Wait, What? Equal Steps Along One Axis Do Not Promise Equal Steps Along the Other
A learner sees a table in which a tested condition increases from 1 to 2 to 3 to 4. The steps are perfectly equal. The measured result rises from 4 to 7 to 9 to 10.
The learner says, “Because the condition increases by one each time, the result should also increase by the same amount each time.”
But it does not.
Equal changes in the condition do not automatically cause equal changes in the response.
Science is full of relationships that increase, decrease, level off, begin only after a threshold, reverse direction, or respond differently across different ranges. A neat number scale does not force the natural system to behave proportionally.
This guide teaches how to read that distinction without importing unnecessary mathematics into Primary Science.
Quick Answer
When the changed condition increases in equal steps, compare the actual output change between each pair of readings before describing the relationship. If the output differences are unequal, you may still have an increasing or decreasing trend, but you do not have evidence that the response is proportional or changing at a constant rate.
Use this route:
READ THE INPUT STEPS → READ THE OUTPUT VALUES → CALCULATE OR COMPARE EACH OUTPUT CHANGE → DESCRIBE THE SUPPORTED PATTERN → SELECT THE SCIENTIFIC MECHANISM → CHECK FOR THRESHOLD, PLATEAU, TURNING POINT OR LIMIT → STATE ONLY THE RELATIONSHIP THE TESTED RANGE SUPPORTS.
The Exact PSLE Science Learning Job This Guide Owns
This guide owns one learner job: how a Primary 5 or Primary 6 learner interprets data where the independent condition changes by equal numerical steps but the measured response changes by unequal amounts.
It does not replace the general guide on scientific trends. It does not replace the guide on rate versus amount. It does not teach formal proportionality, gradients or equations as Secondary Mathematics would. Its job is narrower:
do not mistake a tidy input scale for evidence that the output must respond tidily.
This is also not an official answer template. Different PSLE Science questions require different concepts and wording. The durable skill is accurate relationship reading.
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 applying scientific facts, concepts and principles, making predictions, interpreting and analysing information, evaluating observations and methods, and communicating explanations and reasoning.
Equal-step data are a useful test of those capabilities because a learner has to inspect the relationship instead of assuming one. The numbers may look orderly while the scientific response is not proportional.
First Distinction: Ordered Does Not Mean Proportional
Suppose the condition and result are:
| Condition | Measured result | Change from previous result |
|---|---|---|
| 1 | 4 | — |
| 2 | 7 | +3 |
| 3 | 9 | +2 |
| 4 | 10 | +1 |
The input rises by +1 each time. The output still increases at every step, but the increases become smaller: +3, +2, +1.
A safe description is:
As the tested condition increases from 1 to 4, the measured result increases, but by progressively smaller amounts.
An unsafe description is:
“Every increase of one unit in the condition causes the same increase in the result.”
The data contradict that stronger claim.
Four Relationship Questions to Ask
| Question | What it checks |
|---|---|
| Does the output move in one general direction? | Whether there is an increasing or decreasing trend. |
| Are equal input steps followed by equal output changes? | Whether the change appears constant over the tested range. |
| Does doubling the input double the output? | Whether a simple proportional claim might be supported. |
| Does the relationship change across the range? | Whether there is a threshold, plateau, reversal or another boundary. |
Do not answer all four with one word such as “increases”.
Monotonic Change: One Direction, Unequal Steps
A relationship can move consistently in one direction without changing by equal amounts.
- 2, 5, 7, 8: always increasing, but by +3, +2, +1.
- 20, 14, 10, 8: always decreasing, but by −6, −4, −2.
- 3, 4, 7, 13: always increasing, but the increments grow larger.
At Primary level, the important distinction is verbal rather than formal: same direction does not mean same amount of change.
Why Scientific Responses Are Often Unequal
There are many scientifically legitimate reasons:
- the effect of the changed factor becomes weaker as the system approaches a limit;
- another necessary condition becomes limiting;
- the difference driving a process becomes smaller;
- a threshold must be crossed before the response becomes visible;
- the system enters a different operating region;
- different mechanisms become important at different conditions;
- measurement sensitivity compresses small changes;
- living systems naturally respond non-uniformly.
The graph pattern does not tell you which explanation is correct by itself. Use the scientific context.
