PSLE Science Reality Lab Vol No.091 | “We Used the Same Test Surface for Both” — Did the First Trial Change the Starting Conditions for the Second?

PSLE-SCI-REALITY-0091

Wait, What? Using the same surface can make a comparison less fair.

Two cleaning liquids are demonstrated on the same stained tile. Liquid A is used first. The tile is wiped. Then Liquid B is used on the same spot.

The presenter says, “This is fair because both products were tested on exactly the same tile.”

But after Liquid A, is it still the same starting condition? Some stain may be gone. The surface may be wetter, warmer or scratched. Liquid A may have left residue. The first wipe may have changed the material itself.

The object is physically the same tile. Scientifically, the second test may be starting from a different state.

Reality Lab Vol No.091 teaches one powerful transfer habit: when tests are done one after another on the same specimen, surface or apparatus, ask whether the earlier trial changed the starting conditions for the later trial.

Quick Answer

Identify what the first trial can change.
Describe the starting state needed for a fair comparison.
Check whether that state was restored before the second trial.
Look for carryover: residue, heat, moisture, wear, depletion, contamination or memory in the apparatus.
Prefer fresh comparable specimens when a true reset is impossible.
Where appropriate, reverse or vary the order to test whether “first” and “second” change the result.
Conclude only from trials that begin from scientifically comparable states.

What This Page Owns — and What It Leaves With the Main PSLE Science Guides

This is not a new owner for fair testing, variables or resetting an investigation. Those skills already have homes. This Reality Lab page applies them to a distinct real-world evidence object: a sequential demonstration that looks fair because it reuses the same thing, even though the first test may have changed the second test’s starting state.

Original Reality Lab Case: Two Cleaners, One Tile

This is an original teaching case. The products are fictional and the example is not a claim about any real brand.

A tile has a dark removable coating. A presenter applies Cleaner A for 30 seconds, wipes ten times, then measures the remaining darkness. Without replacing the tile, the presenter applies Cleaner B to the same area for 30 seconds and wipes ten more times.

Stage	Measured darkness score
Before Cleaner A	100
After Cleaner A	62
After Cleaner B on the same area	28

A dramatic advertisement could say, “Cleaner B reduced the darkness from 62 to 28, so it cleaned better.” But that comparison is not symmetrical. Cleaner A met a fresh coating. Cleaner B met a coating that had already been partly removed, wetted and rubbed.

Even the direction of the bias is not obvious. Perhaps the first cleaner loosens the coating and makes the second appear stronger. Or perhaps the first cleaner removes the easy-to-remove part and leaves a stubborn residue, making the second appear weaker. The key problem is that order and starting state are mixed with product identity.

Observed, Claimed and Inferred

Layer	Statement
Observed	The darkness score was 100 before A, 62 after A, and 28 after B was then used on the same area.
Claimed	B is the better cleaner.
Inferred	The second treatment is being treated as though it began from the same kind of state as the first.

Same Object Is Not the Same Starting Condition

Primary Science often asks learners to keep relevant conditions the same. In a sequential real-world test, one of the most important conditions may be hidden in time: what state was the specimen in immediately before each treatment?

If trial 1 changes the specimen, trial 2 inherits that history. A repeated test can therefore contain a carryover effect — information or material from the earlier run influencing the later one.

The Carryover Inventory

Before trusting a sequential comparison, look for changes that can survive from one trial to the next:

Residue: a chemical, powder, liquid or film remains.
Moisture: the surface becomes wetter or drier.
Temperature: rubbing, heating or cooling changes the starting temperature.
Wear: a material is scratched, stretched, compressed or fatigued.
Depletion: the available substance, charge, fuel, stain or stored energy has already been reduced.
Contamination: material from the first sample remains in tubing, a container or sensor chamber.
Biological change: a specimen grows, adapts, is injured or becomes stressed.
Software or instrument memory: an earlier reading influences a later baseline if the system is not reset properly.

Reset, Replace or Redesign?

There is no universal command to “reset everything”. The repair must match the evidence problem.

Reset when the system can genuinely return to a comparable starting state and you can check that it has done so.
Replace with a fresh comparable specimen when the first trial causes an irreversible or uncertain change.
Redesign when neither reset nor replacement can produce a fair comparison.

A reset is evidence, not magic. Saying “we washed it” is not enough if the previous material can remain. Saying “we waited five minutes” is not enough if the system needs twenty minutes to return to baseline.

Worked Case 1: Paper Towels and the Same Spill

Towel A is pressed onto a 20 mL spill first. Towel B is then pressed onto the same area. If B absorbs only 4 mL, that does not fairly show B is less absorbent. A has already removed much of the available liquid. Each towel needs a fresh matched amount of liquid under comparable conditions.

