Primary 6 practical reasoning begins before any data are collected. A strong pupil can turn a scientific question into a workable plan, choose what to change and measure, record results clearly and write a conclusion that matches the evidence. A weak plan can produce neat tables and still fail to answer the original question.
This guide develops practical planning, procedural clarity, data recording, result checking and evidence-based conclusions for Primary 6 and PSLE Science.
Return to the Primary 6 Science Learning Hub.
The practical rule
QUESTION → VARIABLES → APPARATUS → PROCEDURE → RECORDING PLAN → REPEATS → PATTERN → CONCLUSION → LIMIT.
This is an eduKate reasoning routine, not an official SEAB formula.
Part I — Plan from the question backward
Question: “How does surface type affect the distance travelled by a toy car?”
Work backward:
- Need to compare distance.
- Need several surface types.
- Need the same car.
- Need a consistent release.
- Need a way to measure distance.
- Need repeated trials if possible.
- Need a table prepared before testing.
The procedure grows from the scientific relationship, not from whatever apparatus happens to be available.
Part II — Apparatus should have a job
Do not list equipment without knowing why it is there.
| Apparatus | Scientific job |
|---|---|
| Ruler/tape | Measure distance or length |
| Stopwatch | Measure time interval |
| Thermometer | Measure temperature |
| Measuring cylinder | Measure liquid volume |
| Lamp | Provide controlled light condition |
| Switch/cells | Control electrical circuit condition |
If an item has no clear role, it may be irrelevant.
Part III — Procedure writing must be operational
Weak: “Test the car on different surfaces.”
Stronger:
- Place the same car at the same marked starting point.
- Release it using the same method onto Surface A.
- Measure the distance travelled before stopping.
- Repeat the trial several times.
- Repeat the same procedure for the other surfaces.
- Compare the recorded distances.
A good procedure tells another person exactly what to do.
Part IV — Define the start and endpoint
Measurements become inconsistent when pupils disagree about when to start or stop.
Examples:
- Start timing when the switch is closed.
- Stop timing when the liquid reaches 40°C.
- Measure car distance from the release line to the front of the car when it stops.
- Count bubbles for exactly five minutes.
Operational definitions improve repeatability.
Part V — Prepare the table before collecting data
A well-designed table reveals whether the plan measures the right quantities.
Example:
| Surface | Trial 1 distance (cm) | Trial 2 distance (cm) | Trial 3 distance (cm) | Average (cm) |
|---|---|---|---|---|
| A | ||||
| B | ||||
| C |
Headings should contain quantities and units.
Part VI — Independent variable belongs in the first column
A common layout puts the changed condition in the left column and measured results across the table.
This makes the comparison structure visible.
Example: lamp distance in the first column; bubble counts in later columns.
Part VII — Do not mix units inside one column
If one row records centimetres and another metres, comparison becomes error-prone. Convert to a common unit before analysis where appropriate.
Put the unit in the heading rather than repeating it in every cell when the format allows.
Part VIII — Record raw data before averaging
Do not record only the average if individual trials are available. Raw results reveal variation and outliers.
Example: 80, 81, 79 cm is reassuringly consistent.
80, 42, 79 cm reveals a problem that an average could hide.
Part IX — Qualitative observations belong somewhere too
Not all evidence is numerical.
Useful notes may include:
- colour change;
- gas bubbles observed;
- spring did not return to original length;
- plant wilted;
- circuit component became warm;
- surface became wet between trials.
These observations can explain unusual results or method limitations.
Part X — Repeats and averages
Repeat the same valid procedure under the same condition. Then calculate an average if the question requires it.
Averages reduce dependence on one unusual reading but do not repair a confounded design.
Part XI — Decide what to do with an anomalous result
Do not remove it automatically.
Check:
- Was the procedure followed?
- Was the instrument read correctly?
- Did an external condition change?
- Did the apparatus fail?
- Does a repeat give a similar unusual value?
Part XII — Draw the graph only after understanding the table
Choose the horizontal axis for the changed variable and the vertical axis for the measured variable in typical school investigations, unless the question specifies otherwise.
Include:
- axis labels;
- units;
- sensible scale;
- accurately plotted points;
- appropriate line or bars according to data type.
Part XIII — Conclusions answer the question
Weak: “The experiment worked.”
Weak: “Surface A is best.”
Stronger: “Under the tested conditions, the car travelled farther on Surface A than on Surface B, indicating a smaller frictional effect on A.”
