eduKate Singapore

VIEW THIS AS

Auto mode follows the Route Engine until you choose a viewpoint.

YOU ARE HERE

ROUTE CHECK

CONNECTED TO

WHAT NEXT

Use the canonical route for this room, or HELP if you are unsure.

Next useful routeCanonical hubHELP

PG1, PG2 and PG3 SEC Numerical Accuracy | 2027 Significant Figures, Units, Rounding and Mathematical Evidence

—

by

eduKate Asia
Three learners review open books together at a classroom table, with stacks of textbooks, stationery and a whiteboard in the bright room.

How many significant figures should a PG1, PG2 or PG3 student use in the 2027 SEC? The answer depends on the registered subject, not the original Posting Group. For 2027 G1 Mathematics K110, G2 Mathematics K210 and G3 Mathematics K310, SEAB instructs candidates to give non-exact numerical answers to three significant figures, or degree-angle answers to one decimal place, unless the question specifies otherwise. It also expects essential working and correct treatment of units.

The helpful way to teach accuracy is to ask five questions: What quantity is being found? Which units belong to it? Which mathematical model applies? When should rounding occur? What answer form is requested? A student who understands these questions can check a calculator result instead of merely copying all its digits.

This guide builds those habits with original worked examples in Mathematics, Science, statistics and everyday life. It is optimistic for every learner and careful about a crucial distinction: the default precision written in the Mathematics syllabuses should not be invented as a universal rule for every separate Science paper.

Independent eduKateSG education guide, checked against 2027 SEAB K110, K210 and K310 syllabus assessment notes and the official SEC G-level subject catalogues in October 2026. Always check the candidate’s actual paper and question instructions.

The quick accuracy checklist

  • 2027 Maths G1 K110, G2 K210, G3 K310: usually three significant figures for non-exact numerical results.
  • Angles in degrees: usually one decimal place for non-exact results.
  • Question instruction overrides the default: exact form, nearest cent, whole number or stated precision.
  • Keep working more precise: especially when a question asks to show a rounded result.
  • Units: centimetres for length, cm² for area, cm³ for volume, m/s for speed and m/s² for acceleration.
  • Postings: PG1/PG2/PG3 are admission labels, not separate rounding systems.

Read the accuracy question that applies

1. Why precision matters more than extra decimals

A mathematically correct method can lead to a misleading answer if the candidate ignores the required units, rounding convention or quantity being measured. The 2027 SEC Mathematics syllabuses K110, K210 and K310 contain explicit instructions about numerical accuracy. Precision is a small, learnable habit that supports reliable mathematics and science.

Worked example: A student calculates an area correctly but reports the length unit cm instead of cm².

What to practise: Check the meaning and units before formatting the final number.

Common trap: A long calculator display is not automatically more correct than a carefully rounded result.

2. Posting Group is not a precision rule

PG1, PG2 and PG3 describe Secondary 1 entry categories. The examination rules depend on the actual subject, its G-level and the precise paper. Under Full Subject-Based Banding, a child may take approved subjects at different demand levels. Numerical communication remains useful in any of them.

Worked example: A PG2 entrant has Mathematics at G2 but studies another eligible subject at a different level.

What to practise: Identify the K-code on the school’s current subject list.

Common trap: The admission group does not determine a permanent mathematical ability ceiling.

3. The three SEC Mathematics codes

The 2027 SEAB catalogue lists Mathematics G1 K110, G2 K210 and G3 K310. Each syllabus includes a paper-specific scheme and a set of notes about mathematical communication. The three documents agree on important numerical-answer conventions while differing in course scope and paper structure.

Worked example: A learner preparing for G2 K210 finds a worked example labelled K310 online.

What to practise: Use K210 for the relevant current formal requirements.

Common trap: A shared accuracy habit does not mean every G-level examination has identical questions.

4. The actual 2027 Mathematics rounding rule

K110, K210 and K310 direct candidates to give non-exact numerical answers to three significant figures, or to one decimal place for angles in degrees, unless the question specifies a different level of accuracy. Exact answers and special instructions must be treated appropriately. This distinction protects students from overgeneralising.

Worked example: A calculation gives 2.347891, and a Mathematics question asks for the usual numerical answer.

