PG1 PG2 PG3 Estimation and Units 2027

How can PG1, PG2 and PG3 students catch unreasonable answers, wrong units and avoidable calculation mistakes in G1, G2 and G3 Mathematics and Science? Before calculating, make a rough prediction of the size and direction of the answer. Write the quantities with their units and context. After calculating, check whether the result matches the requested quantity, whether the units are consistent and whether the number is plausible. Estimation cannot prove an answer is correct, but it often identifies a misplaced decimal, wrong reference amount or impossible interpretation.

Students sometimes know the formula yet still report 2.5 litres as 2.5 millilitres, use centimetres when the question asks for area, or accept a discounted price greater than the original. These mistakes need more than a generic reminder to ‘be careful’. They require a deliberate sense of what the quantity means.

Under Full Subject-Based Banding, learners study subjects at appropriate G1, G2 or G3 levels. A student may need basic measurement understanding in one subject while tackling more complicated scientific calculations in another. The habit of interpreting units and checking reasonableness is shared across levels, even though the content and formal response requirements differ.

Independent eduKateSG parent guide reviewed 21 October 2026 against MOE Full SBB and SEAB’s official 2027 SEC Mathematics and Science syllabuses. Numerical situations below are original illustrative teaching questions, not released SEC papers. School calculator, rounding, significant-figure and method requirements depend on the actual question and registered subject.

Quick answer: five checks that catch many errors

Check	Question	Example of an error it could reveal
Quantity	What am I finding?	Calculated the discount instead of the final price
Approximate size	What range seems plausible?	Calculator answer is ten times too large
Unit	What does the number measure?	Area labelled cm rather than cm²
Reference amount	Percentage of what original quantity?	Used new price as the old percentage base
Context	Does the answer obey the problem’s limits?	Reported a fractional number of buses

These checks are aids to verification, not a replacement for valid mathematical working or scientific reasoning. An answer can pass a plausibility test while still being numerically wrong.

1. Estimate before pressing the calculator

Suppose a school supply item costs S$246 and the question asks for an 18% discount. An estimate of 20% is useful: one-fifth of S$246 is S$49.20, so the true discount should be somewhat less than S$49.20.

The exact discount is 0.18 × 246 = S$44.28. The estimate supports its general size. If the calculator display instead showed S$442.80, that would contradict the expected range and demand a recheck.

Estimation helps students notice errors; the final answer should still follow the required exact or appropriately rounded calculation.

2. A useful estimate is a range, not necessarily a single guess

Instead of trying to predict S$44.28 precisely, the learner can say 18% lies between 10% and 20%, so the discount must lie between S$24.60 and S$49.20. The exact result fits within that range.

Bounds are especially helpful when arithmetic is cumbersome. They give the child a way to question a calculator output without first repeating every multiplication.

When a problem includes a minimum or maximum value, estimation may also help decide whether a proposed solution can satisfy that restriction.

3. Estimation and rounding are not the same

Estimation is a planning or checking strategy that produces an approximate value before or alongside a calculation. Rounding converts a number to a specified degree of precision. The two can work together but have different purposes.

For example, an exact price of S$61.20 may be estimated mentally as about S$60. A question that asks for a final price to the nearest cent requires S$61.20, not the estimate.

Students should read the required rounding instruction, especially in Science where measurements and reported precision may be governed by the actual assessment context.

4. Read the quantity, not just the number

The numeral 12 can mean 12 people, 12 metres, 12 seconds or 12 dollars. The operation chosen depends on the quantity and its relationship to the question. A student who copies all numbers from a word problem into a calculator has not yet modelled the situation.

Write a short label beside each given value: ‘distance = 120 km’, ‘time = 2 h’, or ‘price before discount = S$80’. The labels make it easier to see what can be combined meaningfully.

If the question asks for a different quantity, such as average speed, the student can then choose a relevant relationship.

5. Speed has distance divided by time as its unit structure

A vehicle travelling 120 km in 2 hours has an average speed of 120 ÷ 2 = 60 km/h. The unit is kilometres per hour because distance was divided by time.

The numerical result 60 without units is incomplete when the question expects a physical speed. If the student reports 60 km, they have labelled a speed as a distance. A unit check can expose the conceptual mistake.

The average speed does not establish that the vehicle travelled at exactly 60 km/h every moment; it summarises total distance over total time.

