PSLE Science Reality Lab Vol No.111 | “The Test Reports Down to 10, but the Claim Needs Below 5” — Can This Method Prove It?

PSLE-SCI-REALITY-0111

Wait, What? A Test Can “Pass” and Still Be Too Blunt to Prove the Claim

A fictional product label says:

Residual Substance Q: below 5 units.

The supporting laboratory report says the method’s reporting limit is 10 units. The result is shown as “<10”.

Can the laboratory result prove the label’s claim of below 5?

Not from that evidence alone. A result of “<10” means the method did not report a reliable numerical amount at or above its stated reporting boundary under those conditions. The real amount could be 9, 6, 4, 1 or something else compatible with the method’s reporting convention. Some of those possibilities satisfy “below 5”; others do not.

The learner job is not to memorise laboratory terminology. It is to compare what the claim needs with what the method can actually distinguish or quantify reliably.

Quick Answer

Identify the threshold the public claim requires.
Identify the method’s detection, reporting or quantification capability relevant to that result.
Ask whether the method can reliably separate values on the two sides of the claim threshold.
Do not treat “not reported above 10” as proof of “below 5”.
If the required threshold is lower than the method can support, ask for a more sensitive suitable method or other valid evidence.

The Exact Learner Job This Page Owns

This page owns one distinct evidence-transfer job: checking whether a scientific method is sensitive enough to support the numerical threshold built into a real-world claim.

It does not replace the main PSLE Science owners for negative results, measurement limits, inequality symbols, evidence selection or conclusion writing. It applies those skills to a laboratory report where the measurement boundary and the advertising or decision boundary are not the same number.

Original Reality Lab Case: The Two Thresholds

This is an original teaching case using fictional Substance Q and invented data.

Object	Number	What it means
Public claim	Below 5 units	The product claim requires the amount to be less than 5.
Method reporting limit	10 units	The method does not routinely report a reliable ordinary numerical result below 10 under the stated convention.
Laboratory result	<10 units	The result is below the method’s reporting boundary, not automatically below 5.

A learner should immediately notice that the method boundary sits above the claim boundary. The test cannot cleanly tell us whether the true value lies at 8 or at 3. That uncertainty matters because 8 fails the claim and 3 passes it.

Observed, Claimed and Inferred

Layer	Statement
Observed/reporting fact	The laboratory reported <10 units.
Claim	The product contains less than 5 units.
Hidden inference	Anything below the reporting limit of 10 must also be below 5.
Problem	Values between 5 and 10 are compatible with <10 but do not satisfy the claim.
Needed evidence	A suitable method capable of supporting the lower threshold or another valid measurement strategy.

The Most Important Diagram Is a Number Line

Draw a line from 0 to 12. Mark the claim threshold at 5. Mark the method’s reporting limit at 10.

Now shade everything below 10. That shaded region contains two different scientific zones:

0 to below 5: values that satisfy the claim.
5 to below 10: values that do not satisfy the claim.

The laboratory result “<10” does not tell us which zone the sample occupies. That is the whole problem.

Detection, Reporting and Quantification Are Related but Not Identical

Laboratories and standards use several related terms for low-level measurement capability. Exact definitions can differ by method and programme. NIST defines a quantitation limit as the minimum amount that can be quantified with acceptable accuracy and precision. EPA and USGS guidance also distinguishes levels used for detection and reporting and warns that uncertainty grows near these low-concentration boundaries.

A Primary learner does not need to memorise acronyms such as LOD, LOQ, MDL or LRL. The transferable question is:

At what level can this method support the kind of statement we want to make?

Why “Not Detected” Is Not the Same as Zero

When a method does not detect or report a target above its boundary, the target may truly be absent. It may also be present below the method’s reliable detection or reporting capability. That is why a non-detection is not automatically a numerical zero.

EPA guidance on data near detection limits explicitly warns against treating non-detects as zero without justification. USGS reporting systems similarly use “less than” conventions because the measurement cannot support an ordinary numerical value in that low region with the same confidence.

The Claim-Threshold Check

Many real-world claims quietly contain a threshold:

“less than 5 units”
“below the specified limit”
“no more than 2%”
“at least 95%”
“within ±1 unit”

A good Reality Lab reader finds that decision boundary first. Then the reader checks whether the method is capable of distinguishing the relevant sides of it.

The Representation Check: How a “PASS” Badge Can Hide the Method Limit

An infographic might show a green PASS symbol beside “<10” while the fine print defines the desired threshold as 5. The symbol encourages a quick conclusion before the reader compares the two numbers.

A better scientific representation would show both:

the method reporting capability, and
the decision or claim threshold.

If the method is not sensitive enough for the decision, the honest result may be inconclusive for that particular claim, even if the laboratory work itself was performed correctly.

A Good Laboratory Result Can Still Be Inadequate for a Stricter Claim

This distinction is important. We are not accusing the laboratory of error. The method may be excellent for its designed purpose. The problem can arise later, when someone uses the result to support a claim requiring finer discrimination than the method provides.

That is why NIST’s idea of fit for purpose matters. Measurement quality is always connected to the job the number must perform.

What Evidence Would Strengthen the Claim?

A suitable method whose reliable reporting or quantification capability lies below the claim threshold.
Validation showing acceptable performance near the threshold that matters.
Appropriate sample preparation that does not raise the effective reporting limit above the claim threshold.
Replicate or quality-control evidence showing the method behaves acceptably at low relevant levels.
A clear report stating whether the laboratory result is a detection, estimate, upper bound or ordinary quantified value.
A public claim whose numerical wording matches what the method can actually support.