What to practise: Report 2.35 to three significant figures, keeping accurate working before that final step.

Common trap: Do not impose this Mathematics convention blindly on all Science syllabuses or on a question that requests another format.

5. A significant figure begins at the first non-zero digit

Significant figures count from the first non-zero digit. Leading zeroes locate the decimal point but do not count as significant digits. This becomes important when interpreting small measurements or financial quantities written in decimal form.

Worked example: 0.004568 contains four significant figures: 4, 5, 6 and 8.

What to practise: To three significant figures, write 0.00457, because the next digit requires rounding.

Common trap: The zeroes before 4 are not counted as meaningful measured digits.

6. Trailing zeroes need context

A zero following a non-zero digit can be significant, particularly when written after a decimal point or in a clear scientific notation. A result like 110 may be ambiguous about how many significant figures are intended.

Worked example: An answer is 109.97 and needs to be shown clearly to three significant figures.

What to practise: Write 1.10 × 10², making the required three significant figures explicit.

Common trap: Do not confuse writing more digits with adding accuracy to the underlying measurement.

7. Decimal places count positions after the decimal point

Decimal places and significant figures are different systems. One decimal place means one digit after the decimal point, regardless of how large or small the number is. Three significant figures count meaningful digits from the first non-zero value.

Worked example: A value of 0.04756 to two decimal places is 0.05, while to two significant figures it is 0.048.

What to practise: State which instruction is being used before rounding.

Common trap: The two results are not contradictory; they answer different accuracy requests.

8. Rounding should occur at the requested point

Intermediate rounding can change later answers, particularly in multi-step problems involving percentages, rates, geometry or financial calculations. Keeping exact fractions or sufficient calculator digits until the final requested step is a useful way to preserve correctness.

Worked example: A student rounds a unit price at the beginning and multiplies it across hundreds of units.

What to practise: Keep the exact value in working and round the final requested amount.

Common trap: Premature rounding can produce a different answer even when every subsequent operation is performed correctly.

9. Why 3.3075 becomes 3.31

When a non-exact Mathematics result is to be given to three significant figures, inspect the fourth significant digit and round accordingly. The rule is mechanical, but students should still understand the calculation and the quantity represented.

Worked example: 2.45 multiplied by 1.35 is 3.3075.

What to practise: To three significant figures, 3.3075 becomes 3.31; check that the unit is preserved if one applies.

Common trap: A correct rounded number without the correct formula or required working may not complete a multistep question.

10. Angles in degrees have a specific convention

The 2027 SEC Mathematics syllabus notes specify one decimal place for non-exact angles in degrees unless another accuracy level is requested. This is a degree-angle instruction, not a rule that all trigonometric or radian results are one decimal place.

Worked example: A calculator gives an angle of 27.56 degrees in a relevant Mathematics problem.

What to practise: State 27.6° under the usual one-decimal-degree convention.

Common trap: A question asking for an exact value, different accuracy or another angle unit must be followed as written.

11. Angle mode is a separate issue from rounding

Calculators may use degree or radian settings. Correctly rounding the output cannot fix a trigonometric calculation performed in the wrong angular mode. The learner should first decide the intended angle unit and verify the calculator setting.

Worked example: A candidate enters sin 30 expecting thirty degrees while the calculator is in radian mode.

What to practise: Confirm the intended unit before using the trigonometric function.

Common trap: A long decimal output can look trustworthy even when it answers a different input.

12. Check the question before using the default

A syllabus-wide default does not override a direct instruction such as nearest whole number, two decimal places, exact form or give the answer in terms of π. Good examcraft starts by reading the specific request before deciding how to format the response.

Worked example: A ticket question requests the minimum number of whole buses required.

What to practise: Interpret the whole-number requirement from the situation rather than returning a decimal number of buses.

Common trap: Rounding 5.2 buses to 5 by the ordinary nearest-integer rule would be wrong if six are needed.

13. Exact fractions should stay exact when appropriate

Fractions and surds may represent exact values. Turning an exact expression into a rounded decimal too early can lose precision or ignore what the question asks. Mathematical answers should use the appropriate representation.

Worked example: A question asks for an exact value and the solution is 7/12.