6. Metres per second and kilometres per hour are different units

Suppose a toy moves 18 metres in 3 seconds. Its average speed is 6 m/s. That is a valid answer expressed in metres per second, not kilometres per hour.

Students should not write ‘6 km/h’ simply because the number 6 appeared. Unit conversion requires applying a valid relationship between the physical measures; it is not a change of label.

Before solving, inspect the units in the question and the requested output. If the problem specifically requires conversion, carry out that step explicitly and check the result.

7. Litres and millilitres: everyday proportional reasoning

A 2.5-litre container holds 2,500 millilitres. If seven people each receive 350 millilitres, they receive 7 × 350 = 2,450 millilitres in total. The remaining amount is 50 millilitres, assuming no spillage.

A student who subtracts 2,450 from 2.5 without converting to compatible units will produce a meaningless result. The quantities represent the same kind of measurement but use different units.

A rough check also helps: seven servings of roughly one-third of a litre approach 2.5 litres, so a small remainder is plausible.

8. Kilograms and grams in a school context

Suppose a project uses 0.75 kg of a material and the remaining package is labelled in grams. The mass is 750 g because one kilogram contains 1,000 grams.

A learner should preserve the decimal place and relationship. Confusing 0.75 with 75 leads to answers that are off by a factor of ten or more, which a basic estimate may detect.

In practical subjects, students must follow school instructions for equipment and safe handling rather than treat every example as permission for unsupervised measurement.

9. Area and perimeter measure different things

Consider a rectangle 7 cm wide and 12 cm long. Its perimeter is 2(7 + 12) = 38 cm. Its area is 7 × 12 = 84 cm². Both calculations are correct for different questions.

Area measures surface coverage and has square units; perimeter measures boundary length and has linear units. A learner who responds ’84 cm’ to an area problem has lost the meaning of the quantity even if the multiplication was correct.

A labelled sketch can help choose the relevant measure before the calculation begins.

10. Volume requires another dimension

A rectangular block 4 cm by 3 cm by 2 cm has volume 4 × 3 × 2 = 24 cm³. The cubic unit reflects three-dimensional space, not simply a convention to memorise.

The same number could be unrelated to a surface area or length, so the student must ask what the question actually measures. This becomes increasingly important in upper-secondary Science, where density and other derived quantities involve volume.

For diagrams not drawn to scale, rely on stated dimensions and geometric relationships rather than physically measuring the printed picture.

11. Density illustrates a derived unit

Suppose an object has a mass of 12 g and occupies a volume of 4 cm³. Its density is 12 ÷ 4 = 3 g/cm³. The derived unit shows mass per unit volume.

A student who writes 3 g or 3 cm³ has changed the physical meaning. Unit structure helps verify that the intended relationship was used.

This is an illustrative basic physical quantity. Formal Physics, Chemistry or Science examination requirements should be checked in the student’s actual syllabus.

12. Fractions and proportions help test plausibility

If a recipe uses 3 parts of one ingredient for every 5 parts of another, the first ingredient represents 3 out of 8 total parts. Its proportion is 3/8, or 37.5% of the total.

A child who reports 60% has used 3/5, the first-to-second ratio, instead of first-to-total. The arithmetic is mathematically valid for another comparison, so a formula check alone will not catch the interpretation error.

A simple bar model with eight equal parts often makes the correct denominator visible.

13. Percentage change depends on its base

A price rises from S$80 to S$100. The change is S$20, which is 20/80 × 100% = 25% increase from the original S$80.

If it later falls from S$100 to S$80, the decrease is S$20 relative to S$100, or 20% decrease. The numerical movement is the same S$20, but the percentage bases differ.

A student should identify the starting value every time. It is unsafe to assume that a 25% rise is undone by a 25% fall.

14. A discount should not create a higher price

Suppose an item costs S$80 and receives a 15% discount. The exact discount is S$12 and the final price is S$68. A result of S$92 would indicate that the wrong operation or interpretation was used.

Checking that the final price is less than the original is a quick contextual test. It does not identify every possible arithmetic error, but it can prevent acceptance of an obviously implausible result.

If tax, delivery fees or other charges are also stated, inspect the actual conditions before concluding that the total must always be lower.

15. Multiple discounts need successive reference amounts

Suppose the S$80 item receives an additional 10% reduction after the first 15% discount. After the first reduction it costs S$68; the second reduction is S$6.80, leaving S$61.20.

It would be incorrect to apply the second 10% to the original S$80 unless the question says so. A learner should represent the sequence of operations before calculating.