Worked Example 1 — Cooling Measured at Equal Time Intervals
Original practice data:
| Time / min | Temperature / °C | Temperature change |
|---|---|---|
| 0 | 70 | — |
| 5 | 56 | −14 |
| 10 | 47 | −9 |
| 15 | 41 | −6 |
| 20 | 38 | −3 |
The measurements are taken every five minutes: equal time steps.
But the temperature decreases by 14°C, 9°C, 6°C and 3°C across successive intervals. The decreases become smaller.
A learner should not write, “The water cools by the same number of degrees every five minutes.”
A stronger description is: “The water’s temperature decreases over time, but the amount of temperature decrease during each five-minute interval becomes smaller.”
If the surroundings remain cooler than the water, the scientific explanation can connect this pattern to the decreasing temperature difference between the water and its surroundings. The exact depth of that explanation should remain appropriate to the question.
Worked Example 2 — Stretch Setting and Toy-Car Distance
A toy car is launched by the same elastic system at four increasing pull-back settings.
| Pull-back setting | Distance travelled / cm | Increase |
|---|---|---|
| 1 | 30 | — |
| 2 | 51 | +21 |
| 3 | 65 | +14 |
| 4 | 72 | +7 |
Across this original practice setup, more pull-back is associated with greater travel distance, but each equal increase in setting produces a smaller additional distance.
The data therefore support an increasing relationship over the tested range. They do not support the claim that every extra setting adds exactly the same distance.
Nor do they prove that setting 8 would produce twice the distance of setting 4. That would be an extrapolation and a proportionality claim beyond the evidence.
Worked Example 3 — Number of Layers and Light Passing Through
Suppose identical translucent sheets are stacked and a light sensor records:
| Number of sheets | Displayed light reading |
|---|---|
| 1 | 80 |
| 2 | 55 |
| 3 | 39 |
| 4 | 29 |
Each added sheet is one equal step in sheet number. The light reading decreases, but not by equal amounts: −25, −16, −10.
Do not say “every sheet blocks exactly 25 units” because only the first added sheet produced a 25-unit difference in these data.
The learner should preserve what the measurement shows: more layers are associated with a lower measured light reading under the tested conditions.
Worked Example 4 — Equal Time Intervals and Plant Height
A young plant is measured once per day:
| Day | Height / cm |
|---|---|
| 1 | 10.0 |
| 2 | 10.4 |
| 3 | 11.0 |
| 4 | 11.3 |
The days are equally spaced. The plant’s measured growth is not: +0.4 cm, +0.6 cm and +0.3 cm.
Biological variation is normal. Do not force a constant daily growth rate because the time intervals are equal.
Worked Example 5 — Equal Increases Produce No Early Response, Then a Response
Conditions 1, 2, 3 and 4 produce outputs 0, 0, 3 and 7.
The equal condition steps do not produce equal changes. The first two measured outputs are the same, then the response appears and grows.
This pattern may indicate a threshold-like observation. The scientific meaning depends on what was measured and how sensitive the detector was.
Worked Example 6 — Equal Steps, Then a Plateau
Conditions 1 through 6 produce outputs 2, 5, 7, 8, 8, 8.
The output first rises by unequal amounts, then becomes constant in the recorded data. A single sentence such as “more condition gives more output” is no longer adequate across the full tested range.
A better description is: “The measured output increases from Conditions 1 to 4, then remains at 8 from Conditions 4 to 6.”
The mechanism must explain both regions if the question asks why.
Equal Inputs Do Not Mean a Straight-Line Graph
When the horizontal-axis values are equally spaced, the graph points will be evenly spaced horizontally. That says nothing about whether they line up vertically into a straight line.
A curved pattern can be scientifically meaningful. Do not “correct” the data into a straight line simply because the x-axis uses equal increments.
Increasing Is Not the Same as Proportional
Consider these two data sets:
| Condition | Data Set A | Data Set B |
|---|---|---|
| 1 | 3 | 3 |
| 2 | 6 | 5 |
| 3 | 9 | 6 |
| 4 | 12 | 6.5 |
Both increase across the tested range. Data Set A increases by +3 every time. Data Set B increases by +2, +1 and +0.5.
Calling both simply “increasing” is acceptable if the question asks only for direction. Calling both “the same relationship” loses important structure.
