not from those two figures alone. You can calculate electrical power delivered to the load, 0.450 W, but efficiency requires an appropriate input power such as the solar power incident on the module. Neither covered-area fraction nor resistor value supplies that energy input. A carefully reasoned refusal to calculate an undefined quantity is better mathematics than inserting an invented denominator.
Practice 18: Why is a straight line through all three points not an exact description?
The interval gradients are −0.792 and −0.648 W per unit shaded fraction, so they are not equal. A single line cannot pass exactly through all three supplied points. A graph can show a trend without asserting an exact physical rule. If an examination asks for a line of best fit, interpret the estimated relationship as a model rather than an established law.
Practice 19: The available time is doubled while power stays constant. What happens to energy?
Energy E = Pt doubles when the constant power stays fixed and duration doubles. For instance, at 0.450 W, ten minutes gives 270 J and twenty minutes gives 540 J. If output varies, the simple proportionality cannot be used without additional assumptions. Always distinguish this conditional relation from a forecast of solar output under real weather.
Practice 20: An observer records current as 0.20 A at 2.4 V on the unchanged nominal 20-ohm load. What should you do?
The ratio V/I gives 12 ohms, not 20. State the inconsistency and recommend checking the meter connections, component and raw record. Do not silently replace the reading. Numerically detecting a conflict is valuable; the true cause cannot be established from the two numbers alone. A well-communicated answer separates diagnosis from speculation.
Practice 21: Explain one reason why a three-condition experiment cannot predict annual school electricity savings.
There is no duration-weighted record of actual sunlight or measured building-scale energy output, and the small demonstration uses one fixed load. A correct power calculation at a single instant does not provide annual energy. More information on operating periods, environmental variation and system configuration would be necessary. This is an evidence-limitation question, not an invitation to guess.
Practice 22: Write a measured conclusion from the table.
The measured load power decreased from 0.648 W without deliberate cover to 0.288 W when approximately half the small panel was covered, under the fictional fixed-load conditions. That is a 0.360 W decrease, or about 55.6% of baseline output. This conclusion names the experiment and avoids promising the same change for every solar installation.
Practice 23: A classmate wants to treat x = 0.75 as if it were directly observed. Respond.
The table contains x values of 0, 0.25 and 0.50, not 0.75. Any estimate at 0.75 would be extrapolation beyond the observed range and require an additional model. Even an algebraic curve fitted exactly to three points is not necessarily a reliable physical law. State the unsupported step before attempting the requested prediction.
Practice 24: How would you check a percentage answer quickly?
Inspect whether the computed value makes sense relative to the raw numbers. From 0.648 to 0.288 W the output loses more than half, so a percentage reduction a little above 50% is plausible. If the calculator gives 155.6%, the ratio or subtraction has likely been reversed. Estimation is not the final calculation, but it is a strong defence against unnoticed keystroke errors.
25. A method for the last ten minutes of revision
Choose one fresh dataset from a school exercise rather than copying these exact numbers. First write the variable definitions and units without calculating. Next derive powers, differences and any required ratios, showing the correct baseline. Then interpret the graph’s direction and state whether the data justify a model. Finally, explain a practical decision using one relevant numerical threshold and one limitation. This sequence trains understanding at each level without relying on memorising a model answer.
If reading multi-step quantities remains difficult, revisit the PSLE Maths word-problem guide for the earlier skill of translating language into quantities before solving. Then return to this chapter’s percentage-change, inequality and time-weighting cases. The same distinction between a number and its meaning appears from upper primary through advanced examinations. Repairing that link can improve many different questions.
26. A final audit before submitting a real-world answer
Check the base of every percentage, the quantity on each graph axis, the units of every computed value and the assumptions behind any interpolation or energy forecast. If a quantity was not measured, do not pretend it was. When an output figure is conditional on a fixed load, identify that load; when a time calculation assumes constant power, say so. A complete solution ends with the requested result, not just an array of unlabelled calculator figures.
The next volume applies this same evidence to English Paper 1 Situational Writing. You will need to explain an honest school demonstration proposal to a real audience without confusing measured load power with energy savings. Mathematics gives the correct comparisons; language ensures that the comparisons retain their meaning when they leave the exercise book.