The estimate provides an extra check: a total reduction slightly less than 25% of S$80 is plausible because the second percentage is applied to a lower base.

16. Significant figures and decimal places are question-dependent

Some questions specify the number of decimal places or significant figures for the final result. Measurements and scientific contexts may also require attention to sensible precision. These instructions should not be invented by the student.

A result such as 7 ÷ 3 = 2.333… has many possible representations: an exact fraction 7/3, a recurring decimal, or a rounded value such as 2.33 to two decimal places. The correct response depends on the task.

Premature rounding during intermediate steps can introduce avoidable error. Retain suitable precision while working, then apply the required final instruction.

17. Exact answers versus approximate answers

An exact answer can be left as a fraction or expression where permitted. An approximate answer should make its approximation status clear when relevant. A student who reports a rounded result as exact may misrepresent the mathematical quantity.

For example, the mean of 25, 29 and 28 is 82 ÷ 3 = 27⅓. If a question requires one decimal place, report approximately 27.3 in the appropriate units. The exact mean and its rounded representation are not two different measurements.

Ask whether the output was a counted integer, a measured quantity or a mathematical expression before deciding how to present it.

18. Rounding people and vehicles requires context

A group of 52 people needs transport in vehicles that each safely hold 12 passengers. Four vehicles provide 48 places, which is not enough, so at least five vehicles are needed, assuming all 52 travel under those capacity conditions.

The arithmetic 52 ÷ 12 ≈ 4.33 does not permit renting 4.33 whole vehicles. A normal rule of rounding to the nearest integer would also give an insufficient four.

This is a context-dependent decision: countable resources with minimum capacity may need rounding up, not the nearest-value rule used in other situations.

19. Check the sign of a result

A calculated negative temperature may be meaningful on a Celsius scale, but a negative number of tickets sold is impossible in an ordinary counting problem. Students should not reject every negative answer automatically; they must interpret the quantity.

Likewise, a negative rate of change can represent a decrease, while a negative physical distance travelled may not match the intended definition. Context determines which signs are legitimate.

A simple question—’What does the negative sign represent here?’—can expose a model error or confirm that the result is reasonable.

20. Check the direction of the change

If a question describes growth, does the calculated amount actually increase? If it describes a discount, does the applicable price component decrease? If two quantities are inversely related under stated conditions, does the result reflect that relationship?

These checks are not universal physical laws for every situation. A real bill can rise despite a discount when other charges are added. Students should read all conditions before applying a simple expectation.

The value of the check is prompting a second look at the question and model.

21. Calculator display errors are not always mathematical errors

A student may press a wrong decimal point, forget brackets or misread scientific notation. The calculator can execute the entered expression accurately while the expression itself is wrong. Estimation provides a way to question that output.

For example, 18% of S$246 should be below 20% of S$246, or S$49.20. A display showing S$442.80 is therefore inconsistent with the expected size.

The remedy is to inspect the expression entered, not simply press the same buttons again without thinking.

22. Scientific notation needs scale awareness

For very large or very small quantities, scientific notation keeps numbers manageable. A student should understand that multiplying by 10³ changes scale by a factor of 1,000 and that 10⁻³ represents one-thousandth.

If a measured quantity of 0.004 is written as 4 × 10⁻³, its magnitude should remain the same. Writing 4 × 10³ instead would change it by a factor of one million relative to the intended quantity.

The exact scientific notation demands differ by subject and level, so focus on the present syllabus while building a sense of powers of ten.

23. Word problems need a representation before estimation becomes meaningful

Estimation is strongest when the learner knows what relationship to expect. If a ratio question is misread as a percentage question, a plausible-looking estimate may still be built on the wrong denominator.

First identify what each number represents and how the quantities relate. Then estimate the expected size, solve accurately and check the units and conditions.

The companion Read, Represent, Solve and Check Word Problems explains that earlier modelling stage.

24. Graph scales can make small changes appear large

Graphs may use truncated axes, different intervals or non-uniform scales. A rising line can look dramatic even when the numerical change is modest relative to the measured quantity. Students should read the title, axes, units and exact values before interpreting the shape.

Suppose values increase from 98 to 102. The absolute increase is 4, and relative to 98 it is about 4.1%. A graph starting at 97 may emphasise the visual difference; a graph starting at zero may make it appear smaller.

Both presentations can be legitimate for different purposes, but the student should describe the actual data rather than rely on visual impression alone.

