PSLE Science Reality Lab Vol No.421 | “Accuracy = ±2% RH at 23°C” — Does That Same Accuracy Claim Apply at Every Temperature?

PSLE-SCI-REALITY-0421

Wait, What? The Big Number Says “±2% RH”, but the Small Words Say “at 23°C”

A digital humidity instrument is shown in a product comparison. The large specification says Accuracy: ±2% RH. A learner reads the number and concludes, “Whenever this instrument is switched on, its humidity reading is accurate to within 2 percentage points.”

Then someone notices the rest of the specification: ±2% RH from 0–60% RH at 23°C, with a different accuracy statement at higher humidity. The same datasheet separately lists a wider operating-temperature range.

Now the evidence job changes. The learner must decide which words belong together. Is “±2% RH” a universal promise for every temperature and humidity, or a performance statement under a stated range and condition?

This is a common scientific-reading trap. A specification can look like one bold number while its actual meaning is a bundle: quantity + range + condition + method + limitations. Removing the condition can create a stronger claim than the source supports.

Quick Answer

No. If an instrument manufacturer or laboratory specification states an accuracy under a particular condition such as 23°C, that evidence supports the claim under the stated conditions. It does not by itself prove that the exact same accuracy applies at every other temperature in the instrument’s operating range.

The instrument may still perform very well elsewhere. The point is narrower: the learner should not invent an unstated guarantee. To extend the claim, look for additional specifications, test data, temperature coefficients, uncertainty information or other authoritative evidence covering the new conditions.

The Exact Learner Job This Page Owns

This Reality Lab owns one real-world evidence-transfer job: evaluating an instrument accuracy statement that is explicitly tied to a reference temperature, range or other condition, without silently extending that statement to all operating conditions.

It does not replace the canonical eduKate Sengkang guides on accuracy, precision, uncertainty, resolution, calibration or measurement. Those pages own the general skills. This article applies them to the communication object learners actually meet: a datasheet, product page or laboratory specification where a headline number is narrower than it first appears.

Rebuild the Communication Object: The Fictional Humidity Meter Card

Classroom teaching specification — HumiCheck S1
Measurement range: 0–95% RH
Resolution: 1% RH
Accuracy: ±2% RH from 0–60% RH at 23°C
Accuracy: ±3% RH from 61–95% RH at 23°C
Operating temperature: 0–50°C
Sample rate: approximately 1 reading/s
This is an original composite example, not a copied commercial datasheet.

A careless reading extracts only “±2% RH”. A stronger reading asks four questions immediately:

What quantity? Relative humidity.
What part of the measurement range? Here, 0–60% RH for the ±2 statement.
Under what stated condition? Here, 23°C.
What is not stated? This card does not tell us that ±2% RH holds at every other operating temperature.

The Most Important Distinction: Operating Range Is Not the Same Thing as an Accuracy Guarantee

An operating range tells you where the device is intended or specified to operate under the manufacturer’s stated conditions. An accuracy specification describes how close results are expected to be to a reference under defined conditions and according to the way the specification is written.

Those two statements can overlap without being identical. A device can be allowed to operate from 0°C to 50°C while one headline humidity-accuracy value is explicitly stated at 23°C. That does not mean accuracy suddenly disappears at 22°C or 24°C. It means the evidence printed in that line is bounded by its condition.

Science rewards this restraint. Do not claim less than the evidence supports, but do not claim more either.

Observed, Claimed and Inferred

Layer	Statement	Scientific status
Printed specification	Accuracy ±2% RH from 0–60% RH at 23°C	Explicit claim
Printed operating range	Instrument can operate over a wider temperature interval	Separate explicit claim
Reasonable use	At 23°C and within the stated humidity range, the ±2 specification is relevant	Bounded interpretation
Stronger inference	The instrument is ±2% RH accurate at 0°C, 35°C and 50°C too	Not established by that line alone
Needed evidence	Additional specifications or test data across temperature	Would support extension

The Qualifier-Binding Test: Which Small Words Belong to Which Big Number?

When students read scientific specifications, they often separate a number from the phrase that limits it. A better habit is to read the whole statement as one evidence unit.

Instead of memorising ±2% RH, say:

±2% RH for the stated humidity range at the stated reference temperature.

The condition is not decorative fine print. It is part of the claim.

Why Temperature Can Matter to an Instrument

Real sensors are physical systems. Their electrical, optical, mechanical or chemical responses can depend on environmental conditions. Temperature can affect sensor materials, electronics, reference elements, equilibrium, compensation algorithms or the quantity being measured itself.

For a Primary 5/6 learner, the key point is not to memorise sensor engineering. It is to recognise a general scientific pattern: performance measured under one condition should not automatically be assumed identical after a relevant condition changes.

