Wait, What? Repeating your own investigation and getting the same answer is useful—but another learner getting a comparable answer can test something different.
PSLE Science students learn early that repetition can strengthen evidence. But “repeat” hides more than one scientific job. If the same learner uses the same apparatus in the same place and repeats the same procedure, the new trials mainly test whether the result is reasonably consistent under similar conditions. If another learner, another set of apparatus or another time produces a comparable result from the same clear method, the evidence begins to test whether the finding survives some change in conditions.
In wider measurement science, these ideas are often described using repeatability and reproducibility. Primary learners do not need to memorise technical metrology definitions. They do need the underlying reasoning: what stayed the same, what changed, and what kind of agreement does the new evidence actually test?
Quick Answer
Use this simple distinction:
- Repeatability question: If we repeat the measurement or trial under closely similar conditions, do the results agree reasonably well?
- Reproducibility question: If the method is followed under a deliberately changed condition—such as another learner, another comparable instrument, another day or another location—does the main result still hold?
Neither kind of agreement proves that a causal explanation is correct. Both are evidence about the reliability and transportability of the measurement or method.
The PSLE Science Learning Job This Guide Owns
This guide owns one distinct inquiry job: deciding whether a new result is another repeat under the same practical conditions or a check that the method survives changed conditions.
It does not replace the pages on repeated trials, writing a reproducible method, specimen choice, averaging or fair tests. Those pages answer different questions. Here, the focus is the scope of the repeat: are we checking consistency inside one measurement situation, or checking whether the method travels beyond it?
Why This Matters in the Current PSLE Science Frame
For examination from 2026, PSLE Science assesses the 2023 Primary Science syllabus. The official assessment objectives include interpretation and analysis, evaluation of observations, information and methods, and communication of explanations and reasoning. A learner who can say only “repeat the experiment” has not yet explained what the repeat is testing. A stronger learner can state which conditions remain fixed, which condition changes, and what agreement or disagreement would mean.
One Method, Two Kinds of Check
Imagine an original investigation that measures how far a toy car travels after being released in a standard way.
Check A: The same learner repeats the trial five times with the same ramp, car, ruler and room conditions. This mainly shows how much the result varies under closely similar conditions.
Check B: Another learner follows the same written method with a comparable set of apparatus and obtains a similar pattern. This tells us something extra: the finding did not depend entirely on one learner’s hidden habits or one exact apparatus set.
The second check is not automatically “better”. It answers a different question.
Worked Example 1: Same Learner, Same Set-Up
A learner measures 41 cm, 42 cm, 41 cm, 43 cm and 42 cm in repeated trials using the same method. The results are reasonably close. That is evidence of repeatability under the tested conditions.
Do not immediately claim that the result will be the same for every learner or every apparatus set. The evidence has not travelled that far yet.
Worked Example 2: Another Learner Follows the Method
Student A writes a method that specifies the starting position, release rule, surface, measuring point and how distance is recorded. Student B uses a comparable set-up and follows the method. Student B obtains 40 cm, 42 cm, 41 cm and 42 cm.
The agreement is useful because Student B did not rely on Student A’s unspoken memory of how the test was run. The method carried enough information for another person to reproduce a comparable measurement situation.
That does not prove every detail is perfect. But it is stronger evidence that the finding is not merely a private result created by one operator’s hidden technique.
Worked Example 3: Another Learner Gets a Different Result
Suppose Student B follows the written method and consistently obtains much larger values. Do not choose a winner immediately. Treat the disagreement as diagnostic evidence.
- Were the apparatus genuinely comparable?
- Was the same quantity measured?
- Did both learners use the same start and end criteria?
- Was the method clear enough to prevent hidden choices?
- Did temperature, surface, specimen or other conditions differ?
- Was one measuring instrument shifted or poorly referenced?
Disagreement can expose a hidden variable or ambiguous method. That makes it scientifically useful even when it is inconvenient.
Reproducibility Is Not “Copy Exactly Everything”
If absolutely everything were identical—including the same learner, same instrument, same place and same short time window—you would mainly be testing repeatability. A reproducibility-style check deliberately allows some condition to change while preserving the scientific question and method well enough for comparison.
The changed condition must be named. “Another group did it too” is incomplete. Ask: another observer? another apparatus set? another day? another classroom? another similar specimen? Each change tests a different part of the method’s robustness.
Do Not Confuse Reproducibility With a Fair Test Variable
Suppose the original scientific question is whether surface type affects sliding distance. The surface type is the variable intentionally tested inside the investigation. Repeating the entire method with another learner is a separate evidence-quality check. Do not treat “different learner” as though it were the scientific variable the experiment was designed to study.
A Comparison Table
| Question | Repeatability-style check | Reproducibility-style check |
|---|---|---|
| Who performs it? | Often same learner | May be another learner |
| Apparatus | Same apparatus where possible | Comparable but possibly different apparatus |
| Time/place | Usually close together | May change |
| Main job | See consistency under similar conditions | See whether result survives stated changed conditions |
| What agreement means | Result is reasonably stable within that setup | Method/result may be less dependent on one exact setup or operator |
| What agreement does not prove | Accuracy or causal truth | Accuracy, universal validity or causal truth |
Why a Clear Method Matters
A reproducibility-style check is only informative if another learner can tell what to do. Hidden choices destroy comparability. A useful method states enough about:
- which apparatus or specimen is used;
- what is changed;
- what is kept comparable;
- where and when measurements are taken;
- what counts as the start and end;
- what units are recorded;
- how repetitions are handled.
