Science tuition for Yishun families should help students match the strength of their language to the strength of the evidence. A careful Science answer does not use proves, always or definitely when the investigation only supports a narrower conclusion.
This rebuilt legacy URL now owns one distinct Science reasoning job: evidence quality → scope → certainty → confidence word → conclusion. Old ranking claims, guaranteed-result language, mixed Primary/Secondary material and implied Yishun-centre wording have been removed.
eduKateSG does not claim a current Yishun branch. Current centres are at 83 Punggol Central, Singapore 828761 and 8 Fourth Avenue, Singapore 268674. Selected Primary Science classes run in focused 3-pax small groups, typically 1.5 hours.
The Direct Answer
Useful confidence language can be arranged roughly from stronger to more cautious:
- shows — directly displayed by the evidence;
- supports — evidence provides good backing for an explanation or claim;
- suggests — evidence points in a direction but does not settle every alternative;
- may/could — one plausible outcome or explanation;
- cannot conclude — evidence is insufficient or confounded.
Good Science writing is not weak when it is cautious; it is accurate about what the evidence earns.
When “Shows” Is Appropriate
Table:
- Plant A grew 6 cm;
- Plant B grew 2 cm.
Safe statement:
“The data shows that Plant A grew 4 cm more than Plant B over the measured period.”
The values directly establish the difference.
When “Supports” Is Better
If Plant A and B were similar and light was the main changed factor, greater growth in A may support the explanation that higher light exposure contributed to greater growth.
The evidence supports a causal explanation because the experimental design matters.
When “Suggests” Is Better
Suppose a food web shows fewer prey and later fewer predators.
“The pattern suggests that reduced prey availability may have contributed to the predator decline.”
Other environmental factors may remain possible.
When “May” Is Better
Prediction:
“If the prey population continues to fall, the predator population may decrease.”
The word may preserves uncertainty because alternative food sources could exist.
When “Cannot Conclude” Is the Strongest Answer
Question: Does water temperature affect dissolving time?
Setup A uses hot water and stirring.
Setup B uses cold water and no stirring.
Because two important factors changed, the correct response may be:
“We cannot conclude that temperature alone caused the difference.”
Refusing to overclaim is scientifically stronger than guessing.
The Evidence-Strength Ladder
| Evidence situation | Useful language |
|---|---|
| Direct measured difference | shows |
| Controlled comparison + mechanism | supports |
| Pattern with alternatives remaining | suggests |
| Possible prediction | may/could |
| Confounded/insufficient evidence | cannot conclude |
Worked Example: Evaporation
Two equal volumes of water are placed in containers with different exposed surface areas under otherwise comparable conditions.
After one hour, the wider container loses more water.
Direct evidence:
“The wider container showed greater water loss.”
Mechanistic claim:
“The result supports the idea that greater exposed surface area increases evaporation under these conditions.”
Worked Example: Light and Shadow
If an object is moved closer to a lamp and its shadow becomes larger on the screen:
“The observation shows that the shadow size increased in this setup.”
“The result supports the straight-line light-path model used to explain the change.”
Worked Example: Population Change
If a predator population falls after a prey decline:
Too strong:
“The prey decline definitely caused the predator decline.”
Better:
“The timing and food-web relationship support reduced prey availability as one explanation, although other factors may also affect the predator population.”
“Proves” Is Rarely Needed
Primary Science questions seldom require students to declare that one classroom test proves a universal law.
A well-controlled investigation can support a relationship strongly without justifying “always”.
“Always” and “Never”
Universal words demand universal evidence or an established definition/law at the relevant syllabus level.
Example:
“All metals are magnetic” is false.
A few magnetic metal samples cannot justify all.
“Likely”
Useful when evidence makes one outcome more probable but not guaranteed:
“The plant is likely to wilt if root damage prevents sufficient water uptake.”
Use the word only when the mechanism and conditions make the prediction reasonable.
“Consistent With”
This phrase is useful when the result fits an explanation without uniquely proving it:
“The data are consistent with the idea that insulation slowed cooling.”
Another explanation might still fit unless the design excludes it.
The Alternative-Explanation Test
Before writing a strong confidence word, ask:
Could another reasonable cause also produce this result?
If yes, reduce certainty unless the design or other evidence rules it out.
The Sample-Size Test
One plant, one trial or one animal provides limited evidence.
Even if the observation is correct, generalisation should remain cautious.
The Range Test
If temperature was tested only from 20°C to 40°C, the result does not automatically justify a claim about 80°C.
Confidence should remain within the tested range.
The Time Test
A seven-day pattern may not continue for seven months.
“Over the seven-day test…” is often better than writing a permanent prediction.
The Population Test
Evidence about one plant species should not automatically become a claim about all plants.
Scope and confidence work together.
Worked Example: “Supports” vs “Suggests”
Supports: a fair test changes one main variable, repeats the measurement and finds a consistent difference.
Suggests: a graph shows an association, but several uncontrolled factors may be changing.
The same topic can deserve different confidence language depending on design quality.
Confidence in MCQ
Distractors often use excessive certainty:
- always;
- only;
- must;
- definitely;
- proves.
Do not reject these automatically. Ask whether the evidence truly warrants them.
Confidence in Open-Ended Answers
Instead of:
“The predator will decrease.”
Use:
“The predator population may decrease because fewer prey are available as food.”
The wording respects the food-web model’s limits.
Confidence in Conclusions
A conclusion should combine:
what changed + under which conditions + how certain the evidence allows us to be.
Example:
“Under the tested conditions, greater exposed surface area was associated with greater water loss and supports faster evaporation as the explanation.”
The Confidence Audit
- Underline the strongest word in the conclusion.
- Ask what evidence supports that level of certainty.
- Look for alternative explanations.
- Check sample, range and time limits.
- Reduce or increase the wording until it matches the evidence.
P3–P6 Progression
P3: “I observed…”
P4: “The result shows…”
P5: “The fair comparison supports…”
P6: “The evidence suggests/may/cannot conclude…” when models and alternatives become more complex.
The exact vocabulary can vary, but the principle remains: certainty should be earned.
Curriculum Context
MOE’s Primary Science syllabus emphasises interpreting evidence, communicating conclusions and evaluating investigations across P3–P6. Official reference: MOE Primary Science Syllabus.
Common Confidence-Language Failure Modes
| Symptom | Problem | Repair |
|---|---|---|
| One trial proves | Certainty too high | Use supports/suggests |
| may rewritten as will | Prediction strengthened | Preserve modality |
| Confounded setup gets causal claim | Evidence too weak | Cannot conclude |
| Specific result becomes all/always | Scope widened | Keep tested boundary |
| Every answer uses “maybe” | Confidence too low | Use stronger wording when data directly shows it |
Why Three Students Helps
One student states the evidence, one chooses the confidence word and the third tries to construct an alternative explanation that would force the wording to become more cautious.
When Yishun Families May Consider This Support
- The child uses “proves” and “always” too easily.
- Predictions are written as certainties.
- Good evidence is described too weakly with “maybe”.
- PSLE open-ended answers lose precision because confidence and scope are mismatched.
What Progress Should Look Like
A stronger learner can match confidence language to evidence quality, explain why a conclusion is strong or tentative and refuse to make claims the investigation cannot support.
Independent location notice: eduKateSG does not claim a current Yishun branch. Current centres: 83 Punggol Central, Singapore 828761 and 8 Fourth Avenue, Singapore 268674. Contact +65 8823 1234.