What to practise: Keep 7/12 rather than giving a decimal rounded to three significant figures.

Common trap: The usual non-exact numerical-answer convention is not a command to decimalise every exact result.

14. π requires attention to the wording

The 2027 Mathematics syllabus notes provide that candidates should use the calculator value of π or π = 3.142 unless a question asks for the answer in terms of π. Those options apply to the relevant syllabus. A student should read which representation is requested.

Worked example: A circle area problem requests the exact answer in terms of π.

What to practise: Leave the expression in terms of π when that is the stated task.

Common trap: Replacing π with 3.14 when an exact π-form is requested can change the answer format.

15. A percentage has a reference whole

A percentage value cannot be interpreted correctly without knowing its base. A difference between 20% and 30% is ten percentage points, while the relative increase from 20 to 30 is 50%. These are distinct mathematical claims that may appear in data interpretation.

Worked example: A school programme’s participation grows from 20% to 30%.

What to practise: Describe the change as 10 percentage points or a 50% relative rise, as the question requires.

Common trap: Do not label every difference between percentages simply as the same kind of percent increase.

16. Money is a practical unit with context

Financial questions may involve amounts stated to cents, installment schedules, exchange rates or other conventions. The expected answer can depend on the question’s specific currency and rounding instruction. Do not force three significant figures onto a money value where that would lose meaningful cents.

Worked example: A cost is calculated as $12.345 and the problem asks for the amount payable to the nearest cent.

What to practise: Give $12.35, explaining the cent rounding if relevant.

Common trap: A generic significant-figure default must yield to a clear nearest-cent requirement.

17. Units belong to the physical quantity

Units describe the kind of measurement: metres for length, square metres for area, cubic metres for volume, seconds for time and so forth. Students should identify the quantity requested before checking the answer’s unit.

Worked example: A rectangle’s area is computed as 24 but written as 24 cm.

What to practise: Correct the unit to cm² when side lengths were measured in centimetres.

Common trap: An accurate number is not a complete measurement if its dimension is misrepresented.

18. SI units are useful reference units

The 2027 Mathematics syllabus notes say SI units will be used in questions involving mass and measures. Science students should also be comfortable identifying the units specified in their own subject materials. Conversions should be stated explicitly where they matter.

Worked example: A problem combines 2.5 metres and 40 centimetres.

What to practise: Convert the quantities to consistent units before adding or comparing.

Common trap: The symbols cm and m differ by a factor of one hundred, not by a cosmetic label.

19. Centimetres to metres requires a scale factor

A length conversion changes the numerical representation while preserving the physical length. One metre contains one hundred centimetres, so 350 cm represents 3.5 m. Keeping units visible makes the operation easier to audit.

Worked example: A student uses 350 cm as 350 m in an area calculation.

What to practise: Write 350 cm = 3.5 m and then use the appropriate dimensions.

Common trap: Changing the unit without changing the numerical value makes the physical result incorrect.

20. Area conversions square the linear factor

Area is measured in square units, so converting centimetres squared to metres squared requires a factor of ten thousand, not one hundred. Understanding the geometry prevents a frequent mechanical error.

Worked example: A tabletop has area 20,000 cm².

What to practise: Convert it to 2 m² because 1 m² is 10,000 cm².

Common trap: Dividing by only one hundred would confuse a two-dimensional quantity with a length.

21. Volume conversions cube the scale factor

Volume uses cubic units. One metre cubed is one million cubic centimetres, since the hundred-fold linear conversion applies across three dimensions. A unit conversion is a relationship, not merely moving a decimal by habit.

Worked example: A container volume is 500,000 cm³.

What to practise: Express it as 0.5 m³ and consider whether litres are the more useful unit in context.

Common trap: A conversion appropriate for area should not be reused for volume.

22. Rates combine different quantities

A rate such as metres per second or dollars per item expresses one quantity relative to another. The expression should be read aloud and interpreted before calculating. Rates are common in Science, transport and financial Mathematics.

Worked example: A student divides time by distance when the question asks for average speed.

What to practise: Write speed = distance / time and check that the unit becomes distance per unit time.

Common trap: A number with an inverted unit often reveals an inverted formula.