Doubling Input Does Not Automatically Double Output
Suppose one battery arrangement produces a certain bulb brightness and a different arrangement uses more batteries. It is unsafe to say the bulb must become “twice as bright” merely because one count doubled. Brightness is an observed output affected by the whole circuit, and Primary Science does not require a simple numerical proportional rule here.
The same caution applies across many contexts: twice the light exposure does not automatically mean twice the plant growth; twice the surface area does not automatically mean exactly twice the mass loss over every condition; twice the force does not automatically produce twice every measured motion outcome.
Use the evidence actually supplied.
Equal Output Differences Are Evidence, Not a Law
If a table happens to show 5, 10, 15, 20 for conditions 1, 2, 3, 4, you may describe equal measured increases over those tested points.
Do not immediately turn that short pattern into a universal rule for every possible condition. A threshold, plateau or turning point may appear outside the tested range.
How to Compare Consecutive Changes
For simple data, write the output differences beside the table:
Output change = later measured value − earlier measured value.
You do not need algebraic notation. You are simply comparing how much the result changed between equal input steps.
If the input intervals are unequal, do not compare raw differences as though the intervals were equal. The question may require rate reasoning instead.
The Constant-Change Test
- Check that the input steps really are equal.
- Calculate each output change.
- Compare those output changes.
- If they are equal or nearly equal within the given measurement resolution, describe that pattern carefully.
- If they differ, do not call the response constant.
- Use the science to explain why the response might strengthen, weaken or level off.
Measurement Resolution Can Make Changes Look Equal
A thermometer reading only whole degrees may show increases of +2, +2, +2 even if the underlying changes differ slightly. A ruler with coarse markings can also hide small differences.
Therefore, “equal recorded changes” means equal within the measurement shown—not necessarily mathematically identical at infinite precision.
Unequal Changes Can Reveal a Boundary
The pattern of differences can itself be evidence:
- +5, +3, +1 may suggest the response is approaching a plateau;
- 0, 0, +4, +6 may suggest a threshold-like onset;
- +4, +2, 0, −3 may signal a turning region;
- +2, +2, +2 may show approximately constant change across the measured range.
The differences are not the explanation. They tell you what scientific feature deserves attention.
Do Not Invent “Efficiency” From Unequal Output
If one step produces a smaller additional output than the previous step, it may be tempting to say the system has become “less efficient”. That word has a specific meaning in many scientific contexts and is not justified by every diminishing response.
Prefer the evidence: “the additional measured increase becomes smaller”. Then explain the mechanism the question supports.
Do Not Invent “Saturation” Unless the Science Supports It
A flattening response may be caused by a limit, another limiting factor, detector range or a genuine system boundary. “Saturation” can be a useful scientific idea in certain fields, but it should not be used as a decorative label for every curve.
What the Question May Actually Be Testing
| Data pattern | Possible learner job |
|---|---|
| Equal input steps, shrinking output increases | Recognise diminishing response / approach to a limit. |
| Equal input steps, zero response then increase | Recognise an observed threshold region. |
| Equal input steps, increase then decrease | Recognise a turning point. |
| Equal input steps, equal recorded output changes | Describe approximately constant change within the tested range. |
| Equal input steps, one strange output | Evaluate possible anomaly or method issue. |
The Earliest-Weak-Link Diagnostic
| Failure signature | Earliest weak link | Repair |
|---|---|---|
| “The condition rises evenly, so the result must rise evenly.” | Input spacing was confused with response behaviour. | Calculate each output difference. |
| “It increases, therefore it is proportional.” | Direction was confused with relationship shape. | Compare whether equal input steps give equal output changes. |
| “Double the input means double the output.” | A proportional rule was assumed without evidence. | Check actual paired values and mechanism. |
| “The graph curves, so the data are wrong.” | Straightness was treated as a quality criterion. | Ask whether the scientific response is expected to be nonlinear. |
| “The last changes are smaller, so the process has stopped.” | Diminishing response was confused with zero response. | Check whether values still change and whether a plateau has truly formed. |
| “The first three steps are equal, so all future steps will be equal.” | Short-range pattern became universal law. | Keep the claim within the tested range. |
Misconception Repair — A Nice Table Does Not Make Nature Linear
Experimenters often choose neat test values because they are easy to control and compare. The system is not obliged to respond in equally neat increments.
Misconception Repair — Equal Time Does Not Mean Equal Change
Taking readings every five minutes controls the observation schedule. It does not impose a constant rate on cooling, evaporation, growth, motion or any other process.