Worked Case 2: Two Batteries, One Device

A toy motor is tested with Battery A for ten minutes, then Battery B. If the motor warms during the first test, the second battery may be tested under a different device temperature. A suitable design would account for that changing state — perhaps by allowing a verified cool-down or using matched devices.

Worked Case 3: Filter Performance

Filter A is placed in a system first, then Filter B is inserted after the first filter has already removed some particles from the circulating fluid. Filter B is no longer facing the same incoming mixture. “Same apparatus” does not mean “same input condition”.

Worked Case 4: Reverse the Order

Suppose the cleaner demonstration is repeated on a fresh matched tile but with B first and A second. If the product used second always looks better, regardless of which product it is, order is a strong alternative explanation. If B looks better both when first and when second, the product explanation becomes stronger — provided the rest of the design is sound.

Why Order Tests Are So Informative

Changing the order is not automatically required in every Primary Science investigation. But in a real-world sequential comparison, it can be a powerful diagnostic question because it separates “this product is different” from “being second is different”.

Scientists use many designs to manage order and carryover in more advanced work. A Primary 5/6 learner does not need the technical vocabulary to use the central idea: if order can change the starting state, order can change the evidence.

What Evidence Would Strengthen the Comparison?

Fresh, matched specimens or surfaces for each treatment.
A measurable baseline showing comparable starting states.
A reset procedure shown to restore the apparatus or specimen sufficiently.
Blank checks where residue or contamination is possible.
Reversed or varied treatment order when order effects are plausible.
Repeated trials across several comparable specimens.
Clear outcome criteria measured in the same way for both treatments.

What Would Weaken It?

The same used specimen is tested repeatedly without a verified reset.
The first treatment visibly changes the surface before the second begins.
There is no baseline measurement before each treatment.
The product used second always appears to win.
Cleaning or waiting is claimed to remove carryover but never checked.
The demonstration confuses convenience — “we used the same object” — with experimental comparability.

Tempting Reasoning That Fails

“Same object means fair.” The object’s state can change.
“Fresh objects are less fair because they are not identical.” Carefully matched fresh specimens may be more comparable than one specimen altered by an earlier trial.
“If we wipe the surface, it is reset.” A reset must actually restore the relevant conditions.
“The second test changed the number more, so the second product caused the larger effect.” The starting state and remaining opportunity for change may be different.

How Far Can the Conclusion Travel?

If two treatments were applied sequentially to one changing specimen, the strongest defensible statement may simply describe that sequence: “After A was applied, the score changed from 100 to 62; after B was subsequently applied, it changed from 62 to 28.”

That is not enough by itself to rank A and B as independent treatments. A stronger product comparison needs comparable starting states.

PSLE-Style Transfer Case

A student compares two absorbent materials. Material P is placed in a dish containing coloured water for one minute and removed. Material Q is then placed in the same dish for one minute. Q absorbs less water.

Question: Why does this not yet prove that Q is less absorbent?

Reasoned answer: P changed the amount of water remaining before Q was tested, so Q did not begin with the same available water condition. The materials should be tested using fresh equal amounts of water, or another design that gives comparable starting conditions.

Explained Practice

Practice A: Two scratch-resistant coatings are tested one after another on the same plastic strip. What should you ask? Whether the first scratch test damaged the strip or coating before the second test.

Practice B: A sensor chamber is rinsed between samples. What would make the reset more convincing? A blank or baseline check showing that the previous sample is no longer affecting the reading.

Practice C: A demonstration is repeated with the product order reversed and the same product still performs better. Is carryover now impossible? No. But a simple order explanation is weaker, especially if fresh matched specimens and other controls also support the result.

Delayed Independent Return: The S-T-A-R-T Check

S — State: What is the specimen like before this trial?
T — Trace: Could the previous trial leave a trace?
A — Apparatus: Has equipment returned to baseline?
R — Replace or reset: Which repair actually restores comparability?
T — Test the order: Would changing the sequence change the result?

Parent and Tutor Teaching Guide

Use a harmless household analogy rather than a real product claim. Ask the learner to imagine tasting two strongly flavoured drinks from the same unwashed cup. Which drink was “really” sweeter? The first drink can leave something behind and change the second experience.

Then move back to scientific investigations: surfaces, containers, sensors and specimens can also carry history. The teaching goal is not to make children distrust demonstrations. It is to teach them to inspect whether each trial truly begins from a comparable state.

Authoritative Sources

The NIST example comes from a far more advanced measurement setting than Primary Science, but the evidence habit transfers cleanly: if a previous sample or trial can remain in the system, later measurements need a method that controls or checks that carryover.

The Quiet Return

The object can be the same.

The starting condition can be different.

Before comparing what happened second, ask what happened first.

Small Group Tutorials