Wait, what? A laboratory report says a reference mass is 50.0 ± 0.5 g. A student immediately draws a box from 49.5 g to 50.5 g and says, “The real value must be inside this box. Anything outside is impossible.” The box feels scientific because the report used a plus-minus sign. But uncertainty is not the same thing as an absolute physical fence.
A measurement result is evidence about a quantity. It is never just a naked number. Good measurement reports also communicate how uncertain that result is and, where expanded uncertainty is used, what coverage interpretation and coverage factor belong to it. The learner’s job is to read the result and its uncertainty as one evidence object.
Quick Answer
Not as an absolute guarantee. If a report states 50.0 ± 0.5 g and defines 0.5 g as an expanded uncertainty, the interval from 49.5 g to 50.5 g expresses a high-confidence range under the stated measurement model, assumptions and coverage factor. In many common cases a coverage factor near 2 is used to give an interval with approximately 95% confidence, but the exact interpretation must come from the report.
The scientifically careful claim is: the reported measurement result is 50.0 g with an expanded uncertainty of 0.5 g under the stated conditions and coverage interpretation. It is too strong to say values outside the interval are physically impossible.
The Owned Learner Job
This article owns one evidence-transfer job: how to interpret a plus-minus expanded uncertainty without converting a confidence-style interval into a hard guarantee.
It does not replace the broader PSLE Science owners for measurement, repeated readings, accuracy, precision, fair tests or graph reading. It applies those habits to a real measurement-report object.
First Ask: What Does ± Mean Here?
The symbol ± can mean different things in different scientific contexts. It might represent expanded measurement uncertainty, one standard deviation, standard error, a manufacturer’s tolerance, a repeatability specification or another defined quantity. The symbol alone is not enough.
Before interpreting 50.0 ± 0.5, ask:
- What quantity was measured?
- What unit is used?
- What does the ± term represent?
- Is a coverage factor k stated?
- Is an approximate coverage probability or confidence level stated?
- What method and conditions define the measurand?
This is the same evidence discipline used throughout Science: never interpret a number before identifying what the number measures.
Rebuild the Evidence Object
Consider a constructed calibration report:
| Report item | Example |
|---|---|
| Measured quantity | Mass of reference object |
| Result | 50.0 g |
| Expanded uncertainty | 0.5 g |
| Coverage factor | k = 2 |
| Coverage statement | Approximately 95% under stated assumptions |
| Conditions | Laboratory temperature and procedure documented |
Now the plus-minus value has a scientific identity. It is not simply “possible error.” It is an evaluated uncertainty attached to a defined measurement result.
Observed, Calculated and Inferred
- Observed: instrument readings and relevant environmental or procedural quantities.
- Calculated: corrections, averages and uncertainty contributions according to the method.
- Reported: a best estimate of the measured quantity plus an uncertainty statement.
- Inferred: the quantity value is believed to lie within a stated interval with the report’s stated level of confidence or coverage interpretation.
- Overclaim: the exact value is guaranteed to be inside and cannot possibly lie outside.
A good PSLE Science learner should recognise that calculated scientific evidence can be strong without becoming absolute certainty.
Why Measurement Uncertainty Exists
Measurement is a comparison between the world and a measurement system. Even careful systems have finite resolution, calibration uncertainty, environmental effects, repeatability limits and other influences. Scientists do not hide those limits. They evaluate them and report them.
Uncertainty is therefore not an admission that the measurement is useless. It is information about the quality and range of values reasonably associated with the result under the stated model.
NIST guidance on expanded uncertainty describes an interval around a measurement result obtained by multiplying combined standard uncertainty by a coverage factor. When appropriate assumptions apply, a factor near 2 is commonly associated with an interval having approximately 95% confidence. NIST also warns that uncertainty does not have perfectly sharp, universal “limits” in the everyday sense.
Confidence Is Not Impossibility Outside the Interval
Suppose an expanded uncertainty interval has approximately 95% coverage under the stated procedure. A student may be tempted to say, “Then there is zero chance outside.” That does not follow. A high-coverage interval is designed to capture the quantity value with high confidence across the defined procedure, not to declare the outside region physically impossible.
The word approximately matters. The method, distribution assumptions, effective degrees of freedom and uncertainty model matter. Good reporting tells the reader how the interval was formed so that the reader can understand what kind of confidence is justified.
The Measurand Check
Another subtle error is to detach the number from the thing being measured. Scientists use the term measurand for the quantity intended to be measured. If the measurand is poorly defined, even a tiny uncertainty number can be misleading for the question you actually care about.
For example, “temperature = 20.0 ± 0.2°C” is incomplete if we do not know whether this is air temperature at one sensor, average water temperature in a tank, surface temperature or another quantity. The uncertainty belongs to the defined measurement result, not automatically to every temperature in the whole system.
Worked Case 1: Two Laboratories
Maren compares two reports for the same reference object:
- Lab A: 50.0 ± 0.5 g
- Lab B: 50.3 ± 0.4 g
She says, “The laboratories disagree because 50.0 is not 50.3.”
That ignores the uncertainty information. The reported intervals overlap substantially. This does not automatically prove the laboratories agree perfectly, but it shows that comparing only the central values throws away important evidence. A proper comparison should consider each result, its uncertainty, method and whether the same measurand was defined.
Worked Case 2: Smaller ± Means Better?