Misconception Repair — More Does Not Always Mean Proportionally More
A condition can have a positive effect while additional increases produce smaller extra effects. “More produces more” and “twice as much produces twice as much” are different claims.
How Equal-Step Questions Appear in Multiple Choice
- Check whether the input steps are equal.
- Compare the output changes.
- Reject options that claim proportionality without support.
- Check for threshold, plateau or turning behaviour.
- Read absolute words such as “always”, “same”, “constant” and “double” carefully.
- Choose the statement that matches both the data and the relevant science.
How Equal-Step Questions Appear in Structured Answers
A useful reasoning shape is:
Although the tested condition increases by equal steps of ______, the measured result changes by ______, ______ and ______. Therefore, the result increases/decreases but not by a constant amount. This is consistent with ______ under the stated conditions.
This is a scaffold for reasoning, not an official phrase to memorise.
Practice Sequence
- Take five tables with equal input intervals.
- Write the output differences beside each table.
- Classify each relationship as constant change, diminishing change, increasing change, threshold-like, plateauing or turning.
- Describe the pattern without explaining it.
- Use the scientific setup to explain the pattern.
- Change the graph scale and check whether your interpretation survives.
- Predict one extra point, then state why it is only a prediction.
- Return several days later with an unfamiliar context.
Unfamiliar Transfer Challenge
A mystery system is tested at settings 2, 4, 6, 8 and 10. The measured outputs are 5, 11, 16, 20 and 23.
The input increases by 2 each time. The output increases by +6, +5, +4 and +3.
What can you say? The output increases across the tested range, but the additional increase becomes progressively smaller.
What can you not say from the numbers alone? You cannot identify the mechanism, claim the response will eventually stop, or prove a universal proportional law.
Delayed Independent Return
Four days later, take a fresh table and answer without notes:
- Are the input steps equal?
- What are the output changes?
- Do they stay equal?
- Does the output move in one direction?
- Does the response strengthen, weaken, plateau or reverse?
- What scientific mechanism fits the changing response?
- What alternative explanation could come from measurement limits?
- What claim is safe only within the tested range?
The Answer-Checking Receipt
- Did I distinguish input steps from output changes?
- Did I calculate or compare consecutive output differences?
- Did I separate “increasing” from “proportional”?
- Did I avoid assuming doubling input doubles output?
- Did I check for threshold, plateau or turning behaviour?
- Did I keep rate separate from amount?
- Did I avoid inventing a mechanism from graph shape alone?
- Did I keep the conclusion within the tested range?
Useful Internal Routes
- How to Tell a Scientific Trend From a Single Comparison in PSLE Science
- How to Separate Rate From Amount in PSLE Science
- How to Read a Plateau in PSLE Science Data
- How to Read a Threshold in PSLE Science
- How to Read a Turning Point in PSLE Science Data
- How to Read PSLE Science Data With Gaps
- Primary Science | Complete P1–P6 and PSLE Science Guide
Parent and Tutor Teaching Guide
When a learner sees equal input steps, ask them to cover the input column and look only at the output sequence. Ask:
“How much did the result actually change each time?”
If the learner can calculate the differences but still calls the relationship proportional, compare two near-miss tables: one with equal output differences and one with shrinking differences. Ask them to explain the distinction without using the word “straight”.
Then change the scientific context while preserving the numerical pattern. A learner who truly understands the relationship should transfer from temperature to distance, plant growth, light measurement or another unfamiliar setting without needing a memorised topic phrase.
Return after a delay. The goal is for the learner to ask “what are the output differences?” automatically before accepting a proportional claim.
Authoritative and Research References
- Singapore Examinations and Assessment Board — PSLE Formats Examined in 2026.
- Singapore Examinations and Assessment Board — PSLE Science syllabus, for examination from 2026.
- Singapore Ministry of Education — Science Teaching and Learning Syllabus, Primary, 2023.
- Research on students’ graph interpretation, proportional reasoning and scientific reasoning is used here as broader learning evidence rather than as PSLE-specific marking policy.
- Dunlosky and colleagues — review of effective learning techniques, including practice testing and distributed practice.
The Quiet Ending
The experimenter chooses the steps.
The system chooses the response.
Good Science never assumes those two rhythms must match.