This is why the separate guide on writing a method another student could follow is an important internal route. Method communication and reproducibility support each other, but they are not the same job.
Failure Signatures
- “We repeated it three times, so it is reproducible.”
- “Another student got the same answer, so the explanation is proven.”
- “Another group used different apparatus, so their result cannot be compared at all.”
- A learner does not record which conditions changed between runs.
- Agreement is celebrated without checking whether both groups made the same systematic error.
- Disagreement is dismissed rather than investigated.
Earliest Weak-Link Diagnosis
- What exact scientific question is being repeated?
- What quantity or relationship is measured?
- Which conditions stayed the same?
- Which conditions changed deliberately between the two checks?
- Are the results directly comparable?
- If they agree, what kind of confidence increases?
- If they disagree, what hidden difference should be investigated first?
Misconception Repair
“Repeatability and reproducibility are synonyms.” They are related evidence-quality ideas but differ in the measurement conditions being repeated or changed.
“If another learner gets the same result, it must be correct.” Two learners can share the same flawed method or reference. Agreement strengthens one kind of confidence but does not replace accuracy and validity checks.
“Different apparatus makes comparison useless.” Not automatically. If the apparatus measures the same quantity appropriately and the changed condition is understood, a changed apparatus set can be part of a useful robustness check.
“Disagreement means someone made a mistake.” Sometimes. But disagreement can also reveal a method ambiguity, environmental effect, natural variation, non-comparable specimen or measurement problem.
The Evidence-Travel Protocol
DEFINE THE QUESTION → NAME WHAT STAYS THE SAME → NAME WHAT CHANGES BETWEEN REPEATS → COMPARE THE RESULT → INVESTIGATE AGREEMENT OR DISAGREEMENT → LIMIT THE CONCLUSION TO THE CONDITIONS TESTED.
Original Practice 1
A learner measures the same object five times with the same ruler in the same lesson. Is this mainly a repeatability or reproducibility check? Explain which conditions stayed similar.
Original Practice 2
Two classes use the same written method and comparable apparatus to investigate the same relationship on different days. Their data show the same direction of change. What extra evidence does the second class provide? What does it still not prove?
Original Practice 3
Two groups obtain different trends. Group A measured from the front edge of an object; Group B measured from its centre. Identify the method ambiguity before deciding which result is better.
Retrieval and Practice Sequence
- Sort examples into same-condition repeats and changed-condition repeats.
- For every example, name the conditions that stayed the same and changed.
- Predict what agreement would strengthen and what it would not prove.
- Study one disagreement and generate three plausible method differences before choosing a cause.
- Return after several days and classify unfamiliar examples without the terms shown.
Unfamiliar Transfer
A plant growth method is followed by a second class using similar plants and another measuring ruler. The second class obtains the same overall pattern but slightly different numerical values. The learner’s job is not to demand identical numbers. It is to decide whether the scientific relationship travels across the changed conditions and whether any difference needs a measurement or specimen explanation.
Delayed Independent Return Test
After a week, present four anonymous study descriptions. The learner passes if they independently identify which checks test same-condition consistency, which test changed-condition reproducibility, and what each result can support.
Answer-Checking Receipt
- I know the scientific question being repeated.
- I know whether the same or changed measurement conditions were used.
- I can name the conditions that changed.
- I have not treated agreement as proof of accuracy.
- I have not treated disagreement as automatic proof of error.
- I can explain what kind of confidence the repeat adds.
- My conclusion stays within the conditions actually checked.
Parent and Tutor Teaching Guide
Use one simple investigation twice. First let the same child repeat it. Then give the written method to another child or adult and ask them to carry it out without extra hints. Compare what changes. This makes the distinction between “I can repeat my own routine” and “my method survives another operator” concrete.
When results differ, avoid asking immediately, “Who did it wrongly?” Ask, “What condition was different?” This keeps multiple explanations alive long enough to diagnose the evidence.
Do not require technical words as answer phrases. The important Primary Science reasoning is to identify conditions, repeated evidence, method clarity and limits.
Useful Internal Routes
- PSLE Science Learning Guide
- How Repeating an Investigation Affects Repeatability, Not Fairness
- How to Write a Method Another Student Could Follow
- How to Tell One Trial, One Measurement and One Specimen Apart
- How to Tell Random Variation From a Systematic Shift
Authoritative References
- MOE — 2023 Primary Science Teaching and Learning Syllabus
- SEAB — PSLE Science, examination from 2026
- NIST Technical Note 1297 — repeatability and reproducibility terminology
- JCGM/BIPM International Vocabulary of Metrology — measurement precision
Quiet Return
Science becomes stronger when a result does not depend on one lucky trial, one person’s hidden habit or one exact set of apparatus. First ask whether the result repeats. Then ask whether it travels. Those are different questions, and learning to keep them separate makes method evaluation much sharper.