23. Speed conversions need time and length

Kilometres per hour and metres per second differ in both length and time units. To convert 80 km/h into m/s, multiply by 1000 and divide by 3600, giving approximately 22.2 m/s to three significant figures.

Worked example: A candidate converts only kilometres to metres and obtains a speed thousands of metres per second.

What to practise: Write 80 × 1000 / 3600 and then round under the relevant question rule.

Common trap: The conversion factor is not simply 1000 because an hour also differs from a second.

24. Acceleration uses seconds squared

Speed and acceleration are different quantities. Acceleration involves the rate of change of velocity with time, so common SI units are metres per second squared. Identifying the correct quantity can prevent confusion between graph height and gradient.

Worked example: A candidate gives an acceleration as 3 m/s after interpreting a velocity–time graph.

What to practise: Check whether the result represents speed or its change per second.

Common trap: A familiar number from a graph does not automatically identify the correct physical quantity.

25. Mass and weight need distinct units

Mass is usually given in kilograms or grams while weight is a force, commonly measured in newtons. Students should distinguish physical definitions when interpreting Science questions and avoid swapping the units.

Worked example: A learner writes that a 2 kg mass has a weight of 2 kg under a question asking for force.

What to practise: Name the quantity and use the appropriate relation and unit.

Common trap: Everyday use of the word weight may be informal, but an examination can require the scientific distinction.

26. Density relates mass to volume

Density has compound units such as g/cm³ or kg/m³. To interpret or compare densities, the mass and volume scales must both be correct. Students should use the given units or convert explicitly.

Worked example: A material has mass 200 g and volume 50 cm³.

What to practise: Calculate density = 4 g/cm³ with the correct compound unit.

Common trap: A density value in grams alone does not represent density.

27. Scientific notation expresses scale clearly

Standard form expresses a number as A × 10^n, usually with 1 ≤ |A| < 10 for a non-zero value. It is useful for very large or small quantities and can clarify significant digits.

Worked example: The quantity 0.0000567 is awkward to compare in a long table.

What to practise: Write it as 5.67 × 10⁻⁵ to highlight the scale and its three significant figures.

Common trap: Changing the power of ten without adjusting the coefficient changes the quantity.

28. Zero is not a rounding instruction

Zero can appear as a measured or calculated value with particular meaning. Zero used as a placeholder differs from a zero reported to a certain precision. Students should not strip zeroes automatically where they communicate the requested decimal places.

Worked example: A result is 2.50 m to three significant figures.

What to practise: Keep the trailing zero when it communicates three significant figures.

Common trap: Writing 2.5 m may be numerically equal but does not visibly preserve the stated precision.

29. Round after a ratio is applied

Ratio problems may involve division into parts or scale conversion. Rounding the value of one part prematurely can make the total fail to reconcile with the original quantity.

Worked example: A budget is divided in a ratio and intermediate values include recurring decimals.

What to practise: Keep exact fractional shares until the necessary final allocation.

Common trap: If a context requires whole items, address the practical remainder rather than quietly discarding it.

30. Reverse percentage needs the correct base

If a price was increased by 15%, the original amount is the new amount divided by 1.15. Rounding the divisor or intermediate amount can distort the result. Identifying the base prevents a much larger conceptual error.

Worked example: A final amount of 120 includes a 15% increase.

What to practise: Compute the original as 120/1.15 before final requested rounding.

Common trap: Subtracting 15% of the new price is not the inverse of a 15% increase on the old price.

31. Round totals, not each item unless instructed

A question may ask for the combined value of several measurements or prices. Rounding each item first can produce a different total from summing accurately and rounding at the end.

Worked example: Three estimated costs are combined and each has an additional decimal digit.

What to practise: Add the unrounded amounts before the final reporting step, unless the problem specifies item-by-item rounding.

Common trap: A difference in cents may come from the stage of rounding rather than the addition method.

32. Graph scales carry numerical meaning

A plotted point is read according to its axis labels, units and interval spacing. Good graph interpretation begins with the scale. A dramatic visual difference can represent a small actual change if the axis begins close to the data.

Worked example: A bar chart starts its vertical axis at 98 rather than zero, exaggerating a change from 99 to 100.