That is the same disciplined thinking used in fair tests. If a condition can affect the outcome, evidence from one condition may need checking before it is extended to another.

Representation Check: A Product Table Can Make Unequal Statements Look Equal

Imagine a comparison table:

Meter	Accuracy
A	±2% RH at 23°C, 0–60% RH
B	±2.5% RH over 10–40°C, 20–80% RH
C	±2% RH

If the table ranks A above B simply because 2 is smaller than 2.5, it may be comparing different scopes. Meter C is even harder to evaluate because the missing conditions prevent a fair comparison.

Before comparing performance numbers, align the conditions. A smaller number under a narrower condition is not automatically a better instrument for every job.

Range Check: One Accuracy Number May Not Cover the Whole Measurement Range

The fictional meter gives ±2% RH up to 60% RH and ±3% RH above that. Suppose the room is at 85% RH. A student quotes the ±2 number because it is the most impressive value on the card.

That is selection, not evaluation. At 85% RH, the learner should use the specification that covers 85% RH. Scientific claims must be matched to the range where they were defined.

Resolution Check: A Small Display Step Is Not the Same as a Small Accuracy Error

A meter may display humidity in 1% steps, or even 0.1% steps, while its stated accuracy is wider. Display resolution tells you the smallest change the display can show. It does not guarantee that the true value lies within one display step.

This matters because product pages sometimes make a highly detailed display look like evidence of highly accurate measurement. The number of digits and the accuracy specification answer different questions.

Worked Case 1: Same Meter, 23°C and 35°C

A classroom uses the composite meter at 23°C and reads 50% RH. The printed ±2% RH specification covers this humidity and temperature. Later the meter is placed in a warm cabinet at 35°C. It reads 50% RH again.

A learner says, “Both are 50, so both are guaranteed within ±2% RH.”

The first claim is supported by the stated specification, assuming the other conditions are met. The second requires additional evidence because the specification was explicitly stated at 23°C. The correct response is not “the 35°C reading is wrong”; it is “the ±2% claim has not yet been established here by the evidence we have.”

Worked Case 2: One Number, Two Humidity Ranges

At 23°C, the device reads 45% RH. The ±2% RH line covers that region. Later it reads 85% RH. The datasheet specifies ±3% RH in the higher range.

A headline saying “±2% accuracy across 0–95% RH” would erase the range boundary. The communication object becomes misleading even though every number originally came from the datasheet.

Worked Case 3: The Instrument Operates at 5°C

The operating-temperature line includes 5°C. A student says, “Because operation is allowed at 5°C, the 23°C accuracy must also apply at 5°C.”

That mixes two different claims. “Can operate here” and “has this exact accuracy here” are not interchangeable statements. A fuller datasheet may indeed specify performance over temperature, but it must be checked rather than assumed.

Worked Case 4: A Review Site Removes the Footnote

A manufacturer table says “±2% RH at 23°C”. A review site copies only “Accuracy: ±2% RH” into a comparison chart. Nothing in the copied number is numerically false, but the scope has changed because the condition disappeared.

This is a strong Reality Lab lesson: misleading communication can happen by deleting context rather than inventing a number.

Worked Case 5: Two Instruments, Different Test Conditions

Meter X claims ±1.8% RH at one reference temperature. Meter Y claims ±2.2% RH across a wider temperature interval. Which is “more accurate”?

Not enough information for a universal ranking. X has the smaller number in its stated condition; Y has a broader stated condition. The appropriate choice depends on the measurement job and the evidence available for the conditions that matter.

Comparison and Baseline Check

Before comparing two accuracy claims, align at least these dimensions:

same measured quantity;
same units;
same part of the measurement range;
same or comparable environmental conditions;
same basis for the accuracy statement;
same confidence or uncertainty convention where relevant;
same sensor configuration, accessories and sampling arrangement if those matter.

If those do not match, a simple “2 is better than 3” comparison may be scientifically weak.

What Evidence Would Strengthen “The Same Accuracy Applies at 35°C”?

The manufacturer explicitly specifies ±2% RH across a temperature interval that includes 35°C.
A temperature-compensation specification and supporting validation cover that condition.
Calibration or verification data at several temperatures show performance consistent with the claim.
An uncertainty budget includes temperature effects and still supports the stated bound.
An independent laboratory evaluation tests the instrument under relevant conditions.
The exact sensor, configuration and range in the evidence match the one being used.

What Would Weaken That Claim?

The only accuracy line is explicitly “at 23°C”.
The new temperature lies near or outside the stated operating range.
The sensor behaves differently at temperature extremes.
Condensation, airflow, self-heating or equilibration creates additional measurement problems.
The comparison quotes a headline number but omits a limiting footnote.
The reading lies in a measurement range with a different accuracy specification.

