PSLE-SCI-REALITY-0092
Wait, What? A calculator can create digits that the experiment never measured.
Three lengths are measured with a ruler and recorded as 12.3 cm, 12.4 cm and 12.3 cm.
A calculator gives the average as 12.333333… cm. A neat infographic then reports: “Average length = 12.333 cm.”
The arithmetic is understandable. But did the ruler suddenly become able to measure thousandths of a centimetre because the numbers passed through a calculator?
No. The calculation may generate more decimal digits, but those digits do not automatically represent new measurement information.
Reality Lab Vol No.092 teaches a durable transfer habit: when a scientific chart, app, product comparison or report shows many decimal places, trace the number back to the original measurement before treating every displayed digit as evidence.
Quick Answer
- Find the original measurements, not just the final calculated number.
- Identify the measuring instrument and the smallest meaningful change it can distinguish.
- Separate measured digits from digits produced by arithmetic, unit conversion or software formatting.
- Check whether the display shows more detail than the measurement process supports.
- Do not confuse more decimal places with greater accuracy.
- Report a result at a level of detail that matches the evidence and purpose.
- Keep uncertainty and method limits visible when small differences matter.
What This Page Owns — and What It Leaves With the Main PSLE Science Guides
This page does not replace the main lessons on accuracy, precision, range, resolution, units or rounding. It applies those skills to one real-world evidence-transfer problem: a communication object whose polished decimal display looks more exact than the underlying measurements justify.
- How to Tell Measurement Precision From Accuracy in PSLE Science Without Assuming Repeated Agreement Means Correct
- How to Choose a Measuring Instrument for PSLE Science That Has the Right Range and Resolution
- How to Read Units, Scales and Measurement Resolution Before Using PSLE Science Data
Original Reality Lab Case: The Three Seedlings
This teaching case is original. It does not reproduce an examination or assessment-book question.
A learner measures three seedling heights using a ruler whose markings make tenths of a centimetre the sensible recorded level.
| Seedling | Recorded height |
|---|---|
| A | 12.3 cm |
| B | 12.4 cm |
| C | 12.3 cm |
The numerical average is 12.333333… cm. That long decimal expansion is a property of the calculation. It is not a record of what the ruler directly distinguished.
A scientifically sensible report might keep the average at a level consistent with the original data and the purpose of the comparison, while preserving the individual measurements when their variation matters.
Observed, Claimed and Inferred
| Layer | Statement |
|---|---|
| Observed | The raw heights were recorded as 12.3 cm, 12.4 cm and 12.3 cm. |
| Calculated | The arithmetic mean has a decimal expansion beginning 12.333333… |
| Claimed | The experiment measured the average height to the nearest 0.001 cm. |
| Inferred | Digits generated by arithmetic are being treated as though they came from finer measurement. |
Measurement Detail and Calculation Detail Are Different
A measurement begins with an interaction between a real object and a measuring system. The instrument has a scale, sensor response, resolution and other limits. A calculation begins with numbers that already exist.
Arithmetic can reorganise information. It can combine repeated measurements, convert units or calculate a ratio. It cannot automatically add physical detail that was never captured.
The Digit Provenance Check
For every suspiciously detailed number, ask where each layer came from:
- Physical quantity: what was actually being measured?
- Instrument: what scale, resolution or sensitivity did it have?
- Raw record: how many digits were meaningfully recorded?
- Calculation: what arithmetic produced the final value?
- Display: did software automatically show a fixed number of decimal places?
- Claim: is the apparent detail being used to imply a difference the measurements can really support?
Why Digital Does Not Mean Infinitely Precise
A digital instrument can show several digits while still having limits on what changes it can reliably distinguish. Display formatting and measurement capability are not the same thing.
This is especially important because digital numbers look authoritative. “12.347” can feel more scientific than “12.3”. Yet scientific strength comes from how the number was measured and checked, not from how many characters appear after the decimal point.
Worked Case 1: Unit Conversion Creates More Digits
A length recorded as 2.4 m is converted by software to 240.000 cm. The conversion has not made the original measurement more precise. The zeros were introduced by formatting, not by a new observation.
Worked Case 2: The App Average
A temperature app receives readings of 24.1°C, 24.2°C and 24.1°C, then displays “mean = 24.133333°C”. The app has performed arithmetic correctly. But reporting six decimal places could falsely suggest the experiment resolved millionths of a degree.
Worked Case 3: Two Products Differ by 0.003 Units
A comparison table lists Product A at 8.421 units and Product B at 8.424 units. The difference is 0.003. Before calling B better, inspect how the original measurements were obtained. If the method can meaningfully distinguish only changes of about 0.1 unit, the three-thousandths difference is not made trustworthy merely by displaying it.