What to practise: Read actual labels and compare numerical differences rather than bar heights alone.

Common trap: A graph’s appearance is not a substitute for measurement.

33. Percentage points should be named

A percentage-point change is the arithmetic difference between two percentages; a relative percentage change compares that difference with a starting value. The distinction can affect Social Studies, Geography, Science and business explanations.

Worked example: A survey reports support rising from 60% to 66%.

What to practise: Say it increased by 6 percentage points or 10% relative to the earlier share, as appropriate.

Common trap: Calling the difference 6% without context can conceal which comparison is intended.

34. Averages require the correct denominator

Mean values depend on how many observations are included. An incorrect group size can create a plausible average with the wrong meaning. Before rounding an average, students should establish the count or total weight used.

Worked example: A class’s total score is divided by 29 when there are 30 valid submissions.

What to practise: Correct the denominator before applying any rounding convention.

Common trap: Better decimal presentation cannot repair a wrong sample count.

35. Weighted averages differ from simple means

When quantities contribute unequally, a weighted mean uses the stated weights. The candidate should understand what each weight represents before calculation and reporting.

Worked example: Two tests contribute 30% and 70% to a course score.

What to practise: Apply the correct percentages and then round to the requested format.

Common trap: Simply averaging the two marks would be incorrect when the weights differ.

36. Probability answers may remain exact

Many probability questions are conveniently expressed as fractions. Where the question requests a fraction or exact answer, rounding to a decimal may be unnecessary or misleading.

Worked example: A probability is exactly 2/7.

What to practise: Use the requested representation and, if a decimal is required, apply its stated accuracy.

Common trap: A decimal such as 0.286 is an approximation, not the exact probability.

37. Trigonometric ratios can be exact expressions

A trigonometric ratio may have an exact form, while a measurement problem may require a non-exact numerical answer. Students should not replace an exact symbolic result by an approximate calculator number when not requested.

Worked example: An algebraic expression simplifies to √3/2.

What to practise: Keep the exact form if the task requests exactness.

Common trap: A calculator decimal is a representation, not a more mathematically exact answer.

38. Show greater precision when proving a rounded answer

The 2027 Mathematics syllabus notes specify that when candidates must show an answer correct to a given accuracy, the result should first be shown to a higher degree of accuracy. This protects the logic of the demonstration.

Worked example: A question asks the learner to show a result rounds to 2.35 to three significant figures.

What to practise: Display an intermediate value such as 2.34789 and then round to 2.35.

Common trap: Writing 2.35 = 2.35 alone does not show that the original calculation justified the rounding.

39. Use a calculator as a checking tool

An approved calculator may be used in both papers of 2027 SEC Mathematics K110, K210 and K310, according to the individual syllabuses. A calculator can improve numerical reliability but it cannot choose the model, unit or interpretation.

Worked example: A student enters the wrong time conversion yet receives a highly precise display.

What to practise: Estimate the order of magnitude before accepting the output.

Common trap: Calculator permission does not imply that raw output alone earns every method or reasoning mark.

40. A permitted calculator is not any device

The examining authority’s approved calculator rules determine which models may be used. A personal phone application, stored notes or external AI assistant is not automatically acceptable.

Worked example: A family practises on a phone and assumes the same app can be used during a national paper.

What to practise: Read SEAB’s approved-calculator guidance for the applicable examination.

Common trap: A successful home practice method may be inappropriate in a supervised exam environment.

41. Physics measurements need subject-specific accuracy

The G3 Physics K323 syllabus includes quantities, units and practical skills. Scientific reporting should follow the question, the instrument and the subject’s instructions; the Mathematics three-significant-figure convention should not simply be imposed without checking.

Worked example: A laboratory value is supplied to a defined instrument resolution.

What to practise: Keep track of the measurement conditions and use the subject’s appropriate representation.

Common trap: Apparent precision beyond the instrument or task can misrepresent what was observed.

42. Chemistry quantities need clear labels

Chemistry questions may involve mass, volume, concentration or temperature. A correct formula can still produce an invalid answer if units are mixed or the wrong substance’s quantity is used.

Worked example: A learner uses 25 cm³ as though it were 25 dm³ in a concentration calculation.