Alternative Explanations When Readings Differ

Two humidity meters disagree more at 35°C than at 23°C. It is tempting to say, “One meter became inaccurate because of temperature.” That is only one possible explanation. Other possibilities include calibration differences, response lag, placement, local humidity differences, airflow, condensation, contamination, different averaging rules or different sensor designs.

The Reality Lab habit is to keep alternatives open until evidence separates them.

How Far Can the Conclusion Travel?

A specification such as ±2% RH at 23°C supports a bounded statement about performance under the stated conditions and range. It may be entirely appropriate for comparing measurements made under those same conditions.

It does not automatically support “±2% everywhere”, “every individual reading is guaranteed to be within exactly 2”, “the device is better than every ±3% device”, or “the same performance holds after every environmental condition changes”. Those stronger claims need their own evidence.

Tempting but Invalid Reasoning

“The largest bold number is the whole specification.” Conditions and ranges can be part of the claim.
“Operating at 50°C means the 23°C accuracy still applies.” Operating range and accuracy scope are distinct.
“The display has many digits, so the reading is highly accurate.” Resolution and accuracy are different.
“±2 is always better than ±3.” Not until the compared conditions and definitions are aligned.
“If the same number appears twice, the measurement quality is the same.” Environmental conditions can change the evidence supporting the reading.

Model and Measurement Limits

Instrument specifications are summaries. They cannot describe every possible installation, ageing state, contamination level, airflow pattern, calibration history or environmental interaction. A specification may be based on particular test procedures and reference equipment. Some uncertainty sources belong to the sensor; others belong to the whole measurement system.

That is why careful scientific communication keeps the specification attached to its conditions rather than turning it into a universal slogan.

PSLE-Style Transfer Case

A sensor is advertised with the statement “Accuracy: ±1°C at 25°C”. Its operating range is 0–60°C. A student uses the sensor at 55°C and writes, “The true temperature must be between 54°C and 56°C because the sensor accuracy is ±1°C.”

Question: Explain why this conclusion is not fully supported by the information given.

Reasoned answer: The ±1°C accuracy statement was specified at 25°C, while the measurement was made at 55°C. The operating range shows that the sensor can be used at 55°C, but it does not by itself prove that the same ±1°C accuracy applies there. More performance information at 55°C is needed.

Delayed Independent Return

Why is “at 23°C” part of an accuracy claim rather than decoration?
What is the difference between operating range and accuracy scope?
Why can one instrument show a smaller accuracy number but not be universally better?
What extra evidence could justify extending an accuracy claim to another temperature?
Why do decimal places not prove accuracy?

Return check: conditions define scope; operation is not the same as identical performance; comparisons need aligned conditions; seek specifications or validation at the new condition; resolution is not accuracy.

Explained Practice

Practice A. “±2% at 20°C” becomes “±2%” in a social-media comparison. What changed?
Answer: The numerical value stayed the same, but the condition defining its scope was removed.

Practice B. A meter operates from −10°C to 50°C. Does that prove equal accuracy throughout?
Answer: No. Look for the accuracy specification across temperature.

Practice C. A reading is within the stated range but the surrounding temperature is outside the condition used for the accuracy claim. Should the reading be thrown away?
Answer: Not automatically. The correct action is to qualify the claim and seek evidence about performance under the actual condition.

Practice D. Two specifications use the same ±2 number but one covers a much wider range. Are the claims identical?
Answer: No. Their scopes differ even though the headline numbers match.

Route to Existing eduKate Sengkang Owners

Parent and Tutor Teaching Guide

Give the learner three pretend product cards. Put the same large accuracy number on all three, but vary the small conditions: one at room temperature only, one across a wide temperature interval, and one with no condition shown. Ask which card gives the strongest evidence for measuring in a hot cupboard.

Then cover the qualifying phrases with sticky notes. Notice how the comparison changes when the conditions disappear. This makes a subtle communication problem visible: evidence can be distorted by cropping away context even when no number is altered.

End by asking the child to read every technical specification as a complete sentence: performance — for what quantity — over what range — under what condition? That habit transfers to product labels, laboratory reports, weather sensors, data loggers and PSLE Science investigations.

Authoritative Sources

NIST’s published instrument pages provide concrete examples of scientific performance statements that are tied to measurement ranges, reference temperatures or sample temperatures rather than existing as free-floating numbers. The MOE and SEAB sources support the deeper learning goal: interpreting information, evaluating methods and assumptions, recognising uncertainty and communicating reasoning without inventing universal rules.

Quiet Return

In science, the small words can be part of the measurement.

Do not carry an accuracy number farther than the conditions that earned it.

Small Group Tutorials