Scientific reasoning becomes stronger when a pupil can ask, “What else could explain this?” A single observation may fit more than one explanation. A new result may contradict the expected trend. An anomaly may come from measurement error, a changed condition or a genuine boundary in the model.
This guide develops alternative explanations, contradiction checking, anomaly resolution and model revision for Primary 6 and PSLE Science.
Return to the Primary 6 Science Learning Hub.
The alternatives rule
OBSERVATION → CANDIDATE EXPLANATIONS → EVIDENCE TEST → CONTRADICTION CHECK → ELIMINATE OR RETAIN → REVISE MODEL.
This is an eduKate reasoning routine, not an official SEAB marking formula.
Part I — One result can have several explanations
Observation: a plant grows less.
Possible explanations:
- less light;
- less water;
- different starting health;
- unfavourable temperature;
- measurement error.
The question is not “Which explanation sounds scientific?” It is “Which explanation best fits the evidence and conditions?”
Part II — Alternative explanations are not distractions
Considering alternatives strengthens causal reasoning because it prevents premature certainty.
If light and water both changed, both remain plausible causes. The design cannot isolate one confidently.
Part III — Use evidence to eliminate possibilities
Suppose two plants received the same water, temperature and soil, but different light levels. The plant in lower light produced fewer bubbles.
Because other major conditions were controlled, a light-related explanation becomes stronger.
Controlled conditions act as elimination evidence.
Part IV — Contradictions matter
If a proposed explanation predicts one outcome but the data show the opposite, the explanation needs review.
Example: hypothesis predicts more cells will make a motor rotate faster, but beyond three cells the motor slows and becomes hot.
The simple relationship has reached a boundary or another effect has become important.
Part V — Do not force data to match the model
A scientific model should be revised when evidence consistently contradicts it.
Changing or ignoring results to preserve the first idea destroys the purpose of the investigation.
Part VI — Anomaly versus contradiction
An anomaly is one unusual result within an otherwise consistent pattern.
A contradiction is evidence that challenges the expected relationship more fundamentally.
One strange trial may be an anomaly. Several repeated reversals may indicate the model is incomplete.
Part VII — Diagnose an anomaly systematically
Ask:
- Was the starting condition the same?
- Was the instrument read correctly?
- Was the procedure followed?
- Did the apparatus change?
- Did an environmental condition shift?
- Was the result repeated?
- Could the model itself be incomplete?
Part VIII — Repeat the suspicious condition
If an anomalous point appears, repeating that condition can help distinguish random variation from a reproducible boundary.
Do not automatically repeat the whole experiment if the weakness is localised to one condition.
Part IX — Alternative explanation in friction
Observation: one car travels shorter distance.
Possible causes:
- surface friction differs;
- release condition differed;
- car mass differed;
- measurement point differed.
A fair test controls the alternatives so surface type becomes the strongest explanation.
Part X — Alternative explanation in photosynthesis
Observation: bubble count falls.
Possible causes:
- less light;
- less carbon dioxide availability;
- temperature changed;
- plant condition changed;
- counting error.
The experiment must tell us which alternatives remain plausible.
Part XI — Alternative explanation in ecology
Observation: predator population decreases.
Possible causes:
- prey decreased;
- habitat changed;
- disease;
- migration;
- weather conditions changed.
A food web alone may not distinguish all these causes.
Part XII — Alternative explanation in cooling
Observation: Cup A cools more slowly.
Possible causes:
- better insulation;
- greater water volume;
- higher starting temperature;
- different container material;
- different room exposure.
Good design removes these alternatives where possible.
Part XIII — Contradictory evidence can reveal a boundary
A spring extends by 2 cm per load unit for several loads, then fails to return to original length.
The new evidence contradicts the earlier simple reversible model.
Conclusion: the earlier relationship applies only within a limited range.
Part XIV — Contradictory evidence can reveal confounding
One trial suggests more light increases growth. Another gives the opposite result.
Before assuming the relationship is false, check whether water, temperature, plant type or starting size differed.
Inconsistent controls can create apparent contradictions.
Part XV — Contradictory evidence can reveal measurement weakness
Bubble count varies wildly while collected gas volume remains stable.
This may show bubble count is a noisy proxy because bubble size differs.
The measurement method, not the scientific process, may explain the contradiction.
Part XVI — Model comparison
Sometimes two explanations fit part of the evidence.
Model A explains the early trend but not the plateau.
Model B explains both the rise and the plateau because it allows another limiting factor.
The better model explains more evidence with fewer unsupported assumptions.
Part XVII — Do not invent advanced mechanisms
At Primary 6, alternative-explanation reasoning should remain within the syllabus and information given.
Do not add advanced molecular or physiological explanations simply because they sound sophisticated.
Original workshop 1 — car anomaly
Surface P trials: 80, 82, 79, 41 cm.
The 41 cm reading is anomalous.
Check release, obstruction, measurement and car condition before averaging.
Original workshop 2 — plant contradiction
Higher light usually produces more bubbles, but one high-light trial produces fewer.
Possible reasons: temperature changed, plant shifted position, counting error or genuine biological variation.
Repeat the condition before revising the whole relationship.
Original workshop 3 — food-web prediction
A predator is expected to decline after prey X falls, but predator numbers stay stable.
Alternative explanation: predator may use prey Y or migrate to another food source.
The original one-prey model was incomplete.
Original workshop 4 — cooling reversal
Insulated Cup A usually stays warmer, but one trial shows it cooler than uninsulated Cup B.
Check starting temperature, water volume, timing and thermometer reading before rejecting the insulation explanation.
Part XVIII — The ALTS test
- A — Alternatives: what else could explain the observation?
- L — Limits: what does the current method fail to rule out?
- T — Test: what new evidence would distinguish the alternatives?
- S — Survive: which explanation still fits after testing?
This is an eduKate teaching mnemonic.
Part XIX — MCQ use
When two options seem plausible, ask which one:
- fits every stated condition;
- requires fewer invented assumptions;
- matches the data direction;
- survives the anomaly or contradiction;
- does not exceed the evidence.
Part XX — Open-ended evaluation frame
“The result could also be explained by [alternative]. Because [condition] was not controlled/measured, the evidence does not isolate [preferred explanation] confidently. A stronger test would [specific improvement].”
Where to connect
- Confidence, Uncertainty, Anomalies & Strength of Evidence
- Models, Assumptions, Limits & Scientific Claims
- Primary 4 Alternative Explanations, Contradictions & Anomalies
Retrieval checklist
- I can generate more than one plausible explanation.
- I use evidence to eliminate alternatives.
- I distinguish anomaly from repeated contradiction.
- I do not delete unusual results automatically.
- I can identify confounding as an alternative explanation.
- I can identify measurement weakness as a source of contradiction.
- I know when repeated evidence should revise a model.
- I can compare two models by explanatory power.
- I stay within Primary-level Science.
- I can propose a new test that distinguishes alternatives.
The quiet return
Scientific maturity begins when the first explanation is treated as a candidate rather than a possession. Evidence decides which explanations survive.
Generate alternatives. Test them. Respect contradictions. Investigate anomalies. Keep the explanation that survives the evidence.
Return to the Primary 6 Science Learning Hub.