What to practise: Convert to the unit required by the actual relationship and name the substance involved.

Common trap: A numerical match without chemically meaningful units is not a robust explanation.

43. Biology rates are measurements too

Biology often uses rates of change, percentages and labelled graphs. Students should know what was measured, over what interval and using which units before comparing results.

Worked example: A photosynthesis graph records bubble counts per minute, while a student writes the value as bubbles.

What to practise: Keep the denominator of the rate visible and interpret it in the context of the experiment.

Common trap: A biology trend does not justify unlimited precision in a claim.

44. Environmental data needs comparability

Geography and environmental Science can compare rainfall, air quality, temperatures or populations. Units, time intervals and location matter. Converting measurements without preserving context can create misleading conclusions.

Worked example: One dataset gives rainfall per day and another gives rainfall per month.

What to practise: Make the reporting periods comparable before calculating a difference.

Common trap: Two figures with the same unit may still refer to different conditions.

45. Financial literacy benefits from precise assumptions

Practical financial Mathematics uses percentages, charges and time periods. Students should distinguish a stated simple-interest rate from compounding, and an estimated cost from an exact contractual amount.

Worked example: A quoted monthly rate is interpreted as an annual charge.

What to practise: Write the rate and time base together before computing.

Common trap: Rounding a financial result cannot repair the wrong interest model.

46. Temperature units require actual conversion rules

Celsius, Kelvin and Fahrenheit have different scales. Where a question requires conversion, the learner must use the appropriate relationship and not treat temperature like ordinary length multiplication.

Worked example: A student assumes doubling a Celsius temperature doubles thermal energy in every context.

What to practise: Follow the subject’s specified temperature model and recognise the role of absolute scale.

Common trap: A Celsius reading does not provide a universal proportional temperature ratio.

47. Sign conventions matter in physical formulas

Direction can be represented by positive and negative values. The meaning depends on the chosen reference and coordinate system. Students should state the convention when it affects interpretation.

Worked example: A vehicle moves in the opposite direction from the chosen positive axis and has negative displacement.

What to practise: Distinguish displacement from total distance and maintain the defined sign system.

Common trap: A negative answer is not necessarily a mistake; it may encode direction.

48. Standard form is useful for large numbers

Scientific notation lets students handle large quantities while keeping their scale obvious. It is especially useful in calculations with many zeros or measurements spanning several powers of ten.

Worked example: A value of 45,600,000 is compared with 4.56 × 10⁷.

What to practise: Recognise that they are numerically equal before interpreting precision.

Common trap: A misplaced exponent is a far larger error than a final-digit rounding slip.

49. Uncertainty belongs in scientific claims

Some measurements are estimates or have limits. Students should not describe a small difference as definitive if the underlying data or instruments cannot support that precision.

Worked example: Two experimental values differ by less than the stated measurement resolution.

What to practise: Report the observation honestly and discuss limitations when the subject requires them.

Common trap: Scientific honesty is more important than inventing additional decimal places.

50. Units can expose an incorrect formula

Dimensional reasoning provides a quick independent check. If a calculated speed has units of seconds per metre, the formula may have been inverted. Learners should use unit meaning as a test of their working.

Worked example: A student calculates time/distance instead of distance/time for speed.

What to practise: Compare the resulting unit with the quantity requested.

Common trap: Not every advanced formula can be proven by units alone, but obvious mismatches often reveal an error.

51. Answer boxes need the final requested form

A question may provide working space and a separate answer line. A learner should ensure the final entry includes the numerical value, correct unit and requested accuracy, as applicable.

Worked example: An entire solution is shown in working but the final box is left blank or uses an unrounded figure.

What to practise: Copy the final result with appropriate notation after completing the working.

Common trap: Readable working helps, but it should not leave the examiner guessing which number is the answer.

52. Read the difference between explain and calculate

A calculation question may require only a number and method, while an explanation question asks why the model applies or what the result means. Accurate presentation should follow the command word.

Worked example: A question asks whether a cost estimate is reasonable and a learner only writes $253.74.

What to practise: Relate the number to the situation and any relevant limit or assumption.

Common trap: A perfectly rounded number does not answer a question about interpretation.