25. A school Science graph needs accurate unit labels

A graph of temperature versus time should distinguish degrees Celsius on the vertical axis from time in seconds or minutes on the horizontal axis. An unlabelled axis cannot support a precise rate or description.

If the temperature rises from 15°C to 27°C over 20 seconds in an illustrative data set, the average rate of change over the interval is 12 ÷ 20 = 0.6°C per second. This does not prove the change was steady at every moment.

The units and time interval make the mathematical interpretation scientifically meaningful.

26. Precision should not exceed the quality of the evidence

If a measurement instrument records values only to the nearest whole unit, reporting a derived result with many decimal places may imply more precision than the original evidence justifies. Students should consider instrument resolution and subject instructions.

This does not mean every answer must be rounded aggressively. Exact mathematical calculations and measured experimental data have different considerations. Follow the relevant method and reporting requirement.

A useful Science habit is to state what was measured, how and to what appropriate precision.

27. A simple three-column unit table

Quantity	Numerical value and unit	What the number tells us
Distance	120 km	Length of journey
Time	2 h	Duration of journey
Average speed	60 km/h	Distance per unit time
Rectangle area	84 cm²	Surface coverage
Block volume	24 cm³	Three-dimensional space
Density	3 g/cm³	Mass per unit volume

The table illustrates different physical meanings. A student should not combine quantities simply because their numbers look convenient. The relationship and units must permit the operation.

28. Errors should be diagnosed by type

Wrong quantity: the student found the discount rather than the final price.
Wrong reference amount: percentage calculated using the new price instead of the original.
Wrong unit: a speed labelled metres rather than metres per second.
Wrong scale: a decimal point or power of ten placed incorrectly.
Wrong context interpretation: a fractional count of whole vehicles.
Wrong scientific inference: claiming causation from an observed pattern alone.

Each error requires a different correction. A blanket instruction to ‘check your work’ gives too little guidance unless the child knows what to check.

29. How to teach a learner to self-correct

Ask the child to estimate before using a calculator, then solve while recording units. After obtaining the answer, ask them to read the question’s final line again and name the requested quantity.

If the answer looks suspicious, compare it with the estimate or substitute it into the original relation. Then determine whether the problem came from the model, operation, unit or interpretation.

The goal is a repeatable checking habit the student can perform in school, not a parent correcting every answer privately.

30. Using a checklist without losing time

A short checklist is useful: quantity, approximate size, unit and context. It should not become a lengthy ritual after every simple arithmetic calculation. Students can develop speed by practising the checks with increasing fluency.

In a time-limited assessment, the actual paper structure determines how checking time is allocated. A high-risk multi-step question may deserve more attention than a basic calculation already verified.

The SEC Paper Time Management guide provides the broader examination context.

31. G1, G2 and G3 use appropriately different applications

A G1 learner may be building reliable numeracy through costs, rates or measurement. G2 students can encounter more complex applications and algebraic relationships. G3 students may need more abstraction and integrated multi-step reasoning. The exact content must be determined by the registered syllabus.

A student’s G-level does not affect whether a centimetre is a length unit or whether 20% of an amount has a particular meaning. Accuracy in foundational quantities supports all levels.

Students should be challenged appropriately but not be judged on another level’s untaught topic.

32. Parents can use everyday questions without setting a mock exam

A supermarket receipt, journey timetable or recipe can provide a simple discussion about quantities and units. The parent might ask what a discount applies to or whether a serving size is described in grams or millilitres.

These conversations should remain accurate and suitable for the child. They are not a substitute for a school’s taught content, and they should not turn every shopping trip into a compulsory worksheet.

The objective is to make mathematical interpretation familiar in ordinary life.

33. Tutors should check reasoning before allowing calculator dependence

A calculator is a tool for computation when permitted; it cannot choose the correct denominator or identify the physical quantity. A good teaching session asks the learner to state an approximate result, explain the model and then calculate.

If the result is implausible, return to the earliest mistaken decision. A tutor should teach why the quantity is wrong rather than simply provide the corrected answer.

Over time the learner should detect such mistakes independently and use checking efficiently.

34. Measurement and practical work must follow school safety

The fact that an article uses examples involving density, volume or electricity does not authorise students to conduct unsupervised laboratory procedures. Schools supply instructions, apparatus and supervision for their practical learning.

A Science calculation can be practised safely on paper using teacher-approved data. If the student missed an assessed practical or coursework component, ask the school about the formal arrangements rather than fabricate measurements.