The conclusion should connect the changed condition to the measured outcome.
Part XIV — Use evidence in the conclusion
When appropriate, include a data comparison.
“The average travel distance was 82 cm on A compared with 46 cm on B.”
This makes the conclusion auditable.
Part XV — Conclusion is not explanation
Conclusion: what relationship the results support.
Explanation: why the relationship occurs scientifically.
Question wording decides whether one or both are needed.
Part XVI — Original practical: lamp distance and bubbles
Question: How does lamp distance affect bubble count from an aquatic plant over five minutes?
Changed variable: lamp distance.
Measured outcome: bubbles counted in five minutes.
Controls: same plant, same water conditions, same duration, same lamp, comparable temperature where relevant.
Recording: table with lamp distance and repeated bubble counts.
Limit: bubble sizes may differ, so bubble count is an indirect proxy for gas volume.
Part XVII — Original practical: spring load
Question: How does load affect spring extension within the tested range?
Measure initial spring length first. Add loads systematically. Measure spring length from the same reference point. Calculate extension by subtracting the initial length.
Do not confuse total length with extension.
Part XVIII — Original practical: cooling cups
Use equal volumes of water at the same starting temperature in identical cups with different insulating materials.
Measure temperature after the same duration or measure time to reach a target temperature—but do not mix the two measurement designs.
Part XIX — Original practical: environmental survey
A survey asks whether insect abundance differs between shaded and open locations.
Use comparable sample areas, similar time periods, repeated locations and a consistent counting method.
Recognise that observational field studies may not isolate cause as cleanly as controlled experiments.
Part XX — Safety and practicality
A valid method also needs to be workable and safe.
Primary Science practicals should avoid unnecessary hazards, unstable apparatus or procedures that require precision beyond the available equipment.
If two methods are equally valid, the simpler reproducible method is often stronger.
Part XXI — Procedure improvement
Target the weak step.
| Weakness | Improvement |
|---|---|
| Release varies | Use a fixed release mechanism |
| Timing subjective | Define start/end points clearly |
| One reading | Repeat the valid procedure |
| Scale too coarse | Use suitable finer-resolution instrument |
| Biological variation | Use several comparable specimens |
| Condition drifts | Monitor/reset relevant condition between trials |
Part XXII — Data integrity
Do not change a result to make the pattern look better.
Record what was observed. Investigate anomalies. Science improves through honest data, not cosmetically perfect tables.
Part XXIII — Practical planning under exam conditions
If asked to design an investigation, write in this order:
- state what changes;
- state what is measured;
- name key controls;
- describe how the change is set;
- describe how the outcome is measured;
- repeat if appropriate;
- state how results are compared.
This produces a complete method without unnecessary storytelling.
Part XXIV — Common practical errors
- No clear measured outcome.
- Two variables changed together.
- No defined starting point.
- Different time intervals.
- Missing units.
- Only averages recorded.
- Irrelevant controlled variables.
- Generic “repeat” without fixing the design.
- Conclusion not tied to data.
- Method too vague for another person to reproduce.
Part XXV — The PRACTICAL checklist
- P — Purpose/question
- R — Relevant variables
- A — Apparatus with a job
- C — Controlled conditions
- T — Timing/start/end definitions
- I — Instrument and units
- C — Clear recording table
- A — Average/repeats when appropriate
- L — Link conclusion to evidence
This is an eduKate teaching mnemonic.
Where to connect
- Confounding Variables, Controls & Experimental Validity
- Measurement, Units, Resolution & Repeatability
- Primary 5 Practical Planning, Data Recording & Conclusions
Retrieval checklist
- I can plan backward from a scientific question.
- I know why each apparatus item is used.
- I can write an operational procedure.
- I define start and endpoint clearly.
- I prepare a table with headings and units.
- I record raw data before averages.
- I keep qualitative observations when useful.
- I investigate anomalous results honestly.
- I write conclusions that answer the original question.
- I can propose a targeted procedural improvement.
The quiet return
A practical investigation is a chain from question to evidence. If every link is clear—variables, apparatus, procedure, measurement, recording and conclusion—the data can answer the question. If one link is weak, more data may only repeat the weakness.
Plan from the question. Measure what matters. Record honestly. Repeat intelligently. Conclude only what the evidence supports.
Return to the Primary 6 Science Learning Hub.