Iona sees one instrument report ±0.2 and another ±0.5. She concludes the first instrument is universally better.
Not enough information. The quantities, ranges, methods, conditions and coverage conventions must be comparable. A smaller uncertainty can be valuable, but only after confirming that the two uncertainty statements mean the same kind of thing for the same measurement job.
This is an important evidence-comparison rule: numbers can only be compared directly when their definitions are aligned.
Worked Case 3: The Rounded Display
Leonie uses a scale that displays 50.0 g. She says, “The last zero proves the mass is exactly 50.0 g.”
A digital display shows the instrument’s reported reading at its resolution; it does not remove uncertainty. A result can be displayed to one decimal place while still having an expanded uncertainty larger than one display step. Precision of display is not identical to certainty of measurement.
Tempting Reasoning That Fails
| Tempting claim | Why it fails | Better claim |
|---|---|---|
| “±0.5 means the true value cannot be outside.” | An uncertainty interval expresses evaluated coverage, not physical impossibility outside. | State the interval and its coverage interpretation. |
| “A smaller ± always means a better instrument.” | The quantities and coverage definitions may differ. | Compare like with like. |
| “The displayed digits are exact.” | Resolution and uncertainty are different concepts. | Use the reported uncertainty. |
| “Uncertainty means the scientist does not know anything.” | Quantified uncertainty is evidence about what is known and how well. | Treat uncertainty as part of the result. |
What Evidence Strengthens the Measurement Claim?
- A clearly defined measurand.
- A documented measurement method.
- Calibration linked to suitable reference standards.
- Repeated measurements showing stable behaviour.
- An uncertainty budget that includes important sources of uncertainty.
- A stated coverage factor and coverage interpretation.
- Environmental conditions kept within the method’s requirements.
What Weakens or Limits the Claim?
- The ± term is not defined.
- The measurement method is unknown.
- The instrument is used outside its stated range or conditions.
- Important uncertainty sources are omitted.
- The measurand is vague.
- The coverage factor is assumed rather than reported.
How Far Can the Conclusion Travel?
A result such as 50.0 ± 0.5 g supports a statement about the defined measured quantity under the stated measurement conditions and uncertainty evaluation. It does not automatically describe every object of the same type, every future measurement, another laboratory’s method, or every environmental condition.
The uncertainty belongs to the evidence chain that produced the result. Change the method, instrument, range, environment or object, and the uncertainty may change too.
PSLE-Style Transfer Case
A sensor report gives a water temperature of 25.0 ± 0.4°C, with the ±0.4°C stated as expanded uncertainty using a documented coverage factor.
A student writes: “The water temperature definitely cannot be 25.5°C.”
Evaluate the statement.
Reasoned answer: The statement is too certain. The report gives 25.0°C as the measurement result with an expanded uncertainty of 0.4°C under the stated coverage interpretation. The interval communicates high-confidence measurement uncertainty; it is not an absolute claim that values outside the interval are physically impossible.
Delayed Independent Return
Now imagine a weather forecast says “24 ± 2 mm of rain.” Does ±2 automatically mean rainfall outside 22–26 mm is impossible? No. The technical meaning may be different from measurement uncertainty, but the reading habit transfers: identify what the uncertainty number means before turning it into a hard boundary.
The symbol stays the same while the scientific meaning changes. That is why evidence literacy begins with definitions, not symbols.
Explained Practice
- A certificate says 10.00 ± 0.08 V but does not explain the ± term. What information must you find before interpreting it?
- Two results are 10.0 ± 0.2 and 10.1 ± 0.1. Why is comparing only 10.0 with 10.1 incomplete?
- An instrument shows six digits. Does that prove six digits of measurement certainty? Explain.
- A report states expanded uncertainty with k = 2. What additional statement should you look for to understand the intended coverage?
Check your thinking: Strong answers identify the measurand, the meaning of the ± term, coverage factor, conditions and the difference between high-confidence evidence and absolute certainty.
For Parents and Tutors: Do Not Teach “Plus-Minus Means Error”
“Error” is often used casually to mean “mistake.” In measurement science, uncertainty is not the same as a blunder. Teach the child to ask what the uncertainty statement represents and how it was obtained. A useful prompt is: “If I remove the ± number, what information about the quality of the measurement disappears?”
Then change the object. Use a ruler, temperature sensor, balance or calibration certificate. Ask the learner to keep the measured quantity, unit, central result and uncertainty together. The goal is not advanced statistics. The goal is disciplined interpretation.
Canonical eduKate Routes
- Observation, inference, prediction and explanation
- Turn diagrams, tables and graphs into evidence
- Use everyday experience without overriding evidence
- Update an explanation when new evidence arrives
Authoritative Sources
- NIST Technical Note 1297 — Expanded Uncertainty
- NIST Technical Note 1297 — Reporting Uncertainty
- NIST Technical Note 1297 — Coverage Factors
- SEAB — 2026 PSLE Science syllabus
- MOE — 2023 Primary Science syllabus
The Quiet Habit
Scientific measurements become more trustworthy when their limits are made visible. Do not erase uncertainty because you want one exact answer, and do not exaggerate uncertainty until the measurement means nothing. Read the central value, the uncertainty, the coverage statement and the conditions together. That is how a plus-minus sign becomes evidence rather than decoration.