What Would Weaken It?

The claim threshold is lower than the method can reliably report or quantify.
The report hides the detection or reporting limit.
“Not detected” is converted to zero without justification.
A less-than result is presented as though it identifies one exact value.
The method’s low-level performance was never evaluated in the relevant sample type.
Sample dilution or another preparation step raises the effective original-sample reporting boundary.

Worked Case 1: <10 Does Not Prove <5

A fictional residue claim requires below 5 units. The laboratory reports <10. The result is compatible with 4, but also with 7. Therefore the evidence does not decide whether the below-5 claim is satisfied.

Worked Case 2: Reporting Limit 1, Claim Threshold 5

Now suppose the method can reliably report down to 1 unit and the result is <1. That is much stronger evidence for a below-5 claim because every value compatible with <1 is also below 5. Other method and sampling questions still matter, but the sensitivity mismatch has been removed.

Worked Case 3: A Very Sensitive Test Used for a Broad Claim

If a method can distinguish down to 0.1 units and the relevant threshold is 20, sensitivity is unlikely to be the limiting issue. The learner should then move to other questions such as representativeness, sample handling, method scope and uncertainty near the measured value.

Worked Case 4: The Threshold Changes but the Test Does Not

A project originally needed to know whether values were below 20. A method reporting reliably to 10 was adequate for many decisions. Later, the public claim becomes “below 5” but the same method is retained. The method did not suddenly become bad. The decision requirement changed, so the old method may no longer be fit for the new purpose.

Tempting Reasoning That Fails

“Not detected means none.” A target can be present below the method’s reliable detection or reporting capability.
“<10 means about 10.” It is an upper-bound style reporting statement, not an ordinary measurement of 10.
“Any result below 10 proves below 5.” Values from 5 to below 10 contradict the claim.
“A more sensitive test is always scientifically better.” Sensitivity is one performance characteristic. The method must still be suitable, accurate enough, selective and fit for the actual purpose.
“If the method cannot prove the claim, the claim must be false.” Inadequate evidence is not proof of the opposite. The correct conclusion may be that the question remains unresolved by this method.

Model and Measurement Limits

Real laboratories use method-specific conventions, and low-level reporting can be more complicated than a single fixed boundary. Some detected values may be reported as estimated below an ordinary reporting level; some programmes use sample-specific limits; and limits can change when a sample must be diluted or when interferences occur.

Reality Lab keeps the transferable logic while avoiding invented universal rules: read the method’s actual reporting convention and compare it with the threshold the claim requires.

How Far Can the Conclusion Travel?

If the method’s reliable low-level capability is comfortably below the decision threshold and the rest of the evidence chain is sound, the measurement can support a threshold claim much more strongly. It still does not establish that every batch, place or time has the same value unless the sampling evidence supports that wider generalisation.

PSLE-Style Transfer Case

A test can reliably report ordinary values down to 8 units. A learner needs to decide whether a sample contains less than 3 units. The laboratory reports “<8”.

Question: Why is the result insufficient to conclude that the sample contains less than 3 units?

Reasoned answer: The value could lie anywhere below 8 according to the reporting statement. Values such as 4, 5 or 7 are below 8 but not below 3, so the method cannot distinguish all values that would pass from those that would fail the claim.

Explained Practice

Practice A: Reporting limit 2; claim threshold 10; result <2. Does sensitivity support the threshold claim? Yes, because values compatible with <2 are also below 10, although other evidence checks still apply.

Practice B: Reporting limit 20; claim threshold 10; result <20. Can the claim be proven? No. The unresolved interval includes values above and below 10.

Practice C: A method was designed for a broad screening threshold but the advertisement uses a much stricter number. What should happen? Re-evaluate whether the method is fit for the stricter claim rather than borrowing the old validation automatically.

Delayed Independent Return: The T-H-R-E-S-H-O-L-D Check

T — Target claim: What numerical boundary must be shown?
H — How low or high can the method reliably report?
R — Region unresolved: Does the method leave values on both sides of the claim threshold possible?
E — Evidence type: Is this a detection, estimate, less-than result or ordinary quantified value?
S — Sample scope: Is the low-level capability valid for this sample type?
H — Handling: Did preparation or dilution change the effective limit?
O — Other evidence: Do QC and validation support performance near the threshold?
L — Limit the conclusion: Say no more than the method can show.
D — Different method if needed: Use a suitable lower-limit method when the claim demands it.

Parent and Tutor Teaching Guide

Draw two number lines. On one, place the claim threshold at 5 and the method limit at 10. On the other, place the method limit at 1. Ask the learner which test can distinguish the two sides of the below-5 claim. This turns an unfamiliar laboratory idea into a visible evidence problem.

Then ask a second question: “If the method is not sensitive enough, does that prove the claim is false?” The learner should answer no. The evidence may simply be unable to decide. That distinction—false versus not established—is one of the most valuable scientific habits a Primary learner can carry into later study.

Authoritative Sources

The current PSLE Science frame requires learners to analyse information, evaluate observations and methods, and communicate reasoning. MOE’s Primary Science syllabus also promotes healthy scepticism and honest communication of uncertainty. Comparing the claim threshold with the method threshold is a direct application of those habits.

The Quiet Return

A laboratory can do everything correctly and still be asked a question its method was never designed to answer.

Before accepting a threshold claim, check whether the measurement is sharp enough to see the threshold.

Small Group Tutorials