The 2027 SEC Science Practical guide explains differences between school learning and national components.

35. Subject-specific marking requirements still matter

A correct answer with units may still need valid mathematical working or scientific explanation according to the task. Conversely, some tasks may specify that units need not be repeated in a particular answer field. The official question’s instructions determine the expected form.

Do not invent method-mark formulas or universal rounding rules. For the 2027 SEC, the registered syllabus and official assessment guidance should be used.

The SEC Show Your Working guide covers the distinction between steps, marks and unsupported assumptions.

36. Career relevance: units protect real decisions

Accurate units and reasonableness matter in engineering drawings, construction measurements, laboratory data, logistics, budgeting and many other activities. An incorrect scale or confused unit can affect a real decision even when the calculator arithmetic itself was executed correctly.

Secondary school builds the foundational habits for such work. It does not automatically certify a profession or create admission eligibility. The learner should continue exploring suitable academic and skills pathways through current official requirements.

The useful early achievement is knowing what a number represents and whether it can plausibly answer a question.

37. Four sample questions to test the habit

Question	Answer	Key check
15% off S$80; final price?	S$68	Lower than original price
7 cm by 12 cm rectangle; area?	84 cm²	Square unit and multiplication
120 km in 2 h; average speed?	60 km/h	Distance divided by time
52 passengers; 12 places per vehicle; minimum vehicles?	5 whole vehicles	Four vehicles are insufficient

These short illustrations check basic interpretation. The student should also explain the relevant method and practise new questions. Memorising the four numerical answers is not the skill.

38. A five-step revision routine

Choose one relevant question from the student’s actual G-level subject.
Before calculating, write an estimate or reasonable bound.
Label every quantity and ensure any conversions are valid.
Solve with clear working and appropriate precision.
Check the requested quantity, units and contextual restrictions before submitting.

If a particular check reveals repeated difficulty, teach that concept deliberately. The child should become more capable of detecting and repairing the same class of error without adult prompting.

39. The 2027 SEC uses subject-specific requirements

SEAB’s 2027 school-candidate directories list G1 Mathematics K110, G2 Mathematics K210 and G3 Mathematics K310, alongside separate Science syllabuses and subject combinations. Students must prepare for their actual registered level and assessment tasks.

The examination may involve written responses, data, diagrams, practical skills or other prescribed components. Estimation and unit awareness help, but they do not substitute for learning the required concepts.

Use the official SEC subject-code guide to avoid choosing an outdated or mismatched paper.

Frequently asked parent questions

Should students estimate even when calculators are allowed?

A quick size estimate can help detect entry errors, but the exact or appropriately rounded result must still follow the question’s instructions.

Does G3 require completely different units from G1?

Basic quantities retain their meanings. The content and complexity of assessed applications differ by subject level.

Is a wrong unit only a careless error?

Not necessarily. It can reflect misunderstanding of the quantity, conversion or formula. Diagnose the actual cause.

Can estimation prove a result is correct?

No. A wrong result may still look plausible. Use accurate working, appropriate checks and the original conditions.

Should every result be rounded to two decimal places?

No. The task, measurement context and official instructions determine appropriate precision.

Why is a bus answer rounded up when normal rounding gives four?

Because the problem requires enough whole vehicles for all passengers. Context matters more than a generic nearest-integer rule.

Does 25% increase followed by 25% decrease return to the original?

No. Percentages apply to different bases. A rise from 80 to 100 is 25%, while a fall from 100 to 80 is 20%.

How do I help a child who repeats decimal errors?

Build place-value understanding, estimate the magnitude and check the expression entered into the calculator before increasing task difficulty.

Can a Science graph establish the cause of a trend?

Not necessarily. Patterns are evidence of association or change, while causal explanation requires appropriate scientific context and controls.

Continue this learning-skills series

Before-Class Preparation and Ten-Minute Preview

Related eduKate learning guides

Official sources

An answer has meaning only when its quantity makes sense

The number on a calculator is the beginning of a final check, not the entire explanation. Students should know what was measured, why the chosen operation applies and whether the result fits the story and units.

Across PG1, PG2 and PG3, this habit supports accurate Mathematics, careful Science reasoning and sensible everyday decisions. Estimate first, calculate deliberately, interpret the unit and verify the answer. That is how students become more independent problem-solvers.

PG1, PG2 and PG3 Estimation and Units 2027 | Check Answer Reasonableness in Maths and Science