Worked Case 4: Extra Digits Can Still Be Useful Internally
Sometimes software keeps extra digits during intermediate calculations to avoid unnecessary rounding at each step. That can be mathematically useful. The important distinction is between carrying computational detail internally and publicly claiming that every digit is independently supported by the measurement.
Worked Case 5: The Instrument Itself Displays 12.347
Suppose a digital instrument directly shows 12.347. Are all those digits now proven? Not automatically. The display tells you what the instrument reports. To understand measurement quality, you still need the instrument’s resolution, calibration or reference information, operating conditions and suitability for the task.
The lesson is not “ignore decimal places”. The lesson is “connect decimal places to measurement capability”.
Precision Is Not the Same as Accuracy
A number can be reported with many digits and still be wrong because of bias, poor calibration or an unsuitable method. Conversely, a well-designed measurement may be accurate enough for its scientific job without displaying many decimal places.
That distinction belongs to the main precision-versus-accuracy guide. In Reality Lab, it protects learners from a communication trick: visual numerical detail is not proof of evidential quality.
What Evidence Would Strengthen a Highly Detailed Number?
- Raw measurements recorded at a compatible level of detail.
- An instrument whose resolution is suitable for the claimed difference.
- Reference or calibration evidence appropriate to the task.
- Repeated measurements showing the fine differences are stable enough to matter.
- A transparent calculation that does not hide coarse source data.
- Uncertainty information when the final comparison depends on small numerical differences.
What Would Weaken It?
- A ruler reading to tenths followed by a reported average to thousandths.
- Software that automatically prints six decimal places for every number.
- A unit conversion presented as though it created new measurement detail.
- A tiny product difference smaller than the measurement system can meaningfully distinguish.
- No raw data or method description showing where the final digits came from.
Tempting Reasoning That Fails
- “The calculator gave the digits, so they must be scientific.” Arithmetic can produce digits that are not new observations.
- “Digital means more accurate.” Digital is a display form, not a guarantee of accuracy.
- “More decimal places always improve a result.” Unsupported detail can make a result look more certain than the evidence.
- “We should throw away all digits after the decimal point.” No. The appropriate detail depends on the measurement process and scientific job.
How Far Can the Conclusion Travel?
If the source measurements are coarse, a highly detailed average can still be used as an intermediate calculation. But the public scientific claim should not pretend that the experiment directly resolved every displayed digit.
If a conclusion depends on a very small difference, the burden of evidence rises: the measurement system must be capable of supporting that difference.
PSLE-Style Transfer Case
A learner measures the mass of three similar objects as 18.2 g, 18.3 g and 18.2 g. A calculator gives an average of 18.233333 g.
Question: Why should the learner be careful about reporting 18.233333 g as though every digit were measured?
Reasoned answer: The original measurements were recorded only to 0.1 g. The extra digits came from averaging the recorded values, not from a balance that measured to millionths of a gram. The result should be reported at a level of detail supported by the measuring method.
Explained Practice
Practice A: A sensor records 7.2 units. A website converts it to another unit and prints 19.43658. Did the conversion improve the original measurement resolution? No.
Practice B: A scale directly displays 3.417 g. What should you inspect before trusting a difference between 3.417 g and 3.418 g? The instrument’s actual resolution, repeatability, calibration/reference evidence and suitability.
Practice C: An average is calculated from ten measurements. Does taking more measurements automatically justify unlimited extra decimal places? No. More evidence can improve a summary, but it does not create unlimited instrument resolution.
Delayed Independent Return: The D-I-G-I-T Check
- D — Data: What were the raw recorded values?
- I — Instrument: What could it distinguish?
- G — Generated: Which digits came from calculation or formatting?
- I — Importance: Does the conclusion depend on those extra digits?
- T — Tell only what the evidence supports: report detail honestly.
Parent and Tutor Teaching Guide
Give the learner three measurements written to one decimal place and let a calculator produce a long average. Ask: “Which digits came from the ruler, and which came from the calculation?”
Then show two product scores separated only in the fourth decimal place. Ask what evidence would be needed before treating that tiny difference as real. The goal is not to teach formal significant-figure rules as a magic template. It is to teach provenance: every digit should have an evidential reason to be there.
Authoritative Sources
- Singapore Examinations and Assessment Board — 2026 PSLE Science Syllabus
- Ministry of Education Singapore — Primary Science Teaching and Learning Syllabus
- NIST/SEMATECH Engineering Statistics Handbook — Measurement Resolution
NIST’s measurement guidance explicitly separates display digits from true resolution: the number of digits shown by an instrument does not by itself establish the instrument’s resolving capability. That is exactly the real-world habit this page translates into Primary 5/6 reasoning.
The Quiet Return
Calculation can create digits.
Only measurement can create measurement evidence.
When a number looks impressively precise, ask where its last digit came from.