53. Retest one error type at a time

When students lose marks from numerical accuracy, identify whether the cause was the formula, unit, conversion, rounding point, calculation or final presentation. Each can be practised separately.

Worked example: A learner repeatedly reports correct numbers with wrong area units.

What to practise: Use three short area questions focused on dimensional meaning, then mix them with new problem types.

Common trap: More random worksheets may hide the recurring cause instead of fixing it.

54. Keep an independent accuracy checklist

A short checklist is most effective when it names specific failure points the learner actually makes. It should be used after solving, not copied as a substitute for understanding.

Worked example: A student often forgets the requested number of decimal places after completing trigonometry.

What to practise: Check model, working, unit, precision and context before moving on.

Common trap: A checklist cannot repair an incorrect scientific explanation unless the learner learns the missing concept.

55. Short practice can build confidence

A family can support one small accuracy task per day or week without turning every meal or outing into an examination. The goal is thoughtful independent checking across different contexts.

Worked example: A child estimates a trip’s speed from a map and compares the result with an exact calculation.

What to practise: Invite the learner to explain why the units and rounding make sense.

Common trap: PG1, PG2 and PG3 are initial school pathways, not ceilings on numerical judgement.

56. A later examination year may revise instructions

The 2027 SEAB documents govern the 2027 SEC sitting. Children entering Secondary 1 in 2027 normally sit examinations in later years and should use the relevant examination-year syllabus.

Worked example: A parent saves a 2027 K310 guide for a younger student and never checks for later changes.

What to practise: Review the official syllabus when the candidate’s examination year approaches.

Common trap: A correct 2027 rule should not be advertised as an immutable requirement for every future cohort.

Numerical accuracy questions parents ask

What precision does 2027 SEC Mathematics normally require?

For non-exact numerical answers, K110, K210 and K310 specify three significant figures unless a question requests something different.

How should degree angles normally be reported?

One decimal place for non-exact angles in degrees unless the Mathematics question specifies otherwise.

Are the same rounding rules automatically used in Science?

No. Use the particular Science syllabus, question and appropriate measurement conventions.

What is three significant figures for 0.004568?

0.00457.

What is three significant figures for 3.3075?

3.31.

What is one decimal place for 27.56°?

27.6°.

Should 7/12 become a rounded decimal?

Not when an exact fractional answer is requested; follow the question.

What if the question asks for cents?

Use the specified financial rounding requirement, commonly to the nearest cent.

Why should intermediate results be kept more accurately?

Early rounding can affect subsequent calculations and produce a different final answer.

Does 1 m² equal 100 cm²?

No. 1 m² equals 10,000 cm².

What is 80 km/h in m/s?

Approximately 22.2 m/s to three significant figures.

What is the unit of acceleration?

Metres per second squared, commonly m/s².

What is the unit of density?

A mass divided by volume, for example g/cm³ or kg/m³.

Do G1, G2 and G3 Mathematics allow approved calculators?

Their 2027 syllabuses permit an approved calculator in both Paper 1 and Paper 2.

Is a phone calculator approved for exams?

Do not assume so. Follow the examining authority’s permitted-calculator rules.

Is 2.50 the same precision as 2.5?

They are numerically equal, but 2.50 visibly shows three significant figures.

Should all money amounts be given to three significant figures?

No. A clear nearest-cent or other financial instruction takes priority.

What if a question asks to show a rounded result?

Show a higher-precision intermediate value before the requested rounded answer.

Does good rounding compensate for an incorrect equation?

No. Precision is one part of solving and communicating the right mathematical model.

What is the best accuracy check?

Confirm quantity, unit, appropriate model, conversion, requested precision and real-world meaning.

Continue through the eduKateSG subject ecosystem

The linked Mathematics guides show the registered G-level syllabuses and practical skills. The related Science, rounding and mistake-diagnosis guides tackle different situations where students can improve their numerical communication.

Official MOE and SEAB references

The useful next step

Take one school problem and have the learner explain why the answer’s unit and precision make sense. Check whether the result fits the context. A child who can do that is learning a transferable skill, not simply memorising a rounding instruction.

Numerical accuracy is learnable in G1, G2 and G3. An initial admission category never limits the student’s long-term ability to reason carefully with quantities.